Compendio de Metodos Microbiologicos [PDF]

Zitiervorschau

Editorial Board

vonne Salfinger joined the Compendium editorial board, representing the Association of Public Health Laboratories (APHL). She graduated from Florida State University (BS) and earned an MS in community health from the University of North Florida. She is pursuing a DrPH from the University of South Florida. After working at the Florida Department of Health for 21 years, she moved to the Florida Department of Agriculture and Consumer Services, initially as the quality manager, to pursue ISO/IEC 17025 accreditation for the Division of Food Safety laboratories; in 2007, accreditation was accomplished. In 2003, she became the chief for the Bureau of Food Laboratories, Division of Food Safety, Florida Department of Agriculture and Consumer Services. She was also involved with the Food Emergency Response Network in several activities relating to the analysis of foods during emergency events, including the establishment of Standard Operating Procedures. After retiring from the Department of Agriculture and Consumer Services in 2012, she relocated to Denver, Colorado, where she now works as a consultant for APHL and the Association of Food and Drug Officials to assist with the implementation of ISO/IEC 17025 across the regulatory food testing laboratories within the United States. Over the years, she has contributed to public health internationally: With the Centers for Disease Control and Prevention (CDC) and the Global Health program of APHL, she worked on the post-Hurricanes Mitch and Georges project in El Salvador. With CDC, she provided technical quality assurance to the tuberculosis laboratory in Ivanovo, Russia.

Y

M

Gary R. Acuff (board), Texas A&M University, College Station Mark Carter (board), MC Squared Enterprises, Flossmoor, Illinois Victor Cook (board), retired from U.S. Department of Agriculture, Washington, D.C. Stephanie Doores (associate editor), Penn State University, University Park David Goldman (board), U.S. Department of Agriculture, Washington, D.C.

Keith Ito (board), University of California, Davis, Dublin Ruth L. Petran (board), Ecolab, St. Paul, Minnesota Yvonne Salfinger (co-editor), retired from Florida Department of Agriculture and Consumer Services, Tallahassee Mary Lou Tortorello (co-editor), U.S. Food and Drug Administration, Bedford Park, Illinois Burton Wilcke (associate editor), The University of Vermont, Burlington

ary Lou Tortorello attended Northern Illinois University (BS) and Loyola University of Chicago (MS, biological sciences). She received a PhD in microbiology from Cornell University where she researched the structure and biosynthesis of lipopolysaccharide endotoxin for her dissertation. At Cornell, she pursued post-graduate work, focusing on bacterial gene transfer mechanisms, improved dairy starter cultures, and rapid assays for detection of pathogens; there, she also taught an undergraduate course, General Microbiology. Since 1991, she has worked as a research microbiologist with the Division of Food Processing Science and Technology, U.S. Food and Drug Administration, in Bedford Park, Illinois. She currently serves as the chief of the Food Technology Branch. Her professional experience also includes work as the product manager of a serum diagnostic test for the HIV/AIDS virus at the Abbott Laboratories Diagnostics Division in Abbott Park, Illinois. She maintains research interests in improved microbiological methods for foodborne pathogens, especially rapid methods, sample preparation, and the behavior and control of microbial pathogens in foods and food processing environments. Furthermore, she is co-editor of the Encyclopedia of Food Microbiology, serves on the Editorial Board of Journal of Food Protection, and is chief editor of Food Microbiology.

| xvii

Authors

Carlos Abeyta Jr, U.S. Food and Drug Administration, Bothell, Washington Gary R. Acuff, Texas A&M University, College Station Vidya Ananth, The Clorox Company, Pleasanton, California Jean E. Anderson, General Mills, Minneapolis, Minnesota Francisco N. Arroyo-Lo´pez, Instituto de la Grasa, Seville, Spain Joaquı´n Bautista-Gallego, Instituto de la Grasa, Seville, Spain Ronald A. Benner Jr, U.S. Food and Drug Administration, Dauphin Island, Alabama Reginald W. Bennett, U.S. Food and Drug Administration, College Park, Maryland Andre´ia Bianchini, University of Nebraska-Lincoln Glenn Black, Grocery Manufacturers Association, Washington, District of Columbia Peter Bodnaruk, Hillshire Brands, Chicago, Illinois Ann Rogers Bontempo, Mondelez International, East Hanover, New Jersey Robert L. Bradley Jr, University of Wisconsin-Madison Pardeepinder K. Brar, University of Florida, Lake Alfred Roger M. Brauninger, American Association for Laboratory Accreditation, Frederick, Maryland Byron Brehm-Stecher, Iowa State University, Ames Fred Breidt Jr, U.S. Department of Agriculture, Raleigh, North Carolina Michael H. Brodsky, Brodsky Consultants, Thornhill, Ontario Ronald W. Buescher, University of Arkansas, Fayetteville Lloyd B. Bullerman, University of Nebraska-Lincoln William Burkhardt III, U.S. Food and Drug Administration, Dauphin Island, Alabama Kevin R. Calci, U.S. Food and Drug Administration, Dauphin Island, Alabama Laurenda Carter, U.S. Food and Drug Administration, Laurel, Maryland Erdogan Ceylan, Silliker, Crete, Illinois Michael C. Cirigliano, Unilever, Englewood Cliffs, New Jersey Dean O. Cliver (deceased), University of California, Davis David Clifford, Nestle´ S.A., Dublin, Ohio Janet E. L. Corry, University of Bristol, United Kingdom Douglas E. Cosby, U.S. Department of Agriculture, Athens, Georgia Nelson A. Cox, U.S. Department of Agriculture, Athens, Georgia Faith J. Critzer, University of Georgia, Griffin, Georgia Sherill K. Curtis (retired), U.S. Food and Drug Administration, Laurel, Maryland Michelle D. Danyluk, University of Florida, Lake Alfred Angelo DePaola, U.S. Food and Drug Administration, Dauphin Island, Alabama

Stephanie Doores, Penn State University, University Park Hari P. Dwivedi, bioMe´rieux, Hazelwood, Missouri Catherine W. Donnelly, The University of Vermont, Burlington Michael P. Doyle, University of Georgia, Griffin Philip H. Elliott, Grocery Manufacturers Association, Washington, District of Columbia Elena Enache, Grocery Manufacturers Association, Washington, District of Columbia George M. Evancho (retired), Lewes, Delaware Marianne K. Fatica, U.S. Food and Drug Administration, College Park, Maryland Ronald Fayer, U.S. Department of Agriculture, Beltsville, Maryland Aamir Fazil, Public Health Agency of Canada, Guelph, Ontario Peter Feng, U.S. Food and Drug Administration, College Park, Maryland Antonio Garrido-Ferna´ndez, Instituto de la Grasa, Seville, Spain Russell S. Flowers, Me´rieux NutriSciences, Chicago, Illinois Anthony J. Fontana, Truesdall Laboratories, Tustin, California Sally Foong-Cunningham, Ecolab, St. Paul, Minnesota Karin Francis, University of California, Dublin Augusto A. Franco (deceased), U.S. Food and Drug Administration, Laurel, Maryland Pina M. Fratamico, U.S. Department of Agriculture, Wyndmoor, Pennsylvania Timothy A. Freier, Cargill, Wayzata, Minnesota Jonathan G. Frye, U.S. Department of Agriculture, Athens, Georgia Richard K. Gast, U.S. Department of Agriculture, Athens, Georgia Narjol Gonzalez-Escalona, U.S. Food and Drug Administration, College Park, Maryland Gopal Gopinath, U.S. Food and Drug Administration, Laurel, Maryland Hassan Gourama, Penn State University, Reading Rodney J. H. Gray, DSM Nutritional Products, Parsippany, New Jersey Gail E. Greening (retired), Institute of Environmental Science & Research, Porirua, New Zealand Linda E. Grieme, Ecolab, St. Paul, Minnesota Christopher J. Grim, U.S. Food and Drug Administration, Laurel, Maryland Joshua B. Gurtler, U.S. Department of Agriculture, Wyndmoor, Pennsylvania Paul A. Hall, AIV Microbiology & Food Safety Consultants, Hawthorn Woods, Illinois Jennifer M. Hait, U.S. Food and Drug Administration, College Park, Maryland

| xix

Compendium of Methods for the Microbiological Examination of Foods |

Roberta M. Hammond, U.S. Food and Drug Administration, College Park, Maryland Linda J. Harris, University of California, Davis Melinda Hayman, Grocery Manufacturers Association, Washington, District of Columbia Jos Houbraken, CBS-KNAW Fungal Biodiversity Centre, Utrecht, The Netherlands Scott K. Hood, General Mills, Minneapolis, Minnesota Kristin Houck, Center for Dairy Research, Madison, Wisconsin Rebecca Illsley, The Clorox Company, Kennesaw, Georgia Lan Hu, U.S. Food and Drug Administration, Laurel, Maryland Tim Jackson, Nestle´ North America, Glendale, California Charlene R. Jackson, U.S. Agricultural Research Service, Athens, Georgia Karen G. Jarvis, U.S. Food and Drug Administration, Laurel, Maryland Rufino Jime´nez-Dı´az, Instituto de la Grasa, Seville, Spain Suzanne D. Johanningsmeier, U.S. Department of Agriculture, Raleigh, North Carolina Eric A. Johnson, Research Institute, Madison, Wisconsin Deana R. Jones, U.S. Department of Agriculture, Athens, Georgia Jessica L. Jones, U.S. Food and Drug Administration, Dauphin Island, Alabama Thomas Jones, DFA of California, Fresno Robin M. Kalinowski, Tyson Foods, Chicago, Illinois Ai Kataoka, Grocery Manufacturers Association, Washington, District of Columbia Jinkyung (Jeannie) Kim, Silliker, Crete, Illinois Phillip H. Klesius (retired), U.S. Department of Agriculture, Auburn, Alabama Dennis J. Kopecko (retired), U.S. Food and Drug Administration, Silver Spring, Maryland Jeffrey L. Kornacki, Kornacki Microbiology Solutions, Madison, Wisconsin Mahendra H. Kothary, U.S. Food and Drug Administration, Laurel, Maryland Ronald G. Labbe, University of Massachusetts, Amherst Katie Laird, De Montfort University, Leicester, United Kingdom Anna M. Lammerding, AML Consulting, Guelph, Ontario Keith A. Lampel, U.S. Food and Drug Administration, Laurel, Maryland Kathleen A. Lawlor, PepsiCo, Valhalla, New York Loralyn Ledenbach, Kraft Foods Group, Glenview, Illinois Sean J. Leighton, The Coca-Cola Company, Atlanta, Georgia Fritz Lembke, Tetra Pak, Stuttgart, Germany J. Eric Line, U.S. Department of Agriculture, Athens, Georgia Susan E. Maslanka, Centers for Disease Control and Prevention, Atlanta, Georgia Giorgio Mastromei, University of Florence, Italy Susan McCarthy, U.S. Food and Drug Administration, Dauphin Island, Alabama Rachel McEgan, University of Guelph, Kemptville, Ontario Wendy McMahon, Silliker, Crete, Illinois Ann Marie McNamara, Jack in the Box, San Diego, California Geoff Mead, Bath, United Kingdom David Melka, U.S. Food and Drug Administration, College Park, Maryland Indaueˆ Ieda Giriboni de Mello, The Coca-Cola Company, Atlanta, Georgia

xx |

Jianghong Meng, University of Maryland, College Park David A. Mills, University of California, Davis Lloyd Moberg, Church & Dwight, Princeton, New Jersey Marirosa Molina, U.S. Environmental Protection Agency, Athens, Georgia Mark Moorman, Kellogg Company, Battle Creek, Michigan Emilia Rico-Munoz, BCN Research Laboratories, Rockford, Tennessee Ranzell Nickelson II, Red Mesa Laboratory Services, Fort Worth, Texas Nenge Azefor Njongmeta, Kraft Foods Group, Northfield, Illinois Brian B. Oakley, U.S. Department of Agriculture, Athens, Georgia Karl E. Olson (retired), Abbott Nutrition, Scottsdale, Arizona Ainsley M. Otten, Public Health Agency of Canada, Guelph, Ontario Greg Paoli, Risk Sciences International, Ottawa, Ontario Mickey E. Parish, U.S. Food and Drug Administration, College Park, Maryland Nina G. Parkinson, University of California, Davis Keila L. Pe´rez, Texas A&M University, College Station Ilenys M. Pe´rez-Dı´az, U.S. Department of Agriculture, Raleigh, North Carolina Ruth L. Petran, Ecolab, St. Paul, Minnesota David H. Pincus, bioMe´rieux, Hazelwood, Missouri Joan M. Pinkas, McCormick, Hunt Valley, Maryland Richard Podolak, Grocery Manufacturers Association, Washington, District of Columbia Julia W. Pridgeon, U.S. Department of Agriculture, Auburn, Alabama Michael S. Ramsey, University of California, Davis Lawrence Restaino, R&F Laboratories, Downers Grove, Illinois Gary P. Richards, U.S. Department of Agriculture, Dover, Delaware Jason Richardson, The Coca-Cola Company, Atlanta, Georgia Steven C. Ricke, University of Arkansas, Fayetteville Emilia Rico-Munoz, BCN Research Laboratories, Rockford, Tennessee Lauren Posnick Robin, U.S. Food and Drug Administration, College Park, Maryland Amy B. Ronner, Silgan Containers Manufacturing Corporation, Oconomowoc, Wisconsin Elliot T. Ryser, Michigan State University, East Lansing, Michigan Dojin Ryu, University of Idaho, Moscow Robert A. Samson, CBS-KNAW Fungal Biodiversity Centre, Utrecht, The Netherlands Venugopal Sathyamoorthy, U.S. Food and Drug Administration, Laurel, Maryland Keith R. Schneider, University of Florida, Gainesville Jean L. Schoeni, Covance, Madison, Wisconsin James D. Schuman, PepsiCo, Barrington, Illinois Shashi Sharma, U.S. Food and Drug Administration, College Park, Maryland Ronald D. Smiley, U.S. Food and Drug Administration, Jefferson, Arkansas Ben A. Smith, Public Health Agency of Canada, Guelph, Ontario L. Michele Smoot, Me´rieux NutriSciences, Chicago, Illinois Les Smoot, U.S. Food and Drug Administration, College Park, Maryland Marianne Smukowski, University of Wisconsin-Madison

|

John N. Sofos, Colorado State University, Fort Collins, Colorado Haim M. Solomon (deceased), U.S. Food and Drug Administration, College Park, Maryland Kent M. Sorrells (retired), Springville, California William H. Sperber, The Friendly Microbiologist, Minnetonka, Minnesota Bradley A. Stawick, Microbac Laboratories, Pittsburgh, Pennsylvania Kenneth E. Stevenson (retired), Clio, California Ben D. Tall, U.S. Food and Drug Administration, Laurel, Maryland Sandra M. Tallent, U.S. Food and Drug Administration, College Park, Maryland Elena Tamburini, University of Cagliari, Monserrato, Italy Peter J. Taormina, John Morrell Food Group, Cincinnati, Ohio T. Matthew Taylor, Texas A&M University, College Station Bradley Tompkins, Vermont Department of Health, Burlington R. Bruce Tompkin (retired), ConAgra Refrigerated Foods, LaGrange Highlands, Illinois

Authors

Suzanne Tortorelli, Campbell Soup Company, Camden, New Jersey Mary Lou Tortorello, U.S. Food and Drug Administration, Bedford Park, Illinois Valentina Trinetta, Ecolab, St. Paul, Minnesota Aaron Uesugi, Kraft Foods Group, Glenview, Illinois Peter S. Unger, American Association for Laboratory Accreditation, Frederick, Maryland Angela M. Valadez, University of Florida, Lake Alfred Purnendu C. Vasavada, University of Wisconsin-River Falls Alissa M. Wesche, Old Orchard, Sparta, Michigan Irene Wesley (retired), U.S. Department of Agriculture, Des Moines, Iowa Charlene Wolf-Hall, North Dakota State University, Fargo Randy W. Worobo, Cornell University, Ithaca, New York Lihua Xiao, Centers for Disease Control and Prevention, Atlanta, Georgia Sung-Sik Yoon, Yonsei University, Wonju, South Korea Guodong Zhang, U.S. Food and Drug Administration, College Park, Maryland

| xxi

Reviewers

Anthony D. Hitchins (retired), U.S. Food and Drug Administration, Rockville, Maryland Larry Beuchat, University of Georgia, Griffin Arun Bhunia, Purdue University, West Lafayette, Indiana Dean Bodager, Florida Department of Health, Orlando Enrico Buenaventura, Health Canada, Ottawa, Ontario Scott L. Burnett, Ecolab, St. Paul, Minnesota Larry Cohen, Magrabar, Morton Grove, Illinois Frederick K. Cook, MOM Brands, Lakeville, Minnesota Victor Cook (retired), U.S. Department of Agriculture, Washington, District of Columbia Maribeth Cousin (retired), Elkhorn, Wisconsin Patricia A. Curtis, Auburn University, Auburn, Alabama James S. Dickson, Iowa State University, Ames Francisco Diez-Gonzalez, University of Minnesota, St. Paul Ralph DiGiacomo, PepsiCo, Valhalla, New York L. Scott Donnelly, scott-donnelly.com, Burlington, Vermont Stephanie Doores, Penn State University, University Park Richard M. Driggs, Nestle´ Purina North America, St. Louis, Missouri Charles Edwards, Washington State University, Pullman Philip H. Elliott, Grocery Manufacturers Association, Washington, District of Columbia Elena Enache, Grocery Manufacturers Association, Washington, District of Columbia Karin Francis, University of California, Dublin Joseph F. Frank, University of Georgia, Athens Kathleen Glass, University of Wisconsin-Madison Margaret D. Hardin, IEH Laboratories & Consulting Group, Lake Forest Park, Washington Melinda Hayman, Grocery Manufacturers Association, Washington, District of Columbia Robert R. Hirst, International Bottled Water Association, Alexandria, Virginia Paul in ‘t Veld, Food and Consumer Product Safety Authority, Utrecht, The Netherlands Wilma Jacobs-Reitsma, National Institute for Public Health and the Environment, Bilthoven, The Netherlands Marlene E. Janes, Louisiana State University, Baton Rouge Crystal N. Johnson, Louisiana State University, Baton Rouge Susanne E. Keller, U.S. Food and Drug Administration, Bedford Park, Illinois David Kingsley, U.S. Department of Agriculture, Dover, Delaware Kalmia Kniel, University of Delaware, Newark

Jeffrey L. Kornacki, Kornacki Microbiology Solutions, Madison, Wisconsin Gayle A. Lancette (retired), U.S. Food and Drug Administration, Atlanta, Georgia Jean Lu, Kennesaw State University, Kennesaw, Georgia Lisa Lucia, Texas A&M University, College Station, Texas Wendy Marcucci, Ball Corporation, Westminster, Colorado Steven Murphy, Cornell University, Ithaca, New York Greg Paoli, Risk Sciences International, Ottawa, Ontario Carol Phillips, The University of Northampton, United Kingdom John Pitt, Food Science Australia, North Ryde, New South Wales, Australia J. Mark Powell, U.S. Department of Agriculture, Madison, Wisconsin Kathleen T. Rajkowski, U.S. Department of Agriculture, Wyndmoor, Pennsylvania Sadhana Ravishankar, University of Arizona, Tucson P. Gopal Reddy, Tuskegee University, Alabama Scott Russell, University of Georgia, Athens Joelle Salazar, U.S. Food and Drug Administration, Bedford Park, Illinois Yvonne Salfinger (retired), Florida Department of Agriculture and Consumer Services, Tallahassee Brian Sauders, New York State Department of Agriculture & Markets, Albany James D. Schuman, PepsiCo, Barrington, Illinois Shashi Sharma, U.S. Food and Drug Administration, College Park, Maryland Joseph R. Shebuski, Cargill, Wayzata, Minnesota Manpreet Singh, Purdue University, West Lafayette, Indiana Gregory R. Siragusa, DuPont, Waukesha, Wisconsin Sterling S. Thompson, The Hershey Company, Hershey, Pennsylvania Suzanne Tortorelli, Campbell Soup Company, Camden, New Jersey Valerie Tournas, U.S. Food and Drug Administration, College Park, Maryland Socrates Trujillo, U.S. Food and Drug Administration, College Park, Maryland Purnendu C. Vasavada, University of Wisconsin-River Falls Benjamin Warren, Land O’Lakes, Arden Hills, Minnesota Burton Wilcke, The University of Vermont, Burlington Charlene Wolf-Hall, North Dakota State University, Fargo Donald L. Zink, U.S. Food and Drug Administration, College Park, Maryland

| xxiii

Preface

T

he globalization of food ensures that foodstuffs are available from every corner of the world, at any time of the year. Consumer demand for fresher foods that are conveniently prepared have extended shelf life, and yearround availability in the market has resulted in an amazing array of novel products and formulations. People can choose from an increasing variety of foods; they no longer need to purchase their foods from a single store, town, or country; and they are increasingly choosing to eat their food outside the home. These changes in the market and in consumer behavior, coupled with changing methods of food production, handling, and distribution, have created emerging microbiological problems that did not exist only a short time ago. Both the food industry and regulatory agencies are facing these challenges: along with providing a seemingly limitless range and availability of food, ensuring the safety of this food supply beyond question is a must. Recent statistics gathered by the Centers for Disease Control and Prevention indicate that approximately 9.4 million episodes of foodborne illness occur annually in the United States, with more than 55,000 hospitalizations and 1,351 deaths attributed to 31 known pathogens. Many more cases—as many as 38.4 million—are estimated to be caused by unspecified or unknown agents. Although foodborne illnesses are a threat to public health, so is an inadequate food supply. Globally, the population of human beings has surpassed the 7-billion mark, and it is expected to reach 8 billion by the year 2025. Food spoilage and waste must be diminished to ensure adequate healthy food for the world. Microbiological methods play a major role in ensuring that the risk of food contamination by both foodborne pathogens and spoilage agents is minimized. The beneficial microorganisms that underlie some of humankind’s oldest technologies (i.e., fermenting and preserving foods) represent the other end of the food microbiology spectrum. Probiotic microorganisms have been increasingly marketed as part of a healthy diet. The methods used for the analysis of beneficial microorganisms in foods must be reliable to ensure that they are functioning as desired in delivering the expected food product characteristics. The Compendium is a primary reference for all food microbiology testing laboratories. Encompassing pathogens, spoilage microorganisms, and beneficial microorganisms, the Compendium is a unique reference for the microbiological analysis of foods. All microorganisms of relevance to food safety and quality—the bacteria, yeasts, molds, viruses,

parasites—are represented in this manual. The Compendium not only collects methods for the individual microorganisms but also covers information relevant to the various food commodities. Since the publication of the fourth edition in 2001, there have been numerous advances in the field of analytical food microbiology and significant changes relating to food safety and public health. In the fifth edition, content has been updated to keep up with the many technological innovations and improvements in the microbiological analysis of foods. The fifth edition has been organized into sections: General Methods, Physiological Groups of Microorganisms, Microbial Genera, Food Commodities, Reference, and Appendix, with each section containing new or updated chapters. Methods for analyzing cell injury and detection of injured cells, featured in earlier editions, have been revisited and updated anew. The number of products incorporating probiotics has increased significantly in the market in recent years, and analytical methods for these microorganisms are included in the new edition for the first time. Cronobacter (formerly known as Enterobacter) sakazakii has emerged as a pathogen and has been given its own chapter. Among the Food Commodities chapters is a new one covering pet foods and animal feed, in recognition of the link between the microbiological safety of these foods and public health. Furthermore, extending its reach into the world of alcoholic beverages, the Compendium includes a new chapter devoted to the microbiological methods for wine and beer. Lastly, this new edition is available both in print and digital formats. The information contained in the fifth edition is the latest available at the time of publication and is presented with the recognition that future endeavors in microbiological methods will surely improve the ability to ensure the safety and quality of the foods we consume. In the words of Marvin Speck, editor of the first edition of the Compendium: ‘‘Shortcomings of different methods are recognized widely … hopefully, research will eliminate these from current analytical methodology.‘‘ The editors thank the members of the editorial board for their dedication to the fifth edition as well as the authors and reviewers for their hard work in providing the highest quality content for these new and updated chapters. The essential work of the APHA staff is gratefully acknowledged; without their organizational efforts and critical publishing knowledge, this new edition would not have unfolded.

| xxv

|

SECTION I

|

General Methods

| 1 |

|

CHAPTER 1

|

Laboratory Quality Management Systems Roger M. Brauninger, Michael H. Brodsky, L. Michele Smoot, and Peter S. Unger

1.1

provide additional benefits to the laboratory. Monitoring and maintaining equipment to ensure proper functioning reduces the risk of operational hazards and reduces the possibility of equipment failure that leads to lost samples and analytical downtime. Another benefit of a laboratory QMS system is the standardization of analytical methods. Record-keeping activities used for laboratory QC and required for assessment audits provide information that helps management evaluate analytical proficiency and consistency. The ability of management and analysts to monitor the quality of work promotes confidence in results and pride in the laboratory’s performance. This chapter introduces the principal concepts of QA for the food microbiology laboratory using the International Organization for Standardization/International Electrotechnical Commission standard (ISO/IEC) 17025:20051 as the basis for the QMS. It is not possible to present specific QA programs to meet the needs of every food microbiology laboratory; rather, this chapter is based on outcomes and not prescriptive. Laboratory QC activities will be addressed, but specific approaches will be referenced rather than discussed in detail. The reader should be mindful that this chapter is intended to serve as a philosophical guide for the design and implementation of laboratory QA programs.

INTRODUCTION

A quality management system (QMS) with both national and international recognition is critical to the successful operation of food testing laboratories. QMS is an organized structure of responsibilities, activities, resources, and events that are integrated to ensure the capability of a laboratory to meet quality requirements. The components of a functional QMS include interrelated documentation of the quality policies or objectives (what to do), procedures (how to do it), and evidence of compliance (records). Quality assurance (QA) is an inherent component of any QMS. The objective of QA is to ensure the reliability of analytical information used in decision-making. Acceptable limits may vary depending on how the information is used, but some variation in results from biological analyses is expected. QA has two major components: 1. Quality control (QC): a specific activity whose purpose is to monitor a discrete laboratory task to ensure that it meets a predefined criterion. Every control point should have written instructions that define the tolerance limits and describe the action(s) required for compliance. 2. Assessment: consists of audit activities whose purpose is to review the efficacy of quality control.

1.2

Laboratory QA is a system of activities encompassed in a QMS, which allows a laboratory to demonstrate its ability to provide measurable high-quality services to its customers. QA is a management tool to ensure that appropriate QC and assessment procedures are performed and documented in a dependable, timely, and economic manner to meet the needs of the customer. A critical aspect of QA is recognizing when QC outcomes do not comply with expectations and taking and documenting the appropriate corrective action. Although the principal objective of a laboratory QA program is to ensure the reliability of data, the systems and procedures required for an effective program of this sort

THE ROLE OF MANAGEMENT IN LABORATORY QA

Direct responsibility for the design and implementation of a laboratory QMS lies with laboratory management. Management must evaluate the risks associated with laboratory variability, including the cost of errors and the time required to address such errors, and the costs and benefits of reducing variability through a QA program. Measurement of the cost of these activities must account for materials, analyst time, and overhead expenses. The first step in developing a QA program for the laboratory is to draft a quality manual that includes a quality policy statement and outlines the requirements for the following: | 3 |

Compendium of Methods for the Microbiological Examination of Foods |

N N N N N N N N N N N N N N

Organization and management Quality systems for media and reagent quality control, audit and review Personnel requirements and training Accommodation and environment critical to the integrity of test results, such as cleaning and sanitation Use and maintenance of equipment and reference materials Calibration of equipment and test materials suitable for the tests being performed, traceable to national and/or international sources Calibration procedures Validated test methods, handling of calibration items (thermometers, weights, reference cultures) Sample handling Records Certificates and/or reports Subcontracting or testing Outside support and supplies Complaints concerning the quality of the data

Specific standard operating procedures (SOPs) should be written for each of the basic components of the analytical framework in a food laboratory. Each SOP should include or reference specific QC activities for critical components. SOPs should be sufficiently detailed and explicit to minimize deviations that could affect the reliability of the data.

1.21

Documents and Records

Today’s laboratory includes a variety of documents available through various media. Examples include standards, test and/or calibration methods, software, instructions, manuals, policy statements, procedures, textbooks, memoranda, flow diagrams, and so on. All documents, whether generated internally or externally, must be controlled to ensure that only current versions are in use, that any changes to documents go through an appropriate review process, that outdated versions are filed for reference, and that changes are clearly identified for notification and training purposes. Records specific to laboratory samples and analytical data must be complete and include all aspects of handling, from the time of collection (or minimally from the time of receipt at the laboratory) until the time of sample disposal. Further, records specific to the analysis must include detailed information on the media/reagents used in performing the test, personnel involved in the testing, timing of each step in the analysis, raw data, and observations, calculations, conclusions, and sample retention. This information may be stored electronically or in bound laboratory notebooks. Regardless of the storage mechanism, dates must be entered permanently and any changes to an entry must allow for observation of the original entry and include an explanation for the change. When computers or automated equipment are used to collect, process, or report data, the system must be validated for their specific use. There must also be procedures for data protection and integrity. Record retention times and manner of storage will depend on the scope of work performed and the nature of the records. 4 |

1.22

Complaint and Non–conformity Processes

Nomenclature to describe failure in meeting QMS or customer requirements may vary from laboratory to laboratory. In general, complaints are often defined as feedback from customers or other non-laboratory parties, and non–conformities are defined as failure to conform to the QMS or previously agreed customer requirements. All laboratories should have active processes in which customer complaints and non–conformities to the QMS are effectively addressed. These processes should identify personnel accountable for addressing the non–conformities, define the responsibilities and authorities of the personnel who are accountable, describe the procedures for responding and correcting the non–conformity, and for the determination of the effectiveness of the corrective action.

1.23

Tool for Continual Improvement

All laboratories should be actively involved in a formal continual improvement process. Continual improvement includes using a formal process of identifying and addressing complaints and non–conformity to the QMS. There are two keys to successful continual improvement. The first involves measurements of current performance. These establish a baseline from which to measure improvement efforts, both short term and long term. The second key is establishing a formal process for problem solving and implementing process changes. Many techniques have been described for addressing complaints, non–conformities, and improvement opportunities. All have some common features that look at processes and outcomes and may be realized through ISO/IEC 17025 accreditation of the laboratory. Identification and documentation of the complaint or non–conformance is the initial step. Identification can come from a variety of sources, including customers, staff observations, process verification data, or review of results and audits. Proper documentation of the process and subsequent investigation is critical. The largest source of ineffective corrective actions often lies with improper problem identification or a poor rootcause analysis. The problem must be clearly and unambiguously defined, thereby allowing team members to address the problem from the same perspective. Including the names of personnel, possible root causes, or potential corrective actions during the problem identification stage will further complicate the issue. Investigation and data gathering are the essential next steps. A common error in this phase is the omission of personnel involved in the non–conformance. It is essential that laboratory personnel are included in the investigation and data-gathering efforts. It is also important to quantify the problem to describe its scope or impact. Objective data take emotion out of the problem-solving process. At this point in the process of root-cause analysis it is important to identify whether the problem is due to a common cause (process) or a special cause. Common cause is defined as a historical, quantifiable variation in a system that leads to non-fulfillment of a requirement. Special cause is defined as an unusual, not previously observed, non-quantifiable variation in a system that leads to a non-fulfillment of a requirement. Many tools are available to identify and evaluate potential root causes. A historical review of non–conformities and

|

previously identified root causes is often valuable at this stage. Identification of potential corrective actions, along with selection and implementation of the final corrective action, should also involve personnel affected by the process change. Selection of the corrective action should also take into account the historical data and trend analysis. Key considerations in this step include the potential for success, impact on laboratory operations and customers, cost of corrective action, and difficulty of implementation. Success requires that all key personnel involved with any changes are in agreement with the corrective action taken. Assigning responsibility for implementation and performance improvement is also critical. The final step—evaluating the effectiveness of the corrective action—is often overlooked or not done thoroughly, but it is critical for effective performance. Establishing the criteria to be used to determine the effectiveness of the corrective action prior to implementation will facilitate proper execution of the evaluation process in an unbiased manner. Many of the principles used in problem resolution appear straightforward but are very difficult to practice effectively. Critical appraisal of the outcome at each stage is essential.

1.24

Control of the Laboratory Process

As indicated, formal categorization of complaints and non– conformities allows tracking and trending analyses in a manner similar to statistical process control (SPC) charting. A review of recurring issues and historical performance provides greater insight into the effectiveness of corrective actions than single, snapshot measures taken a short time after the implementation of a corrective action. Critical review of the effectiveness of the corrective actions is essential. Without this, the documentation of non–conformities can be reduced to a paper shuffle with limited value.

1.3

LABORATORY OPERATIONAL FRAMEWORK AND QC REQUIREMENTS

The following summarizes the basic components of the analytical framework of a comprehensive QA system for a food laboratory:

N N N

Written directions are required for all laboratory-related activities to maximize uniformity. Records or evidence of compliance must be maintained. Specific QC protocols that include measuring, monitoring and analyzing data are required to address the critical components that contribute directly to analytical variability or measurement uncertainty.

1.31

Facilities

1.311 Separation of Activities The laboratory should identify ‘‘high-risk’’ and ‘‘low-risk’’ activities as they relate to the potential for cross-contamination. Activities that involve the handling of cultures, such as transfers, isolation procedures, and biochemical/serological identifications, are considered high risk, whereas low-risk activities include activities prior to sample enrichment such as sample preparation, sample log-in, or media preparation. Once each activity is identified, laboratory operations

Laboratory Quality Management Systems

should be designed to separate high-risk activities from low-risk activities. Additionally, not only should the activities be separated, but the individuals that perform those activities should also be dedicated to specific activities as well. Sharing of equipment and materials should be limited and only considered between activities of similar risk levels. In addition to the characterization of high- and low-risk activities, laboratory management should also characterize the types of samples being processed into high and low microbial load samples. High microbial load samples should be separated by space and time from low microbial load samples in order to prevent cross-contamination.

1.312 Workflow Considerations The laboratory should be designed and organized to contain all of the equipment, materials, and space necessary for the workers to successfully perform the analysis. The location of equipment and materials in the laboratory should be positioned to minimize traffic intended to retrieve supplies or access incubators. The workstations should be stocked with all essential supplies needed by the staff for a given day. The samples should flow in one direction through the laboratory to minimize the potential for cross-contamination. Refrigerators should be readily accessible in order to properly store perishable samples prior to analysis. Workers should have easy access to handwashing stations dedicated to their department. 1.313

General Environment of the Laboratory Conducive to Safety and Proper Practices The laboratory should be air-conditioned and well ventilated to minimize temperature variations. The air-conditioning unit with clean vent filters will reduce the amount of particulates in the air. The laboratory should be designed with worker safety in mind. It needs to be spacious enough to include all necessary equipment and have adequate workbench space for the staff. Adequate storage is needed to minimize clutter, which allows for proper cleaning and sanitization of surfaces. The laboratory should be well lit, with a maintained light intensity of approximately 50–1000 lumens. A dependence on natural light is discouraged during the day owing to high variability in intensity. Direct sunlight should also be avoided as it can negatively affect media, reagents, and organisms. Laboratory conditions should be comfortable for workers. It is recommended that the laboratory atmosphere be at an ambient temperature between 21uC and 23uC, with a relative humidity of 45%–50%. 1.314

Safety Practices in Accordance with Local Ordinances Laboratory facilities should be designed to comply with federal, state, and local building and safety codes. Employees should have easy access to working fire extinguishers, alarms, eyewash stations, and safety showers. Laboratory personnel must be required to wear appropriate personal protective equipment, such as safety glasses, autoclave gloves, face shields, laboratory coats, or other protective clothing. All entrances must be secured, and visitors should sign in | 5

Compendium of Methods for the Microbiological Examination of Foods |

before being granted access to the facility. It is recommended that an employee accompany visitors at all times while they are in the building.

1.315 Housekeeping and Environmental Monitoring A master cleaning schedule and appropriate documentation should be established for the laboratory to ensure that cleaning is documented and can be monitored for effectiveness. Laboratory surfaces should be cleaned prior to sanitization with a cleaning solution that contains surfactants to remove dirt and organic materials. For sterilization there are several disinfectant options, such as iodophors, chlorine, quaternary ammonium, or phenolic disinfectants. To verify the effectiveness of the cleaning schedule, a standard operating procedure for environmental monitoring should be established. The operating procedure should describe the sampling procedure, the locations to be sampled, and the procedure for responding to a positive result for qualitative analysis or out-of-specification data for quantitative analysis. Laboratory personnel should perform the environmental sampling technique consistently, and, although the locations and frequency are detailed in the operating procedure, the actual sites should be chosen randomly.

1.42

Personnel considered analysts should include all nonmanagement roles. Often, laboratory management will also have analyst responsibilities. It is not constructive to the effectiveness of QMS to differentiate between professionals (analysts, technologists, etc.) and non-professionals (assistants, technicians, clerical, etc.) within the framework of the system. Although proper organization and assignment of responsibilities is critical, all personnel must take ownership of the QMS within their area of responsibility and expertise. It is imperative that quality permeates throughout the organizational structure. Critical QMS activities must reside at the bench level. It is not uncommon for laboratories to establish a quality function separate from the laboratory process. This type of organization has value if the separated activities involve non-routine tasks, general system documentation, selected organizational functions, training, and other duties not directly related to the generation of test results. Analysts must be involved in all activities related to process control and verification. Quality must be ‘‘done’’ by the analysts and not by a separate function within the laboratory.

1.43 1.4

PERSONNEL

Laboratory personnel must work within the policies and procedures of the laboratory QMS. The QMS must contain job descriptions for all personnel that define responsibilities related to the performance of tests, adherence to the QMS, required education, experience, skills, and training relevant to duties. Managerial duties should also be defined if applicable. All personnel must be competent in their described responsibilities. Fulfillment of their respective roles by each member of the laboratory team is critical to QMS performance.

1.41

The Role of Laboratory Management

Personnel considered members of laboratory management should not be limited to those with the title of manager or director. Laboratory management is defined as from supervisors up to the highest level of management directly responsible for the outcome of the laboratory efforts. This can and should include executives. Many references discuss the role of laboratory management in QA. Most of these discussions focus on business management, personnel management, process management, and quality management functions. All points made are pertinent to overall laboratory management functions. However, the single most significant action laboratory management can take to ensure the success of its QMS is to take ownership of it. True ownership is an emotional state as much as a state of mind. Ownership is demonstrated in actions, not just words or written policies. If management feels a sense of ownership of the QMS, their actions will result in demonstrable support of it. This greatly reduces the occurrence of conflict between words and actions. Such conflicts are typically viewed by personnel as hypocritical and result in negative consequences for QMS performance. 6 |

Role of Analysts

Evaluation of Personnel

Many literature sources discuss the recruitment, selection, and training of personnel. The performance of the personnel within the criteria of the laboratory QMS is the final determining factor for success or failure of these efforts. Criteria for competency must be based on an expected value, the measured variation associated with the laboratory process, and statistical analysis of multiple data points. This indicates that the laboratory must have measures of and criteria for its process capability for each method performed. The use of published criteria may not be suitable, depending on the comparable laboratory process capability. Laboratory-developed failure rates for qualitative analyses and statistically derived variations for quantitative analyses meet these needs. An excellent approach for quantitative analysis, not historically applied to microbiology, is the use of SPC charting. A control chart is a graphical representation of analytical results with respect to the time or sequence of measurements. The graph includes limits within which results are expected to fall when the analytical process is in a state of ‘‘statistical control.’’ These limits are referred to as upper and lower control limits. In addition to providing a clear boundary between in-control and out-of-control data, control charts illustrate trends and cycles that provide valuable information on the analytical process that may require appropriate corrective action. The X-bar and Rcharts are the most popular of all control charts and are appropriate for quantitative microbiology when control samples are available with known cell levels (Figures 1-1 and 1-2). For a detailed description of the principles of statistical process control charting the reader is referred to Shewart et al.3 In brief, to use SPC charting the laboratory firsts generates data under conditions known to be in control over a specified period of time (usually 20 days or more). These data are used to calculate the parameters of the X-bar (Figure 1-1)

|

Figure 1-1. X-bar chart for aerobic plate count.

and R-charts (Figure 1-2). By using data over a given period, the laboratory is able to define the variation normally associated with their testing process. Whenever the testing process is changed, new charts need to be established. The X-bar, or top portion of the chart, plots the overall process average and the daily points, while the bottom chart, or the R-chart, provides a measure of the precision of the process by plotting the day-to-day variation of the analysis. It is important to note that microbiological data must first be normalized by converting to log10 prior to data manipulation. When an analyst’s results are repeatedly within the expected variation established for a given method and randomly distributed about the sample mean, competency is demonstrated. Participation in a reputable proficiency program can also be used to demonstrate competency to a method. ISO/ IEC 17025:2005 states the following: The laboratory shall have quality control procedures for monitoring the validity of tests and calibrations undertaken. The resulting data shall be recorded in such a way that trends are detectable and, where practical, statistical techniques shall be applied to the reviewing of the results. This monitoring shall be planned and reviewed and may include, but not be limited to, the following: participation in interlaboratory comparison or proficiency testing programs.

For quantitative microbiology, the criterion for determining whether or not the data is acceptable involves assigning a z-value to each submitted result. This is accomplished by converting the results to log10 and calculating the mean and standard deviation for the set of data submitted by all participants for that analyte. The z-value is obtained by subtracting the mean from the individual result and dividing by the standard deviation. Therefore, the z-value is simply the number of standard deviations the result is from the mean. A negative number indicates that the result is below the mean, whereas a

Figure 1-2. R-bar chart for aerobic plate count.

Laboratory Quality Management Systems

positive result means the submitted result is above the mean. Z-value graphs for a specific test represent how many standard deviations from the mean a given result is for a given data set. Z-value graphs are a valuable tool for investigating long-term biases that laboratories may have relative to other laboratories. The rules that apply to the evaluation of data on statistical process control charts are also applicable to z-value graphs. For example, a run of eight or more data points all above or below the mean indicates a systematic error that is probably persistent over a long period of time when it involves proficiency results. A series of seven or more consecutive points in a row moving steadily up or down may reflect a gradually deteriorating situation within the process. Data points outside three z-values would indicate a process that is out of control and would not serve as proof of competency to a given method. For qualitative microbiology, the criteria for determining the acceptability of results include the detection of the organism in positive samples and the absence of the organism in negative samples. When detected, further identification may be initiated and those results are compared to results submitted by the population, as well as the known inoculum. When selecting a proficiency program, it is important to choose one that 1. offers food matrices commonly analyzed by the laboratory, 2. includes target organisms that are characteristically isolated from foods, and 3. uses inoculum levels representative of levels normally present in the matrix being evaluated. There is little value in performing a qualitative analysis on a food sample that has been inoculated with high levels of the target organism. Qualitative test methods are designed to detect low levels of a particular microorganism. Small changes in the testing protocol that might affect recovery in a routine sample would not be uncovered in a check/proficiency sample if high inoculum levels are used. Similarly, the organisms used in a check/proficiency sample program should reflect typical strains isolated from food matrices. The methods by which these samples are being tested were developed and validated specifically for food matrices. Use of an atypical strain is not reflective of the testing process and thus challenges the test method rather than the laboratory’s testing process or an analyst’s performance. Check/proficiency samples should be treated as any other sample being analyzed in the laboratory. Assigning a laboratory number indistinguishable from other sample identification numbers can assist in ensuring that check/proficiency samples follow the same process as routine samples. Regardless of how competency is demonstrated, every analyst must prove competency to each method prior to generating client results. Analyst competency must be continually and regularly verified through the use of known controls and include the entire testing process, not just a single step in the method. | 7

Compendium of Methods for the Microbiological Examination of Foods |

1.44

Evaluation of the Laboratory’s Competence

Just as the previous section discusses the performance criteria for personnel, these same approaches are frequently useful for determining the competence of the laboratory itself. Accreditation bodies use the data obtained from participation in proficiency testing programs in an analogous manner. Criteria for competency must be based on an expected value, the measured variation associated with the laboratory process, and statistical analysis of multiple data points (the difference being that the proficiency data represent the organization as output). But care must be taken to ensure that these activities are performed as close to a routine as possible.

1.5

EQUIPMENT/INSTRUMENTATION

The laboratory is furnished with many items of equipment, including reference materials, required for the correct performance of testing. Regularly scheduled maintenance of this equipment is essential to the smooth operation of the laboratory. A lack of attention to this aspect of the laboratory QMS will lead to unexpected and expensive equipment failures. Besides performing regularly scheduled maintenance of equipment, ISO/IEC 17025:2005 requires that laboratories have a record of the maintenance plan and the maintenance performed. All equipment needs to have a unique identifier that is clearly displayed, as well as an equipment file for each piece of critical equipment and its software and/or operation manual. Critical equipment is defined as any equipment that directly affects the analytical results.

1.51

Traceability to National Standards

Nearly every determination in chemistry and microbiology involves equipment that must be calibrated and/or verified prior to being used to generate data. To ensure the accuracy of the instrument, calibration is performed at installation, whereas verification is performed periodically at a frequency defined in the operating procedure. Wherever possible, the calibration must be traceable to a national standard. The results of this initial calibration are placed in the equipment file. Regular calibration programs for equipment must be established and documented. Reference standards used for calibration and/or verification of instruments and general equipment will be traceable to a primary standard recognized and accepted by the National Institute of Standards and Technology (NIST) (see www.nist.gov), ASTM International (www.astm.org), International Organization for Standardization (ISO) (www.iso.org), United States Pharmacopeial Convention (USP) (www.usp.org), or other standard organizations when feasible or required by the method. Reference materials must be used for calibration purposes only and must not be exposed to the rigors of everyday use unless it can be established that their performance as standards would not be invalidated.

1.52

Established Operating Procedures for Use in Taking Measurements

ISO/IEC 17025 states this requirement as follows: ‘‘Calibration programmes shall be established for key quantities or values of the instruments where these properties 8 |

have a significant effect on the results.’’ Equipment used for testing or sampling must be capable of achieving the accuracy required by the method or procedure. The laboratory should establish operating procedures that specify the criteria, standards, and requirements used to establish, implement and control the calibration and/or verification of equipment and instruments used for standard analytical testing. With a defined procedure in place, the obtained measurements will be independent of the personnel taking those measurements. Some examples of operational equipment requiring verification and monitoring are incubators, ovens, refrigerators, water baths, freezers, and so on. Some examples of measuring equipment requiring calibration and monitoring are balances, thermometers, water activity instruments, semi-automated pipettes, and pH meters.

1.53

Preventative Maintenance and Contracts With Approved Suppliers

The established operating procedures should include preventative maintenance contracts where necessary, as well as contracts with approved suppliers. A system of controlled preventative maintenance and monitoring is used to identify and eliminate potential sources of problems before they result in equipment failure. Such maintenance must also be documented to ensure the program is being followed according to schedule. When feasible, the laboratory performs the calibration/verification of equipment. Some equipment requires an outside service to come in periodically for preventative maintenance and calibration as described in the operations manual. If an outside supplier is employed, it is recommended to have an agreement on the work to be performed, the frequency of calibration, and the calibration method they will be using. When the vendor performs the calibration, the laboratory should verify the traceability of standards used for calibration and the date the vendor’s calibration equipment or standards were last certified. Once the calibration is complete, the vendor should supply a statement to the laboratory stipulating the condition of the instrument, calibration data, comparison of performance versus the acceptance criteria for the instrument, and a certificate of conformance.

1.6

LABORATORY GLASSWARE AND PLASTICWARE

Specifications of laboratory glassware and plasticware should be established and followed. For example, the calibration of newly purchased glass or plastic pipettes should be checked upon receipt at the laboratory. The calibration marks on dilution bottles should be checked with NIST certified volumetric or Class A glassware. Glassware should be made of high-quality, low-alkali borosilicate glass. Glassware composed of soft glass presents problems because of leaching of components and the presence of surface alkali, which may interfere with some analytical procedures.

1.7

SAMPLE MANAGEMENT

The chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis’’ offers specific

|

information on the collection, shipment, and preparation of samples for microbiological analysis. This section discusses the criteria used to determine the acceptability of samples received by the laboratory and the proper handling of samples accepted for analysis.

1.71

Criteria for Acceptance of a Sample

The laboratory should establish acceptability/rejection criteria for samples submitted for analysis. Adequate documentation and identification must accompany samples, including a description of the place of collection, manufacturer, date and time of collection (especially for perishable samples), reason for collection (for example compliance with legal standards or routine surveillance), sampling plan followed, analysis requested, and storage conditions. The original condition of samples and the integrity of sample containers must be maintained from collection until receipt at the laboratory. The manner of shipment must be appropriate for the type of sample. Other considerations for acceptance of samples are as follows: (1) They must be representative of the lot of product and processes sampled. (2) The laboratory must have the capability to perform the requested analyses.

1.72

Handling of Samples in the Laboratory

After receipt, samples must be stored to maintain their original condition until analyzed. Samples should be tested as soon as possible after receipt. Facilities should be available for both short-term storage before and during analyses and, when required for forensic reasons, longterm storage after analyses have been completed. It may be necessary for an individual in the laboratory to have the responsibility as sample custodian to maintain accountability records for samples in the laboratory. This individual may 1. 2. 3. 4. 5. 6. 7.

1.8

receive samples, record the date and time samples are received, initially verify the identity of samples, store samples according to the accompanying instructions, record the date and time when samples are delivered to analyst(s) for examination, and the date and time when they are returned to storage following analysis, maintain a long-term sample storage system, and dispose of samples as necessary.

(www.iso.org), have published collections of validated or ‘‘reference’’ microbiological methods. Consequently, many collaboratively studied methods have been standardized and are available for use by analytical food microbiologists. Despite recent progress in efforts to harmonize microbiological methods internationally, there is no universal recognition of validated or approved methods for specific analyses, however. Regardless, laboratory management should endeavor to use standard, reference, or validated methods to provide more reliable results to clients. When adopting a new method or modifying an existing one, the performance characteristics of the new or revised method should be verified against the previously used one before the change is instituted to show that the laboratory is proficient in the new method. This should be done even when the new method is considered to be a standard, reference, and/or validated method. An SOP for this validation process is necessary to provide consistency for methodologic changes. The selection of methods to be included in the manual will depend on the type of laboratory, that is, government, commercial, or research. Method descriptions should be highly detailed to avoid the need to reference other publications. This section should address necessary controls and checks on materials, media, reagents, positive and negative controls, desired response for each control, and corrective measures that should be taken if a control is incorrect. Where known, the limitations of each test should be included in the manual in addition to a list of advised precautions to be taken. Possible interference should be detailed at length (e.g., natural inhibitory substances in foods that must be diluted out before growth of any organism can occur). All the necessary controls and checks on materials, media, reagents, positive and negative culture controls, the desired response from each control, and corrective measures that should be taken if a control is not correct should be included, or the appropriate SOP referenced. For research laboratories, general SOPs may be established for development of a research proposal, approval of the project, and periodic evaluation of the progress made on the project. Specific SOPs, such as method validation and execution, may need to be modified during a project as new data are obtained and interpreted.

1.9

ANALYTICAL METHODS

Beginning with the 1910 publication of Standard Methods for the Examination of Dairy Products, the scientific community has recognized the need to promote the consistency of laboratory testing results, both nationally and internationally. Since that time, many government and scientific organizations, including AOAC International (www.aoac. org), American Public Health Association (APHA) (www. apha.org), United States Food and Drug Administration (FDA) (www.fda.gov), Health Canada (www.hc-sc.gc.ca), United States Department of Agriculture (www.usda.gov), International Commission on the Microbiological Specification for Foods (ICMSF) (www.icmsf.org), and International Organization for Standardization

Laboratory Quality Management Systems

1.91

CULTURE MEDIA AND REAGENT PREPARATION OR TEST KITS Media/Reagents

Because microbiological media and reagents are critical materials that may affect the quality of analytical data, each new lot of medium or reagent must undergo performance testing. Where feasible, the performance should be verified using national standards, reference cultures, and/or certified reference cultures. All new batches of media, whether made internally or purchased pre-made, should be tested for sterility, productivity, selectivity, and appearance. When possible, media should be tested prior to use; however, if the short shelf-life of some media limits the ability to undertake such testing | 9

Compendium of Methods for the Microbiological Examination of Foods |

prior to use, then performance should be verified at the time of use. Once verified, all media and reagents are properly stored and discarded following manufacturers’ recommendations. All media must be identified and traceable to QC results and ultimately traceable to each test performed. This includes preparation, traceability to media, pH, appearance, sterilization batch (with related records), fill volumes (if appropriate), batch size, and quantity. Small changes in the preparation of media or a decline in the shelf-life owing to less than optimum storage conditions may not be apparent when the verification protocol uses overnight cultures. These cultures contain high cell levels and are relatively robust compared to target organisms on actual laboratory samples. A good complement to regular media verification is the routine use of control samples of known cell levels that are representative of levels in actual samples. When these control samples follow the analytical steps for a given test they serve to detect any change in the testing process that could potentially affect the accuracy of the result, including minor changes to a medium, reagent, or test kit that might not be overtly obvious during the original verification process.

1.92

Media/Reagents Prepared In-House

Sterility testing is performed on all media following autoclaving. Randomly chosen samples of the media are set aside and allowed to cool. The media are then placed at the appropriate temperatures and incubated for the appropriate amount of time. After the desired incubation time is completed, the plate, slant, or broth is checked for growth or turbidity. Sterility testing should also be conducted on media that have been subjected to a filter sterilization process. When the filter is sterilized, the tube or plate chosen for testing should be at the end of the process rather than randomly selected, as the sterilization process is more likely to be compromised as the filtration time increases. All new batches of agars, slants, and broths should be tested for productivity and selectivity. Considering the number of media employed in a microbiology laboratory and the various control cultures associated with each, the verification of media performance can be quite burdensome. Mossel et al.2 describe a simplified technique for verifying media performance. Using this method, cultures of test organisms that are to be detected and those that are to be suppressed are streaked in parallel lines onto a solid medium or a liquid medium that has been solidified by the addition of 15 g agar/L and poured into each section of a quadrant Petri plate. Organisms expected to multiply on the medium should develop in all quadrants, whereas organisms expected to be suppressed should only develop in the first or second quadrants streaked. The pattern of growth— the absolute growth index (AGI)—is calculated for each test organism. The AGIs for new lots of media should be similar to those of standard or control lots. Finally, all media are tested for a final pH and volume and verified as being within the required tolerance of each for the test being performed. All prepared media should be traceable back to the source ingredients, and records pertinent to the prep 10 |

aration, sterilization parameters, and storage conditions should be maintained.

1.10

ACCREDITATION OF TESTING LABORATORIES

Accreditation of laboratories is becoming increasingly important both nationally and internationally. With the increased emphasis on food safety domestically and the globalization of the food marketplace, the need for accurate and reliable microbiological test results has become an essential part of public safety and commerce. The Food Safety Modernization Act of 2010 requires the FDA to recognize programs of accreditation for food safety testing laboratories. The ultimate goal of accreditation is to recognize laboratories with a demonstrated ability to carry out specific tests or types of tests, producing accurate, reliable, and consistent results using validated methods. Any accreditation body to be recognized by government and the food industry should abide by international standards and be recognized under the International Laboratory Accreditation Cooperation (ILAC) Mutual Recognition Arrangement (MRA) (www.ilac.org). As the generic requirements for accreditation, ILAC MRA signatory accreditation bodies must meet ISO/IEC 17011:2004, Conformity Assessment—General Requirements for Accreditation Bodies Accrediting Conformity Assessment Bodies. In addition, they must address additional ILAC policies including measurement traceability,1 proficiency testing, (ISO/IEC 17043:2010, Conformity Assessment–General Requirements for Proficiency Testing), and calibration (ISO/IEC 17025:2005, General Requirements for the Competence of Testing and Calibration Laboratories). The most current versions of these standards may be used. The management system and technical requirements found in ISO/IEC 17025:2005 are intended to be applied to all types of testing laboratories.

1.101

Guidance for Accreditation of Microbiological Laboratories

The Analytical Laboratory Accreditation Criteria Committee (ALACC) of AOAC International has developed Guidelines for Laboratories Performing Microbiological and Chemical Analyses of Foods and Pharmaceuticals, 2010, which is a sector-specific supplement to ISO/IEC 17025: 2005 and provides useful guidance. In Europe, Accreditation for Microbiological Laboratories, 2013, was produced by a joint European cooperation for Accreditation/EURACHEM Working Group. It supplements ISO/IEC 17025 and provides specific guidance on the accreditation of laboratories performing microbiological testing, for both assessors and laboratories preparing for accreditation. ISO/IEC 17025 remains the authoritative document and, in case of dispute, the individual accreditation bodies will adjudicate on unresolved matters. The ISO has also published ISO 7218:2007, Microbiology of Food and Animal Feed Stuffs—General Requirements and Guidance for Microbiological Examinations. The American Association for Laboratory Accreditation (A2LA) has published G108—Guidelines for Estimating Measurement Uncertainty for Microbiological Counting Methods.

|

Laboratory Quality Management Systems

Table 1-1. Websites for References Useful for Maintaining Laboratory QMS and QA Systems American Association for Laboratory Accreditation (A2LA). Guidelines for Estimating Measurement Uncertainty for Microbiological Counting Methods. See www.a2la.org. American Public Health Association (APHA). See www.apha.org. AOAC International (AOACI). Guidelines for Laboratories Performing Microbiological and Chemical Analyses of Foods and Pharmaceuticals. See www.aoac.org. ASTM International. See www.astm.org. CALA Canadian Association for Laboratory Accreditation Measurement Uncertainty Policy, P19. See www.cala.ca. CALA Application of Requirements in ISO/IEC 17025: 2005, PO7. See www.cala.ca. Health Canada (HC). See www.hc-sc.gc.ca. International Commission on the Microbiological Specification for Foods (ICMSF). Microbiology of Food and Animal Feed Stuffs–General Requirements and Guidance for Microbiological Examinations. See www.icmsf.org. International Laboratory Accreditation Cooperation (ILAC) Mutual Recognition Arrangement (MRA). See www.ilac.org. International Organization for Standardization (ISO). See www.iso.org. National Institute of Standards and Technology (NIST). See www.nist.gov. United States Department of Agriculture (USDA). See www.usda.gov. United States Food and Drug Administration (FDA). See www.fda.gov. United States Pharmacopeial Convention (USP). See www.usp.org.

The Canadian Association for Laboratory Accreditation (CALA) has published P19—CALA Measurement Uncertainty Policy and P07—CALA Application of Requirements in ISO/IEC 17025:2005. The above references are useful in preparing microbiological laboratories for accreditation. Table 1-1 is a list of websites for references that are useful for maintaining laboratory QMS and QA systems.

ACKNOWLEDGMENT Fourth edition authors: Richard B. Smittle and Anita J. Okrend.

REFERENCES 1. ISO/IEC 17025:2005. General Requirements for the Competence of Testing and Calibration Laboratories. International Organization for Standardization, Geneva, Switzerland. 2. Mossel, D. A. A., F. Van Rossem, M. Koopmans, M. Hendricks, M. Verdouden, M., and Eelderink, 1. 1980. A comparison of the classic and the so-called ecometric technique. J. Appl. Bacterial. 49: 439-454. 3. Shewart, W.A., and Deming, W.E (eds.) 1986. Statistical Method From the Viewpoint of Quality Control. Dover Publications, Inc., New York, NY. Reprinted, with editing, from the 1939 edition published by the Graduate School of the Department of Agriculture, Washington, DC.

| 11

|

CHAPTER 2

|

Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis T. Matthew Taylor, John N. Sofos, Peter Bodnaruk, and Gary R. Acuff

2.1

INTRODUCTION

The aim of this chapter is to assist sample collectors to obtain representative samples of a food lot (based on an appropriate sampling plan design), prepare samples for proper shipment to a laboratory in a condition that is microbiologically unchanged from the time of sampling, and prepare samples for proper analysis. In general, the procedures described in this chapter deal with collecting, labeling, shipping, transporting, storing, and preparing samples for microbiological analysis. Specific information and discussion regarding sampling and analytical procedures for canned foods is found in the chapters ‘‘Canned Foods—Tests for Commercial Sterility’’ and ‘‘Canned Foods—Tests for Cause of Spoilage.’’ Chapters covering relevant food types and specific microorganisms should be consulted for potential special sampling and preparation requirements.

2.2

GENERAL CONSIDERATIONS

The first priority in the microbiological examination of any food product is a representative sample that is appropriately collected and transported to the laboratory and prepared for examination. The results and the interpretation of laboratory analyses are valid only when appropriate samples have been examined. Samples must be representative of the entire lot of material under evaluation, must be the proper type of sample for the analysis to be conducted, and must be protected against extraneous contamination and/or improper handling, especially at temperatures that may alter the microbiological profile or number of microorganisms present. To prevent the destruction or the growth of organisms in a sample, refrigeration often must be provided for holding and storage. Unfrozen samples with high water activity (aw . 0.64) must be refrigerated, preferably between 0uC and 4.0uC, from the time of collection until the point of analysis. Samples of frozen foods should be collected and shipped solidly frozen. A sealed eutectic coolant in the shipment container is recommended to avoid contacting the product with | 13 |

melting ice or coolant. When dry ice is used as a shipping coolant, the containers should have tight closures to prevent pH changes in the samples caused by their absorption of the carbon dioxide. As a general rule, samples should be examined as soon as possible or within 36 h after sample collection. Perishable items that cannot be analyzed within 36 h should be frozen or retained at refrigeration temperatures for up to 18 h, depending on the type of product, the reason for analysis, and the type of analysis. Unfrozen samples of shellfish should be examined within 6 h after collection, but without being frozen.1 Each sample or shipment must be clearly and completely identified by the following information: (1) sample description; (2) the collector’s name; (3) the name and address of the product manufacturing establishment; (4) the production lot number; (5) the dealer or distributor; and (6) the date, place, and time of sample collection. The temperature of the sample at the time of collection and receipt is frequently useful to the laboratory for proper interpretation of results. It is also often desirable that the reason for testing is provided (e.g., samples may be collected as part of a quality control or surveillance program, as official samples to determine conformity to regulatory standards, or as part of a foodborne disease investigation). Relevant municipal, state, federal, or other agencies (e.g., U.S. Food and Drug Administration) prescribe the specific information that is required for completing the official analysis.31

2.3

EQUIPMENT, MATERIALS, AND REAGENTS

The following devices and implements may be necessary for sample collection, depending upon the type of food, the samples to be collected, and the objectives of sample collection.

N

Instruments for opening containers: Sterile scissors, knives, scalpels, can openers, or other hand tools, as required.

Compendium of Methods for the Microbiological Examination of Foods |

N N

N

N

N N

N N

N

Sample transfer instruments: Sterile single use or multi-use spatulas, scoops, spoons, triers, forceps, knives, scissors, tongue depressors, drills and auger bits, corers, dippers, metal tubes, and swabs, as required. Sample containers: Sterile single use or multi-use containers (large- or small-mouthed design); nontoxic, leakproof, and presterilized polyethylene bags; or other suitable sterile, nontoxic containers, as appropriate. Nonsterile, nontoxic, single-service vials, polyethylene bags, or bottles are acceptable transport containers if they are clean and dry and do not have a viable bacterial count greater than 1 organism per milliliter of capacity in rinse tests. Sterile, vacuum-packaged sampling equipment also may be used. Sterile glass containers are usually undesirable because of possible breakage and consequent glass contamination of the sampling environment. Thermometers: Thermometers should be used that measure from 220uC to 100uC with graduation intervals not exceeding 1uC. A metal dial type or digital electronic unit is preferred to avoid the risk of glass thermometers breaking. Mercury-containing thermometers should be avoided. Thermometers should be sanitized by dipping in a solution of hypochlorite ($ 100 mg/L), or other equivalent microbiocide, for at least 30 s and allowed to dry before being inserted into food. Microbiocide: Medium strength (100 mg/L) hypochlorite solution, 70% ethyl alcohol, 71% isopropyl alcohol, or other approved disinfectant should be prepared in accordance with the manufacturer’s instructions, by using sterile distilled water. Labeling supplies: Pressure-sensitive tapes and labels, tags of adequate size to hold essential sample information, and indelible marking pens or other appropriate labeling materials. Sample shipping containers: For frozen or refrigerated samples, use insulated rigid metal or plastic containers that are equipped with a tight-fitting cover. Each container should have ample space for the refrigerant so that samples will remain at the desired temperature until their arrival at the laboratory. Nonperishable samples do not need refrigeration. Containers for nonperishable samples should be made of sturdy corrugated cardboard or other material capable of withstanding abusive shipping conditions. A refrigerant or dry ice is added as needed for perishable samples but should not come into contact with the samples. Balance: A calibrated balance with a 2000 g capacity and a sensitivity of 0.1 g with a 200 g load is acceptable. Blenders and mixers: Examples of acceptable mixing equipment are mechanical blenders with several operating speeds or variable speed control with sterile glass or metal blending jars and covers, a stomacher, or other equivalent homogenizing device. Diluents: Use sterile Butterfield’s phosphate buffer; sterile 0.1% peptone (w/v) water; or appropriate sterile sodium chloride solutions. For further details, see the chapter ‘‘Microbiological Media, Reagents, and Stains.’’ See specific chapters for special diluents required for specific microorganisms and special analytical conditions.

14 |

2.4

PRECAUTIONS

Adequate precautions should be taken to prevent microbial contamination of samples from external sources such as contamination by the person taking the samples, by factors in the sampling environment (e.g., air, dust, dripping fluids), by sampling devices, by sample containers, and by the shipping vehicle. When foods are packaged in small, sealed containers, the unopened containers should be collected, rather than portions from each container. The sampling operation should be organized in advance with all necessary equipment and sterile containers available. Instruments compatible with the physical state of the food should be used for collecting samples. Sampling instruments should be protected from contamination exposure before and during use. When using sampling equipment to collect samples, the sampling instruments should not be passed over presterilized instruments. The sterile sampling container should be opened sufficiently to admit the sample and then closed and sealed immediately. Do not touch the inside of the sterile container’s lip or lid. Do not allow the open lid to become contaminated. Do not hold or fill a sampling container over the top of a bulk food container when transferring a sample. To prevent overflow and allow proper mixing of sample in the laboratory, the sample container should not be filled more than three-fourths full. Do not expel air when folding or whirling plastic sample bags. An empty sterile sampling container that has been similarly opened and closed should be submitted for analysis as a control. Samples of any presterilized gloves that were used can also be tested. The sample collectors must keep their hands away from their mouth, nose, eyes, face, and other body parts, and other contaminating factors in the environment. Hands must be washed immediately before beginning the sampling and washed during sampling if they become contaminated. Using sterile plastic gloves may limit contamination during the procedure. Contaminated sampling equipment must be placed into proper containers for later disposal and/or sterilization. Labels must never be moistened with the tongue. Directly label the container or use pressure-sensitive labels. The chapter ‘‘Laboratory Quality Management Systems’’ describes additional safety precautions to ensure proper handling and proper preparation of samples for microbiological analysis. Food samples may contain infectious microorganisms or potentially hazardous toxic materials. The best protection against these hazards is by using good sampling techniques and treating each sample as potentially contaminated.

2.5 2.51

PROCEDURES Sampling Plans

In 1923, engineers of the Western Electric Company first developed sampling plans; however, these plans were not adopted until after World War II when the U.S. Department of Defense developed Military Standards for attribute and variables sampling plans.24–27 These plans took into account sampling arrangements, based on the history of the producer’s performance. Sampling levels (i.e., tightened,

|

Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis

normal, and reduced) that reflected previous producer performances were established for these plans. Special sampling plan categories have also been adapted for agricultural products and food.4,8,14,19,32 A comprehensive set of single sampling attribute plans has been published for food microbiology.7,14 Typical points along the food processing continuum in which sampling plans may be applied to foods or ingredients are as follows:

N

1. 2. 3. 4. 5.

N

Raw materials Production line Producer’s warehouse Retail storage or sales outlet International port–export or import

A discussion of sampling plans requires an understanding of certain basic concepts. First, the lot to be sampled must be defined. A ‘‘lot’’ is a quantity of food produced and handled under uniform conditions. 23 ‘‘Sampling’’ is the procedure for surveying a given quantity (or lot) of a product and taking units or portions of this lot for examination (including microbiological analysis). Sampling must be applied in a way that will ensure statistically valid results. Thus, a sampling plan is developed so that the selection of samples taken from a lot is performed in a manner that ensures that each sample has the same probability of being selected for collection. For example, a lot that is defined as ‘‘1,000 packages in a warehouse’’ is sampled. Each package is assigned a number. If the 1,000 items are stacked equally in 10 rows, then items 1 to 100 would be assigned to the units in the first row, items 101 to 200 to the second row, and so on, until all packages are numbered. A random number generator or table would be used to determine which numbered containers should be taken as samples.14,15 In this example, if five packages are sampled, then five typical random numbers (in a lot size of 1,000 units) could be 586, 973, 99, 383, and 737. In addition to a random number selection, a sampling plan must include instructions that specify the number of packages to be collected. The results of the sampling is the basis for accepting or rejecting the lot. As used in this context, a sampling plan would state the number of units that are required to be randomly collected from a lot and would state the criteria for acceptance or rejection. Before using a sampling plan, it may be prudent to consult a professional statistician to ascertain that the lot of food meets the criteria required by the chosen sampling plan. Manufacturers may define a lot one way (e.g., an 8-h shift, a set number of produced units); however, it is important to recognize that regulatory bodies typically define a lot as a product that is produced on a common line (or shared portions of a line) between validated cleaning and sanitization cycles.5,16,23

2.511 Definitions Used in Sampling Plans The following statistical terms are frequently used in the sampling literature.14,22,26,30

N

Acceptance number: The maximum number of defective units in an attribute sampling plan for which a lot will be accepted.

N

N

N

N N N N N N N

N N

Acceptance quality level (AQL): The maximum percent defective (or the maximum number of defects per 100 units) that, for the purposes of sampling inspection, can be considered satisfactory as a process average.25 Lots possessing a quality level equivalent to a specific AQL will be accepted approximately 95% of the time when using sampling plans that are given for that AQL.22 Analytical unit: The amount (i.e., volume or mass) of sample that is actually analyzed. Attribute: A qualitative characteristic that may be measured for a sample unit (e.g., the presence or absence of Salmonella in the rinsate of a poultry carcass).14 Average outgoing quality limit (AOQL): The maximum possible defective percentage that results after completing a rectifying inspection of a lot. In such an inspection, 100% of rejected lots are screened to remove defective items (provided the screening is 100% effective).19 The AOQL levels typically do not exceed 5%. Binomial distribution: The distribution of a population in which a proportion (p) of the units in the population have a certain characteristic and another proportion (q, which is equal to 1 – p) does not have these characteristics. Therefore, each individual unit falls into one of the two categories: p or q. Consumer protection: The ability of a sampling plan to identify and reject unacceptable lots. This is measured as the complement of the probability of acceptance for limiting quality (LQ) lots.22 Consumer’s risk: The consumer assumes the risk that a sampling plan will accept a lot, although the lot does not conform to requirements. It is a type 1 risk and can be analogous to a type 1 error.22,24,25 Continuous distribution: The distribution of a population of measures that assumes a continuum of values. A variables sampling plan may be applicable (Section 2.515). Control chart: A graphic depiction used for monitoring repeated sampling results from a manufacturing or measurement process. Defect: A unit that does not conform to specified requirements.25 Individual sample units may contain multiple defects.22 Destructive testing: A testing method or process that destroys the sample unit, thereby disallowing further testing or sale into commerce. Discrete variate: A random variable consisting of isolated points that can have a finite or infinite number of values. A discrete variate often results from counting (e.g., the number of defective sample units within a lot).6,21 Discrete variate values can be used in an attribute sampling plan (Sections 2.512, 2.513, and 2.514). Estimate: Any value computed from sample data that is used to infer a corresponding population (i.e., lot or a value such as the sample mean). Frequency distribution: The mathematical description of the distribution of the members of a population. The information about the distribution is used to calculate the probability of a lot’s acceptance or rejection. The discrete random variable takes on a countable number of values and the probability distribution is defined by the probability mass function. A | 15

Compendium of Methods for the Microbiological Examination of Foods |

N N N

N

N

N

N

N N

N N

continuous random variable is defined by a density function. Homogeneity of variances: The equality of variances among populations. It sometimes requires determination. Homogeneous: A product has a uniform texture or content. Limiting quality (LQ): The percent defective units or the percent of defects per 100 units.22 A lot having a 10% probability of acceptance—based on the definitions of many common standards—is a lot in which the quality level is equal to the LQ.24,26 Lot: The number of sample units produced in one batch or within a specified period so that the units will have approximately the same quality. Each lot or batch should consist of units of the product of a single type, grade, class, size, and composition. The lot should be manufactured under the same conditions and essentially at the same time, or within a defined continuous period of time. Lot inspection by attribute: An inspection whereby the sample unit is classified as either ‘‘defective’’ or ‘‘nondefective’’ with respect to a requirement or set of requirements (when on a ‘‘defective’’ basis), or an inspection whereby defects in each sample unit are counted with respect to a requirement or set of requirements (when on a ‘‘defective’’ basis).22 Lot quality: A measure of the characteristic being controlled. The results of lot inspection are often expressed as percent defective units. The quality is less frequently expressed by the variable being measured (weight/unit, coliforms/g). Operating characteristic (OC) curve: A graphical representation of the relation of the probability of lot acceptance to the lot quality, usually expressed as percent defective units (e.g., Figure 2-1). This curve will depend on the number of units required in the sampling plan and the acceptance number. The curve also shows the lot quality associated with the consumer’s risk and the producer’s risk. The OC curve describes the consequences of the sampling plan (i.e., the decision rule) for accepting lots that have different qualities. Population: Any finite or infinite collection of individuals, samples, or units on which decisions are to be made. Probability: An estimate of the frequency of the occurrence of an event (e.g., the probability of n sample units from a population being positive for Salmonella). It is expressed as a value ranging from 0 to 1. For a sampling plan, the assumption of a particular probability distribution (e.g., binomial, Poisson, log10 normal) allows the estimation of the computation of the probability of lot acceptance versus the lot quality in an OC curve. Producer’s risk: The risk that a producer takes that a lot will be rejected by a sampling plan even though the lot conforms to requirements. It is set at 5% in many sampling plans.22 Proportion defective units (P): The number of defective units divided by the total units in a lot. Proportion defectives (P) or percent defectives (100% 6 P) are often plotted on the abscissa on an OC curve.

16 |

N N

N N

N

N N

Random sample: A sample that is chosen in such a way that all samples or units in a lot have an equal chance of being selected. This is often achieved with the aid of a random number generator or table. Representative sample: In the broadest sense, a sample that is representative of a population, not merely a portion of it. Regardless of how a representative sample is chosen, it can be considered typical of a population with respect to certain characteristics. To obtain a truly representative sample, one must (1) determine the location of sampling points critical to the population, (2) establish a sampling method representing the population characteristics, (3) select the sample size, and (4) specify the frequency of sampling. Sampling unit: The smallest definable part of a lot; it is also called a ‘‘unit.’’ The term is to be differentiated from the analytical sample unit, which is specified by the analytical method used for lot sampling. Sampling plan: A design that indicates the number of units to be collected from each lot and the criteria to be applied in accepting or rejecting a lot. (1) A single sampling plan requires the lot to be judged on one set of sample units (Section 2.512). (2) A double sampling plan is a sampling inspection in which the inspection of the first sample leads to a decision to accept, reject, or collect a second sample. The inspection of the second sample then leads to a decision to accept or reject the lot. (3) In a procedure called sequential sampling, units are drawn one by one (or in groups), but at any stage the drawing can result in a decision to accept, reject, or continue sampling.22,25 Sample size (n): The number of samples collected from a lot. The sample size should describe the population accurately, and most economically attain a certain level of accuracy. If the analytical method is not destructive to the sample, then it may be possible to sample 100% of a lot. Microbiological analysis methods are often destructive to samples, thereby eliminating the potential for 100% sampling of a lot. Sampling size can be determined by using sampling plan master tables, examples of which are found in the U.S. Department of Defense Military Standard 105E.25 Stratified random sampling: A procedure in which the lot is divided into strata that differ with respect to the characteristics under study. Zero defective tolerance: A system that indicates that a lot must be free of an undesirable characteristic or defect. All lot units must be sampled to guarantee zero defective tolerance. Therefore, zero defective tolerance cannot be applied to microbiological sampling plans.

2.512 Single Sampling (Two-Class) Attribute Plans Single sampling procedures are useful in food inspection since a lot can be sampled and tested only once in many production processes. The results of microbiological tests are also clearly defined as attributes. For example, an attribute such as the presence or absence of a microorganism (e.g., Escherichia coli O157:H7) on ground beef is frequently reported. In other instances, a certain number of organisms may be acceptable. For example, a unit may be acceptable if it contains no more than 3 colony forming units (CFU) of E. coli per gram of sample.

|

Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis

Single sampling attribute plans also have the advantage that the true distribution is not required for the variable (e.g., E. coli) in question. Single attribute sample plans can be evaluated by using hyper-geometric, binomial, or Poisson distributions. The choice of distributions used to compute the probability relationships depends on the number of units (N) in a lot. A Poisson distribution can be assumed and can serve as an approximation of the binomial distribution when N is large, relative to the sample size (n), or when the number of defective samples in a lot is small. The types of lots acceptable for single sample procedures are assumed to be large with a homogeneous quality and satisfy the above conditions. Steps for choosing and applying an attribute sampling plan include the following: 1. 2. 3. 4. 5. 6. 7. 8.

Select the measurements of interest. Define the sampling units that constitute a lot, when necessary. Determine a value of consumer’s and/or producer’s risk to ensure the lot quality desired (e.g., LQ, AQL). Obtain an estimate of the process average. Compute or select a plan that meets the risk and lot quality requirements. Calculate the OC curve. Apply the plan on a group of randomly selected units from a lot. Maintain records on the process average (if you are the producer) and make changes to the plan, as needed.

Figure 2-1 presents the OC curves for seven single sampling plans. These OC curves for defined sampling plans will illustrate the process of choosing a plan. For presentations of complete sets of plans where OC curves for additional values of n or c are given, refer to Dodge and Romig,9 Duncan,11 and the International Commission on the Microbiological Specifications for Foods (ICMSF).14 It is important to recognize that n and c both represent the number of samples, wherein c is a subset of n. In Figure 2-1, the acceptance number (c) is zero (i.e., ‘‘zero tolerance’’), which means that any positive test results will result in the rejection of the tested lot.7,14 For example, a lot (N . 1000 units) is analyzed for coliform MPN/g and a unit with an MPN of 100 or more coliforms per gram is identified as ‘‘defective’’ and a unit with an MPN less than 100 coliforms per gram is identified as ‘‘acceptable’’ or ‘‘nondefective.’’ A sampling plan is needed to define the value of n to sample in which the value of c equals zero and the probability (i.e., consumer’s risk) of accepting lots with 8% or more defective units is equal to 10%. Using these criteria, the sampling plan is n 5 30 and c 5 0 (Figure 2-1). Many microbiological tests are destructive to the sample. Therefore, the cost of sampling may be balanced with the cost of the risk that is to be detected by sampling. For example, a plan may be designed to inspect incoming raw materials or may be designed to sample consumer products before releasing lots for sale. The measurement (i.e., coliform MPN/g) may be the same, but the choice of plan will be impacted by other factors such as processing conditions, potential health hazard of the products,8,15 persistence of organisms under different storage conditions, and type of plan used by regulatory agencies to inspect the same lots.22

Figure 2-1. Single sampling (two-class) attribute plans for n 5 3, 5, 10, 15, 20, 30, and 60, and c 5 0.

In Figure 2-1, plans in which n 5 15, 20, 30, and 60, with c 5 0, are commonly used to test for Salmonella.30 A plan in which n 5 5 and c 5 0 (ICMSF Case 10) may be used to screen raw materials or ingredients.14,17 Some analytical procedures are sufficiently sensitive to detect the presence of a single organism when sampling units are pooled or composited. A positive test indicates that one or more of the units are positive for the analyte from the pooled unit. This positive result produces the same decision for a plan (e.g., n 5 5, c 5 0) as if the units had been analyzed separately. Compositing sample units cannot be completed when c . 0 or when a positive sample is defined quantitatively (e.g., E. coli CFU/g) because of the inability to determine which of the composited portions contained the organisms of interest and thus the homogeneity of the organism’s presence in the lot.9,11 As sample size (n) increases, the OC curve becomes idealized with a probability of acceptance of 100% with 0 defects and falls to a probability of 0% when the fraction of defects is greater than 0. As the acceptance number decreases, the OC curve shifts to the left. This is evidence that the probability of accepting the lot decreases with a given number of defective units.

2.513 Three-Class Attribute Plans Bray et al.7 developed three-class attribute plans in conjunction with the ICMSF8 that is to be used with | 17

Compendium of Methods for the Microbiological Examination of Foods |

methods recommended by ICMSF.14 Bray states that ‘‘the test is concerned primarily with plans that may be applied to lots presented for acceptance at ports or similar points of entry.’’7 It was assumed that very little, if any, information would be known about the quality of lots and that attribute plans, two-class plans (Section 2.512), and three-class plans would thus be applicable. These plans are also useful for inspecting lots within a country or corporation where more information is known about the lot. Three-class attribute plans differ from those described in Section 2.512 by having two microbiological limits that create three classes of product: defective, marginally acceptable, and acceptable. Bray et al.7 noted that the choice of limits (e.g., for coliform MPN, a quality or sanitary condition indicator) is difficult. They nonetheless stated that most scientists can provide two numbers: one below which they have little or no concern and a higher value above which they clearly begin to have concern. If we denote the lower level by m and the larger one by M, then the set of values in the range (m, M) can be considered marginally safe.

Observations falling in the regions defined by m and M are identified as acceptable, marginally acceptable, and unacceptable/defective when they have values equal to or less than m (# m), between m and less than or equal to M (m , x # M), and greater than M (. M), respectively. Three-class attributes plans can be specified by sample size n, the number of units allowed (c1) between greater than m and less than or equal to M, and the number of units allowed (c2) equal to or greater than M. This assumes that all units less than or equal to m are acceptable. However, as stated previously, the value of M is the decision point for this type of sampling plan. The distance between m and M is indicative of the maximum variability that is acceptable in using three-class plans.8 The value of c2 5 0 was set in all plans discussed previously14 and will be incorporated in the following discussion. Thus, the sampling plans are noted as n and c, in which c indicates the number of marginally acceptable samples allowed before a lot is deemed unacceptable. Three-class attributes plans have been previously suggested for a variety of food types such as fish and fishery products, vegetables, dried foods, frozen foods, dairy (i.e., fluid milk, milk products), raw and processed meats, shelf-stable canned foods, and fresh or frozen raw shellfish.14 Different measurements of interest exist for different products and include the aerobic plate count, coliforms, fecal coliforms, Salmonella, Vibrio parahaemolyticus, Staphylococcus aureus, Bacillus cereus, Clostridium perfringens, and C. botulinum. The limits (m, M) chosen for product/measurement combinations and the plans were established on the basis of risk. To aid microbiologists in the selection of a sampling plan, the ICMSF15 has categorized the types of microbiological hazards and conditions to which a lot of food would be exposed, as follows:

N

There is no direct health hazard and utility (e.g., general contamination, reduced shelf life, and/or spoilage).

18 |

N N N N

The health hazard is low and indirect (indicator). The health hazard is moderate, direct, and with limited spread. The health hazard is moderate, direct, and with potentially extensive spread. The health hazard is severe and direct.

These hazards are linked with three risk conditions (i.e., reduction, no change, or increased hazard risk) that reflect how a food is to be handled and consumed after sampling. The combination of these types of hazards and conditions or risk yields 15 ‘‘cases’’ in which sampling plans and limits could be suggested for products, although only a few cases would be realistically applicable for a given product and measurement.14 The approach of ICMSF8 on three-class sampling plans was to define risks, suggest limits for a wide variety of specific cases, and suggest plans for use. This approach differed from other texts in which mathematical tools for computation of sampling parameters are presented and the user is expected to choose the conditions of sampling and interpretation of sampling results.7,25 Operating characteristic contours presented as two-way tables are used rather than OC curves for three-class attribute plans. These contours reflect the fact that the true marginal percent can vary from 0% to 100% with the restriction that the sum of percent acceptable units, percent marginally acceptable units, and percent unacceptable units must be 100%.8 The selection of an ICMSF sampling plan should be as follows: 1. 2.

3.

4.

Specify the food and measurement (e.g., freshwater fish, fecal coliforms). Specify the risk; in the case of freshwater fish and testing for fecal coliforms, ICMSF Case 4 (i.e., low, indirect health hazard)14 is identified with n 5 5 and c 5 3. Choose limits m and M for the product to be sampled. The ICMSF8 recommends m 5 4 CFU/g and M 5 400 CFU/g fecal coliforms for freshwater fish. Once the product and measurement are specified, the complete plan is listed (e.g., n 5 5, c1 5 3, c2 5 0, m 5 4 CFU/g fecal coliforms, and M 5 400 CFU/g fecal coliforms).

To derive a three-class plan, one must consider the following: (1) the assignment of risk to the product; (2) the choice of sampling plan based on the probability of acceptance, compared to the percent marginal units (assuming no defective units are accepted [c2 5 0]); and (3) the values of m and M or the product.8 The choice of risk and the setting of m and M may differ, depending on the purpose of the sampling.

2.514 Variables Sampling Plans A variables sampling plan is used when the probability density function of a measurement is known. Most published plans are computed when the distribution of the variable or its transformation (e.g., log10) is normally distributed (i.e., it has a Gaussian distribution). An advantage of variables plans over single sampling plans is that a lower cost is needed to achieve the same protection as a single attribute plan since fewer samples are required. Disadvantages include the number of calculations involved

|

Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis

in evaluating a lot; each variable (e.g., coliforms, E. coli) requires different calculations; and the probability distribution must be known or assumed for each measurement. The last two requirements may prove the most problematic. A product may require a series of three or four variable measurements. The transformation that ensures normal distribution may be unknown, and the estimates of variance are often different for each measurement. For these reasons, variables sampling plans are not widely used for microbiological measurements in the food industry. Further information about the derivation of variables sampling plans may be found in Duncan,11 Kramer, and Twigg,17 and MIL-STD 414.26

Table 2-1. Number of Samples Needed to Detect a Fraction Positive With a Probability of 0.90, 0.95, or 0.99 in Which at Least One Positive Result Occurs

2.515

Sampling Procedures for Low Contamination Levels A common situation encountered by microbiologists is the need to sample production lots of a product with a low incidence of a pathogen (i.e., a low number of contaminated units within the sample size n such as 1 positive unit in 500 units). The question in such situations is the number of sample units required for testing to have a high probability of detecting a pathogen. For example, if 200 sample units were negative for a pathogen such as E. coli O157:H7, what can be concluded about the lot? The entire lot may clearly not be free of the pathogen. A similar question or concern arises in sterility testing. It is assumed in both situations that the pathogen or contaminant can be detected if it is present in an analytical sample and that the results will be reported as either positive or negative. Other necessary assumptions are given in the discussion of single sampling attribute plans (Section 2.512). Table 2-1 lists the number of sample units (n) needed to detect a positive result at a given level of fraction positive analytical units in a lot. Table 2-1 can be used to determine the number of units that should be analyzed to detect a positive result at a given fraction positive level. For example, if the fraction positive of a pathogen is 0.04, what sample size n is required to find a positive unit with a probability of 0.95? Based on Table 2-1 and given these conditions, the value of n (i.e., required sample size) is 75 units. The situation may also be examined from another point of view. If a producer knows that pathogens in the product have a frequency of 6 units in 1,000 units and the regulator takes 30 sample units per lot, the probability (Pr) that all 30 samples will be negative is Pr 5 e2(30)(0.006) 5 0.84. Thus, the regulator has a 0.16 chance of detecting a positive from a lot with a frequency of defective units of 0.006. Simply stated, 84% of the time that 30 samples are collected from this defective lot, no defects will be found, whereas defective units will be found 16% of the time. Another situation arises when a sample of n units is examined and all are negative for the target pathogen. What can be concluded about the rest of the lot in terms of defective units? Table 2-2 shows one way of expressing the result. The fraction positive (i.e., defective units) per lot can be related to the probability when all examined n sample units are negative. If 100 sample units were negative, then there is only 1 in 10 chances (i.e., Pr 5 0.10) that the fraction positive exceeds 0.023 per lot. However, the probability is 9 in 10 chances that the fraction positive is 0.023 or lower.

Number of Analytical Units Tested (n) Probability (1 – Pr)

Fraction Positive Samples (P)a

0.90

0.95

0.99

1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.001 0.0001

3 3 3 4 4 5 6 10 12 23 26 29 33 39 46 58 77 115 230 2,303 23,026

4 4 4 5 5 6 8 10 15 30 34 38 43 50 60 75 100 150 299 2,996 29,963

4 5 6 7 8 9 12 16 23 46 51 58 66 77 92 115 154 230 461 4,605 46,052

Source: Adapted from Dodge and Romig.9 a The fraction of positive units (e.g., 90 positive units per 100 analytical units).

There is also a probability of 10% (at most) that 230 sample units in 10,000 have a pathogen. Thus, even in a lot size of 10,000, there is some chance (although only a 10% chance) that a positive unit will reach a consumer. The fact that contaminants are not evenly distributed throughout a lot increases the chances of not detecting the contaminants, and thus increases the probability of accepting the defective lot. Therefore, there is an additional, although small, probability that the consumers will receive the unit. If the samples were homogeneous and the organisms assumed as randomly distributed throughout the lot, then the concentration can be estimated (see the chapter ‘‘Culture Methods for Enumeration of Microorganisms’’). For example, assuming that one Listeria monocytogenes cell occurs in each 10 mL of milk if n samples of milk are taken from a well-agitated bulk milk tank, then the concentration of the pathogen can be estimated. If the analytical sample is 1 mL, the sample units required to obtain at least one positive sample with probability 0.95 is 30 (Table 2-1). Further assume that an examination shows that 2 of the 30 results are positive. Based on the results of a single dilution, the estimate of the concentration of L. monocytogenes/mL is ln(n/number of negatives) 5 ln(30/28) 5 0.069. Thus, sampling plans are useful for lot inspection, in determining risk, and in predicting a specified level of organisms associated with a given probability. However, | 19

Compendium of Methods for the Microbiological Examination of Foods |

Table 2-2. Fraction Positive Samples When the Probability Is That All n Samples Are Negative Fraction Positive Samples (P) Probability (Pr)

Number of Analytical Units (n)

0.10

0.05

0.01

3 5 10 15 20 25 30 35 40 45 50 100 200 400 500 1000

0.77{ 0.46 0.23 0.15 0.12 0.092 0.077 0.066 0.058 0.051 0.046 0.023 0.012 0.0058 0.0046 0.0023

1.00 0.60 0.30 0.20 0.15 0.12 0.10 0.086 0.075 0.067 0.060 0.030 0.015 0.0075 0.0060 0.0030

1.50 0.92 0.46 0.31 0.23 0.18 0.15 0.13 0.12 0.10 0.092 0.046 0.023 0.012 0.0092 0.0046

Note: Adapted from Dodge and Romig.9 Rounded to two significant digits.

{

this is limited to lots in which the organism is evenly distributed. In addition, the concentration can also be estimated in some restricted cases.

2.516 Summary This chapter presents a general survey of sampling plans. The discussion of sampling plans in this section has been modified to apply to microbiological measurements, although some of these procedures do not lend themselves directly to microbiological sampling in foods. Single sampling plans are generally best suited to most situations because sampling for microbial analysis is destructive. The results may be delayed for several days because of enrichment and incubation requirements and the foods may be perishable. Two-class and three-class attribute plans are easy to apply and thus should aid in the wider use of statistics in microbiological sampling of foods. The ICMSF14 also presents extensive suggestions of specific plans and microbial criteria to be used on a wide variety of foods. This reduces the amount of work necessary for calculations and time spent in plan selection. The single sampling two-class and three-class attribute plans therefore have the most utility in sampling foods for microbiological analysis. Individuals who wish to derive their own sampling plans are advised to consult Dodge and Romig,9 Duncan,11 and MIL-STD-105E.25 It must be emphasized that no sampling plan—short of one that requires testing 100% of a product lot—is capable of guaranteeing absolute absence of a microorganism.15 2.52

Sampling Procedures

The physical state of the product to be sampled (e.g., dry, semisolid, viscous) and the reason for sampling and testing 20 |

are necessary considerations before obtaining samples for the number of units to be representative and/or statistically adequate for the intended use.14 This section describes general sampling procedures. Unusual or special sampling procedures that may be required for certain foods or analyses are presented in chapters dealing with particular microorganisms or commodities. The chapter ‘‘Molecular Typing and Differentiation’’ discusses specialized sampling procedures for the molecular characterization of foodborne microorganisms from foods. Individuals trained in appropriate methods and proper aseptic technique should perform sampling.

2.521 Finished Products Consumer packages of foods should be sampled from original unopened containers of the target processing lot. Processing information and product codes should be submitted on forms accompanying the samples. The practice of submitting unopened containers prevents contamination that may be introduced by opening and handling at a sampling location. It also allows laboratory examinations to be performed on products and packages or containers as they are offered to the public. 2.522 Bulk Liquid Material If the products are in bulk form or in containers of a size that is impractical for an intact submission, then representative sample portions should be collected as follows: 1.

2.

3. 4.

5. 6.

7. 8. 9. 10.

11.

12.

13.

Before drawing a sample, aseptically mix the food mass to ensure that the sample is as homogeneous as possible. If adequate mixing or agitation of the bulk product is impossible, multiple samples should be drawn from the bulk container. Disinfect the sampling port or opening. Aseptically transfer into sterile, leak-proof containers with appropriate sterile implements (e.g., presterilized funnel), where appropriate. Do not fill sample containers over bulk containers of food. Carefully select sample containers that have sufficient capacity to accommodate the needed sample volume when the container is three-fourths full. Avoid using glass sample containers. Thermometers used in bulk food containers should be sanitized and dried before use.30 Cool perishable samples to 0uC to 4.4uC quickly, if they are not already refrigerated. Metal thermometers are preferred since the breakage of a mercury thermometer would contaminate the product with hazardous and toxic material. When appropriate, a temperature control sample should be prepared and submitted with the samples to be tested. If a temperature control sample is needed for frozen samples, use a container of ethylene glycol or other suitable low freezing point material (e.g., propylene glycol). The sample container should be sealed properly to avoid breakage, leakage, or introduction of extraneous contamination.

|

Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis

14. If the sample is to be examined for a regulatory purpose, the sample container must be sealed so that it cannot be opened without breaking the seal. 15. An empty sterile sample container should also be submitted as a container control.

since product appearance or quality cannot be affected. In some instances, a product rinse in a suitable buffer solution may suffice; when the product size is prohibitive for this method, an assessment of microbial load can be obtained by sampling several sites by using a sterile swab or sponge.3,18 Sterile disposable gloves are necessary when collecting sponge or swab samples.

2.523 Bulk Solids or Semisolids Dry or semisolid foods should be sampled by using sterile triers, spoons, or spatulas. (Sterile tongue depressors may be substituted for spatulas.) Portions from several areas of the food under examination should be obtained to ensure a representative sample. Carefully protect samples from excess humidity. 2.524 Frozen Bulk Materials Frozen bulk foods may be sampled with sterile corers, auger bits, and other sharp sampling instruments. A presterilized auger bit or hollow tube may be used to obtain sufficient material for analysis. Frozen samples should be kept frozen until their arrival at the laboratory. Avoid thawing and refreezing samples. A suitable procedure for obtaining test units of frozen foods (particularly from large samples) is to use an electric drill combined with a funnel.1 A sterile auger bit is inserted through a sterile plastic funnel, which has been cut so that the hole is slightly larger than the bit. The funnel is held against the frozen sample. The frozen shavings are conveyed to the surface and collected in the funnel. The shavings can then be aseptically placed in a sterile sample container. For large solid frozen or unfrozen food samples, test units should be obtained aseptically from several areas by using sterile knives and forceps. These portions should be mixed to provide a composite sample representative of the food to be evaluated. 2.525 Line Samples (In-Process Samples) of Liquids Sterile metal tubes or dippers may be suitable sampling instruments at certain plant locations for liquid food samples. Disposable presterilized plastic transfer pipets can also be used. A special line sampling technique involves using a disposable sterile hypodermic needle and syringe. The needle is inserted into the rubber closure of a stainless steel nipple. The nipple can be clamped on or permanently located at the desired location.12 Sampling cocks on holding tanks and product pipelines may be used; however, they can be a source of contamination. Disinfection and material flowthrough of the sampling cock must consequently be assured before collecting the sample. 2.526 Line Samples of Solids Sampling of solid line samples may be accomplished by using the same equipment and procedures that would be used for bulk solid products or semisolid products. Automatic sampling devices are available for powdered products and other solid products that do not require refrigeration. When automatic samplers are used, manufacturer directions for their use must be strictly followed. 2.527 Nondestructive Sampling When food products are sampled for indicator organisms and/or pathogens, non-destructive samples may be preferred

2.528

Special Purpose Samples (Foodborne Disease Outbreaks, Consumer Complaints) In some instances, samples are analyzed as part of a foodborne disease outbreak investigation or because of a consumer complaint in regulatory, clinical, or public health laboratories. If the record of sample collection and handling is incomplete, or if samples are received in a partially decomposed state, have been stored under abusive conditions, or the chain of custody has not been maintained, then the laboratory results may be of little or no value since these situations may be the subject of legal proceedings and laboratory personnel may be required to testify concerning the results of their examinations. Sampling for disease outbreak investigations should involve collecting samples from all suspect foods. If there are no leftover foods, efforts should be made to obtain samples of items prepared in a similar manner as the suspected food. Ingredients or raw items used in the suspect food should also be collected, if available. All ingredients or raw items should be held under suitable conditions until an analysis of the attack rate data and other epidemiologically gathered data can aid in identifying the suspect food or foods.15 The original containers in which the foods were found should be collected, labeled, and submitted for examination. Other specimens (in addition to the suspect foods) from outbreak investigations are essential. Depending upon the suspected cause of illness, human specimens may include appropriately collected stool, vomitus, and serum. Such samples should be taken by qualified individuals and are outside the scope of this chapter. 2.529

Samples for Water Activity and/or pH Measurement Samples intended for water activity determination should be collected in sealable vapor tight containers.20 Small, unopened, hermetically sealed retail-sized packages should be collected, if possible. Sampling from bulk containers should be performed quickly to minimize changes in the water content of the product. Samples to be tested for their pH and/or titratable acidity should be collected in tightly sealed containers. If the material to be tested is undergoing fermentation or some other gas producing reaction, a vented container or flexible plastic bag with ample space for expansion should be used. 2.53

Storage, Shipment, and Receipt of Samples

When it is necessary to store samples before shipment, a storage area should be available for frozen samples (220uC) and for refrigerated samples (0uC–4.4uC). Direct labeling of containers with waterproof labels is preferred to prevent the loss of labels. Whenever possible, samples | 21

Compendium of Methods for the Microbiological Examination of Foods |

should be submitted to the laboratory in original unopened containers. The product label may also indicate whether refrigeration is required. If the product is in a dry condition or is canned (flat or normal), it need not be refrigerated for shipment. The following are important sample storage, shipment, and handling considerations:

N

N N N N N N N N N N

N N N N N N N

Since laboratory examination of food samples requires preparatory work, the laboratory should be given advance notice, if possible, of the number and types of samples being submitted and the tests to be performed and any appropriate sample dilutions. Samples should be delivered to the laboratory as rapidly as possible. The condition, time, and date of arrival at the laboratory should be recorded. The samples should be packed to prevent breakage, spillage, or changes in temperature. To avoid the possibility of cross-contamination during transit, finished product and raw material samples should be shipped separately or packaged by using additional packaging material to ensure adequate isolation. Swollen containers should be shipped under refrigerated conditions. Samples not requiring refrigeration or freezing may be packed in a cardboard box with appropriate packing material to prevent breakage. Refrigerated products must be transported in an insulated shipping container with sufficient refrigerant to maintain the samples at 0uC to 4.4uC until arrival at the laboratory. Water frozen in plastic containers or cold packs serves well for shipping at 0uC to 4.4uC and should last 48 h under most conditions. Avoid using loose ice since this may cause product contamination if the container breaks or leaks. Dry ice may be used for longer transit times; however, the sample should be separated from the dry ice packing material to avoid freezing, and containers should have secure closures to prevent possible pH changes in the sample caused by the absorption of carbon dioxide.28 Refrigerated products should not be frozen since this will destroy certain microorganisms. Frozen samples can be kept frozen by ensuring that the samples are shipped with dry ice. Frozen samples collected in plastic bags, however, must not come in direct contact with dry ice since the plastic bag will become brittle and will be subject to rupture from the extreme temperature. Use paper or another suitable material to protect the sample. Samples should be transported to the laboratory by the fastest possible means. Mark the shipment of samples as ‘‘Perishable,’’ ‘‘Packed in Dry Ice,’’ ‘‘Refrigerated Biologic Material,’’ or ‘‘Fragile,’’ as appropriate. The shipment should be labeled according to federal postal service rules and Department of Transportation regulations.28,29 When samples are collected in response to a legal dispute, the chain of custody documentation should accompany the samples. Sample containers should be

22 |

sealed with an official seal that will indicate any tampering before analyses are initiated. Samples should be inspected on receipt to determine the transit time, whether any damage or leakage has occurred, and if the samples are at the required temperature. Documentation describing each sample should be cross-checked with the contents of the shipping container.

2.531 Preparation of Sample Homogenates The use of aseptic technique is required. The following are necessary considerations:

N

N

N

N

N N

Samples should be examined promptly. Nonperishable, canned (normal, flat) or low-moisture food samples may be stored at room temperature until ready for analysis. When the initiation of analysis must be postponed, frozen samples should be stored at 220uC until they are ready for examination. Refrigerate unfrozen perishable samples at 0uC to 4.4uC for no more than 36 h. To destroy microorganisms that may later contaminate the sample, before opening the container, swab its exterior area with 70% ethyl alcohol or other appropriate disinfectant (while using proper precautions in regard to flammability and confined space considerations). Frozen samples should be thawed at refrigeration temperatures (# 4.4uC) for no longer than 18 h in the original container in which it was received.13 The sample must be removed aseptically from the original container. Higher temperatures alternatively may be used for a short period, but the temperature must remain low to prevent the destruction of microorganisms (, 40uC for up to 15 min). Frequently shaking samples is necessary when they are thawed by the alternate procedure. A thermostatically controlled water bath with agitator is recommended for rapidly thawing samples. Liquid or semi-liquid samples in containers that have an airspace can be mixed by rapidly inverting the sample container 25 times. Sample containers that are two-thirds to three-fourths full should be shaken 25 times in 7 s in a 30-cm arc. The interval between mixing and removing the test portion should not exceed 3 min. To ensure a homogeneous sample when no airspace is present, aseptically open the container and pour the product from the filled container back and forth into a sterile container three times. Dry samples should be aseptically stirred with a sterile spoon, spatula, or other utensil to ensure a homogeneous sample. Test portions of nonviscous liquid products (i.e., the viscosity is less than that of milk) may be measured volumetrically for dilution by using a sterile pipette (11 mL into 99 mL, 10 mL into 90 mL, or 50 mL into 450 mL). If the pipette becomes contaminated before completing the transfer, replace it with a sterile pipette. Do not insert the pipette more than 2.5 cm below the sample surface. The pipette should be emptied into the diluent by letting the column drain from the graduation mark to the rest point of the liquid in the tip of the pipette from 2 s to 4 s, and then touch the lower edge of the pipette tip against the inside of the neck of the

N N N

N

N

N

N

N

N

|

Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis

dilution container. Do not blow out the last drop or rinse the pipette in the dilution fluid.17 When measuring products having a viscosity similar to milk, the last drop should be blown from the pipette. For viscous liquids, the test portion for the initial dilution should be aseptically weighed (11 ¡ 0.1 g) into a sterile 99 mL dilution blank (or 10 ¡ 0.1 g into 90 mL, or 50 ¡ 0.1 g into 450 mL). This provides a 1:10 dilution. Test portions of solid or semisolid foods should be 50 ¡ 0.1 g. The 50 ¡ 0.1 g test portion should be weighed aseptically (by using sterile forceps or spatula) into a sterile tared blender cup,11,17 and then 450 mL of sterile diluent should be added. Other analytical methods may be used, depending on the analyte and test or validated compositing scheme. A variety of diluents may be used depending upon the nature of the product. Those most commonly used are Butterfield’s phosphate buffer and 0.1% w/v peptone water (see the chapter ‘‘Microbiological Media, Reagents, and Stains’’). When analyzing for specific organisms or matrices, other diluents may be appropriate such as 3% NaCl for Vibrio parahaemolyticus. (Refer to the chapters on specific groups of organisms for other diluents.) When analyzing fatty foods or lump-forming powder, wetting agents/emulsifiers such as Tergitol Anionic-7 (1% w/v) may be included in the diluent to promote emulsification. Blend for 2 min at low speed (,8,000 rpm) to disperse the material.32 The blending time may vary, depending on the type of food.30 Some blenders operate at speeds lower than 8,000 rpm. It is preferable to use a higher speed for a few seconds initially. No more than 15 min should elapse from the time the sample is blended until all dilutions are prepared. Avoid overheating the sample in the flask. As an alternative, if the entire food sample is less than 50 g, weigh approximately one-half of the sample to the nearest 0.1 g portion into a sterile tared blender cup. Add sufficient sterile diluent to make a 1:10 dilution (i.e., add an amount of diluent equal to nine times the weight of the test portion in the blender cup). The total volume in the blender cup must completely cover the blades. Blend as described above. If the sample is not homogeneous, weigh 50 g from a representative portion of the package into a sterile, tared blender cup or analyze each portion of food separately. Proceed as described above. Exercise caution when blending to prevent excessive heating. The amount of heating may vary with foods of different consistencies and may be expected to increase if blending times greater than 2 min are required. Chilled diluent (tempered in an ice-water bath) may be employed to decrease the chances of excessive heating. Stomaching is an acceptable alternative to blending when preparing a food sample homogenate.2,10,19 In this procedure, the food sample with diluent is placed in a clean, preferably sterile, plastic bag. The plastic bag is positioned appropriately within the stomacher and pummeled for 1 to 2 min. Because the sample is contained in a plastic bag, the developers recommend that samples with bones or other sharp or protruding objects should not be prepared by stomaching. Thirty foods were

evaluated by using this procedure to determine its usefulness in a regulatory agency’s laboratory. Results indicate that only certain food homogenates should be prepared by using this procedure.2 In some solid food products, the microbial flora is restricted primarily to the surface area (e.g., properly smokehouse-treated frankfurters). More accurate enumeration of these microorganisms may be obtained by rinsing the sample with sterile diluent, rather than by blending.18 This can be accomplished by placing the sample in a suitable sterile container and adding a volume of sterile diluent equal to the weight of the sample. The container is then shaken in a manner similar to that used for preparing an initial dilution of a liquid food sample. Each milliliter of the rinsate thus prepared represents 1 g of sample.

N

See the chapter ‘‘Mesophilic Aerobic Plate Count’’ for discussions on the preparation of further dilutions and plate count procedures. Use the media recommended in microorganism-specific chapters for organisms of interest. Complete formulations and special preparation procedures for microbiological media are provided in the chapter ‘‘Microbiological Media, Reagents, and Stains.’’

2.532

Preparation of Samples for Water Activity Testing The ideal preparation of samples for water activity measurement is to grind the material to a fine consistency before testing. It is important to avoid heat build-up in the sample and moisture loss or gain during the grinding process. Certain emulsions such as oil/water emulsions may be difficult to measure, unless the water phase can be separated by low temperature cycling or centrifugation. Judgment must be used when evaluating the accuracy of measurements of the water phase of an emulsion. The prepared sample should be quickly added to the test chamber of the water activity meter and avoid an exchange of moisture with the air. If prepared samples will not be tested promptly, they should be stored in vapor-tight, sealed containers. Avoid storing in high- or low-humidity environments. When performing an analysis, transfer sample portions to the test instrument sample holder and follow the manufacturer’s instructions. A reliable reading may take from 5 min to several hours, depending on the type of instruments used. Proper maintenance and calibration of water activity instruments requires some skill and experience. Instrument manufacturers provide high-quality training and support. A complete description of water activity testing can be found in Rockland and Beuchat.20 See the chapter ‘‘Measurement of Water Activity, Acidity, and Brix’’ for further discussion of water activity measurement. 2.533

Preparation of Samples for pH Determinations Many types of liquid samples require very little preparation for pH determination. Semisolid samples, mixtures of solids and liquids, emulsions, and various types of marinated foods in oil require special preparation steps. Semisolid samples can be blended to a thick paste and a small amount of distilled water (# 20 mL/100 g) is added | 23

Compendium of Methods for the Microbiological Examination of Foods |

to provide a more fluid test portion. Mixtures of solids and liquids can be tested by blending the mixture to a paste and measuring directly or by separating the solid and liquid portions by using a U.S. standard #8 sieve. For marinated products in oil, separate the oil from the product, blend the solids (adding a small amount of distilled water if necessary), and test the paste. When attempting to determine the pH of emulsions, it may be necessary to separate the water phase for testing by low temperature cycling or centrifugation. Temperature effects on pH electrode and the actual hydrogen ion activity will modify the pH readings from electronic pH meters. Instruments that have temperature compensation adjust the response of the electrodes. However, the sample cannot be corrected for ionic activity. For accurate results, standardization and the actual determinations should all be performed at the same temperature and within a range of 20uC to 30uC. For further discussion of pH measurement, see the chapter ‘‘Measurement of Water Activity, Acidity, and Brix.’’

ACKNOWLEDGMENT Fourth edition authors: Thaddeus F. Midura and Raymond G. Bryant.

REFERENCES 1. Adams, D. M., and F. F. Busta. 1970. Simple method for collection of samples from a frozen food. Appl. Microbiol. 19:878. 2. Andrews, W. H., C. R. Wilson, P. L. Poelma, et al. 1978. Usefulness of the stomacher in a microbiological regulatory laboratory. Appl. Environ. Microbiol. 35:89-93. 3. Arthur, T. M., J. M. Bosilevac, X. Nou, et al. 2004. Escherichia coli O157 prevalence and enumeration of aerobic bacteria, Enterobacteriaceae, and Escherichia coli O157 at various steps in commercial beef processing plants. J. Food Prot. 67:658-665. 4. Bartlett, R. P., and J. B. Wegener. 1957. Sampling plans developed by United States Department of Agriculture for inspection of processed fruits and vegetables. Food Technol. 11:526-532. 5. Beef Industry Food Safety Council. Guidance document for sampling and lotting of beef products and sample analysis for pathogens. Available at http://www.bifsco.org/cmdocs/ bifsco2/new%20best%20practices/sampling_lotting_and_ sample_analysis_document_final_oct_2010_posted-2.pdf. Accessed April 16, 2015. 6. Boslaugh, S., and P. A. Watters. 2008. Statistics in a nutshell. O’Reilly Media, Sebastopol, CA. 7. Bray, D. F., D. A. Lyon, and I. W. Burr. 1973. Three class attributes plans in acceptance sampling. Technometrics. 15:575-585. 8. Dahms, S., and G. Hildebrandt. 1998. Some remarks on the design of three-class sampling plans. J. Food Prot. 61:757-761. 9. Dodge, H. F., and H. G. Romig. 1959. Sampling inspection tables: Single and double sampling. John Wiley & Sons, New York, NY. 10. Donegan, K., C. Matyac, R. Seidler, and A. Porteous. 1991. Evaluation of methods for sampling, recovery, and enumeration of bacteria applied to the phylloplane. Appl. Environ. Microbiol. 57:51-56. 11. Duncan, A. J. 1974. Quality control and industrial statistics. Richard D. Irwin, Homewood, IL.

24 |

12. Elliker, P. R., E. L. Sing, L. J. Christensen, and W. E. Sandine. 1964. Psychrophilic bacteria and keeping quality of pasteurized dairy products. J. Milk Food Technol. 27:69-75. 13. International Commission on Microbiological Specifications for Foods. 1982. Microorganisms in Foods 1: Their Significance and Methods of Enumeration. University of Toronto Press, Toronto, Canada. 14. International Commission on Microbiological Specifications for Foods. 1986. Microorganisms in Foods 2. Sampling for Microbiological Analysis: Principles and Specific Applications. University of Toronto Press, Toronto, Canada. 15. International Commission on Microbiological Specifications for Foods. 2002. Microorganisms in Foods 7. Microbiological Testing in Food Safety Management. Kluwer Academic and Plenum Publishers, New York, NY. 16. Kornacki, J. L. 2012. Hygiene control in the dry food products industry: the roles of cleaning methods and hygienic indicators. In: Hoorfar J., editor. Case Studies in Food Safety and Authenticity. Woodhead Publishing, Ltd., Philadelphia, PA. 254-266. 17. Kramer, A., and B. A. Twigg. 1966. Fundamentals of Quality Control for the Food Industry. AVI Publishing Company, Westport, CT. 18. Luchansky, J. B., A. C. S. Porto-Fett, F. M. Wallace, and J. E. Call. 2002. Recovery of Listeria monocytogenes from vacuumsealed packages of frankfurters: comparison of the U.S. Department of Agriculture (USDA) Food Safety and Inspection Service product composite enrichment method, the USDA Agricultural Research Service (ARS) product composite rinse method, and the USDA-ARS package rinse method. J. Food Prot. 65:567-570. 19. Nedoluha, P. C., S. Owens, E. Russek-Cohen, and D. C. Westhoff. 2001. Effect of sampling method on the representative recovery of microorganisms from the surfaces of aquacultured finfish. J. Food Prot. 64:1515-1520. 20. Rockland, L. B., and L. R. Beuchat. 1987. Water Activity: Theory and Applications to Food. IFT Basic Symposium Series. Marcel Dekker, New York, NY. 21. Steel, R. G. D., J. H. Torrie, and D. A. Dickey. 1997. Principles and Procedures of Statistics: a Biometrical Approach. McGraw-Hill Companies, New York, NY. 22. U.S. Department of Agriculture. 2011. Standards for Sampling Plans, Title 7, Code of Federal Regulations, Part 43. 101-106. 23. U.S. Department of Agriculture-Food Safety and Inspection Service. Verification activities for Escherichia coli O157:H7 in raw beef products. Directive 10010.1, Rev. 3. Available at http://www.fsis.usda.gov/OPPDE/rdad/FSISDirectives/ 10010.1Rev3.pdf. Accessed December 10, 2012. 24. U.S. Department of Defense. 1954. Administration of Sampling Procedures for Acceptance Inspection (H-105). U.S. Government Printing Office, Washington, DC. 25. U.S. Department of Defense. 1989. MIL-STD-105E. Military Standard: Sampling Procedures and Tables for Inspection by Attributes. U.S. Government Printing Office, Washington, DC. 26. U.S. Department of Defense. 1957. MIL-STD-414. Military Standard: Sampling Procedures and Tables for Inspection by Variables for Percent Defective. U.S. Government Printing Office, Washington, DC. 27. U.S. Department of Defense. 1994. MIL-STD-109C. Military Standard: Quality Assurance Terms and Definitions. U.S. Government Printing Office, Washington, DC. 28. U.S. Department of Transportation. 2010. Carbon Dioxide, Solid (Dry Ice), 49 C.F.R. Section 173. 217. 29. U.S. Department of Transportation. 2010. Hazardous Materials Table, Special Provisions, Hazardous Materials

|

Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis

Communications, Emergency Response Information, Training Requirements, and Security Plans, 49 C.F.R. Section 172. 30. U.S. Food and Drug Administration. Bacteriological Analytical Manual, 8th ed., Rev. A. Available at http://www.fda.gov/ Food/FoodScienceResearch/LaboratoryMethods/ucm2006949. htm. Accessed April 16, 2015.

31. U.S. Food and Drug Administration. Investigations Operations Manual. Available at http://www.fda.gov/ICECI/Inspections/ IOM/default.htm. Accessed April 16, 2015. 32. Wehr, H. M., and J. F. Frank, editors. 2004. Standard Methods for the Examination of Dairy Products, 17th ed. American Public Health Association, Washington, DC.

| 25

|

CHAPTER 3

|

Microbiological Monitoring of the Food Processing Environment Lloyd Moberg and Jeffrey L. Kornacki

3.1

INTRODUCTION

Microorganisms are ubiquitous in the natural environment. The survival and growth of microorganisms in a food processing environment may lead to contamination of the finished product that may reduce its microbiological safety and quality. There is a finite risk of food contamination from the processing environment whenever this food is not biocidally treated in its final container.59 However, it is unrealistic to expect food processing facilities to be sterile. An understanding of the relative risk of contamination from the plant environment must consequently be gained through monitoring, and then the contamination must be controlled. Several variables impact this risk. The most significant of these variables include the proximity of microbial growth niches to the product, the number of microbial niches in the plant environment, the spatial relationship of the niches to the product, the population in such niches, the degree to which niches are disrupted, and the exposure of the product to such niches.39 Microbial growth niches occur whenever the constellation of adequate moisture, nutrition, and time occur at a growthconducive temperature.59 Nonpathogenic microorganisms may promote spoilage with an adverse impact on quality, whereas pathogenic microorganisms may reduce microbiological safety. Most food plants have locations that can support the growth of pathogens and spoilage microorganisms.84 Microorganisms from these locations may be transferred directly onto the product or carried into additional niches. Sources that raise the potential risk of environmental microbial contamination can be grouped into three basic categories: (1) unsanitary operating conditions such as misapplied cleaning and sanitation, (2) unsanitary maintenance and repair practices, and (3) unsanitary design of the equipment and facility.60 In addition, raw materials, insects, and rodents can be a source of contamination to a product and environment. Microbial growth niches may also be established when water is used to clean-dry processing environments that are not designed for wet cleaning or when all points in the equipment are not | 27 |

promptly and completely dried. The chemical composition of the food and the conditions of water activity, pH, temperature, and so on will then determine the ‘‘normal’’ organisms that can grow there. Strong evidence from literature and personal experience indicates that postprocess contamination is the most significant source of processed food contamination.59,79 Behling et al.7 summarized a variety of accounts of environmental contamination events in processed foods. These included Campylobacter jejuni contamination of tuna salad; Salmonella contamination of ice cream, infant formula, soft cheese, cooked sliced ham, chocolate, canned meat, pastry, yeasts, and pasteurized milk; Listeria monocytogenes contamination of butter, hot dogs, luncheon meats, and Mexican-type cheeses; Clostridium botulinum contamination of canned salmon; Staphylococcus aureus contamination of lasagna, crabmeat, and canned mushrooms; Escherichia coli O157:H7 contamination of ground beef from a contaminated meat grinder and processing equipment, minced meat, and flavored yogurt, pasteurized milk, and various products that were handled with restaurant slicers and other equipment; Yersinia enterocolitica contamination of chocolate milk and pasteurized milk; and Bacillus cereus contamination of pasteurized milk. In a recent Centers for Disease Control and Prevention (CDC) publication, nontyphoidal Salmonella and L. monocytogenes were reported as the number one (28%) and number three (19%), respectively, causes of death from contaminated food.81 Salmonella contamination from the environment has historically been associated with foods of animal origin (e.g., meat, poultry, eggs, dairy products).59 The CDC estimates that an ice cream outbreak resulted in nearly a quarter million illnesses.50 However, fruit and vegetable outbreaks have also occurred. More than 25,000 Salmonella contamination cases occurred in a multistate cantaloupe outbreak that may have occurred from contact with unwashed rinds during and after cutting.80 Dry foods have also been implicated in Salmonella contamination with events occurring in the last decade from Salmonella-contaminated

Compendium of Methods for the Microbiological Examination of Foods |

breakfast cereal, peanut butter, spices, hydrolyzed vegetable protein, infant formula, dry milk, dry vegetable snacks, almonds, peanuts, chocolate, and tahini; this results in a major industry focus on controlling Salmonella in the dry food processing environment.83,15,16 Vij reviewed spiceassociated recalls occurring in the United States between 1970 and 2003, and found that 76% (16 of 21 recalls) occurred in the last 4 years of the study.95 The pathogen L. monocytogenes has been isolated from numerous wet processing environments in dairy,13 egg, seafood, vegetables, meat, and poultry plants.4,61 In a potato processing plant, Listeria spp. were isolated from floors and drains, condensed and stagnant water, process equipment, conveyor belts, and wiping cloths.21 In addition, Listeria spp. have been isolated from brine chillers, dehumidifiers, air handling systems, product conveyors, slicers, dicers, spiral freezers, packaging machines, fillers, wet insulation, cracks and crevices of floors, milk case conveyor belts, and crevices of many types of processing equipment in a variety of food processing environments.4,8,36,45,61,64,100 A review of Listeria spp. contamination in a variety of food production of facilities is found in Kornacki and Gurtler.61 Airborne microorganisms and endotoxins have been recovered in herb processing plants,27 potato processing plants,26 and flour mills.6 Biofilms are also an important and potential source of microbial contamination within a processing environment. A biofilm is formed when microorganisms colonize on a surface. Most food industry biofilms consist of microorganisms, and their exocellular polymers intermingle with food residues or mineral deposits. Any surface within a facility that is exposed to water or moist food will support biofilm formation if it is not effectively cleaned at regular intervals.9,87 Environmental surfaces such as drains, floor mats, and equipment exteriors may support the growth of pathogens.39 Many pathogens such as Salmonella and L. monocytogenes form biofilms.39 L. monocytogenes survives well in multispecies biofilms that may accumulate in such environments.55 Worn, abraded, or corroded food contact surfaces tend to accumulate biofilms because they are difficult to clean and sanitize.38,51 In addition, poor equipment design or ineffective cleaning regimens will ultimately lead to biofilm formation. Microorganisms, including L. monocytogenes, are highly resistant to chemical sanitizers when growing in biofilms.37,65 However, if the biofilm is completely disrupted and dissolved by the cleaning process, these microorganisms are readily inactivated by commonly used chemical sanitizers.36 If equipment design does not allow this degree of effective cleaning, then sanitizers that are especially suited for biofilm inactivation may need to be used. To prevent, minimize, or control potential microbiological contamination within a food processing facility, a strategy encompassing multiple tactics would yield the most success. First, the hygienic design and construction ideally would initially prevent microbial niches. Second, adherence to good manufacturing practices in the maintenance of the factory and equipment, and adherence to the hygienic operation of the processes and equipment would then further aid in preventing or minimizing the potential microbial contamination of products. Third, the application of appropriate cleaning (e.g., dry vs. wet) and disinfection procedures would constitute the principal effective approach for the 28 |

control of microbial contamination and growth. To suppress the establishment of microbial growth niches and biofilms, the environment, including the processing equipment, must be designed and fabricated to resist microbial growth or designed to be easily cleanable. Then, as presented in this chapter, the appropriate monitoring and control procedures must be established to ensure prevention or early detection and control of any potential microbiological problem.

3.2

ENVIRONMENTAL SAMPLING STRATEGIES

Microbiological monitoring of the food processing environment may be conducted to meet one or more of the following objectives: (1) verification of the effectiveness of cleaning and disinfection practices, (2) determination of the frequency required for cleaning and disinfection, (3) determination of the presence of foodborne pathogens in the environment, (4) discovery of environmental sources of spoilage organisms, (5) determination of the frequency required for special maintenance procedures (e.g., changing of air filters to reduce airborne mold contamination), and (6) evaluation of hygienic design and fabrication of food processing equipment and facilities. Adenosine triphosphate (ATP) bioluminescence may be used to measure the efficacy of the cleaning of food plant environments and equipment and used to provide a means to validate in real time that effective cleaning has occurred. Programs established to meet these objectives are directed toward preventing or eliminating a potential microbiological hazard in a finished food product. Control of the food manufacturing plant environment is a prerequisite to an effective Hazard Analysis Critical Control Point (HACCP) system.52 A sampling program may be established to verify cleaning and disinfecting procedures for processing, conveying, and packaging equipment, since these procedures can be important control points for preventing postprocessing recontamination (not critical control points). Monitoring of this control can be best accomplished through sensory inspection (i.e., the plant and equipment appear, smell, and feel clean), chemical tests (i.e., measurement of sanitizer concentration and pH), physical tests (i.e., measurement of temperature), and microbiological tests of the equipment and environment. Microbial criteria for the acceptance of the cleanliness of the equipment and the environment can be developed by using a database derived from repeated routine sampling and testing of specific sites.29 However, the time required to obtain microbiological results is usually too long to make sampling and testing an effective tool for day-to-day monitoring. To be effective, monitoring systems must provide information promptly; the development of biosensors holds promise for this type approach (see the chapter ‘‘Rapid Methods for the Detection and Identification of Foodborne Pathogens’’). Four approaches may be used to verify the microbiological acceptability of food processing equipment and environments: (1) collecting and testing nonfood contact samples from the food processing environment; (2) sampling and testing equipment before and after operations; (3) collecting and testing in-process product samples; and (4) measuring microbial loads in food products after the completion of all processing, packaging, and handling.52 Environmental sampling and testing in the environment and

| Microbiological Monitoring of the Food Processing Environment

on process equipment can be an early warning system to detect and eliminate niches of undesirable microorganisms before the risk of product contamination increases significantly. This monitoring program should be designed to measure the occurrence and numbers of the normal spoilage flora and the pathogens that present the greatest risk to the product. The collection of microbiological samples should not be limited to environmental sites that are easily cleaned and sanitized because the results from only these points may not reveal critical hazards and risks. If cleaning and sanitizing are effective, then such easily disinfected sites should yield satisfactory results. The verification procedures should also include collections of food or other organic residue samples from more inaccessible or neglected niches. In general, plant environments are categorized into four zones for the purpose of microbiological monitoring.53 Zone 1 sites are direct food contact surfaces and may include surfaces such as conveyors, tables, holding vats, tanks, utensils, the inside of product pumps, valves, slicers, dicers, freezers, and filling or packaging machines. After cleaning and sanitation, these surfaces are usually microbiologically sampled for sanitary indicators such as aerobic plate count, coliforms, or Enterobacteriaceae, yeasts, and molds. Such testing can provide invaluable evidence—when combined with documented cleaning and sanitization approaches and appropriate product testing of a sanitation ‘‘break point’’—that may be used to establish the acceptability of a product produced between cleaning and sanitization events. Zone 1 surfaces are rarely tested for pathogens. Zone 2 surfaces are close to a product and may include the exterior of equipment such as the nonproduct side of the following: conveyor belt guides, bearings for agitator shafts, bucket elevators, screw conveyors, gasketing inside rotary product valves, and ultrahigh molecular weight plastic inside freezers and coolers (when not directly over or in contact with a product). Some Zone 2 surfaces have a high potential to contaminate product. Zone 3 surfaces are in the processing environment and may include surfaces such as walls, drains, and forklifts. Zone 4 surfaces are even further away from a product and may include hallways, restrooms, locker rooms, and cafeterias within the food production plant. Pathogen monitoring is typically performed in Zones 3 and 4. Pathogen testing is typically not recommended in raw areas of the plant because negative results may provide a false sense of control. To avoid errors of judgment and interpretation in quantitating microbial hazards and risks associated with the equipment and environment and to help establish which microorganisms should be sought in a monitoring or verification program, it is important to understand the microbial ecology of a specific food and its process. The ecological pressures of extrinsic and intrinsic factors, such as heat processing steps, processing and storage temperatures, packaging atmospheres, the oxidation reduction potential of the product, acidity, pH and water activity, and competitive microbiota, should determine which organisms are important in a particular environment. In plants that process dry foods that are not intended to be cooked before consumption (e.g., nonfat dry milk, chocolate, and peanut butter), Salmonella may be a significant hazard, especially when high-moisture– containing microbial growth niches are present.

The environmental (e.g., Zones 3 and 4) monitoring program should be designed to measure the occurrence and the number of normal spoilage flora and the pathogens or the hygienic indicators suggesting their presence that present the greatest risk to the product. To cite one example, dried milk products are used to make chocolate and confectionery products and are a potential source of salmonellae. Destruction of these organisms by thermal processing is unlikely because their heat resistance is very high in low water activity products; therefore, the milk processing environment should be monitored for salmonellae to reduce risks further up the processing chain. This is consistent with the HACCP approach in that if a control step does not exist in the manufacturing of a finished product, control must be exercised upstream at the ingredient stage. An environmental and process equipment monitoring or verification program that includes tests for indicators, such as the aerobic plate count, family Enterobacteriaceae, Listeria spp. or Listeria-like indicator bacteria, and fungi, and tests for the plant’s unique microflora will permit a more accurate assessment of microbial contamination of equipment and the plant, compared to pathogen testing alone. Negative pathogen test results may be misinterpreted as indicating that the site is microbiologically inert. Such pathogen test results merely indicate that the pathogen of interest was not detected in that site at the time of sampling; they do not provide useful information about the general microbiological risks associated with that site.9 Of all the rapid tests currently available, the only test that potentially offers real-time results is ATP bioluminescence (see the chapter ‘‘Rapid Methods for the Detection and Identification of Foodborne Pathogens’’). Adenosine triphosphate bioluminescence has limitations for implying the levels of indicator bacteria or the potential presence of pathogens; however, it may be used to measure the efficacy of the cleaning of food plant environments and equipment and to provide a means of quickly validating that effective cleaning has occurred. If a food is processed as an ingredient for another product, it is prudent to monitor the ingredient’s production environment for those organisms that will be a hazard to the ultimate finished product.

3.3

SAMPLING OF SURFACES: EQUIPMENT AND PHYSICAL PLANT

Sampling sites on equipment should be selected to include all points that are liable to harbor microorganisms that may directly or indirectly contaminate the product. Sampling sites should not necessarily be limited to direct product contact zones because microbial contamination can also be transferred indirectly into a product from condensation, aerosols, lubricants, packaging materials, line workers’ garments, and so on. The distinction between what is and is not a product contact zone is not always easy to determine—especially in open systems in which the product is exposed to the processing environment and is not continuously protected by enclosure in a pipeline or vessel.52 Direct product contact surfaces include pipeline interiors, conveyors, product storage vessels, fillers, utensils, work tables, mixers, and grinders. Nonproduct contact sites include the structural | 29

Compendium of Methods for the Microbiological Examination of Foods |

components of machinery; the exterior of equipment, pipelines, and vessels; walls; motors; bearings; floors and floor drains of buildings; heating, ventilation, and air-conditioning equipment; forklifts; workers’ garments, gloves, and footwear; mechanics’ tools; and cleaning tools. Microorganisms can be transferred from nonproduct contact surfaces to direct product contact surfaces during production, between cleaning and sanitation cycles, and during maintenance procedures. Failure to clean and disinfect all sites that harbor microbial growth will increase the risks of contaminating food contact surfaces and the finished product. Verification that sites are microbiologically acceptable is best accomplished by sampling and testing. Sensory evaluation is useful for detecting environmental conditions that may lead to microbial growth and survival, but visually clean sites may still harbor microorganisms and microbial growth niches. Therefore, verification of cleanliness and microbiological acceptability requires sampling and testing. Sampling and testing merely provide a rough estimate of the quantity of food debris and microbial populations on equipment, but the accumulation of data from repeated tests will permit the development of criteria by which to judge the hygienic condition of specific pieces of equipment.28 In addition, these historical data will provide an indication of the effectiveness of the current cleaning and sanitation process. Repeated findings of bacteria exceeding these criteria would indicate an ineffective cleaning and sanitation process that needs improvement. Preprocess sampling and microbiological testing of equipment by conventional microbiological methods has limitations. Because of the time required after sampling to obtain results, it is not useful for immediately approving production equipment for use. The production will generally begin before sites are identified that have unacceptable numbers or types of microorganisms. However, the application of realtime ATP bioluminescence testing for monitoring hygiene has gained acceptance in the food industry and may offer some evidence of surface cleanliness, but not microbiological suitability of the sampled surfaces, because several authors have shown a lack of correlation between cleaning (e.g., soil removal) and sanitation (e.g., the reduction of microbial populations to acceptable levels).19,44,63 Microbiological analysis of preoperative samples is nevertheless useful to verify and historically track the efficacy of the cleaning and sanitizing procedures to evaluate the performance of the cleaning and sanitizing crew and/or the cleanability of a particular piece of equipment.

3.4

PRINCIPLE OF MONITORING THE MICROBIOLOGICAL FLORA

Natural selection is the underlying scientific principle that applies to the need to monitor the microbiological flora of a food processing environment. Based on the type of product that is produced in the manufacturing environment, microorganisms will be selected that can best adapt and survive in the environmental conditions encountered in the plant, manufacturing equipment, and residual food matrices. Failure to exert control over the selection of these ‘‘normal flora’’ will result in their proliferation with the subsequent deterioration of product quality and a potential increase in the safety hazard of a food. 30 |

3.5

PRECAUTIONS

All sampling should be performed by personnel trained in aseptic techniques to avoid inadvertent sample contamination. Such contamination will result in inaccurate findings leading to misinterpretation of results. In addition, pathogenic microorganisms may occasionally be recovered in samples. Use of appropriate sampling tools, diluents, and neutralizing broths/buffers to inactivate sanitizer residues and validated testing methods are critically important for ensuring the accuracy and value of test results. Appropriate handling of cultures in the laboratory, including disposal (e.g., sterilization), should be practiced to prevent hazard to laboratory personnel and prevent the inadvertent broader distribution of the organism throughout the facility. The broader exposure or contamination of equipment or the environment would also apply to nonpathogenic microorganisms. To avoid such risk, some facilities do not allow on-location microbiology laboratories (where culturing could produce high levels of the microorganisms) to test for pathogenic microorganisms. In this circumstance, testing may be performed offsite.

3.6

LIMITATIONS

Knowledge of the type of potential spoilage or pathogenic microbes of concern in the environment and product is essential; the obtained results may otherwise not be indicative of the potential hazard. The results of the microbiological monitoring of the food processing environment are generally limited by the specific microbiological methods being utilized. Recovery media utilized for the general aerobic or facultative, mesophilic population will indicate only this population. Testing for a strictly anaerobic population would similarly require the use of the appropriate media and conditions. Monitoring and testing for specific pathogenic microorganisms would also require the appropriate selective methods and media for recovery and identification. For example, the aerobic plate count does not imply the presence or the absence of a potential pathogen such as L. monocytogenes or Salmonella. Consultation with an industrial and/or food microbiologist may be beneficial when developing a monitoring program for a facility to understand microorganisms of concern, proper methods and media, environmental, equipment and in-process test points, and interpretation of results. Specific limitations of the individual methods covered in this chapter are provided in the discussion of the respective procedures.

3.7

RINSE SOLUTION METHOD FOR SAMPLING CONTAINERS AND PROCESSING EQUIPMENT SYSTEMS98

3.71

Equipment, Supplies, Solutions, and Media

N N N N

Sterile pipettes Sterile Petri dishes Stock phosphate buffer solution (e.g., Butterfield’s phosphate buffer dilution water stock solution containing 34 g of KH2PO4 dissolved in 500 mL distilled water) Sterile buffered rinse solution comprising appropriately autoclaved or filter-sterilized solution of 1.25 mL of the

| Microbiological Monitoring of the Food Processing Environment

N N N N N N

aforementioned stock solution made with distilled water up to 1 L. Sodium thiosulfate (10%) solution Standard Methods Agar Violet red bile agar m-Endo broth Millipore filter (MF) Nutrient broth (may be used as a substitute for buffered rinse solution) Plate count agar

3.72

Packaging Containers

Remove containers from the conveyor line or container cartons. For 1-L or smaller containers, aseptically add 20 mL of sterile buffered rinse solution into each container; for 1.89-L containers, use 50 mL; and for 3.78-L or larger containers, use 100 mL. After adding the rinse solution, recap the container. Holding the container firmly with its long axis in a horizontal position, shake vigorously 10 times through a 20-cm arc. Turn the container 90u and repeat the horizontal shaking treatment. Turn the containers 90u twice more and repeat horizontal shaking. Swirl the container vigorously 20 times in a small circle with the long axis in the vertical position, and then invert and repeat. Stand the container upright before removing the sample. For small containers, determine the number of bacteria in the rinse solution by distributing 5 mL equally between two sterile Petri dishes. For larger containers, place 2 mL of the rinse solution into a single plate. Determine the number of coliforms and/or aerobic bacteria in the rinse solution by dividing a total of 10 mL of the rinse solution among three sterile Petri dishes. Pour 15–20 mL of the desired medium (e.g., plate count agar for aerobic bacteria or violet red bile agar for coliforms) by using the appropriate incubation conditions. Yeasts and molds, proteolytic bacteria, and other specific microorganisms may be determined by using appropriate differential media and incubation temperatures and times, as described in their respective chapters. To calculate the residual bacteria count per container, multiply the total number of colonies by the volume of the rinse solution and divide the result by the volume of the sample plated. When using 100-mL or greater portions of rinse solutions, follow membrane filtration procedures for analysis (see the chapter ‘‘Mesophilic Aerobic Plate Count’’), particularly if low levels of contamination are expected. Membrane filtration may also be used for the analysis of 20-mL rinse samples. Interpretation of results from the container rinse samples should take into consideration the number and the types of microorganisms. The types of microorganisms may be important in their potential to cause spoilage. The number of microorganisms should normally be very low. In most circumstances, the number of organisms added to the product from the container will be much lower than the number that is indigenous to the product. However, under certain circumstances such as with aseptic packaging, the microbiological condition of the containers is crucial. Application of microbiological guidelines and standards is uncommon for the myriad of containers in use today. However, bacterial standards have been published for multiuse

containers and single-serve containers for packaging pasteurized milk and milk products35 and bottled water.93 These standards require that such containers have a residual bacteria count of one colony or less per milliliter of capacity or no more than one colony per square centimeter of contact surface. No coliform organism may be present. For pasteurized dairy product containers, four containers from any particular day are sampled, and three of the four samples must meet this standard. The standard for bottled drinking water containers states that, at least once every three months, bacteriological swab or rinse samples are to be obtained from at least four containers and closures selected just before filling and sealing.

3.73

Processing Equipment Systems (e.g., Tanks, Pipelines, Fillers)72

Water for large-volume rinse-sampling of equipment should be heat-sterilized or may be treated by chlorinating to a residual concentration of 25 mg per liter, holding for 10 minutes, and then neutralizing by adding an excess of sterile 10% sodium thiosulfate solution. Tap water may be used after sterilizing by membrane filtration, followed by the addition of sterile 10% weight per volume (wt/vol) sodium thiosulfate to inactivate residual disinfectant. A sufficient volume of treated rinse water is added to the system at the upstream end of the assembly and then pumped or allowed to flow by gravity through the assembly. A control sample (, 1 L) of the treated rinse water is obtained before using the water for rinsing. Samples of rinse water are collected from the discharge end of the assembly from the first, middle, and final portions of the rinse water. Samples may also be collected at various points throughout the assembly. The membrane filtration procedure (see the chapter ‘‘Mesophilic Aerobic Plate Count’’) may be used to analyze large volumes of rinse water. Analyses of rinse solutions from clean-in-place processing assemblies and control samples require the use of membrane filtration procedures. Average the number of colonies obtained from rinse samples obtained at the beginning, middle, and end of drainage and subtract the number of colonies (if any) obtained from the control samples. Calculate the ratio of the sample volume to the rinse volume. Multiply the result by the corrected yield to obtain an indication of the number of organisms present in the entire system. The presence of specific types of organisms may be determined by employing appropriate differential media and incubation temperatures. Refer to the specific chapters covering these organisms.

3.8

SURFACE CONTACT METHODS

A meaningful microbiological examination of surfaces requires selection of an appropriate method. The replicate organism direct agar contact (RODAC) procedure, the Petrifilm (3M, St. Paul, MN) aerobic count plate procedure, and the swab procedure are the usual methods of choice for sampling surfaces. Swab techniques should be used for surfaces with cracks, corners, or crevices (i.e., areas having such dimensions that a swab is more effective in recovering organisms from them). Swab procedures should also be used for sampling utensils, tableware, and kitchenware. | 31

Compendium of Methods for the Microbiological Examination of Foods |

Sponge/swab procedures are useful for sampling large areas of food processing equipment and environmental surfaces. The RODAC and 3M Petrifilm procedures should be used only on flat impervious surfaces that are relatively easy to clean and disinfect. Selection of the proper technique is essential to obtain meaningful results.

3.81 3.811

N

N

N

Swab Contact Method

11,91

Equipment and Supplies

Sterile nonabsorbent cotton swabs with the head firmly twisted to approximately 0.5 cm in diameter by 2 cm long on a wooden applicator stick 12–15 cm long may be used. Swabs should be packaged in individual or multiple convenient protective containers with the swab heads away from the closure. Calcium alginate, dacron, and rayon swabs may also be used. Presterilized swabs may be purchased or the swabs may be sterilized in the laboratory. A commercially available test system that includes a swab sampler and various agar recovery media is comparable to conventional swab sampling procedures.24,88 Swabs made of calcium alginate fibers are soluble in aqueous solutions (e.g., rinse, culture media) containing 1% of sodium hexametaphosphate, sodium glycerophosphate, or sodium citrate. All organisms captured on the swab will be liberated from the calcium alginate swab. Presterilized calcium alginate swabs contained in various transport media are commercially available. The transport medium maintains microbial viability while inhibiting multiplication. After autoclaving, small screw-capped vials (7–10 cm long) are prepared to contain 5 mL buffered rinse solution (or 4.5 mL if calcium alginate swabs are used).

When sampling surfaces previously subjected to chemical disinfection, appropriate neutralizers should be incorporated into the rinse solution. A commonly used neutralizer (e.g., Letheen Broth, Neogen Corp, Lansing, MI) is 0.5% polysorbate (i.e., Tween 80) plus 0.07% soy lecithin. Some commercially available neutralizing buffers contain sodium thiosulfate for inactivation of the germicidal effects of chlorine and iodine-based sanitizers. They are buffered to neutralize acid sanitizers. Other neutralizing buffers (e.g., Dey-Engley [DE] neutralizing broth) also contain lecithin and a pH indicator dye (bromocresol purple). The pH indicator turns the purple buffer to yellow should it become contaminated with fermenting microbes that are allowed to grow. Dehydrated media for preparing neutralizing solutions are commercially available. Polysorbate 80 neutralizes some substituted phenolic disinfectants and soy lecithin neutralizes quaternary ammonium compounds. The efficacy of any disinfectant neutralizer should be validated under actual use conditions. Consideration should be provided to the selection of the appropriate neutralizing broth for the recovery of one or more target organism. For example, DE neutralizing broth reportedly neutralizes quaternary ammonium compounds, phenolics, iodine, and chlorine; however, it is not appropriate for recovering coliforms on Petrifilm. Letheen Broth reportedly promotes growth; neutralizes quaternary ammonium compounds, 32 |

phenolics, somewhat neutralizes iodine and chlorine sanitizers. However, it can be used with Petrifilm.67 To inactivate peroxyacetic acid sanitizers, a buffer that contains sodium thiosulfate must be used.

3.812 Sampling Procedure To sample equipment surfaces, open the sterile swab container, grasp the end of a stick while being careful not to touch any portion that may be inserted into the vial, and remove the swab aseptically. Open a vial of buffered rinse solution, moisten the swab head, and press out the excess solution against the interior wall of the vial with a rotating motion. Hold the swab handle to make a 30u angle contact with the surface. Rub the swab head slowly and thoroughly over a surface area of approximately 50 cm2 three times, while reversing direction between strokes. Move the swab on a path 2 cm wide by 25 cm long or other dimensions to cover an equivalent area. Return the swab head to the solution vial, rinse briefly in the solution, then press out the excess. Swab four more 50-cm2 areas of the surface being sampled, as described previously, and rinse the swab in the solution after each swabbing. Remove the excess. After the areas have been swabbed, position the swab head in the vial, and break or cut it with sterile scissors or other device,12 leaving the swab head in the vial. Replace the screw cap, put the vial in a waterproof container packed in a suitable refrigerant, and deliver to the laboratory. Analyze the sample within 48 ¡ 2 hours after collection.34 When sampling utensils such as knives and ladles, moisten the swab with dilution fluid and then run the swab slowly and firmly three times over the significant surfaces of the utensil. Reverse the direction each time. After the utensil has been swabbed, return the swab to the buffered rinse solution by the procedure described previously. When unmeasured surface areas such as pump impellers, gaskets, rings, valve seats, and filler nozzles have been swabbed, the results may be reported on the basis of the entire sampling site instead of a measured area. 3.813 Plating Swab Rinse Solutions At the laboratory, remove the vial from refrigerated storage. Shake it vigorously, making 50 complete cycles of 15 cm in 10 seconds, striking the palm of the other hand at the end of each cycle with the flat lengthwise portion of the vial. Groups of vials may be shaken together to save time. Plate 1-mL and 0.1-mL portions of rinse solution, plus additional dilutions, if deemed necessary. Pour plates with Standard Methods Agar or other appropriate media, depending on the organisms of interest; incubate; count colonies; and then calculate the number of colonies recovered from 50 cm2 (equivalent to 1 mL of rinse). When searching for groups of microorganisms other than the aerobic plate count, plate with appropriate selective/ differential media and incubate, as required. 3.814 Interpretation As a guide, the U.S. Public Health Service (Silver Spring, MD) recommends that adequately cleaned and sanitized food service equipment have not more than 100 colonies per utensil or surface area of equipment sampled.92 Interpretation

| Microbiological Monitoring of the Food Processing Environment

of results obtained from unmeasured surface areas such as utensils, gaskets, and pump impellers should be based on the knowledge of historical data obtained when the surfaces had been documented as being thoroughly cleaned and sanitized. In general, the levels of microorganisms should not exceed more than a few colonies per sampling site. In many cases, the types of microorganisms may be more significant than their numbers alone. For example, the presence of even very low numbers of Saccharomyces bailii and/or Lactobacillus fructivorans on salad dressing processing equipment may be highly significant with respect to potential spoilage of the finished product. Thus, for the spoilage organisms of specific foods, the standards for evaluating sanitation may be much more stringent than when only the total numbers are used. When swabbing is performed for purposes other than evaluating sanitation procedures, interpretation of results must be based on the knowledge of the product, process, and equipment to determine the significance of data. In addition, the objectives of sampling may govern the interpretation of results.

3.82 3.821

N

N

N

Sponge Contact Method86 Equipment, Supplies, Solutions, and Media

Cellulose or polyurethane68 sponges free of antimicrobial preservatives should be cut into approximately 5 cm 6 5 cm pieces, placed in individual Kraft paper bags, and autoclaved. As an alternative, commercially sterile cotton gauze surgical swabs (approximately 10.2 cm 6 10.2 cm) may be used. Sterile plastic bags are suitable for containing the sponges after sampling. Sterile buffered rinse solution, nutrient broth, or 0.1% peptone water may be used as the rinse solution. If the surface to be sampled contains fatty materials, 0.5% to 1.0% Tween 80 or other noninhibitory surfactant solution may be used. For sampling equipment that may contain residual disinfectants, the use of neutralizers in the buffered rinse solution is recommended (see Section 3.811). It is prudent to incorporate neutralizers in all fluids used to collect samples from equipment and the plant. Neutralizing cocktails and transport media are commercially available. Sterile crucible tongs, sterile rubber or plastic gloves, or other means may be used to hold the sponge aseptically during sampling.

3.822 Sampling Procedure Moisten the sponge with approximately 10 mL of the appropriate sampling fluid. While holding the sponge aseptically with tongs or sterile gloves, swab the surface to be sampled by vigorously rubbing the sponge over the designated area. If the surface is flat, the rinse solution may be applied directly to the surface and then taken up into the sponge by the rubbing action. An area of several meters may be effectively swabbed. After sampling, place the sponge aseptically in a sterile plastic bag and transport it to the laboratory under refrigeration. Analyze the sample within 48 ¡ 2 hours after collection.34 3.823 Plating and Analysis Because large areas may be sampled with the cellulose sponge, this technique is particularly useful for detecting

pathogens (e.g., Salmonella or Listeria) or spoilage microorganisms in the food plant environment. For Salmonella or Listeria analyses, the sponge is introduced directly into an enrichment broth, incubated, and then tested by approved methods for Salmonella (see the chapter ‘‘Salmonella’’) and Listeria (see the chapter ‘‘Listeria’’). In contrast to quantitative plating methods, enrichment-based methods provide the best opportunity for detecting the lowest possible levels of Listeria or Salmonella contamination. The sponge sample may be subjected to a variety of microbiological analyses in the same fashion as fabric-tipped swabs. For quantitative analyses, 50–100 mL of diluent are added to the bag containing the sponge. The sponge is then vigorously massaged with diluent for 1 minute or more to release the microorganisms. Aliquots of the diluent are removed from the bag, further diluted if required, and plated into the desired media for the microorganisms in question. After incubation, the number of microorganisms per unit surface can be calculated on the basis of the area swabbed, the amount of diluent used, and the size of aliquot plated. For example, if 50 colonies are obtained from a 1-mL aliquot derived from a sponge in 100 mL of diluent that swabbed 1 m2, then the count per milliliter squared will be 5000.

3.824 Interpretation Interpretation of results from sponge samples obtained from cleaned and sanitized equipment is essentially the same as the interpretation for results obtained from fabric-tipped swabs. The sponge technique has historically been useful in sampling the environment for Salmonella.86 Experience shows it also is useful for Listeria.40 This technique can be used to evaluate the efficacy of cleaning and sanitizing programs for the environment, particularly for foodborne pathogens. Results should obviously always be negative after the application of appropriate cleaning and sanitizing procedures. Sponge swabs can be taken to identify areas that harbor pathogens, and the results can be used to develop a program to control the organisms. The evaluation of results from samples taken of cleaned and sanitized floors and other areas where relatively high residual microbial levels are expected is affected by the history and experience related to particular sites in a plant. As a rule of thumb a 4- to 5-log reduction in the residual microbial level should be obtained on most floor surfaces after cleaning and sanitizing. Another common approach is to assume that large surface areas (e.g., 1 ft2) are unlikely to have 100,000 or more colony-forming units (CFUs) via the aerobic plate count unless microbial growth has occurred. Such freshly sanitized surfaces should contain aerobic plate counts of less than 100 organisms to 1,000 organisms per sponge and should contain no coliforms, Enterobacteriaceae, yeasts/molds, Listeria spp. or Listeria-like indicator bacteria that are easily destroyed by appropriate cleaning and sanitization.59 3.83

Agar Contact Method 5,47,97

An agar contact method such as the RODAC plate method provides a simple, valuable contact technique for estimating the sanitary quality of surfaces. The method is recommended particularly when quantitative data are sought from flat, impervious surfaces. It is not intended to be used for crevices or irregular surfaces, although the RODAC | 33

Compendium of Methods for the Microbiological Examination of Foods |

plate may be useful even if its only purpose is to demonstrate the presence or absence of a specific microorganism. The RODAC plate method should ideally be used on previously cleaned and sanitized surfaces. Samples taken from heavily contaminated areas will result in overgrowth on the plates. If accurate colony counts are desired, the plates should have fewer than 200 colonies. A sufficient number of sites should be sampled to yield representative data. Randomization of site selection may permit additional comparisons and inferences.

3.831

N

N

N

N

Equipment, Supplies, Solutions, and Media

Disposable plastic RODAC plates may be purchased prefilled with test medium or they may be filled in the laboratory. When prepared in the laboratory, the plates should be filled with 15.5–16.5 mL of the appropriate medium. The meniscus of the agar should rise above the rim of the plate to give a slightly convex surface. This is necessary so that the agar makes proper contact with the surface to be sampled. Plate count agar is normally used for aerobic plate counts. However, if qualitative data for specific microorganisms are desired, selective or differential media may be used (e.g., Lactobacillus Section agar for lactic acid bacteria, violet red bile agar for coliforms, or Baird-Parker agar for S. aureus). Dey-Engley neutralizing medium may be used in place of plate count agar. This medium incorporates a variety of ingredients that are capable of neutralizing germicidal chemicals likely to be encountered on surfaces.30 After preparation of the plates, they should be incubated at 32uC for 18–24 hours as a sterility check. They should be used within 12 hours after preparation, unless they are wrapped and refrigerated. In lieu of RODAC plates, two commercially available systems—3M Petrifilm74 or Con-Tact-It (Birko Chemical Corp., Henderson, CO)88—can be used as a medium contact method. Another method is to use mylar adhesive tape (Dynatech Laboratory, Inc., Alexandria, VA), which is transferred to the surface of an appropriate agar plate after being pressed to the surface of the sample site.18

3.832 Sampling Procedure Remove the plastic cover from the RODAC plate and carefully press the agar surface to the surface being sampled. Make certain that the entire agar meniscus contacts the surface by using a rolling uniform pressure on the back of the plate. 3.833 Incubation and Colony-Counting Procedure Replace the cover and incubate in an inverted position under the appropriate time and temperature conditions for the microorganisms in question. Colonies should be counted using a Bactronic (Diversified Equipment Co., Lorton, VA) or Quebec colony counter (Reichert Technologies, Depew, NY) and recorded as the number of colonies per RODAC plate or number of colonies per cm2. 3.834

The Pros and Cons of Different Sampling Methods Advantages and disadvantages associated with swab, sponge, and contact plates have been previously described.59 34 |

Traditional swabs are useful for sampling penetrations, cracks, crevices, and other sandwich areas that agar contact and sponge sampling cannot reach. Enrichments can be performed. Before enrichment, nonenriched aliquots of samples can be diluted for quantitative assays. Sponges work well for large surfaces area and more pressure can be applied with sponges than with swabs. They can also be tested quantitatively and qualitatively. Contact plates are inappropriate for penetrations and it may be difficult to get a representative sample from cracks and crevices. Typically dilutions cannot be performed unless the agar is aseptically scraped, and appropriately macerated in buffer; however, this approach has not been validated and may not be sufficiently consistent to be reliable. Qualitative testing is also impractical. Presterilized tongue depressors or scrapers are valuable in removing encrusted material on environmental samples. An inexpensive alternative approach was published in which presterilized single ply tissues were hydrated with a small amount of buffer; the recovery of L. monocytogenes from surfaces was greater than its recovery by environmental sponge samples, which was greater than the recovery by cotton-tipped swabs, which was greater than the recovery by calcium alginate fiber-tipped swabs. The RODAC plate analysis of surfaces sampled by these approaches were typically overgrown after sponge sampling, were positive after cotton and calcium alginate swab sampling, and were negative after swabbing with the oneply tissue approach.96 In another study, there was no significant differences between the single-ply tissue method in conjunction with the Soleris optical analysis system (Neogen Corp, Lansing, MI) and the conventional U.S. Department of Agriculture environmental sponge enrichment method for inoculated stainless steel and polyethylene surfaces and environmental samples.99

3.835 Multiple Tests per Swab When a surface is sampled, it is altered microbiologically. Because an unknown proportion of the microbial flora is reduced by each sampling, multiple samplings of the same surface is discouraged. Therefore it is often necessary to determine the results of multiple microbiological tests from a single surface sample. Multiple tests can be performed from a single sample, provided an adequate standard quantity of appropriate diluent/buffer is added and a convention is established for determining the dilution associated with the sample. For example, a laboratory may elect to routinely add 25 mL of Butterfield’s phosphate buffer to each sponge sample or 10 mL to each swab sample and call this a ‘‘zero’’ dilution or a 1:10 dilution. However, laboratories should be clear about the convention they use. In this way, there should be adequate sample for multiple quantitative analyses and several qualitative analyses. For example, 1 mL can be removed for plating and for serial dilutions. Serial dilutions can be used across a variety of media. Aliquots of the remaining diluent can be used for multiple enrichments for selected pathogens, subject to vigorous stomaching of the sponge to release cells. The sponge itself can also be enriched. Figure 3-1 below shows an example of this approach in which a laboratory wishes to test for three pathogens (e.g., Salmonella, L. monocytogenes, and a miscellaneous pathogen),

| Microbiological Monitoring of the Food Processing Environment

Figure 3-1. Quantitative and qualitative assays: example of testing approach for multiple organisms per sponge.59 MRS 5 de Man, Rogosa, and Sharpe; PDA 5 potato dextrose agar; TGY 5 tryptone-glucose-yeast extract; VRB 5 violet red bile.

and perform four quantitative assays (e.g., aerobic plate count, yeast and mold count, presumptive coliform count, and acidophile count).

3.84

Petrifilm Aerobic Count Method

The Petrifilm direct-contact method provides a simple means of detecting bacterial contamination on flat surfaces and on curved surfaces. The procedure should not be used for surfaces with cracks or crevices.

3.841 Equipment, Supplies, Solutions, and Media Petrifilm plates are provided by the manufacturer (3M, St. Paul, MN) in sealed foiled pouches. Sealed pouches may be stored at 2uC–4uC until the specified expiration date. Plates must be prehydrated before use. 3.842 Sampling Procedure Prehydrate plates by dispensing 1 mL of sterile dilution water onto the center of the bottom film. If the surface has been treated with sanitizer, incorporate an appropriate neutralizer into the sterile dilution water. Replace the top film down onto the diluent. Distribute the diluent by exerting downward pressure on the center of the plastic spreader. Do not slide the spreader across the film. Remove the spreader and allow 30 minutes for the gel to solidify. To sample the test surface, lift the top film of the prehydrated plate without touching the growth surface. The gel should adhere to the top of the film. Allow the gel and the top of the film to contact the test surface. Firmly rub fingers over the entire film side of the gelled area to ensure good contact with the surface. Lift the film from the surface and rejoin the top and bottom sheets of the plate.

3.843 Incubation and Colony-Counting Procedure Incubate the plates in a horizontal position with the clear side up at 32uC for 48 hours. Count all the red colonies in a 20-cm2 circular growth area. When very high concentrations of colonies on the plate cause the entire growth area to become red or pink, record the plate results as greater than 250. 3.9

MICROBIOLOGICAL AIR-SAMPLING STRATEGIES

In most instances with many nonperishable foods, the quality of the air in a food plant does not directly affect the microbiological safety or the keeping quality. However, air can be an important means by which low numbers of microorganisms can move from high to low pressure areas in a processing facility. If these contaminants land in moist areas, then microbial growth niches may develop. If these moist areas are in areas close to a product, then contamination may result. On the other hand, some perishable products such as fluid dairy products, ready-to-eat meats, and some baked goods are particularly sensitive to airborne contaminants. Environmental air quality, especially in the packaging areas, is a crucial control area for these foods. Aseptically packaged foods may require that the air supplies in packaging rooms have very low microbial loads such as that supplied by air filtered through laminar flow systems. Measurement of the microbial quality of air is useful for assessing the effectiveness of disinfection procedures for air-handling equipment. Microorganisms occur in air as aerosols consisting of single unattached cells or cells in clumps. They can become airborne from environmental sources such as worker activity, sink and floor drains, water spraying, air-conditioning systems that liberate droplets, dust generated from raw | 35

Compendium of Methods for the Microbiological Examination of Foods |

material, and specific food-processing systems. Microorganisms may adhere to a dust particle or may exist as a free-floating particle surrounded by a film of dried organic or inorganic material. Particulates in microbial aerosols may range in size from less than 1–50 mm. Particle size is the major factor influencing aerodynamic behavior. Compared to bacterial and mold spores, vegetative bacteria may be present in fewer numbers in air since they are rapidly injured in dry dust or as moist bioaerosol droplets desiccate. Many vegetative bacterial cells ordinarily will not survive for long in air unless the relative humidity and other factors are favorable or unless the organism is enclosed in some protective matrix. As a rule of thumb, microbial aerosols generated from the environment will be primarily bacterial spores, molds, and yeasts or injured vegetative cells. When personnel are the source of microbial contamination, the primary types are vegetative bacteria—especially staphylococci, streptococci, micrococci, and other organisms associated with the human respiratory tract, hair, and skin. Kornacki and Gurtler61 reported on the unpublished work of Yan and Kornacki that showed aqueous bioaerosols of L. monocytogenes released in a bioaerosol chamber had a settling rate of 1 log10 per hour in quiescent air at 38% and 75% relative humidity, but the bioaerosols appeared to lose viability (3 log10 in 20 minutes) when the air was being recirculated. However, Zhang et al.100 showed no difference in the settling rate of Listeria innocua in quiescent versus mixed air in a food processing environment. Quantitative and qualitative guidelines should be established that relate numbers and types of microorganisms per volume of air to critical levels of product contamination. These guidelines must be established for each plant or process so that data collected in an air sampling program, such as air flow patterns, filtration systems, or personnel density and activity, related to product contamination can be used to make decisions regarding possible sources. Significant increases above an established guideline may indicate a breakdown of standard contamination control barriers. National Aeronautics and Space Administration (NASA) air cleanliness standards 75 (Table 3-1) may be used as a reference point. However, their suitability for application in a particular processing environment will have to be determined experimentally.

3.10

AIR-SAMPLING METHODS2,3,22,25,43,49,73

Viable airborne microorganisms can be determined quantitatively by a variety of methods, including sedimentation,46,69 impaction on solid surfaces,42,62,63,70 filtration,32 centrifugation,76 electrostatic precipitation, impingement in liquids,66 and thermal precipitation. Of these methods, sedimentation and impaction on solid surfaces are most frequently used. Aerosol-sampling methods have been reviewed by Kang and Frank.56–58 Many collecting and culturing media are available for biological aerosol sampling. The selection of nutrient medium will depend on the nutritional requirements of the organisms under study, the type of information desired from the study, the sampling method, and the sampling conditions. When initial collection is in a liquid medium, the microorganisms must remain viable without growth until aliquots are obtained for culture. Some common liquid media used are tryptose saline, buffered gelatin, peptone water, buffered gelatin enriched with brain-heart infusion, buffered saline, and buffered water. These media are also used as diluting fluids to obtain suspensions suitable for plating. Buffered saline and buffered water are used only for collecting spores and other resistant microbial forms. When collection is made directly on solid nutrient medium, a sufficient concentration of agar (1.5%–2.0%) to produce a stable medium that is capable of withstanding the action from a rapidly flowing airstream should be used. Some common solid nutrient media employed for general bacterial air sampling are blood agar, tryptose agar, trypticase soy agar, proteose extract agar, and nutrient agar. These media are also employed for culturing the liquid collecting media by surface-plating methods, the pourplate method, and the membrane filter method. Under certain sampling conditions, it is desirable to incorporate selective agents into a medium to inhibit interfering contaminants. Some commonly used inhibitory agents are crystal violet, brilliant green, potassium tellurite, and cycloheximide. Chemicals should not be used unless preliminary screening has demonstrated that they do inhibit the target organism. Air samplers should be sanitized or sterilized before use. Sieve and filtration-type samplers that have been

Table 3-1. NASA Air Cleanliness Classes75 Class English (Metric) System 100

1,000

10,000

100,000

Test

(3.5)

(35)

(350)

(3500)

Maximum number of 0.5 mm and larger particles per ft3 (per L) Maximum number of 5.0 mm and larger particles per ft3 (per L)

100 (3.5)

1,000 (35)

10,000 (350)

100,000 (3500)

a

a

65 (2.3)

700 (25)

Note: NASA 5 National Aeronautics and Space Administration. NASA Standards for slit sampling of clean rooms and work stations for the microbially controlled environment.75 a Indicates that the results are statistically unreliable, except when a large number of samples are obtained.

36 |

| Microbiological Monitoring of the Food Processing Environment

wrapped in Kraft paper and liquid impingers with cotton plugs inserted in the intake and exhaust ports can be conveniently autoclaved. In actual use, swabbing the sampler with disinfectant before each sampling period is adequate and convenient. Gaseous sterilization techniques can be used to sterilize all samplers. The following air sampling methods are commonly used in environmental microbiology. Six types of commercially available aerosol samplers are impingers, impactors (slit and sieve), filtration samplers, centrifugal samplers, and electrostatic precipitation samplers. It is important to follow the manufacturer’s directions for each sampler and to understand the limitations of each. The methods listed below for air sampling are by no means comprehensive. Laboratory workers should review Public Health Monograph 60 by the U.S. Department of Health, Education, and Welfare (Washington, D.C.) for a detailed discussion of air-sampling principles.48

3.101

Sedimentation Methods

Sedimentation methods are easy to use, inexpensive, and collect particles in their original state. The exposure agar plate and microscopic slide exposure method rely on the force of gravity and air currents to deposit airborne particles on a nonselective or selective agar surface. Results are obtained as CFUs or particles per minute. Particle size distribution may be obtained by direct microscopic observation. The 17th edition of Standard Methods for the Examination of Dairy Products98 recommends 15-minute exposure of standard size Petri plates (90-mm diameter) containing SMA or a selective medium. After exposure, the plates are incubated by the appropriate procedure. In addition, microscope slides coated with agar can be exposed and the particles counted by using a microscope. This technique is only used for total particulate counts. Sedimentation methods have several disadvantages such as their measure of airborne microorganisms quantitatively (i.e., the number of viable particles per cubic unit of air) and their weak correlation with counts obtained by other quantitative methods.78 They are useful only when fallout onto a particular surface is of interest, and they require a relatively long sampling time. Air movement will influence the deposition of the particles. Thus, these methods are heavily biased toward large particles, which would settle more rapidly than smaller particles. Samples may be obtained at (1) openings in equipment subject to potential contamination from organisms transported by air currents, (2) selected points for testing general room air, (3) areas of employee concentration, and (4) process air passages where air is incorporated into products. Because of air turbulence during operating hours, sampling by volumetric methods will be more effective and dependable than sedimentation samples.97

3.102

Impaction Methods

Impaction usually involves collecting microbial aerosols on an agar surface, but dry or coated surfaces may be used for special purposes such as particle size determination. An impactor consists of an air jet that is directed over the impaction plate so that the particles collide with and stick to the surface. Impaction methods give higher particle

recovery than other methods.33,89,90 Impaction results in low sampling stresses after collection, and sample manipulation is not required. Impactors are of two types: slit samplers (e.g., Casella slit sampler) and sieve samplers (e.g., Andersen multistage sieve sampler).

3.1021 Slit Sampler Slit samplers are usually cylindrical and have a slit tube that produces a jet stream when the air is sampled by vacuum. Beneath the slit is a platform that accepts a culture plate and that is rotated by a clock mechanism. The rate of the plate rotation may be varied. These samplers require a vacuum source sufficient to draw a constant flow of air through the sampler, usually 28.3 L (1 ft3) per minute; however, the air flow may be changed by altering the dimensions of the slit. Some of the common characteristics of slit samplers are relatively high collection efficiency, fabrication from metal, ruggedness, portability, simplicity of operation, and relatively high sampling volume. Some slit samplers cannot be sterilized by autoclaving. Gaseous sterilization is desirable, but swabbing with disinfectant is often sufficient. Samplers employing agar are limited to use in temperatures above 0uC, unless some method of heating is provided to avoid freezing the medium. Slit samplers do not discriminate for size of airborne particles and can be used to detect bursts of contamination associated with specific activities at certain times. 3.1022 Sieve Samplers Sieve samplers are operated by drawing air through a large number of small evenly placed holes drilled in a metal plate (i.e., sieve). The suspended particles impact an agar surface located a few millimeters below the perforated plate. There are single stage (e.g., Ross-Microban) and multistage (e.g., Andersen) sieve samplers. A multistage sieve sampler consists of a series of two, six, or eight stacked sieves and plates, each with successively smaller holes. This arrangement causes increased particle velocity as air flows through the apparatus. Large particles impact at the initial stage, and small particles follow the air flow until accelerated sufficiently to impact at a later stage. The commonly used Andersen six-stage sampler consists of sieves with holes ranging from 1.81–0.25 mm. The distance of the agar collecting surface from the sieve, which is critical, is controlled by utilizing a special Petri dish containing 27 mL of medium. However, conventional prefilled disposable Petri plates can be used with newer designs. Air is drawn successively through each of the sieves at increasing velocities so that larger airborne particles (i.e., . 7 mm) impact the medium in the first stage and smaller particles, depending on their sizes and inertia, impact the medium in the later stages. The optimum flow rate is 28.3 L per minute. After sampling, the plates are removed and incubated. Some models have only two stages, which are designed to differentiate nonrespirable particles ($ 5 mm) from respirable particles (, 5 mm). Some units have a single stage, which does not differentiate particle size. The multistage sieve samplers are used to detect the number of viable particles per unit volume of air during a prescribed sampling time and to yield a size profile of the particulate in the microbial aerosol. This information is usually much more important in health care settings than in food processing | 37

Compendium of Methods for the Microbiological Examination of Foods |

environments. As with the slit sampler, no diluting or plating procedures are required. Final assay results are expressed as particles per unit volume. Associated with sieve samplers are the following limitations: Multistage sieve samplers are cumbersome to handle and are relatively expensive in comparison to other samplers. The exact volume of agar must be poured into all plates so that the gap between the sieve and the agar surface meets the manufacturer’s specifications. The inside of the sampler and the outside of the prepoured agar plates should remain sterile until sampling since they can contribute to contamination.

3.103

Centrifugal Samplers

Centrifugal force can be used to propel aerosol particles onto a collection surface. When the aerosol is spun in a circular path at high velocity, the suspended particles impact the collecting surface with a force proportional to the particles’ velocity and mass. Centrifugal samplers do not generate high-velocity jet flow during sampling; therefore, less stress is imposed on airborne microbes, compared with impingement and impaction methods. Centrifugal samplers are simple and easy to operate and may be less expensive than impactor types. Centrifugal samplers can generally rapidly sample a high volume of air, resulting in a more representative sampling. Assay results are expressed as particles per unit volume of air (e.g., CFU/L). Limitations of some centrifugal samplers are associated with their failure to generate sufficient centrifugal force to propel small particles onto the collection surface. The recovery efficiency of these samplers depends on the particle size being sampled and the amount of centrifugal force generated. The Reuter centrifugal air sampler (RCS sampler; Biotest Diagnostics; Denville, NJ) is battery operated, portable, lightweight (2.5 lb), and convenient to use.17 A plastic strip containing a culture medium lines the impeller drum. Air from a distance of at least 40 cm is sucked into the sampler by an impeller. Air enters the impeller drum concentrically from a conical sampling area. The impeller is set in rotation and the aerosol is impacted by centrifugal force onto the agar surface. Air then leaves the sampling drum in a spiral outside the cone of entering air. After the sample has been obtained, the agar strips are incubated and the colonies counted. The sampler has a self-timer for sampling from 30 seconds to 88 minutes. The actual sampling rate is 280 L per minute. However, the manufacturer has published an effective sampling rate or separation volume of 40 L per minute for 4 mm particles, a value that is derived from an attempt to reconcile the actual number of viable particles collected from an air sample with measurements involving airflow direction, air velocity, and available collecting surface area. Clark and Lidwell17 indicate that the effective sampling volume of the RCS sampler will vary widely, depending on the aerosol particle size. The results obtained by using this sampler must consequently be interpreted with considerable caution. Macher and First71 measured the collection efficiency with increasing particle size. Particles larger than 15 mm are almost 100% collected, particles ranging from 4–6 mm are collected at 55%–75% efficiency, and particles smaller than 1 mm pass through the sampler without significant retention. Although the RCS sampler 38 |

does not accurately estimate the total viable particle concentration, Placencia and Oxborrow76 recommended this sampler for investigations of good manufacturing practices. These investigators found that the RCS sampler will collect more viable particles compared with a slit sampler and can detect the difference in the environmental quality of each medical device manufacturing facility tested. In addition, the RCS sampler effectively detects various types of microorganisms.74

3.104

Filtration Methods

Filters are widely used for aerosol sampling because of their low cost and simplicity of operation. The air filtration apparatus consists of cellulose fiber, sodium alginate, glass fiber, gelatin membrane filter (pore size 3 mm) or synthetic membrane filters (pore size 0.45 mm or 0.22 mm) mounted in an appropriate holder and connected to a vacuum source through a flow rate controller (e.g., the critical orifice). After a fiber filter is used, the whole filter or a section of it is agitated in a suitable liquid until the particles are uniformly dispersed. Aliquots of the suspension are then assayed by appropriate microbiological techniques. Membrane filters can be either treated similarly to fiber filters or placed directly on an agar surface and incubated. The gelatin membrane is water soluble so that it can easily be diluted for plating or be solubilized on top of a nutrient medium, thereby resulting in microbial colonies that are easily counted. The hygroscopicity of the gelatin membranes causes difficulty in sampling because of the swelling of the membrane when the relative humidity exceeds 90%.82 The large number of pores in these membranes allows a large volume of air to be sampled during a short time (2.7 L of air per minute per cm2 per 500-mm water column). The technique is effective in certain types of environments,33 although some investigators have cautioned against the drying of the vegetative bacteria on the membrane filter and the consequent difficulty in recovery. Fields and coworkers have shown that recovery rates between membrane filter techniques and slit samplers are comparable for naturally occurring airborne microorganisms that have already survived drying.32,33 Filtration methods are good for enumerating mold or bacterial spores, but they may not be effective for counting vegetative cells because of the stress of dehydration produced during sampling.31 The shorter sampling times used in gelatin membrane filtration may reduce this stress. Filtration methods do not discriminate between particle size.

3.105

Impingement Methods

Impingement methods use a liquid to collect microorganisms from air. When air is dispersed through the liquid, particles in the air are trapped. Quantification of airborne microorganisms is accomplished by plating the collection fluid or by using a membrane filtration plating technique when the expected microbial level load is low. Liquid impingers can be either low-velocity or highvelocity samplers. Low-velocity samplers utilize the air washing principle: airborne particles entering the sampler at low velocity through a large jet, fitted glass dish, or perforated tube are bubbled through and trapped in the liquid collecting medium. Small particles (i.e., less than 5 mm) are

| Microbiological Monitoring of the Food Processing Environment

not efficiently trapped in low-velocity samplers; they remain in air bubbles and are carried out with discharged air. High-velocity samplers draw air through a small jet and direct it against a liquid surface. While these samplers efficiently collect all particle sizes with a diameter greater than 1 mm, the high velocity tends to destroy some vegetative cells. High-velocity collection disperses clumps of cells, thereby producing counts that may be higher than the counts obtained by gentler collection methods. A suitable collecting medium for liquid impingement samplers must preserve the viability of microorganisms while inhibiting their multiplication. The more common collecting media include buffered gelatin, tryptose saline, peptone water, and nutrient broth. Use of an antifoam agent in the collecting medium is suggested if excessive foaming occurs. Acceptable agents are Dow Corning AntiFoam A (Dow Corning, Midland, MI), General Electric Anti-Foam 60 (GE, Trevose, PA), and olive oil. With extended sampling, air impact has a cooling effect on the liquid. If the ambient temperature is 40uF, the collecting liquid is likely to freeze. Use of a low-freezing-point diluent such as glycerol or some means of temperature control is necessary in such a situation. After sampling, an aliquot of the collecting liquid is plated and incubated in a growth medium to obtain a viable count. In quantitative studies, the total air flow must be measured to calculate microorganisms per volume of air. The volume of collecting fluid must also be measured to determine the number of cells collected. This method is not suited to low concentrations of airborne microorganisms. The All-Glass Impinger sampler (AGI-30, Ace Glass, Vineland, NJ) is a high velocity impinger widely used for air sample collection. The jet is held 30 mm above the impinger base and consists of a short piece of capillary tube designed to reduce cell injury. The AGI-30 sampler operates by drawing aerosols through an inlet tube and is curved to simulate the nasal passage.20 This makes it especially useful for studying the respiratory infection potential of airborne microorganisms. The usual sampling rate is 12.5 L per minute. When it is used for recovering total airborne microorganisms from the environment, the curved inlet tube should be washed with a known amount of collecting fluid after sampling since larger particles (i.e., over 15 mm in diameter) collect on the tube wall by inertial force. The glass impinger is relatively inexpensive and simple to operate, but viability loss may result from the amount of shear force involved in the collection. The air stream approaches sonic velocity when particulates impinge on the collection fluid, resulting in almost the complete collection of suspended particles; however, this condition tends to cause the destruction of vegetative cells3 or may result in overestimation because of the dispersion of dust particles and the breaking up of clumps of bacteria.78 Other constraints are that the glassware should be sterilized before each sampling and that the apparatus should be easily broken.

3.106

Electrostatic Precipitation

Electrostatic precipitation samplers impart a uniform electrostatic charge to incoming airborne particles, which are then collected on an oppositely charged rotating disc. A

known volume of air at a given rate is sampled. Electrostatic precipitators may employ a variety of solid collecting surfaces such as glass or agar. A liquid collecting medium with added wetting agent, to aid in uniform distribution, can also be used to wash the collected particles centrifugally into a collecting vessel. These precipitators can sample at a relatively high rate (# 1,000 L/minute) with a high collection efficiency and low resistance to air flow, although they are complex and must be handled carefully. Furthermore, little is known about the effect of electrostatically charged particles on viability and clumping. During ionization of the air sample, oxides of nitrogen and ozone are produced that may be toxic to microorganisms. Several electrostatic precipitators are manufactured specifically for sampling microbial aerosols, although they are not widely used for this purpose.85

3.107

Comparison Studies on Aerosol Samplers

Comparison studies of air-sampling devices indicate that the choice of the correct sampler to use is seldom obvious. A multistage sieve sampler such as the Andersen multistage sieve may be the most efficient sample at viable particle recovery, but it may not be suitable for routine sampling. It also requires a vacuum source. Filter samplers work well for quality control monitoring of molds85 and bacterial spores; however, bacterial recovery efficiency may be less, depending on the extent of dehydration that occurs during sampling.14 In addition, a vacuum source is required. The RCS sampler is convenient to use, creates its own air flow, and recovers bacteria and molds. The RCS sampler does not recover the smallest viable particles, even though it is useful for determining relative air quality on a routine basis.23,76,77 Slit samplers may not be as convenient to use as the RCS sampler, especially if a vacuum source is required. However, slit samplers are more efficient at recovering small particles.

3.11 3.111

ALTERNATIVE METHODS DNA-Based Probe for Mold Bioaerosol

Microbiological monitoring of the food processing environment for mold has generally been of secondary importance. This is primarily because of the time required to obtain results when using traditional plating methods. The traditional method is also based on microscopic observation of mold structures for identification and quantification. In many food products, bacterial growth by nonpathogenic microorganisms also results in product spoilage; the product is discarded before visible evidence of mold spoilage becomes apparent. However, there are foods of lower water activity that may support mold growth while preventing bacterial multiplication. In such products, environmental sampling for mold presence and quantity would indicate the effectiveness of the cleaning sanitation program and the potential hazard for the product. The development of a DNA-based analysis, moldspecific quantitative polymerase chain reaction (MSQPCR), has provided food processing facilities with an alternative over traditional methods of sampling, identification, and quantification.94 The method was developed to address residential indoor air quality concerns, but it is also | 39

Compendium of Methods for the Microbiological Examination of Foods |

applicable to food manufacturing facilities. The DNA sequences unique to each mold species have been determined and are used in the identification of the mold, after quantitative polymerase chain reaction (qPCR) amplification. Environmental Protection Agency (EPA) scientists have designed and tested probes and primers for more than 100 molds and have designated and patented the resulting technology MSQPCR.94 Dust or air samples are collected, filtered, and placed in an extraction tube; spiked with an external reference; and extracted by a rapid mechanical bead-milling method at 5000 rpm for 1 minute. The DNA is purified with a commercial kit (for 30 minutes). The MSQPCR are species specific. The analysis can be performed on any of several DNA sequencers and microbial concentrations can be determined in 20–40 minutes. The results of assays are compared to standard curves generated from spore suspensions of a known concentration of the target mold. Assays are sensitive to a single spore or a few spores per sample. The EPA has licensed this technology to many of the large commercial laboratories including TestAmerica, EMSL, Mycometrics, Forensics Analytical, and Amtek. If a food processor wanted to bring the technology in-house, they would need to have a commercially available DNA sequencer.

3.12 3.121

AEROSOL SAMPLING AND MEASUREMENT GUIDELINES Standard Methods for Examining Dairy Products

The 17th edition of the Standard Methods for the Examination of Dairy Products98 lists no Class A standard method for testing the microbiological quality of air and dairy environments, although there are methods designated as Class D and Class B. Favero et al.31 introduced air-sampling strategies and various air-sampling methods in a previous edition of this compendium. They indicated that the first and most important decision is determining whether air sampling at any level is required. If it is, then quantitative and qualitative guidelines should be established that relate numbers and types of microorganisms per volume of air to critical levels of product contamination.

3.122

NASA Air Cleanliness Standards

Favero et al.31 also suggest using the NASA air cleanliness standards as a reference point after experiments to determine suitability. The NASA Contamination Control Requirements Manual75 defines four air cleanliness classes (Table 3-1). According to the standards, the collection methods must conform to Handbook for the Microbial Examination of Space Hardware,75a (NHB 5340.1 or revisions thereof), which specifies the use of a slit sampler.

3.123

Federal Standard 209 C41

Federal standard 209C (Clean Rooms and Work Station Requirements, Controlled Environment) established standard classes of air cleanliness for airborne particulate levels in clean rooms and clean zones. These classes were based only on particle enumeration and place more 40 |

emphasis on small particles that are not necessarily viable.43 This standard consequently was not useful for food plant applications.

3.124

ISO 14644-1 Clean Rooms and Associated Controlled Environments: Classification of Air Cleanliness54

The International Organization for Standardization (ISO) has published the official document on classification of air cleanliness for clean rooms and associated controlled environments. This document replaces the Federal standard 209C, which is no longer in effect.41 Table 3-1 in the ISO document lists the 9 ISO classes and corresponding maximum concentration limits for sizes from 0.1–5 mm. These classes are based only on particle enumeration and place more emphasis on small particles that are not necessarily viable. This consequently is not useful for food plant applications.

3.125

Standard Reference Samplers

Brachman et al.10 recommend the AGI-30 sampler as a standard reference sampler because of its historical use, economics, availability, and simple design. On the other hand, the American Conference of Governmental Industrial Hygienists Committee on Bioaerosols used the Andersen multistage air sampler as the reference sampler for its committee activities and reports.3 In the pharmaceutical industry, the slit sampler is the most widely used device for monitoring sterile manufacturing and quality control environments.1

ACKNOWLEDGMENT Fourth edition authors: George M. Evancho, William H. Sveum, Lloyd J. Moberg, and Joseph F. Frank.

REFERENCES 1. Akers, M. J. 1985. Sterility testing. In Parenteral Quality Control. Marcel Dekker, New York, NY. 1. 2. American Conference of Governmental Industrial Hygienists (ACGIH). 1978. Air sampling instruments for evaluation of atmospheric contaminants, 5th ed. Cincinnati, OH. 3. American Conference of Governmental Industrial Hygienists (ACGIH). 1986. Committee on Bioaerosols. ACGIH committee activities and reports. Appl. Ind. Hyg. 1:R19. 4. American Meat Institute (AMI). 1988. Interim guideline: microbial control during production of ready-to-eat meat products. Controlling the incidence of Listeria monocytogenes. 2nd ed. Washington, D.C. 5. Angelotti, R. J., L. Wilson, W. Litsky, and W. G. Walter. 1964. Comparative evaluation of the cotton swab and RODAC methods for the recovery of Bacillus subtilis spore contamination from stainless steel surfaces. Health Lab. Sci. 1:289296. 6. Awad, A. H. A. 2007. Airborne dust, bacteria, actinomycetes and fungi at a flour mill. Aerobiologia 23:59-69. 7. Behling, R. G., J. Eifert, M. C. Erickson, J. B. Gurtler, J. L. Kornacki, E. Line, R. Radcliff, E. T. Ryser, B. Stawick, and Z. Yan. 2010. Selected pathogens of concern to industrial food processors: infectious, toxigenic, toxico-infectious, selected emerging pathogenic bacteria. In Principles of Microbiological Troubleshooting in the Industrial Food

| Microbiological Monitoring of the Food Processing Environment

8.

9.

10.

11.

12.

13.

14.

15.

16.

17. 18.

19.

20. 21.

22.

23. 24.

25.

26.

Processing Environment. J. L. Kornacki, (ed.), Springer, New York, NY. 5–61 Beresford, M. R., P. W. Andrew, and G. Sharma. 2001. Listeria monocytogenes adherence to many materials found in food processing environments. J. Appl. Microbiol. 90:1000-1005. Blackman, I. C., and J. F. Frank. 1996. Growth of Listeria monocytogenes as a biofilm on various food-processing surfaces. J. Food Prot. 59:827-831. Brachman, P. S., R. Ehrlich, H. R. Eichenwald, V. J. Gabelli, T. W. Kethley, S. H. Madin, J. R. Maltman, G. Middlebrook, J. D. Morton, I. H. Silver, and E. K. Wolfe. 1964. Standard sampler for assay of airborne microorganisms. Science 144:1295. Buchbinder, L., T. C. Buck Jr., P. M. Phelps, R. V. Stone, and W. D. Tiedeman. 1947. Investigations of the swab rinse technique for examining eating and drinking utensils. Am. J. Public Health 37:373-378. Buck, T. C. Jr., and E. A. Kaplan. 1944. A sterile cutting device for swab vial outfits utilizing wood applicators. J. Milk Technol. 7:141-142. Charlton, B. R., H. Kinde, and L. H. Jensen. 1990. Environmental survey for Listeria species in California milk processing plants. J. Food Prot. 53:198-201. Chatigny, M. A. 1978. Sampling airborne microorganisms, p. E1. In Air sampling Instruments for Evaluation of Atmospheric Contaminants. 5th ed., American Conference of Governmental Industrial Hygienists, Cincinnati, OH. Chen, Y., V. N. Scott, T. A. Freier, J. Kuehm, M. Moorman, J. Meyer, T. Morille-Hinds, L. Post, L. Smoot, S. Hood, J. Shebuski, and J. Banks. 2009. Control of Salmonella in lowmoisture foods II: hygiene practices to minimize Salmonella contamination and growth. Food Prot. Trends 29:435-445. Chen Y., V. N. Scott, T. A. Freier, J. Kuehm, M. Moorman, J. Meyer, T. Morille-Hinds, L. Post, L. Smoot, S. Hood, J. Shebuski, and J. Banks. 2009. Control of Salmonella in lowmoisture foods III: process validation and environmental monitoring. Food Prot. Trends. 29:493-508. Clark, S., and O. M. Lidwell. 1981. The performance of the Biotest RCS centrifugal air sampler. J. Hosp. Infect. 2:181. Cordray, J. C., and D. L. Huffman. 1985. Comparison of three methods for estimating surface bacteria on pork carcasses. J. Food Prot. 48:582-584. Costa, P. D., N. J. Andrade, S. C. C. Branda˜o, and F. J. V. Passos. 2006. ATP Bioluminescence assay as an alternative for hygiene-monitoring procedures of stainless steel milk contact surfaces. Brazilian J. Microbiol. 37:345-349. Cox, C. S. 1987. The aerobiological pathway of microorganisms. John Wiley & Sons, New York, NY. Cox, L. J., T. Kleiss, J. L. Cordier, C. Cordellana, P. Konkel, C. Pedrazzini, R. Beumer, and A. Siebenga. 1989. Listeria spp. in food processing, non-food and domestic environments. Food Microbiol. 6:49-61. Curtis, S. E., R. K. Balsbaugh, and J. G. Drummond. 1978. Comparison of Andersen eight-stage and two-stage viable air samplers. Appl. Environ. Microbiol. 35:208-209. Delmore, R. P., and W. N. Thompson. 1981. A comparison of air-sampler efficiencies. Med. Device Diagn. Ind. 3:45-48, 53. Devenish, J. A., B. W. Ciebin, and M. H. Brodsky. 1985. Evaluation of Millipore swab-membrane filter kits. J. Food Prot. 48:870-874, 878. Dimmick, R. L., and A. B. Akers. 1969. An introduction to experimental aerobiology. Wiley-Interscience, New York, NY. Dutkiewicz. J., E. Krysinska-Traczyk, C. Skorska, G. Cholewa, and J. Sitkowska. 2002. Exposure to airborne microorganisms and endotoxins in a potato processing plant. Ann. Agric. Environ. Med. 9:225-235.

27. Dutkiewicz, J., E. Krysinska-Traczyk, C. Skorska, J. Sitkowska, Z. Prazmo, and M. Golec. 2001. Exposure to airborne microorganisms and endotoxins in herb processing plants. Ann. Agric. Environ. Med. 8:201-211. 28. Eifert, J. D., and F. M. Arritt. 2002. Evaluation of food processor environmental sampling data and sampling plans. Dairy, Food and Environmental Sanitation 22:333-339. 29. Elliott, R. P. 1980. The microbiology of sanitation. In Principles of Food Processing Sanitation. A. M. Katsuyama and J. P. Strachan (eds.), The Food Processors Institute, Washington, D.C. 35–60. 30. Engley, F. B., and B. P. Dey. 1970. A universal neutralizing medium for antimicrobial chemicals. In Proceedings of the 56th Meeting of the Chemical Specialties Manufacturers Association. New York, NY. 31. Favero, M. S., D. A. Gabis, and D. Vesley. 1984. Environmental monitoring procedures. In Compendium of Methods for the Microbiological Examination of Foods. 2nd ed. M. L. Speck (ed.), American Public Health Association, Washington, D.C. 47. 32. Fields, N. D., G. S. Oxborrow, C. M. Herring, and J. R. Puleo. 1973. An evaluation of two microbiological air samplers, abstract E11, In Abstracts of the Annual Meeting of the American Society for Microbiology. Washington, D.C. 2. 33. Fields, N. D., G. S. Oxborrow, J. R. Puleo, and C. M. Herring. 1974. Evaluation of membrane filter field monitors for microbiological air sampling. Appl. Microbiol. 127:517-520. 34. Food and Drug Administration (FDA). 1998. Bacteriological analytical manual. 8th ed., Revision A. FDA, Washington, D.C. 35. Food and Drug Administration (FDA). 2011 Revision. Grade A pasteurized milk ordinance department of health and human services. U.S. Government Printing Office, Washington, D.C. 36. Food and Drug Administration (FDA) and Milk Industry Foundation/International Ice Cream Association. 1988. Recommended guidelines for controlling environmental contamination in dairy plants. Dairy Food Sanit. 8:52-56. 37. Frank, J. F., and R. A. N. Chmielewski. 1997. Effectiveness of sanitation with quaternary ammonium compound or chlorine on stainless steel and other domestic food-preparation surfaces. J. Food Prot. 60:43-47. 38. Frank, J. F., and R. A. Koffi. 1990. Surface-adherent growth of Listeria monocytogenes is associated with increased resistance to surfactant sanitizers and heat. J. Food Prot. 53:550-554. 39. Gabis, D., and R. E. Faust. 1988. Controlling microbial growth in food processing environments. Food Technol. 42:81-82, 89. 40. Gabis, D. A., R. S. Flowers, D. Evanson, and R. E. Faust. 1989. A survey of 18 dry dairy product processing plant environments for Salmonella, Listeria, and Yersinia. J. Food Prot. 52:122-124. 41. General Services Administration (GSA). 1987. Federal standard 209C. Clean room and work station requirements, controlled environment. Federal Supply Service, U.S. Government Printing Office, Washington, D.C. 42. Greene, V. W., D. Vesley, R. G. Bond, and G. S. Michaelsen. 1962. Microbiological contamination of hospital air. I. Quantitative studies. Appl. Microbiol. 10:561-566. 43. Gregory, P. H. 1973. Air sampling technique. In The Microbiology of the Atmosphere. 2nd ed., John Wiley & Sons, New York, NY. 126. 44. Griffith, C. 2005. Improving surface sampling and detection of contamination. In Handbook of Hygiene Control in the Food Industry. H. L. Lelieveld, M. Mosterte and J. Holah. Woodhead Publishing, Cambridge, UK. 588–618. 45. Gudbjornsdottir, B., M.-L. Suihko, P. Gustavsson, G. Thorkelsson, S. Salo, A.-M. Sjoberg, O. Niclasen, and S. Bredholt. 2004. The incidence of Listeria monocytogenes in

| 41

Compendium of Methods for the Microbiological Examination of Foods |

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

42 |

meat, poultry and seafood plants in the Nordic countries. Food Micro. 21:217-225. Hall, L. B., and H. M. Decker. 1960. IV. Procedures applicable to sampling of the environment for hospital use. Am. J. Public Health 50:491-496. Hall, L. B., and M. J. Hartnett. 1964. Measurement of the bacterial contamination on surfaces in hospitals. Public Health Rep. 79:1021-1024. Health, Education, and Welfare (HEW). 1959. Sampling microbiological aerosols. Public health monograph 60, Public health services publication no. 686. Department of Health, Education, and Welfare, U.S. Government Printing Office, Washington, D.C. Heldman, D. R., and T. I. Hedrick. 1971. Air-borne contamination control in food processing plants. Res. Bull. 33. Mich. State Univ. Agric. Exp. Sta., East Lansing, MI. Hennessey, W. H., C. W. Hedberg, L. Slutsker, K. E. White, J. M. Besser-Wiek, M. E. Moen, J. Feldman, W. W. Coleman, L. M. Edmonson, K. L. MacDonald, and M. T. Osterholm. 1996. A national outbreak of Salmonella enteritidis infections from ice cream. New Engl. J. Med. 334:1281-1286. Holah, J. T., and R. H. Thorpe. 1990. Cleanability in relation to bacterial retention on unused and abraded domestic sink materials. J. Appl. Bacteriol. 69:599-608. International Commission on Microbiological Specifications for Foods (ICMSF). 1988. Cleaning and disinfecting. In ‘‘Microorganisms in Foods 4: Application of the Hazard Analysis Critical Control Point (HACCP) System to Ensure Microbiological Safety and Quality.’’ pp. 93-116, Blackwell Scientific Publications, Palo Alto, CA. International Commission on Microbiological Specifications for Foods (ICMSF). 2002. Sampling to assess control of the environment. In Microorganisms in Foods 7: Microbiological Testing in Food Safety Management. Kluwer Academic Publishers, New York, NY. 199–224. International Organization for Standardization (ISO). 1999. ISO-14644-1. Cleanrooms and associated controlled environments. Part 1: classification of air cleanliness. International Organization for Standardization. Geneva, Switzerland. Jeong, D. K., and J. F. Frank. 1994. Growth of Listeria monocytogenes at 10uC in biofilms with microorganisms isolated from meat and dairy processing environments. J. Food Prot. 57:576-586. Kang, Y. J., and J. F. Frank. 1989. Biological aerosols: a review of airborne contamination and its measurement in dairy processing plants. J. Food Prot. 52:512-524. Kang, Y. J., and J. F. Frank. 1989. Comparison of airborne microflora collected by the Andersen sieve sampler and the RCS sampler in a dairy processing plant. J. Food Prot. 52:877-880. Kang, Y. J., and J. F. Frank. 1989. Evaluation of air samplers for recovery of biological aerosols in dairy processing plants. J. Food Prot. 52:655-659. Kornacki, J. L. (ed.). 2010. Principles of Microbiological Troubleshooting in the Industrial Food Processing Environment. Springer, New York, NY. Kornacki, J. L. 2011. Practical sampling plans, indicator microorganisms, and interpretation of test results from trouble-shooting, In Rapid Detection, Characterization and Enumeration of Foodborne Pathogens. J. Hoorfar (ed.), ASM Press, Washington, D.C. 373–379. Kornacki, J. L., and J. B. Gurtler. 2007. Incidence and control of Listeria in food processing facilities. In Listeria, Listeriosis and Food Safety. E. T. Ryser and E. H. Marth (eds.), CRC Press, Boca Raton, FL. 681–766. Kraidman, G. 1975. The microbiology of airborne contamination and air sampling. Drug Cosmet. Ind. 116:40-43.

63. Krysinski, E. P., L. J. Brown, and T. J. Marchisell. 1992. Effect of cleaners and sanitizers on Listeria monocytogenes attached to product contact surfaces. J. Food Prot. 55:246-251. 64. Lee, S., F. Cetinkaya, and G. E. Soyutemiz. 2009. Occurrence of Listeria spp. in the processing stages of frozen pepper. J. Food Safety 27:134-147. 65. Lee, S-H., and J. F. Frank. 1991. Inactivation of surfaceadherent Listeria monocytogenes hypochlorite and heat. J. Food Prot. 54:4-6, 11. 66. Lembke, L. L., R. N. Kniseley, R. C. Van Nostrand, and M. D. Hale. 1981. Precision of the All-glass Impinger and the Andersen Microbial Impactor for Air Sampling in Solid-waste Handling Facilities. Appl. Environ. Microbiol. 42:222-225. 67. Linton, R. H., W. G. Eisel, and P. M. Muriana. 1997. Comparison of conventional plating methods and PetriFilm (3M Medical-Surgical Division) for the recovery of microorganisms in a ground beef processing facility. J. Food Prot. 60:1084-1088. 68. Llabres, C. M., and B. E. Rose. 1989. Antibacterial properties of retail sponges. J. Food Prot. 52:49-50, 54. 69. Loughhead, H. O, and J. A. Moffett. 1971. Air-sampling techniques for monitoring microbiological contamination. Bull. Parenter. Drug Assoc. 25:261. 70. Lundholm, I. M. 1982. Comparison of methods for quantitative determinations of airborne bacteria and evaluation of total viable counts. Appl. Environ. Microbiol. 44:179-183. 71. Macher, J. M., and M. W. First. 1983. Reuter centrifugal air sampler: measurement of effective air flow rate and collection efficiency. Appl. Environ. Microbiol. 45:1960-1962. 72. Marshall, R. T. (ed.) 1993. Standard Methods for the Examination of Dairy Products. 16th ed. American Public Health Association, Washington D.C. 73. May, K. R. 1967. Physical aspects of sampling airborne microbes, In Airborne Microbes, 17th Symposium of the Society for General Microbiology. Cambridge University Press, New York, NY. 60-80. 74. McGoldrick, K. F., T. L. Fox, and J. S. McAllister. 1986. Evaluation of a dry medium for detecting contamination on surfaces. Food Technol. 40:77-80. 75. National Aeronautics and Space Administration (NASA). 2000. Contamination control requirements manual. Revision D. U.S. Government Printing Office, Washington, D.C. Available at http://paso.esa.int/5_training_materials/training_07_contam%20control.pdf. Accessed December 22, 2013. 75a. National Aeronautics and Space Administration (NASA).. 2010. Handbook for the Microbial Examination of Space Hardware. NASA-HDBK-6022. NASA, Washington, D.C. 76. Placencia, A. M., and G. S. Oxborrow. 1984. Technical Report. Use of the Reuter centrifugal air sampler in good manufacturing practices investigations. U.S. Food and Drug Administration, Sterility Research Center, Minneapolis Center for Microbiological Investigations, Minneapolis, MN. 77. Placencia, A. M., J. T. Peeler, G. S. Oxborrow, and J. W. Danielson. 1982. Comparison of bacterial recovery by Reuter centrifugal air sampler and slit-to-agar sampler. Appl. Environ. Microbiol. 44:512-513. 78. Radmore, K., and H. Luck. 1984. Microbial contamination of dairy factory air. S. Afr. J. Dairy Technol. 16:119. 79. Reij, M. W., and E. D. Den Antrekker. 2004. Recontamination as a source of pathogens in processed foods. Int. J. Food Microbiol. 91:1-11. 80. Reis, A. A., S. Zaza, C. Langkop, R. V. Tauxe, and P. A. Blake. 1990. A multistate outbreak of Salmonella Chester linked to imported cantaloupe. Abstract. Interscience Conference on Antimicrobial Agents and Chemotherapy. American Society for Microbiology, Washington, D.C. 238.

| Microbiological Monitoring of the Food Processing Environment

81. Scallan, R. R., M. Hoekstra, F. J. Angulo, R. V. Tauxe, M-A. Widdowson, S. L. Roy, J. L. Jones, and P. M. Griffin. 2011. Foodborne illness acquired in the United States—Major pathogens. Emerg. Infect. Dis. 17:7-15. 82. Scheurrman, E. A. 1972. The gelatin membrane filter method for the determination of airborne bacteria. Pharm. Ind. 34:756. 83. Scott, V. N., Y. Chen, T. A. Freier, J. Kuehm, M. Moorman, J. Meyer, T. Morille-Hinds, L. Post, L. Smoot, S. Hood, J. Shebuski, and J. Banks. 2009. Control of Salmonella in low-moisture foods I: minimizing entry of Salmonella into a processing facility. Food Prot. Trends 29:342-354. 84. Shale, K., and J. F. R. Lues. 2007. The etiology of bioaerosols in food environments. Food Rev Int. 23:73-90. 85. Silas, J. C., M. A. Harrison, J. A. Carpenter, and J. B. Floyd. 1986. Comparison of particulate air samplers for detection of airborne Aspergillus flavus spores. J. Food Prot. 49:236-238. 86. Silliker, J. H., and D. A. Gabis. 1975. A cellulose sponge sampling technique for surfaces. J. Milk Food Technol. 38:504. 87. Simoes, M., L. C. Simoes, and M. J. Vieira. 2010. A review of current and emergent biofilm control strategies. LWT–Food Sci. Tech. 43:573-583. 88. Stinson, C. G., and N. P. Tiwari. 1978. Evaluation of quick bacterial count methods for assessment of food plant sanitation. J. Food Prot. 41:269-271. 89. Sullivan, J. J. 1979. Air microbiology and dairy processing. Aust. J. Dairy Technol. 34:133-138. 90. Sunga, F. C. A., D. R. Heldman, and T. I. Hedrick. 1966. Characteristics of airborne microorganism populations in packaging areas of a dairy plant. Mich. Agric. Exp. Stn. Q. Bull. 49:155-163. 91. Tiedman, W. D. (chair). 1948. Technique for the bacteriological examination of food utensils. Committee report, In American Journal of Public Health Yearbook 1947-48

92.

93.

94.

95.

96.

97. 98.

99.

100.

(Part 2). American Public Health Association, Washington, D.C. U.S. Department of Health, Education, and Welfare (HEW). 1967. Procedure for the bacteriological examination of food utensils and/or food equipment surfaces. Public health service publication, no. 1631. Technical Information Bulletin, No. 1. HEW, Washington, D.C. U.S. Food and Drug Administration (FDA). 2013. CFR. Title 21. Part 129, Processing and bottling of bottled drinking water, FDA, Silver Spring, MD. Vesper, S. 2011. Traditional mould analysis compared to a DNA-based method of mould analysis. Crit. Rev. Microbiol. 37:15-24. Vij, V., E. Ailes, C. Wolyniak, F. J. Angulo, and K. C. Klontz. 2006. Recalls of spices due to bacterial contamination monitored by the U.S. Food and Drug Administration: the predominance of salmonellae. J. Food Prot. 69:233-237. Vorst, K. L., E. C. D. Todd, and E. T. Ryser. 2004. Improved quantitative recovery of Listeria monocytogenes from stainless steel surfaces using a one-ply composite tissue. J. Food Prot. 67:2212-2217. Walter, W. G., and J. Potter. 1963. Bacteriological field studies on eating utensils and flat surfaces. J. Environ. Health 26:187. Wehr, M., and J. F. Frank (ed.). 2004. Standard Methods for the Examination of Dairy Products. 17th ed. American Public Health Association, Washington, D.C. Yan, Z., K. L. Vorst, L. Zhang, and E. T. Ryser. 2007. Use of one-ply composite tissues in an automated optical assay for recovery of Listeria from food contact surfaces and poultry-processing environments. J. Food Prot. 70:1263-1266. Zhang, G., L. Ma, O. A. Oyarzabeal, and M. P. Doyle. 2007. Aerosol studies with Listeria innocua and Listeria monocytogenes. J. Food Prot. 70:1857-1865.

| 43

|

CHAPTER 4

|

Microscopic Methods Byron Brehm-Stecher and Mary Lou Tortorello

4.1

INTRODUCTION

The microscope is the only scientific instrument that defines a biological group, that is, the microorganisms, which are organisms too small to be seen by the unaided human eye. Renowned for its role in the discovery of the microbial world in the 17th century,198 the microscope has very old origins; however, it is by no means an obsolete instrument. Although increasingly sophisticated methods continue to be developed for microbiological analysis— from stunningly high-throughput gene sequencers to exquisitely discriminatory mass spectrometers—the microscope is still the most accessible tool that allows us to observe microorganisms directly. The microscope is attractive for its speed of analysis, producing nearly immediate results by visual examination of the specimen. The food microbiologist considers both qualitative and quantitative determinations of microorganisms in foods, and the microscope can function in both aspects. The determinations may involve microbial detection, identification, or characterization (qualitative), as well as enumeration (quantitative). Often the first step in microbial identification, the Gram stain depends on microscopy as the essential analytical tool.23 Among its many other applications in food microbiology, the microscope is needed for detecting, and to some extent for differentiating, the endospores of species within Bacillus and Clostridium,125,179,186 for identifying parasite eggs and protozoan parasites,24,147 and for determining the invasiveness of enteropathogens by observing their entry into mammalian tissue culture cells.10 Dark field and phase contrast microscopy are recommended for differentiating Bacillus species by motility186 and for observing the characteristic corkscrew motility of Campylobacter99 and the tumbling motility of Listeria.7,95 Fluorescence microscopy is useful for fluorescent antibody-based identification of pathogens12 and for visualization of the autofluorescent oocysts of Cyclospora.20,147 The details of these and other types of microscopic analysis can be found in the relevant chapters for the microorganisms or commodities in this Compendium. Microscopy can also be a quantitative technique. When used in combination with concentration methods and | 45 |

diagnostic staining techniques, the microscope can provide specific detection and quantitation of microbial cells very rapidly. Direct microscopic counts are among the standard methods for microbiological examination of eggs13 and for grading of milk.158 Mold9 and yeast194 contamination of foods may be quantified by bright field microscopy and fluorescence microscopy, respectively. The direct epifluorescent filter technique (DEFT) combines membrane filtration with epifluorescence microscopy for sample concentration and determination of total microbial cell counts.155 In addition to its use in detection, identification, and enumeration microscopy also has been essential in many studies aimed at understanding factors affecting survival of microorganisms in foods, food contact surfaces, and the food processing environment.79,206 The variety of microscopic technologies and applications is vast. This chapter provides details for only those microscopic techniques with routine practical applications in the microbiological examination of foods (Section 4.5). Other optical and non-optical microscopic or related technologies exist; these have been used to greatly increase our knowledge of the microbiology of foods but are more appropriate in research investigations because of applicability, cost, and technical expertise required. Among these are confocal laser scanning microscopy, electron microscopy, atomic force microscopy, and flow cytometry, all of which are discussed briefly.

4.2

GENERAL CONCEPTS

In order to be useful instruments of analysis, microscopes must provide two functions: magnification and resolution. Magnification is the enlargement of the image of the microorganism, relative to its actual size, and is achieved through the microscope’s lens system. The commonly used compound microscope has two lenses: the objective lens, which is the one nearest to the specimen; and the ocular lens, located in the eyepiece. The total magnification achieved is usually expressed as the product of the two lenses. For example, if the ‘‘low power’’ objective lens achieves an image magnification of 406 and the ocular lens provides an additional 106 magnification, 4006 is the

Compendium of Methods for the Microbiological Examination of Foods |

total magnification of the lens system. Most compound microscopes used in food microbiology can magnify 1,000–1,5006 by using the ‘‘high power’’ objective, and digital images taken using these can be electronically zoomed to provide additional factors of apparent or ‘‘virtual’’ magnification. However, magnification is of little use if the microscope cannot produce a clear image. Resolution is the ability of the microscope to reveal fine detail. The ‘‘sharpness’’ of the image is a function of the resolution and is dependent on the quality of the lenses. Resolution is often described as the ability to distinguish two objects as distinct and separate or as the smallest distance between two points at which they are still seen as two distinct objects, rather than one blurred object. The resolution depends on the numerical aperture of the lens system, which can be regarded as light-gathering efficiency. The human eye can resolve approximately 100 mm (the size of the largest known bacterium, Thiomargarita namibiensis), a typical compound microscope can resolve nearly 0.2 mm, and an electron microscope can resolve approximately 0.2 nm. Cleanliness is an essential practice for success with any type of microscopy and is especially critical for dark field microscopy, where even the smallest non-cellular particles can give rise to a bright signal. Microscope lenses, slides, and coverslips should be clean and free from dust and residues. Because oil immersion lenses are commonly used in microbiological examination, it is important to clean the lenses after every use and to avoid leaving residues of oil, which can become dry and obscure visibility later on. It is also important to avoid oily contamination of low-power, non-immersion lenses. Dried oil residues can be removed from lenses with lens paper and small amounts of a solvent such as xylene. For routine cleaning of lenses and microscope surfaces, commercial alcohol- and ammoniafree glass cleaners such as Sparkle are often used. Ammonia- or acid-containing cleaners should not be used, as they may damage anti-reflective optical coatings, if present. Proper illumination is also critical. The position of the condenser lens and the field diaphragm (if present) should be adjusted to achieve optimal lighting and specimen contrast (Ko¨hler illumination). Procedures for optimizing illumination vary with the type of microscope and are available from the manufacturer. Room lighting should also be considered, especially in fluorescence and dark field microscopy, where a darkened room is best for viewing. Finally, it is important to set up the microscope for the physical comfort of the microscopist. The interocular distance of a binocular microscope must be adjusted for the distance between the eyes so that a single field of view results, and the adjustable ocular lens should be focused for the sightedness of the microscopist. Chair height also should be positioned so that the neck and back muscles are in a relaxed state. In recent years, ergonomic principles have received increased attention at the instrument design level. Practical tips on cleanliness and instructions for illumination and setup are usually available from the microscope manufacturer, and general principles pertaining to these topics have been published.36,142 46 |

4.3 4.31

TYPES OF MICROSCOPY USEFUL IN FOOD MICROBIOLOGY Bright Field Microscopy

Many common microscopy procedures for food microbiology involve the use of bright field microscopy for examination of the sample. Because most samples do not have sufficient contrast to visualize the microorganisms by bright field microscopy, various stains are used to increase contrast (Section 4.4). Classical staining procedures typically require fixation, resulting in inactivation of the microorganisms and perhaps some distortion of features due to structural changes. Microorganisms in a viable state cannot be observed using bright field microscopy. Nevertheless, with appropriate staining, bright field microscopy allows quick determinations of basic morphology (e.g., rods, cocci), visualization of cell structures (e.g., endospores, flagella), and differentiation of fundamental groupings (e.g., Gram stain reaction).

4.32

Phase Contrast Microscopy

The phase contrast microscope uses specialized optics to increase contrast of specimens for observation of microscopic details. The optical components convert differences in refractive index of materials into obvious variations in light intensity; thus microorganisms can be visualized without the use of stains. The ability to maximize contrast makes the phase contrast microscope useful for observing bacterial structures such as endospores and intracellular inclusions such as poly-b-hydroxybutyrate. Because there is no need to use stains to provide contrast, microorganisms can be visualized in a viable state. For example, a common use of phase contrast microscopy in food microbiology is the demonstration of motility for microbial identification. Procedures for phase contrast microscopy are ultimately simple: a ‘‘wet mount’’ is prepared by placing a loopful of culture in liquid suspension under a coverslip on a glass microscope slide, and the cells are then observed using an objective lens providing an appropriate magnification.

4.33

Fluorescence Microscopy

Fluorescence microscopes are used along with fluorochromes to allow sensitive analyses of microbial cells. Fluorochromes (also referred to as fluorophores) are substances that absorb short-wavelength light, often in the ultraviolet range, and then reemit the absorbed energy in the form of longer wavelength light, typically in the visible spectrum. Fluorochromes may be used either as labels, where they act as a fluorescent tag attached to a bioaffinity ligand such as an antibody or DNA probe, or as stains, where they interact physically with macromolecules or cellular structures or serve as substrates for enzymes. Two special filters are components of the fluorescence microscope: an exciter filter and a barrier filter. The exciter filter is positioned between the light source and the specimen and allows only the short wavelengths of light to pass through to the specimen. The barrier filter is located between the objective lens and the ocular lens and blocks the passage of short-wavelength excitation light and allows passage only of the longer wavelengths emitted from the

| Microscopic Methods

fluorophore(s). The exciter and barrier filters are matched to the fluorescence properties of the particular fluorochrome used to label the specimen. Most fluorescence microscopes now in use are designed with the light source mounted above the specimen, so that the short-wavelength light is bounced off of the surface of the specimen (known as incident light excitation, or epifluorescence). The design enables examination of specimens that are relatively thick compared with the thin preparations required for fluorescence microscopes that transmit light through the specimen. Epifluorescence microscopes are also easier to use and provide brighter images. Various light sources are utilized for fluorescence microscopes. The important feature of any light source is the production of an adequate amount of light in the range necessary for exciting the fluorochromes of interest. Xenon, mercury, and metal halide lamps are common excitation sources; but recently, light emitting diodes (LEDs) have gained in popularity, due to their ease of use, low power requirements, high light flux, longevity, and lack of hazardous waste disposal issues. Portable, battery-powered LED-based microscopes with ‘‘plug-and-play’’ USB digital imaging capabilities are now commercially available. These were originally designed for field-based tuberculosis (TB) and malarial diagnoses but may also be used for other microbiological investigations. Under conditions of high illumination, fluorochromes may absorb more energy than they can reemit, resulting in physical destruction of the molecule through breakage of covalent bonds—a process known as photobleaching. Some fluorochromes, such as fluorescein, are intrinsically susceptible to photobleaching and will irreversibly lose fluorescence upon constant illumination. Antioxidants and free radical-scavenging molecules, such as n-propyl gallate, can be added to the sample to slow the photobleaching process. Typically, a cocktail of such compounds, often in a glycerol base, is used. Such antifade cocktails are available commercially, with some also containing a curing mountant (one that hardens upon exposure to air), further retarding the action of free radicals by slowing their diffusion. A chromosomal counterstain, such as 49,6diamidino-2-phenylindole (DAPI), may also be a component of such mixtures. Through chemical modification of existing fluorophores such as aminocoumarin or rhodamine—dyes spanning the spectrum of useful excitation and emission value but having improved properties such as greater photostability, insensitivity to pH, and increased brightness—have also been obtained.

4.34

Confocal Laser Scanning Microscopy

In conventional ‘‘widefield’’ fluorescence microscopy, the entire sample is illuminated by the light source. Widefield illumination of samples thicker than the focal plane results in out-of-focus or ‘‘stray’’ fluorescence that may blur the image and obscure finer details. The confocal laser scanning microscope (CLSM) is an instrument which eliminates such blur, allowing resolution of detail in thicker specimens or those having complex surface topographies.200 This capability is derived in part from use of a laser light source, which has high penetration and low divergence compared with non-collimated sources, and from the

optical components of the system, which include lenses and pinhole apertures to reduce diffusion of the light and to illuminate the sample in and collect fluorescence from a defined focal plane. Scanning is accomplished with a system of mirrors that rasters the laser point (0.25–08 mm in diameter, 0.5–1.25 mm in depth) across the sample along the x- and y-axes. Fluorescence emitted from the sample is collected through the objective lens and directed through a system of beamsplitters and emission filters to detectors assigned to each channel. Changes in light intensity resulting from rastering of the laser over the sample are registered and converted into an electrical signal of varying voltage, which is then converted into digital pixel information. The digital image is constructed through assembly of information collected at each rastered laser point and is displayed on the monitor and stored to a hard drive. Uses of the CLSM in the study of biological systems generally require fluorescent dyes, markers, or probes for detection. Most of the common fluorescent dyes have been employed, for example, fluorescein, DAPI, rhodamine, the SYTO series of stains, etc., depending on the spectral characteristics required of the fluorochrome for the specimen being examined. The green fluorescent protein (GFP) marker has been a popular method of conferring intrinsic fluorescence on microorganisms, allowing observation of their growth and behavior on plant surfaces or in food matrices (Section 4.45). While the resolution of the system is similar to that of a conventional microscope, the ability of the CLSM to visualize microbial cells in situ at multiple focal planes is an outstanding advantage. Multiple optical sections (or ‘‘slices’’) may also be stacked in the z-plane to yield a threedimensional reconstructed image of the original sample. Three-dimensional volumetric rendering of CLSM data using specialized software allows viewing of images from varying angles, providing a greater understanding of the sample’s spatial arrangement. Many studies have involved the CLSM as a principal tool to study the localization and internalization of bacterial67,84,116,117 and protozoon133 foodborne pathogens in fruits and vegetables. Other diverse applications in food microbiology have included determinations of physiologically active foodborne pathogens in foods,39,44 removal of foodborne pathogens from fresh produce,202 distribution of bacterial populations in dairy products16,102,130 and other foods,15 permeabilization and lysis of starter cultures in Gouda cheese,38 in vitro attachment of foodborne pathogens to meat proteins,211 spatial and temporal determinations of foodborne pathogens in biofilms,164 and comparison, growth, and determination of fungal hyphae.145 The use of confocal laser scanning microscopy in food research has been reviewed.184

4.35

Electron Microscopy

The electron microscope uses a beam of electrons, rather than visible light, to create an image of the specimen. Electromagnetic lenses control the path of the electron beam, either by directing them through a resin-embedded, thinly sliced specimen [as in transmission electron microscopy (TEM)] or by scanning across the surface of a specimen coated with electron-dense materials [as in | 47

Compendium of Methods for the Microbiological Examination of Foods |

scanning electron microscopy (SEM)]. In comparison with the light microscope, which can magnify an object approximately 1,5006, an electron microscope can magnify nearly 100,0006, owing to the much shorter wavelengths capable with an electron beam. While the spatial resolution of a light microscope is approximately 0.2 mm, the electron microscope can typically resolve 0.2 nm, with specialized instrumentation capable of resolving features smaller than 50 pm—atomic resolution.72 Electron microscopes are research instruments, and because of their significant capital and maintenance costs, the amount of training needed for their operation, and the sophisticated skills needed for productive use, they are not generally used in routine procedures for the microbiological analysis of foods. Nevertheless, electron microscopy has provided fundamental information in diverse areas related to food microbiology. TEM allows detection of virus particles and can provide details of subcellular structures of bacteria and their hosts. TEM applications have revealed the structural changes accompanying microbial stress responses46 as well as inactivation during processing or after application of antimicrobials.41,132 SEM offers a threedimensional depth of view, and applications have led to insights on microbial attachment, colonization, and survival in situ, for example, in biofilms76,162 and in food matrices,76,127 and on factors affecting their distribution, for example, on stainless steel surfaces or produce.206 An example of the richness of visual information available through SEM analysis is shown in the image of a naturally occurring biofilm present on alfalfa sprouts (Figure 4-1). Limitations of standard TEM and SEM include the need to dessicate samples, which can lead to the collapse of highly hydrated structures such as the biofilm matrix shown in this image. Environmental SEM (ESEM) instruments, which operate under low vacuum and use water as a chamber gas, are able to image samples in their native form, that is, without any preparative steps such as dessication, fixation, or coating with electron-dense materials. Another approach for such ‘‘wet’’ SEM involves the use of specialized chambers equipped with a thin, electrontransparent window for viewing hydrated samples in their natural state, a strategy that has been applied to the study of live cells of the yeast Schizosaccharomyces pombe with a spatial resolution of ,32 nm.150,187

4.36

Atomic Force Microscopy

Atomic force microscopy, developed in the 1980s, has become an important addition to the repertoire of technologies for high-resolution study of microbial cells. Capable of resolving structures at the nanometer level, the atomic force microscope (AFM) is recognized for its high quality imaging of discrete topographical features of microbial cells, including crystalline S-layers and the peptidoglycan fibers that constitute the cell wall. Newer applications include measuring the forces of interaction between microbial cells and their environments. These force measurements allow investigations of the physicochemical or biomechanical properties of cells, including surface hydrophobicity, cell turgor, elasticity, microrheology, or charge interactions that may be involved in attachment of cells to surfaces. Importantly, AFM imaging 48 |

Figure 4-1. Scanning electron microscope (SEM) image of a naturally occurring biofilm in alfalfa sprouts. Alfalfa sprouts were purchased from a local grocery store and prepared for SEM with a 15 min fixation in electron microscopy-grade glutaraldehyde. A drop of the fixed sample was applied to a poly-L-lysine-treated silicon chip, allowed to adhere for 5 min, then samples were fixed further in 1% osmium tetroxide, followed by dehydration in an ethanol series, sputter coating and viewing via SEM using an Hitachi S-3400N microscope. Alfalfa and other seed sprouts support a complex ecosystem comprised of bacteria (,108 to 109 CFU/g sprouts) and yeasts or other fungi (, 106 CFU/g).76 With such a plentiful food supply available, grazing protozoa are also sometimes observed. Unpublished data from B. Bisha and B. F. Brehm-Stecher.

and measurement may be carried out on living cells under aqueous conditions. AFM-based observations have revealed fundamentally new biological observations, such as the discovery of regular nanomechanical oscillations of the cell wall of Saccharomyces cerevisiae. By placing the AFM tip on the yeast cell surface, Pelling et al.151 were able to measure periodic oscillations ranging in frequency from 0.8 to 1.6 kHz, with amplitudes of , 3 nm. These authors concluded that this vibratory motion stemmed from internal metabolic activity of the cell, including the actin of molecular motors such as dynein, myosin, and kinesin.151 These oscillations can be translated into audible sound, allowing us to hear yeast cells ‘‘sing.’’151 The AFM consists of a probing tip, which is mounted on a flexible cantilever and used to scan the surface of the specimen in the x-y plane. The probing tip (analogous to a microscopic phonograph needle) follows the contours of the surface as it moves across the specimen. The tip and the cantilever (which, continuing the analogy, can be thought of as the phonograph’s arm) are deflected in the z-plane by ‘‘atomic’’ forces between the tip and the specimen, including electrostatic, electrosteric, and van der Waals forces. The deflection of the cantilever is registered by a laser-based optical system, and data are translated into a topographical map/image of the specimen’s surface. Other types of tip-surface interactions may also be used to interrogate the sample, including full-contact mode, where the tip is ‘‘dragged’’ across the sample’s surface, and

| Microscopic Methods

‘‘tapping’’ mode, there the tip makes only intermittent contact. The forces required for the movement of the probing tip during scanning or for stretching and breaking bonds between tip-tethered macromolecules and their binding partners can also be measured. The force-distance curves generated can provide detailed information on the physicochemical features of microbial surfaces, with piconewton sensitivity. Use of the AFM tip to push, poke, or puncture cells can allow measurement of a single bacterium’s turgor pressure or enable gene delivery via ‘‘cellular nanosurgery.’’34 Interestingly, Salmonella cells have been shown to remain viable and able to divide even after multiple punctures, providing insight into the selfsealing dynamics of the cell wall and membrane.183 Several excellent review articles describe the fundamentals and general applications of AFM.65,66,68 Recent advances include high-speed or ‘‘video rate’’ AFM, which can provide data acquisition , 1,0006 faster than conventional AFM and has even been used to follow dynamic molecular events such as myosin ‘‘walking’’ along an actin filament or ephemeral phenomena such as the progressive roughening of Eschericia coli surfaces resulting from exposure to an antimicrobial peptide.106 AFM is beginning to be applied to study questions of interest to food microbiologists: effects of food contact surface characteristics on attachment of foodborne pathogens, for example, E. coli O157:H7,86 Listeria monocytogenes,138,177 Campylobacter,144 Salmonella,47 Staphylococcus,118 and Bacillus137; changes in microbial cell structure, including endospores, after antimicrobial treatments56,74,124; role of tomato surface composition on resistance to microbial infection101; and spatial distribution of probiotic bacteria within gel networks for immobilization.63

4.37

contrast (DIC) microscopy is another tool for resolution of discrete features in unstained or minimally stained cells. DIC is an interferometric contrasting technique that enables visualization of gradients in refractive index occurring in a sample. DIC allows resolution of otherwise invisible cell features and results in images having a three-dimensional aspect. Burt et al.40 used this method to evaluate the impact of the essential oil component carvacrol on flagellation in E. coli O157:H7. Flagella were easily visualized by DIC after exposure to a dilute tannic acid-crystal violet stain. A similar technique was used to show the flagella of Salmonella in Figure 4-2. Episcopic DIC microscopy (EDIC) is a variant approach that uses reflected, instead of transmitted light. It is therefore suitable for examination of solid or poorly transmissive samples, including biofilms or opaque materials to which bacteria are attached.204 The enhanced z-plane dimensionality of EDIC allows real-time analysis of samples having surface curvature and topography not easily imaged with using standard microscopic techniques, as demonstrated with spinach or watercress leaves.204 In light of these capabilities, EDIC has been forwarded as a rapid alternative to SEM or CLSM for the analysis of such topographically challenging samples.204

4.38

Related Techniques: Flow Cytometry

Flow cytometry (FCM) is a rapid method for flow-through optical analysis of individual cells, including microbial cells. Several basic instrument architectures exist, including stream-in-air systems, cuvette-based systems, capillarybased systems, and instruments built around an inverted microscope design, the latter of which provides a direct linkage between traditional microscopy and FCM. In FCM, a liquid sample is aspirated from a test tube and the cells

Additional Modes of Microscopy Useful in Food Microbiology

Light microscopy is a mature technology, and scientists have devised many inventive variations on this basic theme. Although these are too numerous to catalog fully, some may have advantageous applications in food microbiology and are discussed briefly here. Many biological subjects, including microbes, are difficult to resolve as they are transparent and have refractive indices similar to the aqueous media in which they are suspended. Dark field microscopy is a method for high-contrast viewing of unstained, living cells. In dark field microscopy, light is not transmitted up through the sample and directly into the objective. Instead, it is directed so that the only light collected is that which is reflected or refracted by the sample. As a result, cells appear brightly lit against a dark background, an effect that has been likened to the (literal) difference between night and day when viewing stars.210 dark field microscopy was used as early as 1909 to observe nanoscale colloidal particles.210 More than 100 years later, dark field microscopy remains a staple for characterization of nanomaterials or for detection of nano-optical labels. Enhanced dark field microscopy, a related technique available on standard light microscopes through use of a specialized condenser, has been used for direct observation of cell-nanoparticle interactions and is capable of resolving features as small as 90 nm.197,205 Differential interference

Figure 4-2. Differential interference contrast (dic) microscopy of Salmonella spp. Suspensions of Salmonella cultures grown in trypticase soy broth were spotted onto polylysine or APEScoated slides and imaged using DIC microscopy using a Nikon Eclipse 80i microscope outfitted for DIC. Panel A is an image of unstained cells. Panel B is an image of cells stained indirectly with a crystal violet-tannic acid flagellar stain. A small drop of the stain was placed on the edge of the coverslip and allowed to diffuse across the microscope slide and interact with the cells.40 Adherence of the dilute stain complex to flagella provided enough contrast to allow clear resolution of flagella using DIC. DIC was also sensitive to staining of particles and glass surface imperfections by the flagellar stain, resulting in the higher background seen in Panel B; two-dimensional blind deconvolution was used to sharpen the image (AutoQuant X software, ver. 2.2; Media Cybernetics, Silver Spring, MD). This figure highlights the three-dimensional quality characteristic of DIC images.

| 49

Compendium of Methods for the Microbiological Examination of Foods |

are passed individually in front of an intense illumination source such as an arc lamp, a water- or air-cooled laser, or a laser diode. Data on scattered light and intrinsic or probeconferred fluorescence are collected for each ‘‘event’’—a term applied to any object whose passage generates a detectable signal. Through combined analysis of scatter and fluorescence signals, cells can be differentiated from the background ‘‘noise’’ typical of food samples. With data collected on many thousands of cells, FCM allows individual-level characterization of large event populations and subpopulations according to user-specified analysis criteria. Although most instruments were developed for analysis of relatively larger mammalian cells, FCM has been used successfully for the analysis of various microbial cell types, including bacterial or fungal spores, vegetative bacterial cells, yeasts, and protozoa.26,35,59,77 Apart from the versatility of the technique, a key feature of FCM is its rapidity. Compared with other detection methods such as plating, imaging, quantitative PCR, and DNA sequencing, FCM can deliver results in minutes, versus weeks, days, or hours.193,203 Typical FCM instruments are capable of analyzing 10,000 events per second at flow rates ranging between 10 and 60 ml/min. Although the details differ among specific instruments, most flow cytometers operate on similar principles. Briefly, some form of hydrodynamic focusing is used to align the cells so that they may be interrogated and analyzed individually. This may be accomplished by forming a nonturbulent (laminar) core sample stream within a flow of buffered saline (sheath fluid), or in cytometers based on an inverted microscope design, by flowing the sample across the surface of a glass coverslip at a precise angle of incidence.128 High-intensity light is directed through a lens, focusing it to a small spot that is used to illuminate the cells as they pass. A system of mirrors and optical filters is used to collect small-angle (forward scatter) and wide-angle (side scatter) light scatter and fluorescence signals, each of which is directed to an appropriate detector such as a photodiode (for the relatively strong scatter signals) or a photomultiplier tube (for the weaker fluorescence signals). These detectors convert photonic signals into electronic output, yielding voltage pulses that differ in shape and height according to the physical properties of the cell that generated the signal. A key strength of FCM is the ability to simultaneously collect multiple measurements on each cell. Parameters collected typically include forward and side scatter and up to eight fluorescence colors, although polychromatic (17 color) and hyperspectral cytometers capable of providing full spectral analysis of each cell have been reported.88,152 Some instruments are also able to physically sort cells of interest through use of fluorescence-activated deflection of charged droplets containing target cells. These instruments can be used to recover specific cells for further growth or analysis or to physically enrich cells having particular characteristics. Cells of interest can be sorted to test tubes, agar plates, or microtiter plates.35,59,141 Non-sorting instruments able to collect forward and side scatter and one or two fluorescence channels are sufficient for most tasks in applied food microbiology such as detection of specific pathogens or enumeration of cells, provided that the system is sensitive enough to register the relatively small 50 |

microbial cells. While cells may be indirectly enumerated using a calibrated bead set (a suspension of polystyrene beads of known concentration), some FCM systems can provide an absolute cell count without reliance on beads. Most applications of FCM in food microbiology require the use of exogenously applied fluorescent stains. Dyes useful for labeling and detection of all key classes of cellular biomolecules (e.g., DNA and other nucleic acids, proteins, lipids, storage polymers) are commercially available.34 Fluorescent Gram staining can be used to characterize the microbial flora of foods such as milk.97 Fluorescent respiratory or enzyme substrates, intracellular redox indicators, and reporters of membrane integrity may be used to assess cellular activity, exposure to applied stresses, or cell viability.34,140 Detection of specific organisms can be achieved through labeling with fluorescently labeled antibodies, rRNA-targeted probes, or nucleic acid aptamers.26,34,69,141,193 Examples of FCM applications in food microbiology include monitoring of food fermentations, detection and quantification of food spoilage, evaluation of starter culture or probiotic activity,52 detection and enumeration of pathogens in foods,25,26,64,90,193 assessing the impact of antimicrobial treatments or common food processing stressors on the physiology and viability of foodborne microbes,85,108,140 analysis of drinking water,203 and differentiation of antibioticsensitive and antibiotic-resistant pathogens.176 Figure 4-3 illustrates flow cytometric detection of Salmonella in 10 hr enrichments of contaminated peanut butter.

4.4

MEDIA, REAGENTS, AND STAINS

See also the chapter ‘‘Microbiological Media, Reagents, and Stains.’’

4.41

Direct Staining

Most cells do not have sufficient contrast for visualization by bright field microscopy (notable exceptions being relatively large cells containing intracellular pigments such as carotenoids or chlorophyll). Chemical stains are used to provide the contrast needed for observing most microbial cells by bright field microscopy. Classical methods for direct staining generally destroy the viability of the cells, however. If it is necessary to observe viable cells, for example, in determining motility, then the instrument of choice is the phase contrast microscope, which increases the contrast of the cells against the background so that they become visible without staining. Direct staining can help to generally classify microbial cells (e.g., bacteria, yeasts, molds, protozoa) and to determine cellular morphology (i.e., rods, cocci, spirals, and filamentous forms) and groupings of cells (e.g., chains or other arrangements). Although bacterial endospores may be difficult to stain, they can be easily discerned using phase contrast microscopy, appearing as refractile bodies within or separate from stained vegetative cells. Heat fixation is commonly done prior to direct staining to promote cell adherence to the microscope slide, although slides may also be pre-treated with chrome alum-gelatin (‘‘gelatin subbing’’), coated with the positively charged polymer poly-L-lysine, or silanized with 3-aminopropyltriethoxysilane (APES), the latter which results in covalent

| Microscopic Methods

gentle rinsing with a stream of water and drying between sheets of blotting paper, the slide is ready for examination.

4.42

Figure 4-3. Combination of flow cytometry and rRNA-targeted probes for detection of Salmonella in peanut butter. Peanut butter samples were contaminated with a low level (0.4 CFU/g) of Salmonella and enriched for 10 hr in a non-selective broth (TB Dry, MO BIO Laboratories, Inc., Carlsbad, CA). Within 30 min of homogenization, enrichments partitioned into three phases: a particulate-rich sediment, an aqueous phase, and a fat phase. The aqueous fraction was collected and stained via fluorescence in situ hybridization using a cocktail of Cy5labeled DNA probes specific for Salmonella. Stained samples were analyzed by flow cytometry (FACSCanto, BD Biosciences, San Jose, CA). The resulting dot plot shows three main populations. Population A is comprised of various non-target events, including particulate matter, fat micelles, and nonSalmonella bacteria (most likely from spores present in the peanut butter). Population B results from the simultaneous passage of probe-labeled Salmonella and non-target events in front of the detector, yielding a composite result combining Salmonella-specific fluorescence with a broad distribution of side scatter values. Population C stems from probe-labeled Salmonella passing individually in front of the detector. Unpublished data from B. Bisha and B. F. Brehm-Stecher.

attachment of cells to the glass. Such treatments minimize cell loss during staining and washing procedures, without damaging cellular structure. In all cases, a thin film of cells, rather than a thick deposit, generally works best. Encircling the area to be smeared with a wax pencil provides a visual indicator of where the cells are and also creates a hydrophobic barrier for containment of the stain. If working from a broth culture, a loopful of the culture is deposited onto the slide, air-dried, and then, when plain glass slides are used, heat-fixed by quickly passing the underside of the slide over a Bunsen burner flame two to three times. If working from an agar culture, a needle is touched to the colony and then into a droplet of water on the microscope slide. The needle is used to mix the culture into the water droplet; then the sample is air-dried and heat-fixed as described above. The stain is then applied to cover the entire area of the smear. Commonly used stains are methylene blue, crystal violet, and carbol fuchsin, all of which can provide adequate staining in 1 min or less. After

Differential Staining

Differential stains allow microbial cells to be distinguished based on a particular property or characteristic. The most important differential stain for the food microbiologist is the Gram stain, which categorizes most bacteria as either Gram positive or Gram negative, depending on their reaction to the staining procedure. The primary reason for the distinction lies within the structure of the cell wall, which determines whether the procedure’s decolorization step will cause the stain-mordant complex to be washed out of the cells. Gram staining reagents are available commercially or may be prepared from component dyes and solvents. It is best to use freshly grown cultures, because older cultures (especially Gram-positive bacteria) may show variable reactivity in the staining procedure. Apart from cell age, other variables affecting Gram staining quality may include overheating during heat fixation, uneven staining, or decolorization from using too many cells, too much or too little decolorization, and use of expired reagents. A common example in this latter category is inorganic iodine, which is prone to oxidation. Commercial kits therefore typically contain iodine that is chemically stabilized, through complexation with L-polyvinylpyrrolidone, for example. The staining protocol generally consists of four steps: the initial stain (crystal violet), the mordant (iodine), decolorization (alcohol, acetone, or a mixture of these), and the counterstain (safranin). Before staining, the sample is prepared by heat-fixing on a microscope slide, as described above (Section 4.41). Although a number of variations exist, a typical Gram staining protocol involves crystal violet staining of the heat-fixed smear for 30 s, application of the iodine mordant (30 s), decolorization (15 s), and safranin counterstaining (30 s), with gentle rinsing with water after each step. A fluorescent Gram stain using SYTO 13 and hexidium iodide has been described.135 Alternate combinations of dyes (SYTO 9, SYTO 13, or SYTOX Green; Texas-Red-labeled wheat germ agglutinin or hexidium iodide) have also been reported or are also available in kit form. Although different dye chemistries are used among these fluorescent protocols, they are functionally similar to each other in that they all allow differentiation of Gram-negative and Gram-positive bacteria based on structure-based differences in dye uptake or exclusion between these cell types.

4.43

Special Structural Stains

A number of stains for special structures are of use to the food microbiologist. These include staining for bacterial endospores, storage materials, protozoan cysts, and parasite eggs. They are briefly described here but are also covered in relevant chapters of this Compendium. Bacterial endospores are not easily stained because of their impermeable nature. Heat must be applied to weaken the endospore wall and allow uptake of the stain. The Schaeffer-Fulton procedure (also referred to as the WirtzConklin method) involves staining of a fixed smear with malachite green on a rack above a boiling water bath for | 51

Compendium of Methods for the Microbiological Examination of Foods |

5 min. After rinsing with water, a counterstain, such as safranin, is applied for 30 s to stain the vegetative portion of the cell red and provides contrast for the green-stained endospore. This stain has also been used to probe exposure to conditions leading to spore germination, as spores exposed to germination enhancers have been shown to stain pink within as little as 30 min after treatment with these compounds, resulting from germination-associated changes in spore physiology.89 Endospore staining helps in speciation, because the location and shape of the endospores within the cell are often diagnostic; for example, Bacillus cereus endospores appear ellipsoidal and are located in a central to subterminal position,186 while Clostridium botulinum produces a characteristic ‘‘tennis racket’’ appearance.179 Before testing a culture for the presence of C. perfringens enterotoxin, which is produced during sporulation, the concentration of endospores should be determined microscopically, so that adequate levels of the enterotoxin are ensured for the assay.163 Certain bacterial species accumulate polymers as storage materials, for example, glycogen, poly-b-hydroxybutyrate, and polyphosphate. These storage materials can be observed microscopically through the use of special stains.23 Food microbiologists are generally not concerned with the staining of storage material, except in the case of Bacillus speciation, where it becomes important in distinguishing between the foodborne pathogen B. cereus and the closely related insect pathogen B. thuringiensis. The two species are identical for many phenotypic characteristics, but a primary distinguishing trait is the production of insecticidal protein toxin by B. thuringiensis.125,186 The protein toxin is visible as diamond-shaped crystals after sporulation of the species. Heat-fixed smears are treated with methanol for 30 s, then air-dried. The smear is covered with carbol fuchsin or basic fuchsin, then heated gently over a flame until steaming of the dye is observed. After cooling the slide for 1 to 2 min, the heating is repeated, followed by rinsing of the slide with water. An alternate staining approach for B. thuringiensis insecticidal toxin using the protein stain Coomassie Blue (0.133% in 50% acetic acid, applied after heat fixation, followed by rinsing, drying, and observation) has been described.161 Protozoan cysts and parasite eggs may be identified by their microscopic morphology. After separation and concentration from the food matrix, the cysts and eggs may be stained with Lugol’s iodine24 or visualized by their autofluorescence or by reaction with fluorescent antibody reagents and/or rRNA-targeted nucleic acid probes87,147 (see Section 4.44).

4.44

Specific Probes

Microbial cells may be distinguished by the use of specific probes, notably antibodies and nucleic acid binding probes. Probes may be designed for target cell recognition at various levels of specificity, for example genus, species, or strain. Labeled with fluorescent dyes and used in conjunction with the fluorescence microscope, antibodies and nucleic acid probes can be direct, specific identifiers of microbial cells. Antibodies are protein molecules produced as part of the immunological response to foreign substances (antigens) in 52 |

animals. The proteins, the polysaccharides, and their derivatives that make up microbial cell structure are antigenic and promote production of antibodies, which, in turn, can be used as specific probes of the microbial cells. Identification of a target antigen that is unique to a microbial group (e.g., genus, species, strain) is the first step in the design of an antibody probe. Antigens usually are present in the cell in amounts sufficient for detection by microscopy; however, certain antigens may be influenced by environmental conditions and may not be expressed by the cell. In fact, antibody probes have been used to study regulation of antigen expression by tracking their appearance microscopically.92 Fluorescent antibodies have been used for microscopic identification of Salmonella in foods for many years,168 and the Fluorescent Antibody Screening Method for Salmonella has undergone the extensive testing required of an official method through AOAC International.12 Its use is limited to screening, that is, presumptive identification rather than confirmation, due to cross-reactivity of the commercial antibody preparation with other members of the Enterobacteriaceae. Immunofluorescence microscopy has been used for identification of many other microbial species, including bacterial foodborne pathogens (e.g., L. monocytogenes,111 E. coli O157:H7189) and protozoa (e.g., Cryptosporidium, Giardia, and Cyclospora87,148) Fluorescent antibody binding has also been used to enhance the detectability of viruses in food matrices.188 Emerging approaches for the generation of non-antibody binders for the detection of specific cells include the use of DNA or RNA aptamers or phage-displayed peptides and single chain variable fragment (scFv) antibodies. Sequence-specific nucleic acid probes recognize diagnostic stretches of DNA or RNA, and several types of probes are available. Oligonucleotides are short (, 18–21 bp) nucleic acid polymers that bind to specific segments of DNA or RNA by complementary base pairing. Technically, either DNA- or RNA-based probes could be used, but DNA is often the default choice, as it is both less expensive and inherently more stable. Peptide nucleic acids (PNAs) are synthetic DNA mimics. PNAs have an uncharged, peptide-like backbone onto which the four standard nucleobases (A, T, G, C) are grafted; the spacing of the nucleobases on the backbone is such that PNAs are able to recognize and bind to natural nucleic acids via Watson-Crick pairing.70 Because of their inherently higher melting temperatures (Tm), PNA oligomers are typically around 15 nucleobases in length. PNA has several practical advantages over DNA due to its uncharged backbone, high Tm, and the relatively short probe lengths used. These include enhanced cell permeability, faster hybridization kinetics, and enhanced mismatch discrimination. Additionally, the ribosome-denaturing reaction conditions used (low salt, high temperature, high pH) allow PNAs to access sites that may be physically unavailable to DNA probes.33 Unlike the antigens targeted by immunofluorescence techniques, diagnostic genetic sequences may be present at copy numbers too low for practical detection by fluorescence microscopy. This has been addressed through advances in probe design or labeling, including use of polynucleotide probes capable of single-copy gene detection,

| Microscopic Methods

dendrimer-based labels carrying hundreds of fluorophores, enzyme-catalyzed deposition of fluorophores at the probe binding site, and use of infrared labels that fluoresce in spectral regions apart from background noise. 50,82,212 Amplification of target sequences within whole microbial cells is also possible using various methods such as in situ PCR, in situ reverse transcription, in situ rolling circle amplification, in situ loop-mediated isothermal amplification, or bacterial chromosome painting.32 However, these approaches for detecting low-copy genetic sequences are specialized and technically demanding; it is often more practical to detect high-copy targets, such as ribosomal RNA (rRNA). The ribosome is a naturally amplified molecule, with most actively growing bacteria containing between 103 and 105 ribosomes per cell.5 Because rRNA sequences reflect evolutionary differences among organisms, fluorescent oligonucleotides complementary to rRNA sequences can be thought of as ‘‘phylogenetic stains’’ for microbial cell detection.60 This approach is commonly referred to as fluorescence in situ hybridization (FISH). At its simplest, FISH is a ‘‘shake and bake’’ procedure, no more difficult than labeling with antibody probes, although cells must first be chemically fixed with formalin or ethanol to allow probe entry, and additional proteinase digestions may also be required for some cell types, depending on which probe technology (e.g., DNA or PNA) is used. While bacterial endospores are recalcitrant structures, they do contain enough rRNA to enable FISH-based detection. As with chemical stains, endospores treated with germinants such as L-alanine become permeable to FISH probes within as little as 20 min, allowing the detection of viable spores (as demonstrated by their germination) with this method (J. Hyldig-Nielsen, personal communication). While the use of nucleic acid probes for direct microscopic analysis had its origins in microbial ecology studies, FISH has since been readily adopted for use in food microbiology using both DNA probes6,71,73,190 and PNA probes.2,3,126,153,154 Review articles on cell-specific probes and their use in microscopy have been published.4,5,28,34,129

4.45

Fluorescent Biomarkers

Autofluorescent molecules such as the GFP from the jellyfish Aequoria victoria are an important category of expressible biomarkers. The importance of GFP as a fundamental tool in biology was recognized with the award of the 2008 Nobel Prize in Chemistry to its inventors. Cloned into a microorganism of interest, the GFP construct allows immediate visualization of the microorganism by fluorescence microscopy, enabling in vivo microscopic studies of specific cell types. Although not a routine component of food microbiological analysis, GFP is mentioned here because of its general usefulness and impact on the study of microorganisms in foods by microscopy, including studies of survival,67 transfer,58 attachment, 2 0 1 localization, 1 0 9 and penetration. 1 3 1 Derivatives of or alternatives to GFP that have different functional properties, for example, spectral characteristics, stabilities, maturation rates, etc., are commercially available.49 GFP and related fluorescent proteins have proven to be fundamental tools in the study of microbial biofilms, allowing visualization of specific cell types in multispecies

films,149 and some have been applied in food matrices or environments.27 Food-related applications include GFP for the study of fungal colonization of grapes,54 red-shifted GFP to study E. coli O157:H7 interactions on lettuce and cauliflower,185 mCherry (derived from Discoma spp. mushroom corals) to demonstrate Pseudomonas in biofilms and on tomato roots,120 and cyan fluorescent protein to show interactions of Salmonella with lettuce leaves.117 Spectral changes in GFP fluorescence at different pH values enable its use as an indicator of intracellular pH (pHi), allowing non-invasive measurement of pHi over a range of 5.5-8.5, an approach that could be used to monitor the effects of organic acid exposure in food systems.34 Finally, GFP has also been used to follow pathogen survival after antimicrobial exposure, although viability of GFP-labeled cells cannot be determined through direct microscopic observation. As an example, this approach was used to monitor the impact of ethyl pyruvate treatment of green onions and baby spinach on E. coli O157:H7, as determined by counting of fluorescent colonies after plating. As a note of caution, however, a study of isogenic GFP-expressing and nonfluorescent pairs of E. coli and Pseudomonas putida determined that incorporation of GFP can lead to increased susceptibility to several classes of antibiotics, suggesting a generalized physiological impact on the fitness of GFP transformants.1 Therefore, appropriate care should be used when interpreting antimicrobial susceptibility data collected using GFP-labeled strains.

4.46

Indicators of Physiological Activity

Certain fluorescent dyes have been used in methods aimed at directly distinguishing between physiologically active and inactive microbial cells. The earliest methods were based on dye exclusion: active cells with impermeable membranes excluded the dye, while cells with leaky membranes allowed it to pass, rendering the cells stained. Propidium iodide is an example of a dye that is excluded by intact membranes. A dye that freely passes through intact cell membranes, for example, DAPI, may be introduced simultaneously to the cell suspension for differentiation of the active population of cells. So-called live-dead staining has been applied in diverse studies,14,103,143 and reagents that have been marketed as commercial kits (LIVE/DEAD BacLight kits)139 have been widely used.119,121,134,170 Live-dead staining has been combined with specific probes, for example, oligonucleotides, to determine viability within populations of specific cell types during mixed culture fermentations.29 Cellular enzyme activity is also used to indicate cell viability or physiological functionality. An example is redox (dehydrogenase) activity associated with cellular respiration, an indicator of the presence of a functional electron transport chain. Colorless or non-fluorescent redox indicators that accumulate in the cell as chromogenic209 or fluorescent169 precipitates upon reduction have been used to detect electron transport activity in direct microscopic assays. Viable microbial populations have been quantified with such redox indicators in various matrices, including bottled water75 and milk.22 Redox indicators have also been multiplexed with other cellular labels. For example, fluorescent antibody staining has been coupled with the | 53

Compendium of Methods for the Microbiological Examination of Foods |

red-fluorescing 5-cyano-2,3-ditolyl tetrazolium chloride (CTC) to obtain viable counts of specific pathogens in water, including E. coli O157:H7 and Salmonella spp.160 CTC has also been combined with DNA-FISH for the analysis of respiring Pseudomonas spp. in milk112; CTC staining of GFPexpressing Campylobacter allowed detection of active populations of this pathogen and their distribution on poultry.44 Fluorescein diacetate (FDA) is a viability stain that relies on the presence of both an intact membrane and intracellular enzyme activity for labeling of active cells.195 FDA is a cellpermeant, non-fluorescent compound that accumulates in viable cells. Dead cells do not accumulate the compound due to their leaky membranes. Inside the cell, the FDA molecule is cleaved by esterase activity to produce free fluorescein, which results in fluorescent staining of the cells. The strategy has been applied to staining of microcolonies produced after filtration and incubation of membrane-filtered samples.19 However, a drawback to the use of FDA is the poor permeability of the Gramnegative outer membrane to such hydrophobic compounds, which results in a staining bias toward Gram-positive cells with this reagent.62,182 Additionally, the fluorescein that is liberated by esterase activity, despite its negative charge, can leak out of the cells rapidly, leading to poor staining of cells.182 FDA derivatives such as carboxyfluorescein diacetate (cFDA) have been developed to address this; the cleavage product of cFDA contains additional negative charges, resulting in greater intracellular retention.182 Another derivative, chloromethyl fluorescein diacetate (CMFDA) undergoes a secondary enzymatic transformation after the initial esterase reaction, producing a fluorescein-thioether adduct that labels cells tenaciously, yet does not interfere with vital processes such as cell division.139 With half of the original label apportioned into daughter cells after each division, CMFDA can be used to track cells and subsequent generations. This label has also been used to follow ingestion of Campylobacter jejuni by amoebae, which have been implicated as vectors for dissemination of pathogens in food environments. Quenching of CMFDA in the acidic food vacuole also allowed measurement of ciliate digestion rates and microscopic identification of digestion-resistant Campylobacter within the amoebae.77 In the presence of nalidixic acid and nutrients, cells that are capable of undergoing cell division increase in length but do not complete the formation of new cell walls; thus elongated cells indicate physiologically active cells in this direct viable count (DVC) assay.23,115 The assay compares 2% formaldehyde-fixed cells with unfixed cells suspended in yeast extract (0.025%) and nalidixic acid (0.002%). The cells are incubated for 6 hr to allow growth processes to proceed, collected on a membrane filter, stained with acridine orange, and examined by epifluorescence microscopy. The use of inhibitors other than nalidixic acid has extended the technique to a variety of bacteria, including Gram-positive cells.37,80,173 The DVC assay also has been combined with fluorescent antibody staining to specifically enumerate active cells of Vibrio and Salmonella in surface or wastewaters,30,61 and with FISH to enumerate members of the family Enterobacteriaceae in drinking water.18 54 |

When incubated in the presence of nutrients, physiologically active cells can form microcolonies, which are detectable microscopically.166,167 The microcolony assay involves membrane filtration of a sample, then incubation of the filter on selective media for several hours to allow bacterial cell division and formation of microcolonies on the filter surface. A stain is then applied, and the filter is examined by epifluorescence microscopy. Acridine orange166,167 or fluorescent antibody167,174 staining may be used to provide total and specific viable counts, respectively. The microcolony assay has the same limitation as plate counting, that is, it presumes provision of appropriate growth conditions for the cells. If the cells are injured, their ability to form colonies (even microcolonies) on selective media may be compromised. Detergent and enzyme treatments that are routinely used to allow filtration of the food sample were found to inhibit microcolony formation,166 and resuscitation measures have been incorporated into the procedure to overcome the inhibition.167 Fluorescent oligonucleotide hybridization to microcolonies has also been used to detect specific viable bacterial populations.175 Remarkably, the use of a microcolony method for enumeration of bacteria in milk was reported as early as 1915.81 In this work, the author reported that results similar to those obtained after 48 hr with standard plating methods could be obtained after only 6 hr using the microcolony approach. Messenger RNA (mRNA) has a fast turnover rate, and the half-lives of these molecules are typically measured in minutes. Therefore, the presence of mRNA, as detected using the reverse transcriptase-polymerase chain reaction, has also been used as an indicator of physiological activity in bacteria.113 Despite their popularity and attractiveness for indicating physiologically active cells, the microscopic methods should not be assumed to be generally applicable; they must be validated for specific circumstances of use. The limitations and pitfalls for all of these methods have been examined and presented.17,21,53,107,182

4.5 4.51

QUANTITATIVE APPLICATIONS Introduction

Rapid, direct quantitation of microbial populations may be obtained by microscopy. The sensitivity or limit of detection often cannot match that of agar plate counts, but depends on the procedure, especially the preparative steps, such as concentration. Without concentration, it can be challenging to detect cells from suspensions having as many as 105 CFU/mL, depending on the magnification used and the volume spotted to the slide.5 Counting chambers, dried films on microscope slides, and membrane filters are some of the accessories needed to quantify microbial populations by microscopy.

4.52

Counting Chambers for Bacteria

Numerous counting chambers are available including the Helber, the Hawksley, the Petroff-Hausser, and the hemocytometer.114 All are similar: each consists of a grid of etched squares of a given area and is covered with a glass slip that is positioned a fixed distance from the etched surface. Counts are usually made at about 4006 magnification, although

| Microscopic Methods

some chambers such as the Hawksley permit the use of an oil immersion lens. The volume of liquid within the etched square equals the area of the square times the depth of the film. The average cell count per square multiplied by the reciprocal of the volume in milliliters (i.e., the chamber factor) will equal the concentration of the microorganisms in the diluent. Chamber factors commonly range from 4 6 106 to 2 6 107; thus the procedure is most applicable to foods that contain large microbial populations. Major sources of error are the difficulty of accurate filling of chambers and the adsorption of cells to glassware surfaces.114 When using a microscope in which the focus dial reads directly in micrometers, the exact chamber thickness can be measured by focusing on cells attached to the coverslip and to the bottom of the slide. Adsorption can be reduced by using anionic detergent diluents and plastictipped pipettes. The material to be analyzed should be prepared in diluent, for example, 0.1% peptone water containing 0.1% lauryl sulfate, so that the concentration of bacteria will equal 5 to 15 cells per small square of the counting chamber grid. The diluted material is added to fill the counting chamber and allowed to settle for about 5 min. Using phase contrast, a sufficient number of squares is counted to give a total count of about 600 cells. The number of microbial cells per gram or milliliter is calculated by multiplying the average count per small square by the chamber factor by the dilution factor.

4.53

Howard Mold Count

The Howard mold count was established over a century ago98 to ensure that ketchup would be made from tomatoes that were relatively free of visible rots. Although most widely applied to tomato products (e.g., ketchup, juice, paste, sauce, canned tomatoes, soup, pizza sauce), the Howard mold count has also been used to assure the quality of other foods such as frozen berries, cranberry sauce, citrus and pineapple juices, fruit nectars and purees, and pureed infant food. The mold count is a standardized procedure9 to determine the percentage of microscopic fields containing mold filaments whose combined lengths exceed one-sixth the diameter of the field. The U.S. Food and Drug Administration has established for many food products regulatory action guidelines that include Howard mold count criteria.196 The analyst must be familiar with the microscopic appearance of sound food tissue and with the morphology of the more common molds in order to distinguish mold filaments from other fibers. Methodologies for preparation of individual food products for Howard mold counting have been established.9 The food sample is placed on the center of a Howard slide and spread evenly with a scalpel. The coverslip is lowered rapidly so that the material is distributed evenly over the center of the slide but not drawn across the moat. Proper contact between the coverslip and the slide can be confirmed by observation of colored bands known as Newton’s rings.9 Twenty-five fields from two or more mounts should be counted. Positive fields are those in which (1) a single filament

exceeds one-sixth of the field diameter, (2) a filament plus the length of its branches exceeds one-sixth of the field diameter, or (3) an aggregate of not more than three filaments exceeds one-sixth of the field. Results are calculated as percentage of positive fields.

4.54

Geotrichum Count

The incidence of filaments of the mold Geotrichum candidum in canned and frozen fruit and vegetable products is an indicator of the hygienic condition of the food processing equipment. Geotrichum can grow on the surfaces of processing equipment207 and thus has been termed ‘‘machinery mold.’’ The method involves microscopic counting of typical filaments using a rot fragment slide at 306 to 456 magnification.8,48 Low filament counts, under one per gram, often do not correlate well with the aerobic plate counts on frozen vegetables.180 Specific procedures for various types of foods have been established, and the following procedure describes the determination of Geotrichum mold in canned vegetables, fruits, and juices.11 The net weight of the can contents is determined, then drained on a No. 8 sieve positioned over a pan. The food is removed from the sieve and discarded, and the container and sieve are washed with about 300 mL water. The liquid and washings are transferred to a 5 in. No. 16 sieve that rests on a 2 liter beaker. The residue is washed with 50 mL water, and the residue is discarded. The liquid and washings are transferred to a 5 in. No. 230 sieve, tilted at a 30u angle. The solids are flushed to the sieve edge with a wash bottle, transferred to a 50 mL graduated centrifuge tube, and diluted to 10 mL. If the initial volume of the residue exceeds 10 mL, it should be concentrated by centrifugation. One drop of crystal violet is added to the 10 mL of residue, the tube is mixed, and then 10 mL of stabilizer solution is added to bring the total volume to 20 mL. After thorough mixing, 0.5 mL is transferred as a streak approximately 4 cm long to a rot fragment counting slide. Duplicate slides should be prepared and counted. The entire surface of the slide is examined at 306 to 456, using transmitted light. Geotrichum fragments that contain three or more hyphal branches at 45u angles from the main filament are counted. The Geotrichum count per 500 g of food is equal to (S/V) 6 (500/W) 6 20, where S equals the average number of fragments per slide, V is the total volume counted (0.5 mL per slide multiplied by the number of slides counted), and W equals the net weight of sample in grams.

4.55

Dried Films

The microscopic examination of a thin film of food dried onto a slide is one of the simplest microbiological techniques available. The method can be used to determine the morphological types of bacteria within a food sample, for example, staphylococci36 or endospore formers,123 but it can also be used for quantitation of microbial populations in a food. Although a quantitative dried film procedure for examination of milk was originally developed a century ago31 and has been widely used for grading of milk,78 similar methods have been applied for the examination of liquid eggs and other foods.13,36 The general principle is that a known quantity of food is spread over a prescribed area | 55

Compendium of Methods for the Microbiological Examination of Foods |

of a microscope slide. After the film is dried and stained, the average number of organisms per microscope field is determined. This count can be converted into numbers per gram or milliliter of food based on the area of the microscope field. Advantages of the method are that it is rapid and that the slides may be retained for later reference. A disadvantage is that it is applicable only to foods containing large populations of microorganisms. The microscope slides used for dried films have one or more circular 1 cm2 (diameter 11.28 mm) areas circumscribed by either painted or etched rings, which are used to contain the material for analysis in a defined area. The material (0.01 mL) is transferred into the circle with a pipette, spread uniformly over the area with a bent needle, and air-dried at 40uC to 45uC. The film is fixed by immersion in 95% ethanol. If the material contains a high fat content, the film may be defatted by immersing in xylene for 1 to 2 min followed by washing in methanol and drying. After fixation, the film is stained. North’s aniline oil methylene blue, which is recommended for liquid eggs, may be used to stain the film for 10 to 20 min. The number of microbial cells in 10 to 100 fields is counted, and the average count per field is determined. The next task in the calculation of microbial cell concentration in the food is to determine the microscope factor (MF). Each microscope lens will have a different MF; for typical bacterial counts, the 1006 lens is used. The MF is the number of microscope fields in the 1 cm2 slide area, divided by the volume of material (0.01 mL) applied to the slide. The area of the microscope field is determined by viewing a stage micrometer through the microscope (1006 objective lens in place) and measuring the diameter of the field in millimeters to the third decimal place. The field area in centimeters squared is calculated by applying the formula for the area of a circle (pr2), that is, dividing the field diameter by 2 to get the radius, squaring the radius, and multiplying by 3.1416; and then dividing by 100 to convert millimeters squared to centimeters squared. The MF is obtained by dividing the slide area (1 cm2) by the field area to obtain the number of microscope fields in the slide area and then dividing this quantity by 0.01, which is the volume in milliliters spread over the slide area. Therefore, in condensed form, MF 5 10,000 divided by pr2 where r equals the radius of the microscope field in millimeters. Finally, the microbial cell concentration in the food is determined by multiplying the average count per field by the MF. If a dilution of the food material was made prior to analysis, the calculated concentration should be multiplied by the reciprocal of the sample dilution. Results may be expressed as the number of microbial cells per milliliter of food.

4.56

Direct Epifluorescent Filter Technique

Membrane filtration may be used to concentrate material for analysis by collecting it on the filter surface. Concentrating the material, including the microbial cells present within it, provides an effective way to lower the limit of detection of an analytical method, thus increasing its sensitivity. Membrane filtration has been combined with 56 |

microscopy in the DEFT155 to increase sensitivity by several orders of magnitude over other microscopic methods. The membrane filter not only concentrates the cells but also provides the surface upon which the microscopic examination is made. Incident light as in epifluorescence, rather than transmitted light, is the type of illumination required for examination of the membrane filter surface. Procedures have been developed for analysis of milk and other beverages,94,165,194 diluted food homogenates,157 and surface rinses100 by DEFT. Although DEFT has primarily been used for enumeration of single cells and cell clumps, the transfer of membranes to selective agar media following filtration also permits the counting of microcolonies.61,146 The methods differ depending upon the food type, the microorganisms of interest, and whether cells or microcolonies are to be enumerated. Various food suspensions, for example, will require different pre-treatments or enzymatic digestions to allow passage through the membrane filter. Acridine orange is the stain most commonly used in the DEFT for quantitation of the total microbial population in a food sample. Binding of this fluorescent dye to singlestranded RNA or double-stranded DNA results in redorange or green fluorescence, respectively, and because growing cells have an abundance of RNA, the characteristic red-orange fluorescence originally was associated with cellular viability96; however, the differentiation depends on the matrix in which the bacterial cells are suspended and does not hold under a wide variety of conditions.110 Acridine orange is now recognized to be more appropriately used as a general stain rather than an indicator of physiologically active cells. Fluorescent antibodies have also been applied for specific enumeration of microbial cells,181,191,192 and a hybrid membrane-filtration/microcolony approach for PNA-based in situ chemiluminescent detection of E. coli, Pseudomonas, and S. aureus has been described.154 The following procedure describes the enumeration of microbial cells in raw milk.43,156 All reagents should be filter-sterilized through 0.22 mm pore size membrane filters before use in the DEFT. Somatic cells and lipid micelles in milk are lysed by adding 0.5 mL rehydrated Bacto-trypsin (BD-Difco) and 2 mL of 0.5% Triton X-100 to 2 mL of milk. The mixture is incubated at 50uC for 10 min, then added to a previously warmed filter assembly (Millipore) holding a 25 mm diameter, 0.6 mm pore size black Nuclepore polycarbonate membrane (shiny side up). A vacuum is applied to filter the digested milk, and the assembly is rinsed with 5 mL pre-warmed 0.1% Triton X-100. The vacuum is disconnected, and the membrane is overlayed with 2 mL of acridine orange stain for 2 min. The membrane is rinsed under vacuum with 2.5 mL 0.1 M citrate-NaOH buffer, pH 3, followed by 2.5 mL 95% ethyl alcohol. After air-drying, the filter is mounted on a slide in a drop of nonfluorescent immersion oil, and a coverslip is applied. It is examined using an epifluorescence microscope fitted with an appropriate fluorescence filter combination for acridine orange. Orange fluorescent microbial cells are counted in randomly selected microscope fields around the filter. The number of fields that should be counted depends on

| Microscopic Methods

the microbial cell density per field. For fields with 0 to 10 cells, 15 fields are counted; for 11 to 25 cells, 10 fields; for 26 to 50 cells, 6 fields; for 51 to 75 cells, 3 fields; and for 76 to 100 cells, 2 fields. If there are more than 100 cells per field, the sample should be diluted before analysis. As in the dried film procedure (Section 4.54), the number of cells per milliliter is obtained by multiplying the average number of cells per field by the MF; in the DEFT, the MF calculation is based on the area of the membrane filter. The membrane filter microscope factor (MFMF) is calculated by determining the area of the membrane through which the sample was filtered (Figure 4-4A), in millimeters squared, and the area of the microscope field of view (Figure 4-4B), in millimeters squared (Section 4.55). The MFMF is determined by dividing A by B. The microbial cell concentration is calculated by multiplying the average number of cells per field by the MFMF and dividing by the volume of material filtered. If a dilution of the original material was made prior to filtration, the reciprocal of this dilution should be multiplied into the calculation.

4.6

IMAGE PROCESSING AND ANALYSIS

Microscopy is an inherently visual technique. Much can be accomplished simply by looking through the eyepiece or at photographs taken with a microscope. The microscopist must then interpret the image, based primarily on experience. However, in the age of digital imaging, evaluation of microscopy data is no longer limited to only such qualitative analyses. Digital images contain a wealth of information that can be extracted and analyzed quantitatively and exported for additional examination. The history and technical details of digital imaging technology have been reviewed extensively elsewhere.93,105 This section provides a general description of image processing software currently available for analysis of digital images. Most microscopes today come equipped with at least a basic version of the manufacturer’s image handling software. These programs enable image acquisition and simple processing of raw digital images, including alteration of brightness and contrast, cropping, cutting, and labeling. Basic analytical operations, such as quantitative measurement of features in an image are accomplished by highlighting regions of interest, provided the system has been calibrated with a stage micrometer. More sophisticated means for extracting data from digital images typically involve serial manipulations, including preprocessing steps such as conversion to grayscale and contrast or intensity adjustments, followed by edge detection, background subtraction, pixel analysis, and other iterative steps involved in feature detection and measurement.34,136 These actions would be tedious, prone to error, and could only be performed on limited numbers of images if carried out manually. Commercial software designed for robust extraction and analysis of data from digital images is available but expensive. As an alternative, a number of software packages that support automated processing of digital images can be freely downloaded. The more versatile packages include ImageJ,51 CellProfiler,122 and daime.57 Others include CellC172 and bioImage_L,45 which have fewer capabilities or are designed for more limited tasks. The features of these software packages and

examples of how some have been applied in food microbiology are discussed further below. ImageJ is an updated version of NIH Image, a Macintosh-based program developed at the National Institutes of Health. Because it is programmed in the Java language, ImageJ can run on Macintosh, Windows, or Linux computers. As an open-source program, users in the ImageJ community have contributed over 400 freely available plugins—subprograms that add functionality to the base software.51 Plugins are available for all aspects of digital imaging, from acquisition to processing and analysis. Plugin-supported operations range from recognition of proprietary image formats and associated metadata containing key information, such as camera or illumination settings and pixel size, to cell counting or tracking and nanoparticle localization within cells. Additional plugins support more advanced manipulations, such as deconvolution, an algorithmic method for correction of image blurriness caused by stray light.51,159 Deblurring of a fluorescence microscopic image of Candida albicans through the use of digital imaging software is illustrated in Figure 4-4. Variants of ImageJ are available; they come prebundled with plugins commonly used by life scientists, such as Fiji (http://fiji.sc), and a next-generation version of ImageJ with improved abilities to interface with other imaging programs such as CellProfiler is now in beta development. CellProfiler is another open source program that runs on multiple platforms (Macintosh, Windows, Linux). Like ImageJ, CellProfiler was developed to further ease the

Figure 4-4. Deblurring of a fluorescence microscopy image using two-dimensional (2d) blind deconvolution Cells of Candida albicans stained with a C. albicans-targeted DNAFISH probe were photographed using a consumer-grade digital camera attached to a Leitz Laborlux S fluorescence microscope equipped with a 636 oil immersion lens (N.A. 1.4). The resulting photographs (Panel A) are blurry and lacking in sharp detail, due to scattering of the probe-conferred fluorescence signal, as is typical of widefield fluorescence images. Panel B shows the impact on image clarity of 60 iterations of 2D blind deconvolution, performed using AutoQuant X software (ver. 2.2; Media Cybernetics, Silver Spring, MD). Blind deconvolution uses an iterative mathematical process that allows image deblurring without extensive prior knowledge of how the image was collected. Artifacts that can be corrected include haze due to scattered light and thermal noise from chargecoupled device cameras. Adapted from Bisha, Kim, and BrehmStecher.26

| 57

Compendium of Methods for the Microbiological Examination of Foods |

adoption of advanced image analysis by the scientific community. The program is user friendly, due to its graphical user interface and modular design.42 Individual modules within the program each perform a specific function. Example modules include ColorToGray, MeasureObjectIntensity, EnhanceEdges, ClassifyObjects, and ExportToSpreadsheet. Users select and arrange modules to build custom ‘‘pipelines’’ for image processing and analysis. Due to its modular nature, no knowledge of programming is required of the user. Together, these features make CellProfiler a flexible and highly accessible tool for advanced image analysis. For more complex analyses involving multiple measurements and images (i.e., hundreds of features per cell, millions of cells) the same group has developed an additional program, CellProfilerAnalyst.104 CellProfilerAnalyst allows the user to explore relationships among image data for large populations of cells, creating flow cytometry-like outputs such as histograms, scatter plots, and density plots.104 Other freely available image analysis programs include bioImage_L (Windows), designed for dual-color studies of biofilm viability and metabolic activity,45 and daime (Linux), whose acronymic title stands for ‘‘digital image analysis in microbial ecology.’’57 As its title suggests, daime was created for analysis of microbial communities stained with rRNAtargeted and other fluorescent probes often used by environmental microbiologists. Special emphases of the program include three-dimensional visualization of confocal microscopy data and quantitation of cellular fluorescence intensity or spatial arrangement of specifically stained cells.57,171 CellC (Windows) is a task-dedicated program that compares images of bacterial cells stained with both non-specific (DAPI) and specific (rRNA-targeted probe) labels. Images of the same microscopic field of view are acquired using the total count (TC) and specific count (SC) stains. The software then performs a series of sequential operations on each image, including correction of image background for variations in brightness and thresholding of cell-associated pixels. The result is a binarized image of white cells against a black background.172 A segmentation algorithm is applied to separate clustered cells from each other, and objects one-tenth the size of all mean objects are automatically discarded to remove staining artifacts. User-defined thresholds for artifact removal may also be set. The software compares the TC and SC images, identifies objects that were visible in both images, and calculates the dimensions (length, width, approximate volume) of these cells. Cell count and dimension data are exported to a comma-separated values (CSV) file accessible with Excel or other spreadsheet software, and the program can be run in batch mode, allowing analysis of multiple images without operator intervention.172 As with ImageJ and CellProfiler, CellC is an open-source program whose MATLAB-based code may be modified by the end user. CellC has been used to investigate biofilm formation in C. perfringens,199 for quantification of spore harvest purity in B. subtilis91 and to characterize the role of calcium ions in induction of the Yersinia type 3 secretion system.208 The wide availability of sophisticated, yet user-friendly, freeware for advanced image analysis promises to extend 58 |

the capabilities of the microscope beyond its traditional use as a qualitative or descriptive tool, enabling researchers to access new reservoirs of quantitative information encoded in digital microscopy images.

4.7

CONCLUSIONS AND FUTURE PERSPECTIVES

Although microscopy has been a staple analytical tool since the dawn of microbiology, microscopic techniques continue to evolve. Advances in optical methods such as two-photon or light sheet microscopies enable imaging of specimens thicker than is possible with CLSM, and various superresolution microscopies now allow resolution of features smaller than the wavelength of light. While these are specialized techniques that may offer more than is needed for the routine detection and examination of pathogens in foods, many of these approaches have been commercialized, opening the possibility for their future use by food microbiologists in much the same way that electron microscopy, once an esoteric technique, is now widely used. In the short term, more modest advances, such as the development of portable, automated, and/or more compact and user-friendly instrumentation or analysis software will impact food microbiologists and their work most directly. As microscopy technology continues to evolve, options for the microbiological analysis of foods may extend to tools such as lensless on-chip optofluidic microscopy,55 cell phone-based imaging, cell counting and data transmission platforms,178 and high-throughput microfluidic imaging systems that blur distinctions between microscopy and flow cytometry and allow analysis rates of 100,000 particles per second.83 Continued advances in microscopy-compatible reagents and detection methods will complement these technological advances.

ACKNOWLEDGMENT Fourth edition authors: Jeffrey W. Bier, Don F. Splittstoesser, and Mary Lou Tortorello.

REFERENCES 1. Allison, D. G., and M. A. Sattenstall. 2007. The influence of green fluorescent protein incorporation on bacterial physiology: a note of caution. J. Appl. Microbiol. 103:318-324. 2. Almeida, C., N. F. Azevedo, C. Iversen, S. Fanning, C. W. Keevil, and M. J. Vieira. 2009. Development and application of a novel peptide nucleic acid probe for the specific detection of Cronobacter genomospecies (Enterobacter sakazakii) in powdered infant formula. Appl. Environ. Microbiol. 75:2925-2930. 3. Almeida, C., N. F. Azevedo, R. M. Fernandes, C. W. Keevil, and M. J. Vieira. 2010. Fluorescence in situ hybridization method using a peptide nucleic acid probe for identification of Salmonella spp. in a broad spectrum of samples. Appl. Environ. Microbiol. 76:4476-4485. 4. Amann, R., and B. M. Fuchs. 2008. Single cell identification in microbial communities by improved fluorescence in situ hybridization techniques. Nat. Rev. Microbiol. 6:339-348. 5. Amann, R. I., W. Ludwig, and K.-H. Schleifer. 1995. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59:143-169. 6. Angelidis, A. S., I. Tirodimos, M. Bobos, M. S. Kalamaki, D. K. Papageorgiou, and M. Arvanitidou. 2011. Detection of

| Microscopic Methods

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

Helicobacter pylori in raw bovine milk by fluorescence in situ hybridization (FISH). Int. J. Food Microbiol. 151:252-256. Anonymous. 2009. Isolation and identification of Listeria monocytogenes from red meat, poultry, egg, and environmental samples. Chapter 8.07. In: B. P. Dey and C. P. Lattuada (Editors). Microbiology Laboratory Guidebook. U.S. Department of Agriculture/Food Safety and Inspection Service. http://www.fsis.usda.gov/PDF/MLG_ 8_07.pdf. Accessed August 23, 2012. AOAC International. 2005. Geotrichum mold counting. Official Method 984.30. In: W. Horwitz and G. W. Latimer, Jr. (Editors). Official methods of analysis. 18th ed. AOAC International, Gaithersburg, MD. AOAC International. 2005. Howard mold counting, general instructions. AOAC Official Method 984.29. In: W. Horwitz and G. W. Latimer, Jr. (Editors). Official methods of analysis. 18th ed. AOAC International, Gaithersburg, MD. AOAC International. 2005. Invasiveness of mammalian cells by Escherichia coli, microbiological method. AOAC Official Method 982.36. In: W. Horwitz and G. W. Latimer, Jr. (Editors). Official Methods of Analysis. 18th ed. AOAC International, Gaithersburg, MD. AOAC International. 2005. Mold in vegetables, fruits and juices (canned), Geotrichum mold count. AOAC Official Method 974.34. In: W. Horwitz and G. W. Latimer, Jr. (Editors). Official Methods of Analysis. 18th ed. AOAC International, Gaithersburg, MD. AOAC International. 2005. Salmonella in foods, fluorescent antibody screening method. AOAC Official Method 975.54. In: W. Horwitz and G. W. Latimer, Jr. (Editors). Official methods of analysis. 18th ed. AOAC International, Gaithersburg, MD. AOAC International. 2005. Techniques for eggs and egg products, microbiological methods. AOAC Official Method 940.37. In: W. Horwitz and G. W. Latimer, Jr. (Editors). Official methods of analysis. 18th ed. AOAC International, Gaithersburg, MD. Autio, K., and T. Mattila-Sandholm. 1992. Detection of active yeast cells (Saccharomyces cerevisiae) in frozen dough sections. Appl. Environ. Microbiol. 58:2153-2157. Auty, M., G. Duffy, D. O’Beirne, A. McGovern, E. Gleeson, and K. Jordan. 2005. In situ localization of Escherichia coli O157:H7 in food by confocal scanning laser microscopy. J. Food Prot. 68:482-486. Auty, M. A., G. E. Gardiner, S. J. McBrearty, E. O. O’Sullivan, D. M. Mulvihill, J. K. Collins, G. F. Fitzgerald, C. Stanton, and R. P. Ross. 2001. Direct in situ viability assessment of bacteria in probiotic dairy products using viability staining in conjunction with confocal scanning laser microscopy. Appl. Environ. Microbiol. 67:420-425. Barer, M. R., and C. R. Harwood. 1999. Bacterial viability and culturability. In: R. K. Poole (Editor). Advances in Microbial Physiology. Vol. 41. Academic Press, London, U.K., 93-137. Baudart, J., J. Coallier, P. Laurent, and M. Pre´vost. 2002. Rapid and sensitive enumeration of viable diluted cells of members of the family Enterobacteriaceae in freshwater and drinking water. Appl. Environ. Microbiol. 68:5057-5063. Baumstummler, A., R. Chollet, H. Meder, F. Olivieri, S. Rouillon, G. Waiche, and S. Ribault. 2010. Development of a nondestructive fluorescence-based enzymatic staining of microcolonies for enumerating bacterial contamination in filterable products. J. Appl. Microbiol. 110:69-79. Berlin, O. G. W., J. B. Peter, C. Gagne, C. N. Conteas, and L. R. Ash. 1998. Autofluorescence and the detection of Cyclospora oocysts. Emerg. Infect. Dis. 4:127-128. Berney, M., F. Hammes, F. Bosshard, H. U. Weilenmann, and T. Egli. 2007. Assessment and interpretation of bacterial

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

viability by using the LIVE/DEAD BacLight kit in combination with flow cytometry. Appl. Environ. Microbiol. 73:32833290. Betts, R. P., P. Bankes, and J. G. Banks. 1989. Rapid enumeration of viable microorganisms by staining and direct microscopy. Lett. Appl. Microbiol. 9:199-202. Beveridge, T. J., J. R. Lawrence, and R. G. E. Murray. 2007. Sampling and staining for light microscopy. In: C. A. Reddy, T. J. Beveridge, J. A. Breznak, G. Marzluf, T. M. Schmidt, and L. R. Snyder (Editors). Methods for General and Molecular Bacteriology. 3rd ed. American Society for Microbiology, Washington, D.C., 19-33. Bier, J. W., G. J. Jackson, A. M. Adams, and R. A. Rude. 2001. Parasitic animals in foods. In: Bacteriological analytical manual. U.S. Food and Drug Administration. http://www.fda.gov/ Food/FoodScienceResearch/LaboratoryMethods/ucm071468. htm. Accessed April 16, 2015. Bisha, B., and B. F. Brehm-Stecher. 2009. Flow-through imaging cytometry for characterization of Salmonella subpopulations in alfalfa sprouts, a complex food system. Biotechnol. J. 4:880-887. Bisha, B., H. J. Kim, and B. F. Brehm-Stecher. 2011. Improved DNA-FISH for cytometric detection of Candida spp. J. Appl. Microbiol. 110:881-892. Bisha, B., A. Perez-Mendez, M. D. Danyluk, and L. D. Goodridge. 2011. Evaluation of modified Moore swabs and continuous flow centrifugation for concentration of Salmonella and Escherichia coli O157:H7 from large volumes of water. J. Food Prot. 74:1934-1937. Bottari, B., D. Ercolini, M. Gatti, and E. Neviani. 2006. Application of FISH technology for microbiological analysis: current state and prospects. Appl. Microbiol. Biotechnol. 73:485-494. Branco, P., M. Monteiro, P. Moura, and H. Albergaria. 2012. Survival rate of wine-related yeasts during alcoholic fermentation assessed by direct live/dead staining combined with fluorescence in situ hybridization. Int. J. Food Microbiol. 158:49-57. Brayton, P. R., M. L. Tamplin, A. Huq, and R. R. Colwell. 1987. Enumeration of Vibrio cholerae O1 in Bangladesh waters by fluorescent antibody direct viable count. Appl. Environ. Microbiol. 53:2862-2865. Breed, R. S. 1911. The determination of bacteria in milk by direct microscopic examination. Zentralbl. Bakteriol. II. Abt. 30:337-340. Brehm-Stecher, B. F. 2008. Methods for whole cell detection of microorganisms. In: Structure, Interaction and Reactivity at Microbial Surfaces. T. Camesano and C. Mello (Editors). American Chemical Society, Washington, D.C., 29-51. Brehm-Stecher, B. F., J. J. Hyldig-Nielsen, and E. A. Johnson. 2005. Design and evaluation of 16S rRNA-targeted peptide nucleic acid probes for whole cell detection of the genus Listeria. Appl. Environ. Microbiol. 71:5451-5457. Brehm-Stecher, B. F., and E. A. Johnson. 2004. Single-cell microbiology: tools, technologies and applications. Microbiol. Mol. Biol. Rev. 68:538-559. Brehm-Stecher, B. F., and E. A. Johnson. 2012. Isolation of carotenoid hyperproducing mutants of Xanthophyllomyces dendrorhous (Phaffia rhodozyma) by flow cytometry and cell sorting. Methods Mol. Biol. 898:207-217. Bryce, J. R., and P. L. Poelma. 2001. Microscopic examination of foods and care and use of the microscope. Chapter 2. In: Bacteriological Analytical Manual. U.S. Food and Drug Administration. http://www.fda.gov/Food/ FoodScienceResearch/LaboratoryMethods/ucm063344.htm. Accessed April 16, 2015.

| 59

Compendium of Methods for the Microbiological Examination of Foods |

37. Buchrieser, C., and C. W. Kaspar. 1993. An improved direct viable count for the enumeration of bacteria in milk. Int. J. Food Microbiol. 20:227-236. 38. Bunthof, C. J., S. van Schalkwijk, W. Meijer, T. Abee, and J. Hugenholtz. 2001. Fluorescent method for monitoring cheese starter permeabilization and lysis. Appl. Environ. Microbiol. 67:4264-4271. 39. Burnett, S. L., and L. R. Beuchat. 2002. Comparison of methods for fluorescent detection of viable, dead and total Escherichia coli O157:H7 cells in suspensions and on apples using confocal scanning laser microscopy following treatment with sanitizers. Int. J. Food Microbiol. 74:37-45. 40. Burt, S. A., R. van der Zee, A. P. Koets, A. M. de Graaff, F. van Knapen, W. Gaastra, H. Haagsman, H, and E. J. A. Veldhuizen. 2007. Carvacrol induces heat shock protein 60 and inhibits synthesis of flagellin in Escherichia coli O157:H7. Appl. Environ. Microbiol. 73:4484-4490. 41. Cameron, M., L. D. McMaster, and T. J. Britz. 2008. Electron microscopic analysis of dairy microbes inactivated by ultrasound. Ultrason. Sonochem. 15:960-964. 42. Carpenter, A. E., T. R. Jones, M. R. Lamprecht, C. Clarke, I. H. Kang, O. Friman, D. A. Guertin, J. H. Chang, R. A. Lindquist, J. Moffat, P. Golland, and D. M. Sabatini. 2006. CellProfiler: image analysis software for identifying and quantifying cell phenotypes. Genome Biol. 7:R100. doi:10.1186/gb-2006-7-10-r100. 43. Champagne, C. P., N. J. Gardner, J. Fontaine, and J. Richard. 1997. Determination of viable bacterial populations in raw milk within 20 minutes by using a direct epifluorescent filter technique. J. Food Prot. 60:874-876. 44. Chantarapanont, W., M. Berrang, and J. F. Frank. 2003. Direct microscopic observation and viability determination of Campylobacter jejuni on chicken skin. J. Food Prot. 66:22222230. 45. Cha´vez de Paz, L. 2009. Image analysis software based on color segmentation for characterization of viability and physiological activity of biofilms. Appl. Environ. Microbiol. 75:1734-1739. 46. Chen, S. Y., W. N. Jane, Y. S. Chen, and H. C. Wong. 2009. Morphological changes of Vibrio parahaemolyticus under cold and starvation stresses. Int. J. Food Microbiol. 129:157-165. 47. Chia, T. W., R. M. Goulter, T. McMeekin, G. A. Dykes, and N. Fegan. 2009. Attachment of different Salmonella serovars to materials commonly used in a poultry processing plant. Food Microbiol. 26:853-859. 48. Cichowicz, S. M., and W. V. Eisenberg. 1974. Collaborative study of the determination of Geotrichum mold in selected canned fruits and vegetables. J. Assoc. Off. Anal. Chem. 57:957-960. 49. Clontech Laboratories, Inc. 2012. Selection guide: fluorescent proteins. Accessed September 24, 2012. http://www. clontech.com/US/Products/Fluorescent_Proteins_and_ Reporters/Fluorescent_Proteins/Fluorescent_Proteins_ Selection_Tool?sitex510020:22372:US. 50. Coleman, J. R., D. E. Culley, W. B Chrisler, and F. J. Brockman. 2007. mRNA-targeted fluorescent in situ hybridization (FISH) of Gram-negative bacteria without template amplification or tyramide signal amplification. J. Microbiol. Methods 71:246-255. 51. Collins, T. J. 2007. ImageJ for microscopy. BioTechniques 43:S25-S30. 52. Comas-Riu, J., and N. Rius. 2009. Flow cytometry applications in the food industry. J. Ind. Microbiol. Biotechnol. 36:999-1011. 53. Creach, V., A. C. Baudoux, G. Bertru, and B. L. Rouzic. 2003. Direct estimate of active bacteria: CTC use and limitations. J. Microbiol. Methods. 52:19-28.

60 |

54. Crespo-Sempere, A., M. Lo´ pez-Pe´rez, P. V.Martı´nezCulebras, and L. Gonza´lez-Candelas. 2011. Development of a green fluorescent tagged strain of Aspergillus carbonarius to monitor fungal colonization in grapes. Int. J. Food Microbiol. 148:135-140. 55. Cui, X., L. M. Lee, X. Heng, W. Zhong, P. W. Sternberg, D. Psaltis, and C. Yang. 2008. Lensless high-resolution onchip optofluidic microscopes for Caenorhabditis elegans and cell imaging. Proc. Natl. Acad. Sci. U. S. A. 105:1067010675. 56. Cui, Y., Y. J. Oh, J. Lim, M. Youn, I. Lee, H. K. Pak, W. Park, W. Jo, and S. Park. 2012. AFM study of the differential inhibitory effects of the green tea polyphenol (2)-epigallocatechin-3-gallate (EGCG) against Gram-positive and Gramnegative bacteria. Food Microbiol. 29:80-87. 57. Daims, H., S. Lu¨cker, and M. Wagner. 2006. daime, A novel image analysis program for microbial ecology and biofilm research. Environ. Microbiol. 8:200-213. 58. Danyluk, M. D., M. T. Brandl, and L. J. Harris. 2008. Migration of Salmonella Enteritidis phage type 30 through almond hulls and shells. J. Food Prot. 71:397-401. 59. Davey, H. M., and D. B. Kell. 1996. Flow cytometry and cell sorting of heterogeneous microbial populations: the importance of single-cell analyses. Microbiol. Mol. Biol. Rev. 60:641-696. 60. DeLong, E. F., G. S. Wickham, and N. R. Pace. 1989. Phylogenetic stains: ribosomal RNA-based probes for the identification of single cells. Science. 243:1360-1363. 61. Desmonts, C., J. Minet, R. Colwell, and M. Cormier. 1990. Fluorescent antibody method useful for detecting viable but nonculturable Salmonella spp. in chlorinated wastewater. Appl. Environ. Microbiol. 56:1448-1452. 62. Diaper, J. P., K. Tither, and C. Edwards. 1992. Rapid assessment of bacterial viability by flow cytometry. Appl. Microbiol. Biotechnol. 38:268-272. 63. Doherty, S. B., V. L. Gee, R. P. Ross, C. Stanton, G. F. Fitzgerald, and A. Brodkorb. 2010. Efficacy of whey protein gel networks as potential viability-enhancing scaffolds for cell immobilization of Lactobacillus rhamnosus GG. J. Microbiol. Methods. 80:231-241. 64. Donnelly, C. W., and G. J. Baigent. 1986. Method for flow cytometric detection of Listeria monocytogenes in milk. Appl. Environ. Microbiol. 52:689-695. 65. Dorobantu, L. S., G. G. Goss, and R. E. Burrell. 2012. Atomic force microscopy: a nanoscopic view of microbial cell surfaces. Micron 43:1312-1322. 66. Dorobantu, L. S., and M. R. Gray. 2010. Application of atomic force microscopy in bacterial research. Scanning 32:74-96. 67. Duffy, E. A., L. Cisneros-Zevallos, A. Castillo, S. D. Pillai, S. C. Ricke, and G. R. Acuff. 2005. Survival of Salmonella transformed to express green fluorescent protein on Italian parsley as affected by processing and storage. J. Food Prot. 68:687-695. 68. Dupres, V., D. Alsteens, G. Andre, and Y. F. Dufrene. 2010. Microbial nanoscopy: a closer look at microbial cells surfaces. Trends Microbiol. 18:397-405. 69. Dwivedi, H. P., R. D. Smiley, and L.-A. Jaykus. 2010. Selection and characterization of DNA aptamers with binding selectivity to Campylobacter jejuni using whole-cell SELEX. Appl. Microbiol. Biotechnol. 87:2323-2334. 70. Egholm, M., O. Buchardt, L. Christensen, C. Behrens, S. M. Freier, D. A. Driver, R. H. Berg, S. K. Kim, B. Norden, and P. E. Nielsen. 1993. PNA hybridizes to complementary oligonucleotides obeying the Watson-Crick hydrogen-bonding rules. Nature 365:566-568. 71. Ercolini, D., F. Villani, M. Aponte, and G. Mauriello. 2006. Fluorescence in situ hybridisation detection of Lactobacillus

| Microscopic Methods

72.

73.

74.

75.

76.

77.

78.

79.

80.

81. 82.

83.

84.

85.

86.

87.

plantarum group on olives to be used in natural fermentations. Int. J. Food Microbiol. 112:291-296. Erni, R., M. D. Rossell, C. Kisielowski, and U. Dahmen. 2009. Atomic-resolution imaging with a sub-50-pm electron probe. Phys. Rev. Lett. 102:96-101. Fang, Q., S. Brockmann, K. Botzenhart, and A. Wiedenmann. 2003. Improved detection of Salmonella spp. in foods by fluorescent in situ hybridization with 23S rRNA probes: a comparison with conventional culture methods. J. Food Prot. 66:723-731. Fernandes, J. C., P. Eaton, A. M. Gomes, M. E. Pintado, and F. X. Malcata. 2009. Study of the antibacterial effects of chitosans on Bacillus cereus (and its spores) by atomic force microscopy imaging and nanoindentation. Ultramicroscopy. 109:854-860. Ferreira, A.-C., P. Vasconcellos Morais, and M. S. da Costa. 1994. Alterations in total bacteria, iodonitrophenyltetrazolium (INT)-positive bacteria, and heterotrophic plate counts of bottle mineral water. Can. J. Microbiol. 40:72-77. Fett, W. F., T. J. Fu, and M. L. Tortorello. 2006. Seed sprouts: the state of microbiological safety. In: K. R. Matthews (Editor). Microbiology of Fresh Produce. ASM Press, Washington, D.C., 167-219. First, M. R., N. Y. Park, M. E. Berrang, R. J. Meinersmann, J. M. Bernhard, R. J. Gast, and J. T. Hollibaugh. 2012. Ciliate ingestion and digestion: flow cytometric measurements and regrowth of a digestion-resistant Campylobacter jejuni. J. Eukaryot. Microbiol. 59:12-19. Fitts, J. E., and D. Laird. 2004. Direct microscopic methods for bacteria or somatic cells. Chapter 10. In: M. Wehr and J. F. Frank (Editors). Standard Methods for the Examination of Dairy Products. 17th ed. American Public Health Association, Washington, D.C., 269-280. Frank, J. F. 2001. Microbial attachment to food and food contact surfaces. In: S. Taylor (Editor). Advances in Food and Nutrition Research. Vol. 43. Academic Press, London, U.K., 319-370. Frank, J. F., M. A. Gassem, and R. A. N. Gillett. 1992. A direct viable count method suitable for use with Listeria monocytogenes. J. Food Prot. 55:697-700. Frost, W. D. 1915. Rapid method of counting bacteria in milk. Science 42:255-256. Gerhart, J., M. Baytion, J. Perlman, C. Neely, B. Hearon, T. Nilsen, R. Getts, J. Kadushin, and M. George-Weinstein. 2004. Visualizing the needle in the haystack: in situ hybridization with fluorescent dendrimers. Biol. Proced. Online 6:149-156. Goda, K., A. Ayazi, D. R. Gossett, J. Sadasivam, C. K. Lonappan, E. Sollier, A. M. Fard, S. C. Hur, J. Adam, C. Murray, C. Wang, N. Brackbill, D. Di Carlo, and B. Jalali. 2012. High-throughput single-microparticle imaging flow analyzer. Proc. Natl. Acad. Sci. U. S. A. 109:11630-11635. Golberg, D., Y. Kroupitski, E. Belausov, R. Pinto, and S. Sela. 2011. Salmonella typhimurium internalization is variable in leafy vegetables and fresh herbs. Int. J. Food Microbiol. 145:250-257. Gou, J., H.-Y. Lee, and J. Ahn. 2010. Inactivation kinetics and virulence potential of Salmonella typhimurium and Listeria monocytogenes treated by high pressure and nisin. J. Food Prot. 73:2203-2210. Goulter-Thorsen, R. M., E. Taran, I. R. Gentle, K. S. Gobius, and G. A. Dykes. 2011. Surface roughness of stainless steel influences attachment and detachment of Escherichia coli O157. J. Food Prot. 74:1359-1363. Graczyk, T. K., B. H. Grimes, R. Knight, A. J. Da Silva, N. J. Pienbiazek, and D. A. Veal. 2003. Detection of Cryptosporidium parvum and Giardia lamblia carried by synanthropic flies by combined fluorescent in situ hybridi-

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

99.

100.

101.

102.

103.

zation and a monoclonal antibody. Am. J. Trop. Med. Hyg. 68:228-232. Gre´gori, G., V. Patsekin, B. Rajwa, J. Jones, K. Ragheb, C. Holdman, and J. P. Robinson. 2012. Hyperspectral cytometry at the single-cell level using a 32-chanel photodetector. Cytometry A 81:35-44. Hamouda, T., A. Y. Shih, and J. R. Baker, Jr. 2002. A rapid staining technique for the detection of the initiation of germination of bacterial spores. Lett. Appl. Microbiol. 34:8690. Harkins, K. R., and K. Harrigan. 2004. Labeling of bacterial pathogens for flow cytometric detection and enumeration. Curr. Protoc. Cytom. 11.17.1-11.17.20. Harrold, Z. R., M. R. Hertel, and D. Gorman-Lewis. 2011. Optimizing Bacillus subtilis spore isolation and quantifying spore harvest purity. J. Microbiol. Methods 87:325-329. Harry, E. J., K. Pogliano, and R. Losick. 1995. Use of immunofluorescence to visualize cell-specific gene expression during sporulation in Bacillus subtilis. J. Bacteriol. 177:3386-3393. Hazelwood, K. L., S. G. Olenych, J. D. Griffin, J. A. Cathcart, and M. W. Davidson. 2007. Entering the portal: understanding the digital image recorded through a microscope. In: S. L. Shorte and F. Frischknecht (Editors). Imaging Cellular and Molecular Biological Function. Springer-Verlag, Berlin, Germany, 3-44. Hermida, M., M. Taboada, S. Menendez, and J. L. RodriguezOtero. 2000. Semi-automated direct epifluorescent filter technique for total bacterial count in raw milk. J. AOAC Int. 83:1345-1348. Hitchins, A. D., and K. Jinneman. 2011. Detection and enumeration of Listeria monocytogenes. Chapter 10. In: Bacteriological analytical manual. U.S. Food and Drug Administration. http://www.fda.gov/Food/ FoodScienceResearch/LaboratoryMethods/ucm071400.htm. Accessed April 16, 2015. Hobbie, J. E., R. J. Daley, and S. Jasper. 1977. Use of nucleopore filters for counting bacteria by fluorescence microscopy. Appl. Environ. Microbiol. 33:1225-1228. Holm, C., and L. Jespersen, 2003. A flow-cytometric gramstaining technique for milk-associated bacteria. Appl. Environ. Microbiol. 69: 2857-2863. Howard, B. J. 1911. Tomato catsup under the microscope with practical suggestions to insure a cleanly product. U.S. Department of Agriculture Bureau of Chemistry. Circular No. 68. Hunt, J. M., C. Abeyta, and T. Tran. 2001. Campylobacter. Chapter 7. In: Bacteriological analytical manual. U.S. Food and Drug Administration. http://www.fda.gov/Food/ FoodScienceResearch/LaboratoryMethods/ucm072616.htm. Accessed April 16, 2015. Hunter, A. C., and R. M. McCorquodale. 1983. Evaluation of the direct epifluorescent filter technique for assessing the hygienic condition of milking equipment. J. Dairy Res. 50:916. Isaacson, T., D. K. Kosma, A. J. Matas, G. J. Buda, Y. He, B. Yu, A. Pravitasari, J. D. Batteas, R. E. Stark, M. A. Jenks, and J. K. Rose. 2009. Cutin deficiency in the tomato fruit cuticle consistently affects resistance to microbial infection and biomechanical properties, but not transpirational water loss. Plant J. 60:363-377. Jeanson, S., J. Chadoeuf, M. N. Madec, S. Aly, J. Floury, T. F. Brocklehurst, and S. Lortal. 2011. Spatial distribution of bacterial colonies in a model cheese. Appl. Environ. Microbiol. 77:1493-1500. Jenkins, M. B., L. J. Anguish, D. D. Bowman, M. J. Walker, and W. C. Ghiorse. 1997. Assessment of a dye permeability assay for determination of inactivation rates of

| 61

Compendium of Methods for the Microbiological Examination of Foods |

104.

105.

106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

119.

120.

121.

62 |

Cryptosporidium parvum oocysts. Appl. Environ. Microbiol. 63:3844-3850. Jones, T. R., I. H. Kang, D. B. Wheeler, R. A. Lindquist, A. Papallo, D. M. Sabatini, P. Golland, and A. E. Carpenter. 2008. CellProfiler Analyst: data exploration and analysis software for complex image-based screens. BMC Bioinformatics 9:482. doi:10.1186/1471-2105-9-482. Joubert, J., and D. Sharma. 2011. Light microscopy digital imaging. Curr. Protoc. Cytom. 58:2.3.1-2.3.11. doi:10.1002/ 0471142956.cy0203s58. Katan, A. J., and C. Dekker. 2011. High-speed AFM reveals the dynamics of single biomolecules at the nanometer scale. Cell. 147:979-982. Kell, D. B., A. S. Kaprelyants, D. H. Eichart, C. R. Harwood, and M. R. Barer. 1998. Viability and activity in readily culturable bacteria: a review and discussion of the practical issues. Antonie van Leeuwenhoek. 73:169-187. Kennedy, D., U. P. Cronin, and M. G. Wilkinson. 2011. Responses of Escherichia coli, Listeria monocytogenes, and Staphylococcus aureus to simulated food processing treatments, determined using fluorescence-activated cell sorting and plate counting. Appl. Environ. Microbiol. 77:4657-4668. Kenney, S. J., S. L. Burnett, and L. R. Beuchat. 2001. Location of Escherichia coli O157:H7 on and in apples as affected by bruising, washing and rubbing. J. Food Prot. 64:1328-1333. Kepner, Jr, R. L., and J. R. Pratt. 1994. Use of fluorochromes for direct enumeration of total bacteria in environmental samples: past and present. Microbiol. Rev. 58:603-615. Khan, M. A., A. Seaman, and M. Woodbine. 1977. Immunofluorescent identification of Listeria monocytogenes. Zentralbl. Bakteriol. Org. A 239:62-69. Kitaguchi, A., N. Yamaguchi, and M. Nasu. 2005. Enumeration of respiring Pseudomonas spp. in milk within 6 hours by fluorescence in situ hybridization following formazan reduction. Appl. Environ. Microbiol. 71:2748-2752. Klein, P. G., and V. K. Juneja. 1997. Sensitive detection of viable Listeria monocytogenes by reverse transcription-PCR. Appl. Environ. Microbiol. 63:4441-4448. Koch, A. L. 2007. Growth measurement. In: C. A. Reddy, T. J. Beveridge, J. A. Breznak, G. Marzluf, T. M. Schmidt, and L. R. Snyder (Editors). Methods for Feneral and Molecular Bacteriology. 3rd ed. American Society for Microbiology, Washington, D.C., 172-199. Kogure, K., U. Simidu, and N. Taga. 1979. A tentative direct microscopic method for counting living marine bacteria. Can. J. Microbiol. 25:415-420. Kroupitski, Y., R. Pinto, E. Belausov, and S. Sela. 2011. Distribution of Salmonella typhimurium in romaine lettuce leaves. Food Microbiol. 28:990-997. Kroupitski, Y., R. Pinto, M. T. Brandl, E. Belausov, and S. Sela. 2009. Interactions of Salmonella enterica with lettuce leaves. J. Appl. Microbiol. 106:1876-1885. Kuda, T., T. Iwase, C. Yuphakhun, H. Takahashi, T. Koyanagi, and B. Kimura. 2011. Surfactant-disinfectant resistance of Salmonella and Staphylococcus adhered and dried on surfaces with egg compounds. Food Microbiol. 28:920-925. Laflamme, C., S. Lavigne, J. Ho, and C. Duchaine. 2004. Assessment of bacterial endospore viability with fluorescent dyes. J. Appl. Microbiol. 96:684-692. Lagendijk, E. L., S. Validov, G. E. M. Lamers, S. de Weert, and G. V. Gloemberg. 2010. Genetic tools for tagging Gramnegative bacteria with mCherry for visualization in vitro and in natural habitats, biofilm and pathogenicity studies. FEMS Microbiol. Lett. 305:81-90. Lahtinen, S. J., M. Gueimonde, A. C. Ouwehand, J. P. Reinikainen, and S. J. Salminen. 2006. Comparison of four

122.

123.

124.

125.

126.

127.

128.

129.

130.

131.

132.

133.

134.

135.

136.

methods to enumerate probiotic bifidobacteria in a fermented food product. Food Microbiol. 23:571-577. Lamprecht, M. R., D. M. Sabatini, and A. E. Carpenter. 2007. CellProfiler: free, versatile software for automated biological image analysis. BioTechniques. 42:71-75. Landry, W. L., A. H. Schwab, and G. A. Lancette. 2001. Examination of canned foods. Chapter 21A. In: Bacteriological analytical manual. U.S. Food and Drug Administration. http://www.fda.gov/Food/FoodScienceResearch/ LaboratoryMethods/ucm109398.htm. Accessed April 16, 2015. La Storia, A., D. Ercolini, F. Marinello, R. Di Pasqua, F. Villani, and G. Mauriello. 2011. Atomic force microscopy analysis shows surface structure changes in carvacroltreated bacterial cells. Res. Microbiol. 162:164-172. Lattuada, C. P., and D. McClain. 1998. Examination of meat and poultry products for Bacillus cereus. Chapter 12. In: Microbiology Laboratory Guidebook. 3rd ed. U.S. Department of Agriculture/Food Safety and Inspection Service. http://www.fsis.usda.gov/wps/wcm/connect/ 7aa4 1946 -bd8 9-4 ba9-9 1cf -7ea 72e1 5e677 /Mlgchp12. pdf?MOD5AJPERES. Accessed April 16, 2015. Lehtola, M. J., C. J. Loades, and C. W. Keevil. 2005. Advantages of peptide nucleic acid oligonucleotides for sensitive site directed 16S rRNA fluorescence in situ hybridization (FISH) detection of Campylobacter jejuni, Campylobacter coli and Campylobacter lari. J. Microbiol. Methods 62:211-219. Liao, C.-H., P. H. Cooke, and B. A. Niemira. 2010. Localization, growth and inactivation of Salmonella Saintpaul on Jalpen˜o peppers. J. Food Sci. 75:M377-M382. Lindmo, T., and H. B. Steen. 1979. Characteristics of a simple, high-resolution flow cytometer based on a new flow configuration. Biophys. J. 28:33-44. Lipski, A., U. Friedrich, and K. Altendorf. 2001. Application of rRNA-targeted oligonucleotide probes in biotechnology. Appl. Microbiol. Biotechnol. 56:40-57. Lopez, C., M. B. Maillard, V. Briard-Bion, B. Camier, and J. A. Hannon. 2006. Lipolysis during ripening of Emmental cheese considering organization of fat and preferential localization of bacteria. J. Agric. Food Chem. 54:5855-5867. Lorca, T. A., M. D. Pierson, J. R. Claus, J. D. Eifert, J. E. Marcy, and S. S. Sumner. 2002. Penetration of surfaceinoculated bactgeria as a result of hydrodynamic shock wave treatment of beef steaks. J. Food Prot. 65:616-620. Lu, X., B. A. Rasco, J. M. Jabal, D. E. Aston, M. Lin, and M. E. Konkel. 2011. Investigating antibacterial effects of garlic (Allium sativum) concentrate and garlic-derived organosulfur compounds on Campylobacter jejuni by using Fourier transform infrared spectroscopy, Raman spectroscopy, and electron microscopy. Appl. Environ. Microbiol. 77:5257-5269. Macarisin, D., M. Santin, G. Bauchan, and R. Fayer. 2010. Infectivity of Cryptosporidium parvum oocysts after storage of experimentally contaminated apples. J. Food Prot. 73:18241829. Martı´nez-Abad, A., G. Sanchez, J. M. Lagaron, and M. J. Ocio. 2012. On the different growth conditions affecting silver antimicrobial efficacy on Listeria monocytogenes and Salmonella enterica. Int. J. Food Microbiol. 158:147-154. Mason, D. J., S. Shanmuganathan, F. C. Mortimer, and V. A. Gant. 1998. A fluorescent Gram stain for flow cytometry and epifluorescence microscopy. Appl. Environ. Microbiol. 64:2681-2685. Meijering, E., and G. van Cappellen. 2007. Quantitative biological image analysis. In: S. L. Shorte and F. Frischknecht (Editors). Imaging cellular and molecular biological function. Springer-Verlag, Berlin, Germany, 45-70.

| Microscopic Methods

137. Mercier-Bonin, M., A. Dehouche, J. Morchain, and P. Schmitz. 2011. Orientation and detachment dynamics of Bacillus spores from stainless steel under controlled shear flow: modelling of the adhesion force. Int. J. Food Microbiol. 146:182-191. 138. Meylheuc, T., C. Methivier, M. Renault, J. M. Herry, C. M. Pradier, and M. N. Bellon-Fontaine. 2006. Adsorption on stainless steel surfaces of biosurfactants produced by gramnegative and gram-positive bacteria: consequence on the bioadhesive behavior of Listeria monocytogenes. Colloids Surf. B Biointerfaces. 52:128-137. 139. Molecular Probes. 2004. Live/Dead BacLight Bacterial Viability Kit. Molecular Probes Handbook: a Guide to Fluorescent Probes and Labeling Technologies. 11th ed. http://www.invitrogen.com/site/us/en/home/ References/Molecular-Probes-The-Handbook/Assays-forCell-Viability-Proliferation-and-Function/Viability-andCytotoxicity-Assay-Kits-for-Diverse-Cell-Types.html. Accessed September 18, 2012. 140. Mols, M., M. Ceragioli, and T. Abee. 2011. Heat stress leads to superoxide formation in Bacillus cereus detected using the fluorescent probe MitoSOX. Int. J. Food Microbiol. 151:119122. 141. Mu¨ller, S., and G. Nebe-von-Caron. 2010. Functional singlecell analyses: flow cytometry and cell sorting of microbial populations and communities. FEMS Microbiol. Rev. 34:554587. 142. Murray, R. G. E., and C. F. Robinow. 2007. Light microscopy. In: C. A. Reddy, T. J. Beveridge, J. A. Breznak, G. Marzluf, T. M. Schmidt, and L. R. Snyder (Editors). Methods for General and Molecular Bacteriology. 3rd ed. American Society for Microbiology. Washington, D.C., 5-18. 143. Narisawa, N., S. Furukawa, T. Kawarai, K. Ohishi, S. Kanda, K. Kimijima, S. Negishi, H. Ogihara, and M. Yamasaki. 2008. Effect of skimmed milk and its fractions on the inactivation of Escherichia coli K12 by high hydrostatic pressure treatment. Int. J. Food Microbiol. 124:103-107. 144. Nguyen, V. T., M. S. Turner, and G. A. Dykes. 2011. Influence of cell surface hydrophobicity on attachment of Campylobacter to abiotic surfaces. Food Microbiol. 28:942950. 145. Nopharatana, M., D. A. Mitchell, and T. Howes. 2003. Use of confocal scanning laser microscopy to measure the concentrations of aerial and penetrative hyphae during growth of Rhizopus oligosporus on a solid surface. Biotechnol. Bioeng. 84:71-77. 146. Ootsubo, M., T. Shimizu, R. Tanaka, T. Sawabe, K. Tajima, and Y. Ezura. 2003. Seven-hour fluorescence in situ hybridization technique for enumeration of Enterobacteriaceae in food and environmental water sample. J. Appl. Microbiol. 95:1182-1190. 147. Orlandi, P. A., C. Frazar, L. Carter, and D.-M. T. Chu. 2004. Detection of Cyclospora and Cryptosporidium from fresh produce: isolation and identification by polymerase chain reaction (PCR) and microscopic analysis. Bacteriological Analytical Manual. U.S. Food and Drug Administration. http://www.fda.gov/Food/FoodScienceResearch/ LaboratoryMethods/ucm073638.htm. Accessed April 16, 2015. 148. Ortega, Y. R., C. R. Roxas, R. H. Gilman, N. J. Miller, L. Cabrera, C. Taquiri, and C. R. Sterling. 1997. Isolation of Cryptosporidium parvum and Cyclospora cayetanensis from vegetables collected in markets of an endemic region in Peru. Am. J. Trop. Med. Hyg. 57:683-686. 149. Pamp, S. J., C. Sternberg, and T. Tolker-Nielsen. 2009. Insight into the microbial multicellular lifestyle via flow-cell technology and confocal microscopy. Cytometry A 75:90103.

150. Peckys, D. B., P. Mazur, K. L. Gould, and N. de Jonge. 2011. Fully hydrated yeast cells imaged with electron microscopy. Biophys. J. 100:2522-2529. 151. Pelling, A. E., S. Sehati, E. B. Gralla, J. S. Valentine, and J. K. Gimzewski. 2004. Local nanomechanical motion of the cell wall of Saccharomyces cerevisiae. Science. 305:1147-1150. 152. Perfetto, S. P., P. K. Chattopadhyay, and M. Roederer. 2004. Seventeen-colour flow cytometry: unravelling the immune system. Nature. 4:648-655. 153. Perry-O’Keefe, H., S. Rigby, K. Oliveira, D. Sorensen, H. Stender, J. Coull, and J. J. Hyldig-Nielsen. 2001. Identification of indicator microorganisms using a standardized PNA FISH method. J. Microbiol. Meth. 47:281292. 154. Perry-O’Keefe, H., H. Stender, A. Broomer, K. Oliveira, J. Coull, and J. J. Hyldig-Nielsen. 2001. Filter-based PNA in situ hybridization for rapid detection, identification and enumeration of specific microorganisms. J. Appl. Microbiol. 90:180-189. 155. Pettipher, G. L. 1986. Review: the direct epifluorescent filter technique. J. Food Technol. 21:535-546. 156. Pettipher, G. L., R. Mansell, C. H. McKinnon, and C. M. Cousins. 1980. Rapid membrane filtration—epifluorescent microscopy technique for direct enumeration of bacteria in raw milk. Appl. Environ. Microbiol. 39:423-429. 157. Pettipher, G. L., and U. M. Rodrigues. 1982. Rapid enumeration of microorganisms in foods by the direct epifluorescent filter technique. Appl. Environ. Microbiol. 44:809-813. 158. Pitts, J. E., and D. Laird. 2004. Direct microscopic methods for bacteria or somatic cells. In: H. M. Wehr and J. F. Frank (Editors). Standard methods for the examination of dairy products. 17th ed. American Public Health Association, Washington, D.C., 269-292. 159. Pontier-Bres, R., F. Prodon, P. Munro, P. Rampal, E. Lemichez, J. F. Peyron, and D. Czerucka. 2012. Modification of Salmonella typhimurium motility by the probiotic yeast strain Saccharomyces boulardii. PLoS ONE. 7:e33796. doi:10.1371/journal.pone.0033796. 160. Pyle, B. H., S. C. Broadaway, and G. A. McFeters. 1995. A rapid, direct method for enumerating respiring enterohemorrhagic Escherichia coli O157:H7 in water. Appl. Environ. Microbiol. 61:2614-2619. 161. Rampersad, J., and D. Ammons. 2005. A Bacillus thuringiensis isolation method utilizing a novel stain, low selection and high throughput produced atypical results. BMC Microbiol. 5:52. doi:10.1186/1471-2180-5-52. 162. Raulio, M., A. Wilhelmson, M. Salkinoja-Salonen, and A. Laitila. 2009. Ultrastructure of biofilms formed on barley kernels during malting with and without starter culture. Food Microbiol. 26:437-443. 163. Rhodehamel, E. J., and S. M. Harmon. 2001. Clostridium perfringens. Bacteriological Analytical Manual. U.S. Food and Drug Administration. http://www.fda.gov/ Food/FoodScienceResearch/LaboratoryMethods/ ucm070878.htm. Accessed April 16, 2015. 164. Rieu, A., R. Briandet, O. Habimana, D. Garmyn, J. Guzzo, and P. Piveteau. 2008. Listeria monocytogenes EGD-e biofilms: no mushrooms but a network of knitted chains. Appl. Environ. Microbiol. 74:4491-4497. 165. Rodrigues, U. M., and R. G. Kroll. 1985. The direct epifluorescent filter technique (DEFT): increased selectivity, sensitivity and rapidity. J. Appl. Bacteriol. 59:493-499. 166. Rodrigues, U. M., and R. G. Kroll. 1988. Rapid selective enumeration of bacteria in foods using a microcolony epifluorescence microscopy technique. J. Appl. Bacteriol. 64:65-78. 167. Rodrigues, U. M., and R. G. Kroll. 1989. Microcolony epifluorescence microscopy for selective enumeration of

| 63

Compendium of Methods for the Microbiological Examination of Foods |

168.

169.

170.

171.

172.

173.

174.

175.

176.

177.

178.

179.

180.

181.

182.

64 |

injured bacteria in frozen and heat-treated foods. Appl. Environ. Microbiol. 55:778-787. Rodrigues, U. M., and R. G. Kroll. 1990. Rapid detection of salmonellas in raw meats using a fluorescent antibodymicrocolony technique. J. Appl. Bacteriol. 68:213-223. Rodriquez, G. G., D. Phipps, K. Ishiguro, and H. F. Ridgway. 1992. Use of a fluorescent redox probe for direct visualization of actively respiring bacteria. Appl. Environ. Microbiol. 58:1801-1808. Santarelli, M., M. Gatti, C. Lazzi, V. Bernini, G. A. Zapparoli, and E. Neviani. 2008. Whey starter for Grana Padano cheese: effect of technological parameters on viability and composition of the microbial community. J. Dairy Sci. 91:883-891. Schillinger, C., A. Petrich, R. Lux, B. Riep, J. Kikhney, A. Friedmann, L. E. Wolinsky, U. B. Go¨bel, H. Daims, and A. Moter. 2012. Co-localized or randomly distributed? Pair cross correlation of in vitro grown subgingival biofilm bacteria quantified by digital image analysis. PLoS ONE. 7:e37583. doi:10.1371/journal.pone.0037583. Selinummi, J., J. Seppa¨la¨, O. Yli-Harja, and J. A. Puhakka. 2005. Software for quantification of labeled bacteria from digital microscope images by automated image analysis. Biotechniques 39:859-862. Servis, N. A., S. Nichols, and J. C. Adams. 1995. Development of a direct viable count procedure for some gram-positive bacteria. Lett. Appl. Microbiol. 20:237-239. Sheridan, J. J., I. Walls, J. McLauchlin, D. McDowell, and R. Welch. 1991. Use of a microcolony technique combined with an indirect immunofluorescence test for the rapid detection of Listeria in raw meat. Lett. Appl. Microbiol. 13:140-144. Shimizu, S., M. Ootsubo, Y. Kuboswa, I. Fuchizaqa, Y. Kawai, and K. Yamazaki. 2009. Fluorescent in situ hybridization in combination with filter cultivation (FISHFC) method for specific detection and enumeration of viable Clostridium perfringens. Food Microbiol. 26:425-431. Shrestha, N. K., N. M. Scalera, D. A. Wilson, B. BrehmStecher, and G. W. Procop. 2011. Rapid identification of Staphylococcus aureus and methicillin resistance by flow cytometry using a peptide nucleic acid probe. J. Clin. Microbiol. 49:3383-3385. Skovager, A., K. Whitehead, H. Siegumfeldt, H. Ingmer, J. Verran, and N. Arneborg. 2012. Influence of flow direction and flow rate on the initial adhesion of seven Listeria monocytogenes strains to fine polished stainless steel. Int. J. Food Microbiol. 157:174-181. Smith, Z. J., K. Chu, A. R. Espenson, M. Rahimzadeh, A. Gryshuk, M. Molinaro, D. M. Dwyre, S. Lane, D. Matthews, and S. Wachsmann-Hogiu. 2011. Cell-phone-based platform for biomedical device development and education applications. PLoS ONE. 6:e17150. doi:10.1371/journal.pone. 0017150. Solomon, H. M., and T. Lilly, Jr. 2001. Clostridium botulinum. Chapter 17. In: Bacteriological analytical manual. U.S. Food and Drug Administration. http://www.fda.gov/Food/ FoodScienceResearch/LaboratoryMethods/ucm070879.htm. Accessed April 16, 2015. Splittstoesser, D. F., M. Groll, D. L. Downing, and J. Kaminski. 1977. Viable counts versus the incidence of machinery mold (Geotrichum) on processed fruits and vegetables. J. Food Prot. 40:402-405. Stewart, D., K. Reineke, J. Ulaszek, T. Fu, and M. Tortorello. 2001. Growth of Escherichia coli O157:H7 during sprouting of alfalfa seeds. Lett. Appl. Microbiol. 33:95-99. Strauber, H., and S. Muller. 2010. Viability states of bacteria— specific mechanisms of selected probes. Cytometry A. 77:623634.

183. Suo, Z., R. Avci, M. Deliorman, X. Yang, and D. W. Pascual. 2009. Bacteria survive multiple puncturings of their cell walls. Langmuir. 25:4588-4594. 184. Takeuchi, K., and J. F. Frank. 2001a. Confocal microscopy and microbial viability detection for food research. J. Food Prot. 64:2088-2102. 185. Takeuchi, K., and J. F. Frank. 2001b. Expression of redshifted green fluorescent protein by Escherichia coli O157:H7 as a marker for the detection of cells on fresh produce. J. Food Prot. 64: 298-304. 186. Tallent, S. M., J. Rhodehamel, S. M. Harmon, and R. W. Bennett. 2012. Chapter 14. Bacillus cereus. In: Bacteriological analytical manual. U.S. Food and Drug Administration. http://www.fda. gov/Food/FoodScienceResearch/LaboratoryMethods/ ucm070875.htm. Accessed April 16, 2015. 187. Thiberge, S., A. Nechushtan, D. Sprinzak, O. Gileadi, V. Behar, O. Zik, Y. Chowers, S. Michaeli, J. Schlessinger, and E. Moses. 2004. Scanning electron microscopy of cells and tissues under fully hydrated conditions. Proc. Natl. Acad. Sci. U. S. A. 101:3346-3351. 188. Tian, P., D. Yang, and R. Mandrell. 2011. Differences in the binding of human norovirus to and from Romaine lettuce and raspberries by water and electrolyzed waters. J. Food Prot. 74:1364-1369. 189. Tison, D. L. 1990. Culture confirmation of Escherichia coli serotype O157:H7 by direct immunofluorescence. J. Clin. Microbiol. 28:612-613. 190. Tortorello, M. L., and K. F. Reineke. 2000. Direct enumeration of Escherichia coli and enteric bacteria in water, beverages and sprouts by 16S rRNA in situ hybridization. Food Microbiol. 17:305-313. 191. Tortorello, M. L., K. F. Reineke, and D. S. Stewart. 1997. Comparison of antibody-direct epifluorescent filter technique with the most probable number procedure for rapid enumeration of Listeria in fresh vegetables. J. AOAC Int. 80:1208-1214. 192. Tortorello, M. L., and D. S. Stewart. 1994. Antibody-direct epifluorescent filter technique for rapid, direct enumeration of Escherichia coli O157:H7 in beef. Appl. Environ. Microbiol. 60:3553-3559. 193. Tortorello, M. L., D. S. Stewart, and R. B. Raybourne. 1998. Quantitative analysis and isolation of Escherichia coli O157:H7 in a food matrix using flow cytometry and cell sorting. FEMS Immunol. Med. Microbiol. 19:267-274. 194. Tournas, V., M. E. Stack, P. B. Mislivec, H. A. Koch, and R. Bandler. 2001. Yeasts, molds and mycotoxins. Chapter 18. In: Bacteriological analytical manual. U.S. Food and Drug Administration. http://www.fda.gov/Food/ FoodScienceResearch/LaboratoryMethods/ucm071435.htm. Accessed April 16, 2015. 195. Tsuji, T., Y. Kawasaki, S. Takeshima, T. Sekiya, and S. Tanaka. 1995. A new fluorescence staining assay for visualizing living microorganisms in soil. Appl. Environ. Microbiol. 61:3415-3421. 196. U.S. Food and Drug Administration. 2012. Foods, colors and cosmetics. Chapter 5. In: Compliance policy guides. Accessed September 25, 2012. http://www.fda.gov/ICECI/ ComplianceManuals/CompliancePolicyGuidanceManual/ ucm119194.htm. 197. Vainrub, A., O. Pustovyy, and V. Vodyanoy. 2006. Resolution of 90 nm (l/5) in an optical transmission microscope with an annular condenser. Opt. Lett. 31:2855-2857. 198. van Leeuwenhoek, A. 1684. Microscopical observations, about animals in the scurf of the teeth, the substance call’d worms in the nose, the cuticula consisting of scales. Philos. Trans. R. Soc. Lond. 14:568-574.

| Microscopic Methods

199. Varga, J. J., B. Therit, and S. B. Melville. 2008. Type IV pili and the Ccpa protein are needed for maximal biofilm formation by the gram-positive anaerobic pathogen Clostridium perfringens. Inf. Immun. 76:4944-4951. 200. Vodovotz, Y., E. Vittadini, J. Coupland, D. J. McClements, and P. Chinachoti. 1996. Bridging the gap: use of confocal microscopy in food research. Food Technol. 50(6):74-82. 201. Wachtel, M. R., L. C. Whitehand, and R. E. Mandrell. 2002. Association of Escherichia coli O157:H7 with preharvest leaf lettuce upon exposure to contaminated irrigation water. J. Food Prot. 65:18-25. 202. Wang, H., H. Feng, W. Liang, Y. Luo, and V. Malyarchuk. 2009. Effect of surface roughness on retention and removal of Escherichia coli O157:H7 on surfaces of selected fruits. J. Food Sci. 74:E8-E15. 203. Wang, Y., F. Hammes, K. De Roy, W. Verstraete, and N. Boon. 2010. Past, present and future applications of flow cytometry in aquatic microbiology. Trends Biotechnol. 28:416-424. 204. Warner, J. C., S. D. Rothwell, and C. W. Keevil. 2008. Use of episcopic differential interference contrast microscopy to identify bacterial biofilms on salad leaves and track colonization by Salmonella Thompson. Environ. Microbiol. 10:918-925. 205. Weinkauf, H., and B. F. Brehm-Stecher. 2009. Facile detection of metal nanoparticle interactions with Candida albicans hyphae via dark field microscopy. Biotechnol. J. 4:871-879.

206. Whitehead, K. A., L. A. Smith, and J. Verran. 2010. The detection and influence of food soils on microorganisms on stainless steel using scanning electron microscopy and epifluorescence microscopy. Int. J. Food Microbiol. 141 Suppl. 1:S125-S133. 207. Wildman, J. D., and P. B. Clark. 1947. Some examples of the occurrence of machinery slime in canning factories. J. Assoc. Off. Agric. Chem. 30:582-585. 208. Wiley, D. J., R. Rosqvist, and K, Schesser. 2007. Induction of the Yersinia type 3 secretion system as an all-or-none phenomenon. J. Mol. Biol. 373:27-37. 209. Zimmermann, R., R. Iturriaga, and J. Becker-Birck. 1978. Simultaneous determination of the total number of aquatic bacteria and the number thereof involved in respiration. Appl. Environ. Microbiol. 36:926-935. 210. Zsigmondy, R. 1909. Colloids and the ultramicroscope: a manual of colloid chemistry and ultramicroscopy (Transl. by J. Alexander). John Wiley & Sons, New York, NY, 101-103. 211. Zulfakar, S. S., J. D. White, T. Ross, and M. L. Tamplin. 2012. Bacterial attachment to immobilized extracellular matrix proteins in vitro. Int. J. Food Microbiol. 157:210217. 212. Zwirglmaier, K. 2005. Fluorescence in situ hybridization (FISH)—the next generation. FEMS Microbiol. Lett. 246:151158.

| 65

|

CHAPTER 5

|

Cultural Methods for the Enrichment and Isolation of Microorganisms William H. Sperber, Mark A. Moorman, and Timothy A. Freier

5.1

INTRODUCTION

This chapter describes the general principles and methods for the enrichment and isolation of microorganisms. It does not describe the detailed requirements for the enrichment and isolation of specific microorganisms. These are presented in other chapters of the Compendium addressing each microorganism. The principles and methods described here are important for the isolation of Salmonella, Listeria, other foodborne pathogens, and indicator microorganisms. There are several reasons for the use of enrichment methods. Primary among these is the need to grow a detectable population of cells from a very low initial level. It is not unusual for the target microorganism to be present in foods at levels of about one cell per 100 g of food. With enrichment methods, as little as one cell per 500 g of food have been detected.11 Enrichment techniques are also used for the recovery of injured microorganisms. Microorganisms in foods are typically in a stressed condition. They often lack the optimal nutrients for growth, they may be in an environment (e.g., pH, temperature) that will not support their growth, or they may be damaged by sub-lethal stresses (e.g., osmotic or temperature shock) imposed during the processing of food. These microorganisms will require a period of time in the appropriate conditions so that cellular damage can be repaired and metabolic pathways activated. Last, enrichment methods permit the proliferation of the target microorganism to detectable levels while repressing the growth of competing non-target microorganisms.13 Unlike quantitative microbial recovery, as described in the chapters ‘‘Mesophilic Aerobic Plate Count,’’ ‘‘Rapid Methods for the Detection and Identification of Foodborne Pathogens,’’ and ‘‘Detection and Enumeration of HeatResistant Molds,’’ enrichment methods are qualitative. They will indicate the presence or absence of the target microorganism, but not its numbers. Enrichment methods can, however, yield quantitative results when they are used in conjunction with the most probable number (MPN) technique described in the chapter ‘‘Culture Methods for Enumeration of Microorganisms.’’ | 67 |

The isolation methods described in this chapter are necessary to obtain pure cultures for the biochemical, serological, and genomic identification of the target microorganisms.

5.2

ENRICHMENT METHODS

Enrichment methods determine the presence or absence of a target organism: they are not conducted to determine the level or quantity of that organism. Many direct plating or quantitative methods exist for the recovery and enumeration of target microorganisms (see the chapter ‘‘Culture Methods for Enumeration of Microorganisms’’). However, quantitative methods are not appropriate in several situations. 1. The permissible level of the organism is less than the maximum sensitivity of the quantitative procedure. 2. The organism is surrounded by large numbers of competing microorganisms. 3. The suspending food is inhibitory to the target organism. The goal of the enrichment method is to permit the growth of the target microorganism, and if enrichment is selective, to suppress or inhibit the growth of competing microorganisms. There are various types of enrichment protocols that, either individually or combined, permit the growth of the target organism to levels necessary for detection or recovery by diagnostic or selective plating procedures, respectively.

5.3

PRE-ENRICHMENT

The purpose of pre-enrichment is to allow the stressed target microorganism to resuscitate in either a non-selective or a moderately selective environment. Although the microorganism may resuscitate, very little growth may occur during the pre-enrichment step (Figure 5-1). These pre-enrichment media are either nonselective or are designed to be moderately selective against competing microorganisms. If the media contain any selective components, these must be balanced to permit the growth of the target organism and repress the growth of competing microorganisms in the food sample. A pre-enrichment procedure may not be necessary

Compendium of Methods for the Microbiological Examination of Foods |

Figure 5-1. Growth of Salmonella and competitive microflora during pre-enrichment and selective enrichment. Courtesy of Silliker Laboratories Group, Inc.

when the enrichment medium has been validated to support the resuscitation and growth of the target microorganism. The liquid version of Baird Parker Agar (Liquid Baird Parker), without agar, has been successfully used to recover Staphylococcus aureus in foods.14 The formulation of the pre-enrichment medium will depend upon the level of competing microorganisms and the ability of the food to inhibit microorganisms. The US Food and Drug Administration (FDA) Bacteriological Analytical Manual describes different pre-enrichment media for the recovery of Salmonella in foods.7 These range from sterile deionized water for recovering Salmonella from nonfat dry milk, to Brilliant Green-supplemented milk for the recovery of Salmonella from chocolate. These media use either the nutrients from the food or contain ingredients (e.g., casein) to neutralize the inhibitory nature of the food (e.g., chocolate).3 Lactose is commonly used as the preenrichment medium for Salmonella, even though few species of this genus are capable of metabolizing this carbohydrate. The presence of lactose is not directly essential for the recovery of Salmonella.5,13 It has been suggested that some selectivity of pre-enrichment media containing lactose is generated by the reduction in media pH when the lactose is fermented by the mixed competitor flora.10

units in one pre-enrichment, thereby reducing the number of samples to be analyzed in the laboratory and the cost associated with testing large numbers of individual samples. Composite analysis for Salmonella has been shown to not reduce the method’s sensitivity.11 The increase in sample weight will increase the volume of the preenrichment medium, often necessitating a larger incubator depending on the number of samples to be tested.

5.31

5.314 Mechanics of Testing The objective of any microbiological assay is to recover the microorganisms present in the food sample, not those from the laboratory environment. The food sample must be handled aseptically, with the test portion being dispensed into the enrichment container without introducing external contaminants. Good laboratory practices require that neither the test container nor the dispensing utensil falls below the plane of the pre-enrichment container. After dispensing, the pre-enrichment container can be shaken to dissolve powders or homogenized by other means to adequately release those microorganisms trapped within the food matrix. Typically this homogenization can last 1 to 2 minutes. Surfactants may be added to aid the release of the trapped microorganisms. After homogenization, the pre-enrichment pH should be verified to be within the appropriate range for recovery of the target organism. Many foods will affect medium pH,

Pre-enrichment Considerations

5.311 Sample Weight The amount of sample analyzed is either defined in the analytical procedure or based upon a sampling plan. The sample weight for pre-enrichment can vary from as little as 1 g to as much as 1500 g. Typically these food samples are diluted in the pre-enrichment medium at a 1:9 ratio of food to medium. The pre-enrichment may be prepared by weighing the sample or by conducting a rinse, such as that for poultry or other meat carcasses. 5.312 Compositing The enrichment protocol is intended to recover viable microorganisms in the tested sample. The sample analyzed may be a single unit, or represent a composite of multiple units dispensed into the sample prior to enrichment. Composite analysis enables the laboratory to examine many 68 |

5.313 Pre-enrichment Temperature Incubation temperatures used for pre-enrichment or enrichment are typically near the ideal temperature for the growth of the target microorganism. Incubation temperatures may be used that provide a selective advantage for the target organism. Listeria monocytogenes, for example, a psychrotrophic pathogen, was initially enriched at refrigeration temperatures. The temperature of the pre-enrichment medium prior to inoculation will have an effect on the recovery of the target microorganism. Large-volume pre-enrichment requires an extended time to equilibrate at the appropriate incubation temperature following the addition of samples. For preenrichment at mesophilic temperatures these large volumes of medium may require equilibration at room temperature or higher prior to inoculation.

| Cultural Methods for the Enrichment and Isolation of Microorganisms

requiring the addition of acid or alkali to re-establish the appropriate pH for the pre-enrichment.

5.315 Incubation Time The pre-enrichment period must be long enough for effective resuscitation of stressed or injured microorganisms. If this time is too short, the target microorganism may not recover, resulting in a false negative test result. Theoretically, if the pre-enrichment time is excessive, overgrowth by competing organisms may occur, again resulting in a false negative test result. D’Aoust found that increased recovery of Salmonella with prolonged incubation was not due to Salmonella growth but rather to the higher rate of death of competing microorganisms, owing to their greater sensitivity to elevated temperatures and toxicity of selective media.5 The appropriate incubation period must be validated based on the unique nature of the food sample, the enrichment medium, and the incubation time. 5.4

SELECTIVE ENRICHMENT

The pre-enrichment will result in resuscitation of the target microorganism and moderate levels of proliferation. Selective enrichment furthers the growth of the target microorganism while suppressing or inhibiting that of competing microorganisms (Figure 5-1). The selectivity of the medium is provided by agents or conditions which are antagonistic or inhibitory to competing microorganisms. These selective agents include temperature, antimicrobials, salts, acids, and metals. The pre-enrichment step can also serve to dilute or minimize interfering agents in the food sample that can impair the selective capability of the selective medium. Taylor and Silliker determined that lactose pre-enrichment prior to selective enrichment in tetrathionate and selenite cystine increased the recovery of Salmonella from albumin.13 Furthermore, direct inoculation of food samples with large numbers of competing microorganisms into selective enrichments may result in false negative results owing to the reduction in the medium’s selectivity.5 Conversely, food material in a pre-enrichment transferred to a selective enrichment will positively affect media efficacy. Abbiss demonstrated that the presence of food material in buffered peptone water pre-enrichment enhanced the recovery of Salmonella typhimurium from selective enrichments. This improved recovery was due to amelioration of the initial inhibitory environment Salmonella encounters when transferred to the selective medium. In the case of Salmonella, neither tetrathionate nor selenite cystine enrichment broths alone will support the growth of all strains of Salmonella.1 Therefore to reduce the risk of a false negative, many selective enrichment protocols will employ more than one medium following the pre-enrichment. To ensure the availability of nutrients to the target microorganism, enrichments may be shaken during incubation. Facultatively anaerobic microorganisms may experience shortened lag phase and generation times when oxygen is added during the enrichment. The oxygen will support an aerobic metabolism that yields higher energy than anaerobic metabolism. Duffy et al.,6 however, studied the growth kinetics of L. monocytogenes and found that aeration of the selective enrichment did not alter the length of the lag phase or the growth rate of L. monocytogenes.

5.41

Mechanics of Transfer

The transfer of the pre-enrichment to selective enrichment requires the aseptic transfer of an aliquot of pre-enrichment to the selective enrichment. This step is highly prone to laboratory contamination and must be done with great care. Typically, ratios of 1:10 pre-enrichment inoculum to selective enrichment are attained. The pre-enrichment should be gently shaken or stirred prior to transfer.

5.42

Motility Enrichment

Motility enrichment media support the growth of the target organism while immobilizing this strain through the use of antisera specific to the target organism. This protocol has been successfully used as a selective agent and diagnostic aid in the recovery and identification of Salmonella.12

5.43

Selective Enrichment

Some media favor the growth of the target microorganism while possessing none of the inhibitory effects of selective agents. An example of such a medium is M Broth, which is termed elective enrichment.12 Originally formulated to promote antigen development after selective enrichment, M Broth contains sodium citrate and D-mannose as the only fermentable carbohydrates. Both compounds are fermented by salmonellae, thereby serving as anaerobic energy sources and providing a further competitive advantage over microorganisms that cannot ferment either carbohydrate. It is conceivable that other applications of the elective enrichment phenomenon could be developed to improve the sensitivity of procedures for the detection of the target microorganism.

5.44

Selective and Differential Isolation Methods

Although enrichment methods increase the proportion of the target microorganisms, the cultures still contain similar competing microorganisms that often must be eliminated before the target microorganism can be isolated and identified. These results are accomplished by the use of agar plating media and biochemical tests. Many selective and differential agents are used in these tests.2

5.441

Selective Agents

5.4411 Antibiotics. Many antibiotics are available for the selective isolation of microorganisms from foods. Those commonly used are polymyxin B, ampicillin, moxalactam, novobiocin, oxytetracycline, D-cycloserine, vancomycin, trimethiprim, and cycloheximide. 5.4412 Other Chemicals. It is also possible to inhibit the growth of non-target microorganisms in selective plating media by using chemicals other than antibiotics. Those commonly used are dyes such as Brilliant Green, sodium selenite, bile salts, potassium tellurite, and sodium lauryl sulfate. 5.4413 Anaerobiosis. The exclusion of oxygen can sometimes provide a selective advantage for the target microorganism. This is usually accomplished by the physical or chemical removal of oxygen inside a sealed incubation chamber. A similar selective effect can be achieved by the use of respiratory inhibitors such as sodium azide and potassium | 69

Compendium of Methods for the Microbiological Examination of Foods |

cyanide in the culture medium. In general, these prevent the growth of catalase-positive microorganisms while permitting the growth of catalase-negative microorganisms. Oxygen can be removed from tubed media by boiling and tempering these media just before inoculation. Diffusion of oxygen into these media can be retarded by the use of sterile 3% agar or mineral oil overlays.

5.4414 pH. Acidified media are commonly used to select particular groups of microorganisms. Media acidified to pH 3.5 are used for the isolation of yeasts, molds, and Alicyclobacilli. Similar use of pH control can be applied for the selection of other aciduric bacteria. 5.4415 Water Activity. The water activity of selective media can be lowered by the addition of numerous solutes, such as sodium chloride, glucose, ethanol, and propylene glycol. This approach is effective when the target microorganism can tolerate the chemical properties of the solute and the reduced water activity. 5.4416 Temperature. As described above for enrichment media, the incubation temperature can be a very effective means to select particular target microorganisms. 5.442 Differential Agents Several differential agents are useful for the screening and presumptive identification of the target microorganism. 5.4421 pH Indicators. The indication of pH change is one of the most common differential methods for the determination of a particular metabolic activity. The fermentation of carbohydrates is accompanied by the production of acids that lower the medium’s pH value. The decarboxylation of amino acids and the hydrolysis of urea are accompanied by the production of ammonia or amines, which raise the medium’s pH value. Commonly used pH indicators in microbiological media are phenol red, methyl red, and bromocresol purple. 5.4422 H2S Indicators. Some microorganisms produce hydrogen sulfide (H2S) as a byproduct of sulfur-containing amino acid metabolism. The production of H2S can be detected by the use of iron salts such as ferrous citrate, ferric ammonium citrate, or ferric ammonium sulfate. These combine with H2S to form ferrous sulfide (FeS), a black compound that is produced anaerobically in the deep portion of tubed media, or under colonies on agar surfaces. The solubility of FeS allows it to diffuse throughout the medium. 5.4423 Egg Yolk Reaction. Egg yolk is added to some microbiological media to assist in the recovery of injured microorganisms. Some of these produce a characteristic reaction that consists of a zone of clearing and/or a light precipitate around the colony, depending on the type of lipolytic enzymes excreted by the microorganism. 5.4424 Blood Hemolysis Reactions. Agar plates containing various species of blood can be used to differentiate pathogens such as S. aureus, Streptococcus pyogenes, and L. monocytogenes. The growth of the organism results in a 70 |

visible hemolytic reaction that varies depending on the particular pathogen and the species of blood that is used.

5.443 Agar Plates Enrichment cultures are streaked onto the surface of agar in Petri dishes to obtain isolated colonies so that pure cultures of the target microorganism will be available for identification tests. It is essential that this step be done well, so that the isolated colonies originate from single cells of the target microorganism. 5.4431 Media Handling and Preparation. Although commercially prepared media are often available, many of the media used for the selective and differential isolation of microorganisms are dehydrated and are reconstituted in the laboratory. It is important that these media be handled and prepared in accordance with the good laboratory practices described in the chapter ‘‘Laboratory Quality Management Systems.’’ Plates of differential media should be poured deep enough to permit observation of the differential characteristic. This can be accomplished by pouring 20 mL of molten agar into a 15 6 100 mm Petri dish. This technique tends to reduce the diffusion of excreted metabolic products and the masking of a particular colony’s reaction by neighboring colonies. The shelf lives of prepared media vary depending on the stability of the components. Manufacturers’ instructions or other reference materials should be consulted. In general, prepared media should not be stored for longer than 1 month. Some media require refrigeration and/or storage in the dark. The degree of hydration of plating media is often critical. If agar surfaces are too wet, spreading and swarming will prevent the isolation of a pure culture. If the surface is too dry, the microorganisms may grow poorly or not at all. Dehydration of prepared media can be prevented by storage in sealed bags or containers. 5.4432 Streaking Technique. To ensure the isolation of a pure culture of the target microorganism, it is essential that the streaking technique will provide isolated colonies. The streaking technique can be varied by skilled technicians because highly selective media are more ‘‘forgiving’’ than less selective media. For general purposes, however, the following technique should be used (Figure 5-2). 1. Using a sterile inoculating loop, transfer a loopful of the enrichment culture to the surface of the agar plate, near the edge of the plate. Streak the loop back and forth over the top quarter of the plate about five to ten times in a tight ‘‘Z’’ fashion. The streaking lines should not cross each other. 2. Resterilize the inoculating loop and allow it to cool. Streak the right quarter of the plate by passing the loop through the original area of inoculation (top quarter of plate) and streaking in a tight ‘‘Z’’ about five to ten times. The loop should not contact the original area of inoculation after the first pass. 3. Resterilize the inoculating loop and allow to cool. Streak the remainder of the plate in the same fashion, beginning with a single pass through the second streaked area (right quarter of plate). If plastic presterilized inoculating loops are used, a separate loop must be used for each step of this procedure.

| Cultural Methods for the Enrichment and Isolation of Microorganisms

Figure 5-2. Recommended streaking technique to obtain isolated colonies.

If isolated colonies are not obtained after incubation of the plates, or if the purity of a culture is in doubt, a portion of the growth on the plate can be restreaked onto a fresh agar plate. It is sometimes necessary to streak a culture several times in order to obtain pure isolated colonies.

5.4433 Picking Colonies. The proper colony-picking technique can help ensure the isolation of a pure culture (Figure 5-3). A sterile inoculating wire should be used. Never use a loop to pick a colony. The tip of the wire should be touched only to the top and center of the colony so that a miniscule amount of the colony is picked. This technique is important when colonies are being picked from mixed cultures on selective plates. The selective plating medium will permit the survival, and probably some growth, of nontarget microorganisms. Some of these may be on the agar surface beneath growing colonies of the target microorganism. The use of this picking technique may permit the isolation of a pure culture from a mixed colony. 5.4434 Tubed Media. Picked colonies are usually inoculated into one or more tubed solid or liquid media for the determination of biochemical characteristics and the amplification of antigens or genetic material for serological or genetic testing.

good laboratory practices described in the chapter ‘‘Laboratory Quality Management Systems.’’ The tubed media should be prepared deep enough so that anaerobic reactions can occur and be observed. In particular, agar slants should be prepared with 10–12 mL of agar in each tube, so that after solidification the butt portion of the tube will contain at least 5–6 mL of agar.

5.4436 Inoculation of Tubed Media. A sterile inoculating wire is used to pick a colony, as described above. The tubed media are inoculated by stabbing the butt and streaking the surface of agar slants, and by gentle twirling in liquid media. It is possible to inoculate multiple tubes without returning to the original colony to obtain more cells. The need to inoculate multiple tubes sometimes induces the inexperienced technician to take too much growth from the colony, even using the needle or an inoculating loop to scoop the entire colony from the agar surface. This practice increases the chance of getting an impure culture. The minuscule—even invisible—amount of growth obtained by the proper colony-picking technique described above is sufficient to inoculate many tubes. 5.5 5.51

5.4435 Media Handling and Preparation. All media should be handled and prepared in accordance with the

QUALITY ASSURANCE OF ENRICHMENT AND ISOLATION METHOD Temperature Control or Management

Temperature control is a critical selective or elective component of many microbiological enrichment protocols. The

Figure 5-3. Recommended technique for picking colonies to obtain pure cultures.

| 71

Compendium of Methods for the Microbiological Examination of Foods |

first step in the quality assurance process of temperature control is the purchase of the appropriate equipment. Most inexpensive gravity flow connection air incubators can maintain incubation temperatures within about ¡ 3uC of the desired temperature. For protocols requiring more stringent control, forced air or water-jacketed incubators, or shaking or circulating water baths, should be used. Incubators and water baths are usually equipped with temperature controls and indicators. These ‘‘built-in’’ devices should never be relied upon as the sole means of temperature verification. Incubators can have hot and cold spots, so a calibration should be performed annually to establish a relationship between the temperature reading of the internal device and the true temperature at key locations within the chamber. This can be done using a recorder equipped with thermocouples that have been calibrated against a reference thermometer. Thermometers of the proper type (partial, total or complete immersion) and of sufficient accuracy and precision should be permanently placed within incubators and water baths. These thermometers should be calibrated at least once per year against a reference thermometer whose accuracy is certified to be traceable to a National Institute of Standards and Technology (NIST) thermometer. Large incubators should have at least two thermometers, one located towards the top and one towards the bottom. If calibration indicates a correction factor is necessary, this correction factor must be used each time the temperature is noted. NIST-traceable temperature monitoring devices are now available that monitor the temperature continuously.

5.52

Media Management

Each lot of media used for enrichment and isolation, whether prepared from individual ingredients, purchased as a dehydrated blend, or ready to use, should be subjected to quality assurance before use. All incoming media should have the date received written on the package, along with the opened date. An inventory system should be developed that documents the type of medium, supplier, supplier’s lot number, internal lot number, expiry date, size or weight, date received, date opened, date discarded, and the initials of the person responsible for this documentation. First-in/ first-out stock rotation should be practiced. If the manufacturer supplies a certificate of analysis, this should be kept on file. If no certificate is available, the medium should be subjected to quality assurance testing before it is used for analysis. These tests should include tests for performance for intended use, selectivity (if applicable), sterility, appearance and pH (see the chapter ‘‘Laboratory Quality Management Systems’’). If media are prepared in the laboratory, a media preparation log book should be developed. This should include the date made, supplier or internal lot number, the final pH after sterilization, batch size or quantity, autoclave used, autoclave load number, fill volume, pre-sterilization pH (where applicable, i.e., buffers), the initials of the person making the media, and an approval signature. Nonselective/ nondifferential media should be tested for sterility and performance for intended use and/or differentiation. 72 |

5.521 Sterility All new batches of media should be tested for sterility immediately after sterilizing. A randomly chosen tube, agar plate, or bottle is set aside and allowed to cool. The media are then placed at the appropriate temperature and incubated for the appropriate amount of time. After the desired incubation time is completed, the medium is checked for growth, discoloration, or turbidity. Results should be documented and any necessary corrective action taken. Ideally, media should be ‘‘quarantined’’ until the results of quality assurance testing are known. If media must be used before results are known, a trace-back system must be developed in case the quality assurance testing indicates problems. 5.522 Performance for Intended Use Media used for enrichment and isolation need to be tested for performance to help insure against false-negative results. Productivity analysis verifies that the medium, as formulated and prepared, will support the growth of the target microorganism(s). Unfortunately, it is very difficult to detect batches of media that are slightly more inhibitory than normal. Inoculation with freshly passaged laboratoryadapted cultures may result in growth, whereas slight increases in inhibitory properties may not allow the growth of injured target organisms from food or environmental samples. This problem can be partially overcome by using very low-level inocula in liquid media and by using the ecometric technique for agar plate media.9 Appropriate stock cultures should be chosen; media should be incubated at the appropriate temperatures and times and then checked for typical reactions. 5.523 Selectivity Selectivity testing need only be done on media that have selective or differential properties. Selectivity analysis verifies that the medium, as formulated and prepared, will prevent the growth of competing microorganisms. Microorganisms that produce a ‘‘negative’’ reaction (they do not grow or do not produce typical reactions) should be chosen. The newly prepared media should be inoculated with the selectivity control, then incubated at the appropriate temperature and for the appropriate time. The growth or reaction should be documented. Many enrichments require the addition of sterile components after the base medium has been sterilized. Great care should be taken to prevent contamination during this step. Reagents should be tested for sterility by transferring a small amount (usually 1 mL) to a nonselective medium such as Standard Methods Agar or Typticase Soy Broth, then incubated for sufficient time (usually 48 h) to prove sterility. Reagents should be divided into containers in amounts sufficient for a single use. The use of large containers of reagents that are used multiple times should be avoided, as each use increases the chances for contamination. 5.53

Laboratory Environment Management

Enrichment techniques are designed to be as efficient as possible at detecting extremely low levels of target microorganisms. This means that they are extremely sensitive to accidental contamination from the laboratory environment. In addition to meticulous aseptic technique, it is often

| Cultural Methods for the Enrichment and Isolation of Microorganisms

desirable to have a controlled ventilation system to reduce the potential for contamination.15 Laboratories handling certain foodborne infectious agents should meet the general requirements of at least a Biosafety Level 2.4,15 Laboratories should be designed with physical separation between critical areas such as sample check-in, storage, pre-enrichment set-up area, enriched culture transfer areas, and media preparation and sterilization areas. Hands-free wash stations should be conveniently located and stocked with soap. Disposable paper towels should be available. Analysts who work with enriched cultures or highly contaminated samples should do this only in designated areas that are separated from other areas, and should wash their hands and change their laboratory coats before entering other areas of the laboratory. In short, many procedures for limiting cross-contamination that are considered good manufacturing practices in the food manufacturing plant should also be applied to the food microbiology laboratory. The air supply for food laboratories conducting enrichments and isolations should reduce the levels of contamination, lower humidity, and control temperature. Airborne microbiological contamination should be controlled by using filters, and the air quality should be verified by microbiological monitoring (air sampling devices, air settling plates, surface swabs). Typically, total bacteria or yeasts and molds are monitored, but monitoring for specific target organisms may be appropriate. Critical work surfaces should be routinely monitored for the presence of the target organisms being enriched or isolated (see the chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis’’). Great care needs to be taken when transferring preenrichment or enrichment cultures. Analysts should be trained in the elimination of aerosols and microdroplets, and in all aspects of aseptic technique. A useful technique for training new analysts is to place brown paper towels on the laboratory bench during transfer practice exercises. After transfers are complete, the towels can be checked for droplets. Micropipettors should be used only with extreme caution, and the use of micropipette tips containing a filter should be considered. Mixing test tube contents by using a vortex mixer before transfer is not necessary in most cases and can be a source of cross-contamination. Receptacles for contaminated pipettes and micropipettor tips should be located as close as possible to the operation being performed to reduce the potential for dripping. If testing indicates that results may have been compromised by the laboratory environment, policies and procedures must be in place that allow for interpreting, evaluating and reporting equivocal results. Strict documentation at every step provides valuable information for the investigation of equivocal results. This documentation information may include analyst, enrichment time, transfer time, sample order, and rack order.

5.54

Positive Control Cultures

The use of positive control cultures is absolutely necessary to verify that the analysis will detect the target organism. The presence of large numbers of healthy target organisms in the positive control can be a source of laboratory crosscontamination. One way to lessen the consequences of this is to choose specific organisms for the positive control that

are easily distinguishable from typical sample isolates. This can be accomplished by choosing ‘‘rare’’ organisms. An example is the use of Salmonella Abaetetuba for the positive control in the Salmonella assay. Another approach is to use a control strain that has been genetically altered with an easy-to-detect characteristic, such as antibiotic resistance, luminescence, or fluorescence.8

ACKNOWLEDGMENT Fourth edition authors: William H. Sperber, Mark A. Moorman, and Tim A. Freier.

REFERENCES 1. Abbiss, J. S. 1986. A study of the dynamics of selective enrichment of Salmonella. The British Food Manufacturing Industries Research Association. Number 565. 1–27. 2. Atlas, R. M. 2010. Handbook for microbiological media. CRC Press, Inc., Boca Raton, FL. 3. Busta, F. F., and M. L. Speck. 1968. Antimicrobial effect of cocoa on salmonellae. Appl. Microbiol. 16: 424––425. 4. Centers for Disease Control and Prevention. 2009. Biosafety in microbiological and biomedical laboratories, 5th ed. US Government Printing Office, Washington, DC. 5. D’Aoust, J-Y. 1981. Update on pre-enrichment and selective enrichment conditions for detection of Salmonella in foods. J. Food Prot. 44: 369–374. 6. Duffy, G. J. L., Sheridan, R. L. Buchanan, D. A. McDowell, and I. S. Blair. 1994. The effect of aeration, initial inoculum and meat microflora on the growth kinetics of Listeria monocytogenes in selective enrichment broths. Food Microbiol. 11: 429–438. 7. Food and Drug Administration. 2011. Chapter 5. Bacteriological Analytical Manual. Available at: http: //www.fda.gov/Food/ FoodScienceResearch/LaboratoryMethods/ucm070149.htm. Accessed May 30, 2013. 8. Fratamico, P. M., M. Y. Deng, T. P. Strobaugh, and S. A. Palumbo. 1997. Construction and characterization of Escherichia coli O157: H7 strains expressing firefly luciferase and green fluorescent protein and their use in survival studies. J. Food Prot. 60: 1167–1173. 9. Mossell, D. A. A., F. Van Rossem, M. Koopmans, M. Hendricks, M. Verdouden, and I. Eeldrink. 1980. Quality control of solid culture media: A comparison of the classic and the so-called ecometric technique. J. Appl. Bacteriol. 49: 439–454. 10. North, W. R. 1961. Lactose pre-enrichment method for isolation of Salmonella from dried egg albumen. Its use in a survey of commercially produced albumen. Appl. Microbiol. 9: 188–195. 11. Silliker, J. H., and D. A. Gabis. 1973. International Commission on Microbiological Specifications for Foods methods studies. I. Comparison of analytical schemes for detection of Salmonella in dried foods. Can. J. Microbiol. 19: 475–479. 12. Sperber, W. H., and R. H. Deibel. 1969. Accelerated procedure for Salmonella detection in dried foods and feeds involving only broth cultures and serological reactions. Appl. Microbiol. 17: 533–539. 13. Taylor, W. I., and J. H. Silliker. 1961. Isolation of Salmonellae from food samples. IV. Comparison of methods of enrichment. Appl. Microbiol. 9: 484–486. 14. Van Doorne, H., R. M. Baird, D. T. Hendricks, D. Margaretha, D. M. Van der Kreek, and H. P. Pauwels. 1981. Liquid modification of Baird Parker’s medium for the selective enrichment of Staphylococcus aureus. Antonie van Leeuwenhoek. 47: 267–278. 15. World Health Organization. 2004. Chapter 3. Laboratory Biosafety Manual. Available at: http://www.who.int/csr/ resources/publications/biosafety/Biosafety7.pdf. Accessed May 30, 2013.

| 73

|

CHAPTER 6

|

Culture Methods for Enumeration of Microorganisms Ruth L. Petran, Linda E. Grieme, and Sally Foong-Cunningham

6.1

INTRODUCTION

Many analyses performed in food microbiology laboratories involve enumeration of microorganisms present in a sample. Although light microscopy can be used to enumerate microorganisms, the technique suffers from three significant limitations. First, it is difficult to differentiate live from dead cells. Second, it is almost impossible to observe bacteria under light microscopy at cell densities less than 106 per mL. Third, solid materials such as food particles cannot be viewed without mechanical disruption under high-power light microscopy.

6.2

PRINCIPLE

This chapter describes the basic principles and techniques used to perform cultural methods for enumeration of microorganisms. Specific details on media, equipment used, incubation conditions, and interpretation of results are contained in subsequent chapters for specific tasks. Procedures described in the chapters ‘‘Laboratory Quality Management Systems’’ and ‘‘Microbiological Media, Reagents, and Stains’’ should be used for accurate results. The basic principle of these methods is that, following one of the protocols below, the population of microorganisms present in the original sample can be estimated by counting the number of colonies or tubes showing evidence of growth, then multiplying by a dilution factor. By varying the growth medium and incubation conditions, different microorganisms can be enumerated by any of these basic methods. The optimum medium and conditions for determining the colony count may vary from one food to another. However, once a procedure for a given microorganism in a particular food is determined, it can be very useful for routine microbial analysis of the food.

6.3

GENERAL DESCRIPTION

Three general methods employed to estimate the number of viable microorganisms present in the samples are agar plate count procedures, most probable number (MPN) procedures, and the membrane filtration plate count method. The initial stages of these procedures are the same. A portion of sample | 75 |

is measured by weight or volume; a series of dilutions is prepared; then aliquots are added to an agar medium (plate count) or tubes of liquid media (MPN), or they are passed through a membrane filter that retains microorganisms and is then placed on the surface of a growth medium. The agar plate count is the simplest and most commonly used of the three methods. It may be utilized for both liquid and solid food homogenates that have a wide range of microbial counts, if accompanied by sufficient serial dilutions. Depending upon the amount of initial dilution needed for plating, sensitivity of agar plate counts requires that samples contain a minimum of 1 to 100 colony-forming units (CFU; see Section 6.522) per gram or milliliter to result in detectable populations on plates. If the sample microbial load is ,10 CFU per gram or milliliter, the MPN method may be a more informative test. It can also be used for both liquids and solid food homogenates and allows for analysis of large-volume samples. The membrane filter plate count is most applicable to large volumes of liquids with low microbial numbers. It is limited in that it cannot be used if any component of a sample or sample dilution clogs the membrane filter.

6.4

PRECAUTIONS

Because minor variations in procedures can alter the results obtained with the colony count15 and other enumeration methods, the competency and accuracy of the analysts are very important. Knowledge of aseptic techniques is critical. Sterility of media, materials, and equipment is also important.

6.5 6.51

PROCEDURES Dilutions

6.511 Basic Principles Enumeration of microorganisms requires dilution of samples to achieve a population that is countable by the chosen method. Generally, decimal or 10-fold dilutions are used for ease of calculation of final results. A variety of diluents are available or can easily be prepared in the

Compendium of Methods for the Microbiological Examination of Foods |

laboratory including phosphate buffer and 0.1% peptone water. Sterile, distilled water should be used in the preparation of diluents. However, use of plain distilled or deionized water as a diluent is inappropriate, due to the potential for osmotic stress on diluted cells. In fact, use of diluents with high levels of sugar is needed for osmophilic yeast tests. Refer to the appropriate chapter to determine the correct diluent for the organism under consideration.

6.512 Liquids Test portions of non-viscous (i.e., viscosity not greater than milk) liquid products or homogenates may be measured volumetrically using a sterile pipette. Do not insert the pipette more than 2.5 cm below the surface of the sample. Empty the pipette into the diluent (e.g., phosphate-buffered water or 0.1% peptone water) by letting the column drain from the graduation mark to the rest point of the liquid in the tip of the pipette within 2 to 4 sec. Promptly and gently expel the last drop when pipetting the undiluted sample31 or when using a pipette designed to be blown out. Do not rinse the pipette in the dilution water. If the pipette becomes contaminated before completing transfers, replace it with a sterile pipette. Use a separate sterile pipette for transfers from each dilution. Dilution blanks should be at room temperature (15–25uC) when used. Caution: Do not prepare or dispense dilutions or pour plates in direct sunlight. When removing sterile pipettes from the container, do not drag tips over exposed exteriors of the pipettes remaining in the case, because exposed ends of such pipettes are subject to contamination. Do not wipe or drag the pipette across the lips and necks of dilution bottles. Draw test portions above the pipette graduation, and then raise the pipette tip above the liquid level. Adjust to the mark by allowing the pipette tip to contact the inside of the container in such a manner that drainage is complete and excess liquid does not adhere when pipettes are removed from sample or dilution bottles.5 Do not flame pipettes. Pipetter aids, assists, or automatic pipetters that are accurately calibrated and that comply with pipette standards can be used instead of traditional pipetters. Apply all precautions identified for routine pipetting when using automatic pipetters or pipetting devices. 6.513 Solid Sample Homogenates For viscous liquid products or food homogenates, the test portion for the initial dilution should be aseptically weighed (e.g., 11 ¡ 0.1 g into a sterile 99-mL dilution blank, 10 ¡ 0.1 g into 90 mL, 25 ¡ 0.1 g into 225 mL, or 50 ¡ 0.1 g into 450 mL). This provides a 1:10 dilution. Vigorously shake all dilutions 25 times in a 30 cm arc in 7 sec17 and pipette up and down to resuspend cells. Optionally, a mechanical shaker may be used to shake the dilution blanks for 15 sec.31 When trying to remove strongly attached cells from a surface, sonication is found to be effective.6 However, sonication can impact microbial viability by causing cell lysis, depending upon the type of organism and sonication conditions11; therefore it may be desirable to validate microbial recovery by sonication compared with other methods.

76 |

6.52

Plating Techniques

6.521 Basic Principles The introduction of agar media in the late 1800s allowed the development of methods to enumerate microorganisms by colony count. Such methods are used extensively for determining approximate viable microbial populations in foods. These procedures are based on the assumption that each microbial cell in a sample will form a visible, separate colony when immobilized on or mixed with an agar or other solid medium and permitted to grow. Since microorganisms in foods often represent a number of populations with many different growth requirements, some organisms may not be capable of growth under conditions used in colony count methods. Additionally, not all microorganisms exist as single cells, and closely associated clumps or chains of organisms will appear as a single colony. Consequently, the counts are at best an estimate and should not be reported as absolute. The aerobic plate count (see the chapter ‘‘Mesophilic Aerobic Plate Count’’) is the major, but not the only application of the colony count method. A more descriptive evaluation of the microorganisms present in the food sample may be obtained by using several non-selective media and incubating under more than one set of conditions (e.g., temperature, atmosphere). Special consideration should be taken when looking for a specific microorganism in products that contain a large number of competing organisms. Specific microorganisms can be enumerated using selective media, conditions, or both. Bacteria, yeasts, and molds can grow either on or within a nutrient-rich substrate. As such, both a pour plate method and a spread plate method can be used to enumerate microorganisms. 6.522 Precautions and Limitations Colony count methods provide an estimate of the number of viable microorganisms in food according to the medium employed and the time and temperature of incubation. Microbial cells often occur as clumps or groups in foods. Whereas shaking samples and dilutions may uniformly distribute the clumps of bacteria, this may not completely disrupt the clumps themselves. Mixing the initial dilution in a mechanical blender may provide better breakdown of the clumps. However, this does not ensure that the microorganisms will be distributed as single cells. Consequently, each colony that appears on the agar plates can arise from a clump of cells or from a single cell and should be referred to as a CFU. Pipetting up and down can also assist with resuspending cells. Precision is defined as the likelihood of obtaining similar results when the same person or other analysts make repetitive counts. Accuracy is the minimizing of difference between the count obtained and the ‘‘true’’ count. When considering the entire procedure and the results obtained, both are important. The failure of some microorganisms to form visible colonies on the agar medium limits the accuracy of a colony count method. This failure can result from nutritional deficiencies of the medium, unfavorable oxygen tension, unfavorable incubation temperature, or cell injury.

|

Incubation time and temperature also may be factors. The presence of inhibitory substances on glassware or in diluents, or produced by competitive microorganisms in the agar, may adversely affect some microbial cells and limit their ability to form colonies. Another factor that affects apparent counts is the analyst’s ability to see colonies distinctly. This depends on colony separation and morphology. Procedures that enhance colony growth and improve size, shape, contrast, and distribution should be used. The analyst’s eyesight and fatigue may reduce the reliability of the count. Automated plate readers may be available for rapid enumeration of colonies on agar plates. Other factors that may influence the accuracy of the colony count include 1. 2. 3. 4. 5. 6. 7. 8. 9.

improper sterilization and protection of sterilized diluents, media, and equipment; inaccurate measurement of sample and dilutions; improper distribution of the sample in or on the agar medium; unsatisfactory working areas that permit contamination; erratic mixing or shaking of sample or dilution; inaccurate determination of colonies because of the presence of artifacts such as food particles in low dilutions and scratches on plates; improper evaluation of spreaders or pinpoint colonies; plating of the wrong dilution; and other errors in counting and computing counts.

Although there are some inherent limitations in enumerating microorganisms by the colony count method, many of the errors can be minimized if the analyst follows directions carefully and exercises extreme care in making all measurements.5 Consistently accurate and meaningful results can be obtained from the routine examination of a food only if the same procedures are used to analyze each sample of that food. This includes sampling procedures, sample preparation, preparation of dilutions, plating medium, incubation conditions, and counting procedures.

6.523

Pour Plate Method

6.5231 Sample Preparation. The bench area should be cleaned and sanitized. All possible sources of contamination should be removed or reduced to a minimal level. Refer to the chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis’’ for complete details on sample preparation. For viscous or solid foods, an initial 1:10 dilution is usually prepared. High-fat foods such as butter may require use of warm (40uC) diluent to facilitate mixing. 6.5232 Labeling. Label all Petri plates, tubes, and bottles where necessary, with the sample number, dilution, date, and any other desired information. 6.5233 Dilutions. For an accurate count, dilutions should be selected to ensure that plates containing the appropriate number of colonies will be produced. Different ranges for the appropriate number of colonies on plates may be applicable for certain procedures because of the crowding of colonies and other factors. For many methods such as the aerobic

Culture Methods for Enumeration of Microorganisms

colony count, plates should contain between 25 and 250 colonies for accurate counts.28,29 If the count is expected to be in the range of 2,500 to 250,000 per milliliter or gram, prepare plates containing 1:100 and 1:1,000 dilutions. Figure 6-1 shows a schematic drawing of examples for preparing dilutions using a single plate for each dilution (left part refers to the Pour Plate Technique; right part refers to the Spread Plate Technique). For increased accuracy, two or more plates per dilution should be employed.

6.5234 Melting and Tempering Media. Melt agar media in a flowing steam or boiling water, avoiding prolonged exposure to high temperatures. Temper the melted media promptly and maintain between 44 and 46uC until used. Set a thermometer into the water or medium in a separate container similar to that used for the test medium; this temperature control medium must have been exposed to the same heating and cooling as the test medium. Do not depend upon the sense of touch to indicate the proper temperature of the medium when pouring agar. Cold gelling agents may be substituted for agar if previously shown to be equivalent. 6.5235 Plating. When measuring diluted samples of a food into petri plates, lift the cover of the Petri plate just high enough to insert the pipette. Hold the pipette at about a 45u angle with the tip touching the inside bottom of the petri plate. Deposit the sample away from the center of the plate to aid in mixing. If a 1.0- or 1.1-mL pipette is used, allow 2 to 4 sec for the sample to drain from the 1-mL graduation mark to the rest point in the tip of the pipette. Then, holding the pipette in a vertical position, touch the tip once against a dry spot on the plate. Do not blow out. When 0.1-mL quantities are measured, hold the pipette as directed and let the diluted sample drain from one 0.1-mL graduation point down to the next 0.1-mL mark. Do not retouch the pipette to the plate when only 0.1 mL is delivered.31 Replicate plates may be prepared for each dilution plated. Roll tubes, screw-cap tubes, bottles, or other containers may be used as alternatives to Petri plates if all appropriate standardization is made and precautions are considered to assure equivalency. 6.5236 Pouring Agar. After removing tempered agar medium from the water bath, blot the bottle dry with clean towels to prevent water from contaminating the plates. Pour 12 to 15 mL of liquefied medium at 44 to 46uC into each plate by lifting the cover of the Petri plate just high enough to pour the medium. Avoid spilling the medium on the outside of the container or on the inside of the plate lid when pouring. This may require holding the bottle in a near horizontal position or refraining from setting down the bottle between pouring steps. As each plate is poured, thoroughly mix the medium with the test portions in the Petri plate, taking care not to splash the mixture over the edge. This can be accomplished by rotating the plate first in one direction and then in the opposite direction, by tilting and rotating the plate, or by using mechanical rotators. Allow agar to solidify (no longer than 10 min) on a level surface. | 77

Compendium of Methods for the Microbiological Examination of Foods |

Figure 6-1. Preparation of dilutions from a nonviscous liquid food sample.

Select the number of samples to be plated in any one series so that not more than 20 min (preferably 10 min) elapse between diluting the first sample and pouring the last plate in the series.2,18 Should a continuous plating operation be conducted by a team, plan the work so that the time between the initial measurement of a test portion into the diluent or directly into a dish and the pouring of the last plate for that sample is not more than 20 min. Avoid stack pouring unless the plates are distributed singly on a cooling surface immediately after mixing.20 Note that to obtain countable plates for foods having low colony counts, low dilutions must be used. For some foods this results in the presence of many food particles in the plate, which makes it difficult to distinguish the colonies easily for accurate counting. This problem often can be overcome by adding 1 mL of 0.5% (wt/vol) 2,3,5triphenyltetrazolium chloride (TTC) per 100 mL of melted agar medium just prior to pouring the plates. Most bacteria form red colonies on an agar medium containing TTC. Counts should be made initially with and without TTC to determine if the TTC has any deleterious effect on the count. The TTC should be prepared as an aqueous solution and sterilized by passage through a sterilizing filter. To avoid decomposition, the solution must be protected from light and must not be exposed to excessive heat. Sterility controls of medium, diluents, and equipment are recommended. Pour control plates for each lot of dilution blanks, medium, Petri plates, and pipettes.

6.5237 Incubation. After solidification, invert the plates to prevent spreaders as appropriate, and promptly place them in 78 |

the incubator. Incubation time should be sufficient for a single organism to form a visible colony on the media selected. Plates for enumeration of yeasts and molds are not inverted during incubation; refer to the chapter ‘‘Detection and Enumeration of Heat-Resistant Molds’’ for further information. Incubation conditions for specific methods and commodities are presented in appropriate chapters. Agar within the plates should equilibrate to incubation temperature within 2 hr. Slower equilibration caused by excessive height of stacked plates or crowded incubators must be avoided. Avoid excessive humidity in the incubator to reduce the tendency for spreader formation, but prevent excessive drying of the medium by controlling ventilation and air circulation. Agar in plates should not lose weight by more than 15% during 48 hr of incubation. Under some conditions, humidity control may become essential.

6.5238 Counting Colonies. Count colonies with the aid of magnification under uniform and properly controlled artificial illumination, using a tally. Routinely use a colony counter25 equipped with a guide plate ruled in square centimeters. Examine plates in subdued light. Try to avoid mistaking particles of undissolved medium, sample, oil droplets, or precipitated matter in plates for pinpoint colonies. Examine doubtful objects carefully, using higher magnification, if necessary, to distinguish colonies from foreign matter. A stereo microscope or magnifying glass may be useful for this examination. Carelessness, impaired vision, or failure to recognize colonies can lead to erroneous results. It is generally suggested that laboratory workers who cannot duplicate their own counts on the same plate

|

within 5% and the counts of other analysts within 10% should discover the cause(s) and correct such factors.9 However, others indicate that these percentages should be 7.7% intra-analyst and 18.2% inter-analyst.12 Schedules of the laboratory analyst should be arranged to prevent eye fatigue and the inaccuracies that inevitably result from this. Count all colonies on selected plates containing the appropriate number of colonies promptly after the incubation period.28 Refer to Section 6.5239 for guidelines on selecting plates and computing counts. If impossible to count at once, the plates may be stored, after the required incubation, at approximately 0 to 4.4uC for less than 24 hr.12 This should not be a routine practice.

6.5239 Computing and Reporting. Record the dilution used and the number of colonies counted or estimated Table 6-1. Selected MPN Estimates and 95% Confidence Limits8 for Fermentation Tube Tests When Three Tubes With 0.1 g, 0.01 g, and 0.001 g Sizes Useda No. of Positive Tubes/3 Tubes

95% Confidence Limits b

0.1 g

0.01 g

0.001 g

MPN/g

0 0 1 1 1 1 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3

0 1 0 0 1 2 0 0 1 1 2 0 0 1 1 2 2 2 3 3 3 3

0 0 0 1 0 0 0 1 0 1 0 0 1 0 1 0 1 2 0 1 2 3

,3 3+ 4 7+ 7 11+ 9 14+ 15 20+ 21 23 39 43 75 93 150 210+ 240 460 1,100 .1,100

Lower

Upper

— ,1 ,1 2 2 4 2 5 5 7 8 9 10 10 20 30 50 80 90 100 300 —

— 17 21 27 28 35 38 48 50 60 62 130 180 210 280 380 500 640 1,400 2,400 4,800 —

a Normal results, obtained in 95% of tests, are not followed by a plus (+). Less likely results, obtained in only 4% of tests, are followed by a plus (+). Combinations of positive tubes not shown occur in less than 1% of tests, and their frequent occurrence indicates that technique is faulty or that assumptions underlying the MPN estimate are not being fulfilled. MPN estimates for combinations that are not shown may be obtained by extrapolation (or by Thomas’ formula, Section 6.534) to the next highest combination that is shown in the table. For example, a result of 20-2 has an MPN of approximately 20, which is the MPN for a more likely result of 2-1-1. b All values under MPN/g in this table may be multiplied by 100 for reporting MPN/100 g.

Culture Methods for Enumeration of Microorganisms

on each plate. To compute colony counts, multiply the total number of colonies per plate (or the average number of colonies from replicate plates if the same dilution is used) by the reciprocal of the dilution used. To avoid giving false ideas of precision and accuracy when computing colony counts, record only the first two left-hand digits. Raise the second digit to the next highest number only when the third digit from the left is 5, 6, 7, 8, or 9; use zeros for each successive digit to the right of the second digit (Table 6-1). Report counts (or estimates thereof) as CFU per gram or milliliter, as applicable. When counts on duplicate plates or consecutive dilutions are averaged, round off counts to two significant figures only at the time of conversion to the CFU per gram (Example 2G). The appropriate number of colonies to count on a plate is a function of colony size, plate size, and size of differential properties produced on the medium. Typically, 25 to 250 colonies per plate yield reliable results.28,29 Use this as a guide unless an alternate range is indicated for specific methods. The following guidelines or ‘‘rules’’ should be used for selecting plates and calculating the CFU per gram or milliliter, as applicable: 1. One plate with 25 to 250 colonies: Select a plate with 25 to 250 colonies (Examples 1A and 1B) unless excluded by spreaders or lab accidents (Examples 1C and 1D; see also Rule 8). Count all colonies, including those of pinpoint size, and record the dilution used and the total number of colonies counted. Example 1: Colonies Counted Dilution 1:100 1:1,000

Colony Countb (CFU/g or mL)

Count Ratioa Common application, one plate from each of two dilutions: 1A 234c 23 — 23,000 1B 305 42 — 42,000 1C Sprd 31 — 31,000 1D 243 LAe — 24,000 a

Count ratio is the ratio of the greater to the lesser plate count, as applied to plates from consecutive dilutions having between 25 and 250 colonies. b All counts should be made in accordance with instructions in Section 6.5238, as well as any other rules listed or given in the text. c Underlined values used to calculate count. d Spreader (Spr) and adjoining area of repressed growth covering more than one-half of the plate. e LA, laboratory accident. 2. Duplicate plates: Count 25 to 250 colonies and average the counts to obtain the colony count (Example 2A). If only one plate of a duplicate pair yields 25 to 250 colonies, count both plates (unless excluded by spreaders), and average the counts (Examples 2B and 2C). When counting duplicate plates from consecutive decimal dilutions, compute the count per gram for each dilution and proceed as in Rule 3 (Examples 2D through 2G). | 79

Compendium of Methods for the Microbiological Examination of Foods |

Example 2: Colonies Counted Dilution Colony Countb a 1:100 1:1,000 Count Ratio (CFU/g or mL) Procedure where two plates per dilution are poured: 2A 175 16 — 19,000 208 17 2B 239 16 — 28,000 328 19 2C 275 24 — 30,000 280 35 2D 228 28 1.2 25,000 240 26 2E 138 42 2.4 15,000 162 30 2F 228 28 1.1 24,000 240 23 2G 224 28 1.4 24,000 180 Spr a,b

See Rule 1.

3. Consecutive dilutions with 25 to 250 colonies: If plates from two consecutive decimal dilutions yield 25 to 250 colonies each, compute the count per gram for each dilution and report the arithmetic average as the CFU per gram (Example 3A; also 2D, 2F, and 2G), unless the higher computed count is more than twice the lower one. In that case, report the lower computed count as the CFU per gram (Examples 3B and 2E).

5. All plates have fewer than 25 colonies: If plates from all dilutions yield fewer than 25 colonies, record the actual number of colonies on the lowest dilution (unless excluded by spreaders) and report count as est. CFU per gram (Examples 5A and 5B). Example 5: Colonies Counted Dilution Colony Countb a 1:100 1:1,000 Count Ratio (CFU/g or mL) Common application, one plate from each of two dilutions: 18 2 — 1,800 est. 5A Procedure where two plates per dilution are poured: 5B 18 2 — 1,700 est. 16 0 a,b

See Rule 1.

6. Plates with no colonies: Inhibitory substances in a sample may be responsible for the lack of colony formation. The analyst may suspect the presence of inhibitory substances in the sample under examination when plates show no growth or show proportionately less growth in lower dilutions. Such developments cannot, however, always be interpreted as evidence of inhibition, and unless inhibition is demonstrated, should be reported as laboratory accident. If plates from all dilutions have no colonies and inhibitory substances have not been detected, report the estimated count as less than one times the corresponding lowest dilution (Examples 6A and 6B). Example 6: a,b

See Rule 1.

Example 3: Colonies Counted Dilution 1:100 1:1,000

Colony Countb (CFU/g or mL)

Count Ratioa Common application, one plate from each of two dilutions: 3A 243 34 1.4 29,000 3B 140 32 2.3 14,000

Colonies Counted Dilution Colony Countb a 1:100 1:1,000 Count Ratio (CFU/g or mL) Common application, one plate from each of two dilutions: 6A 0 0 — ,100 est. Procedure where two plates per dilution are poured: 6B 0 0 — ,100 est. 0 0

a,b

See Rule 1.

4. No plate with 25 to 250 colonies: If there is no plate with 25 to 250 colonies and one or more plates have more than 250 colonies, select plate(s) having nearest to 250 colonies and count as in Rule 7 for crowded plates. Report count as the estimated (est.) CFU per gram (Examples 4A and 4B). Example 4: Colonies Counted Dilution Colony Countb a 1:100 1:1,000 Count Ratio (CFU/g or mL) Common application, one plate from each of two dilutions: 4A 325 20 — 33,000 est. Procedure where two plates per dilution are poured: 4B 287 23 — 28,000 est. 263 19 a,b

See Rule 1.

80 |

7. Crowded plates (.250 colonies): If the number of colonies per plate exceeds 250, count colonies in portions of the plate that are representative of colony distribution to estimate the aerobic colony count. If there are fewer than 10 colonies per square centimeter, count the colonies in 12 cm2, selecting six consecutive squares horizontally across the plate and six consecutive squares at right angles, being careful not to count a square more than once. When there are more than 10 colonies per cm2, count the colonies in four representative squares. In both instances, multiply the average number of colonies per square centimeter by the area of the plate to determine the estimated number of colonies per plate. Individual laboratories should determine the area of the plate and the proper factor for multiplication; however, the area of a standard 15 6 100 mm plastic Petri plate is approximately 56 cm2 and therefore the appropriate factor is 56. For an example using an average count of 15 colonies per

|

square centimeter on a 56 cm2 plate, see Example 7A. Do not report counts on crowded plates from the highest dilution as ‘‘too numerous to count’’ (TNTC). Where bacterial counts on crowded plates are .100 colonies/ cm2, report as greater than the plate area multiplied by 100, multiplied by the highest dilution plated. For example, for a 56 cm2 plate, the count would be 5,600 times the highest dilution plated. Report as estimated CFU per gram (Example 7B). When all colonies on a plate are accurately counted and the number exceeds 250, report as estimated CFU per gram (Examples 4A and 4B). Example 7: Colonies Counted Dilution Colony Countb a 1:100 1:1,000 Count Ratio (CFU/g or mL) Common application, one plate from each of two dilutions: 7A TNTC 840 — 840,000 est. 7B TNTC 7150 — .5,600,000 est.

6.524 Surface or Spread Plate Method Methods of plating designed to produce all surface colonies on agar plates have certain advantages over the pour plate method.19 The use of translucent media is not essential with a surface or spread plate but is necessary with the pour plate to facilitate location of colonies. The colonial morphology of surface colonies is easily observed, improving the analyst’s ability to distinguish between different types of colonies.23 Organisms are not exposed to the heat of the melted agar medium, so higher counts may be observed in some situations.3,9,24,30 On the other hand, since relatively small volumes (0.1–0.5 mL) of the sample must be used, the method may lack precision for samples containing few microorganisms. 1.

2.

a,b

See Rule 1.

8. Spreaders: There are three distinct types of spreaders.31 The first type is a chain of colonies, not too distinctly separated, that appears to be caused by disintegration of a bacterial clump when the inoculum is dispersed in or on the plating medium. If one or more chains appear to originate from separate sources, count each as one colony. Do not count each individual colony in such chain(s) as separate colonies. The second type of spreading colony develops in a film of water between the agar and the plate.9 The third type forms in a film of water at the edge or over the surface of the agar. These two types develop mainly because of moisture accumulation at the point from which the spreader originates, and these spreaders may repress the growth of individual colonies. When dilution water is uniformly distributed throughout the medium, bacteria rarely develop into spreading colonies. Steps to eliminate spreaders of this type should be taken if 5% of a laboratory’s plates have spreaders covering 25% of the plate. If spreaders occur on the plate(s) selected, count colonies on representative portions thereof only when colonies are well distributed in spreader-free areas and the area covered by spreader(s), including the total repressed growth area if any, does not exceed 50% of the plate area. Calculate the estimated count by multiplying the average count per square centimeter by the area of the plate. Where the repressed growth area alone exceeds 25% of the total area, report as spreaders (Spr; Example 8A). Refer also to Example 1C for additional information when handling spreaders. Example 8: Colonies Counted Dilution Colony Countb a 1:100 1:1,000 Count Ratio (CFU/g or mL) Common application, one plate from each of two dilutions: 8A Spr Spr — Spr a,b,d

See Rule 1.

Culture Methods for Enumeration of Microorganisms

3. 4. 5.

6.

7. 8.

Prepare sample (see the chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis’’). Pour approximately 15 mL of the agar into sterile Petri plates. To facilitate uniform spreading, the surface of the solidified agar should be dried by holding the plates at 50uC for 1.5 to 2 hr. Plates may also be dried at a lower temperature (25–35uC) for longer periods (18– 24 hr), or in a laminar flow hood, with covers ajar, for 0.5 to 1 hr. Label all Petri plates as in pour plate method. Prepare 10-fold serial dilutions following general procedures described previously. Measure 0.1 mL of diluted sample onto agar surface using a sterile pipette (graduated into 0.1-mL divisions). Larger volumes may be appropriate under certain situations, but take precautions to ensure that the liquid does not remain on the agar surface to promote spreaders. For example, 0.2 mL of a 1:10 dilution can be delivered to each of five plates to get the equivalent of 1 g of food for CFU determinations. Spread the diluted sample on the surface of the agar medium with a sterile bent glass rod (hockey stick), sterile cell spreaders, or equivalent, as quickly and carefully as possible. Use a separate sterile cell spreader, sterilize the glass rod in between each plate, or spread the plates for a given sample starting with the most dilute plate and proceed to the least dilute plate in series with aseptic technique throughout. Allow the plates to dry at least 15 min prior to inversion. Yeast and mold plates do not require inversion. Incubate plates. Compute and record colony counts as in pour plate method (Sections 6.5238 and 6.5239).

6.53

Most Probable Number (MPN) Techniques

6.531 Basic Principles As a sample is serially diluted, some of the aliquots eventually contain such small amounts of sample that they will contain no microorganisms. The MPN method is based on diluting out of the microorganisms, and therefore may be described as the ‘‘multiple tube dilution to extinction method.’’ The most satisfactory information is obtained when

| 81

Compendium of Methods for the Microbiological Examination of Foods |

all of the tubes with the large sample portions show growth and the tubes with the smaller portions show no growth. The MPN dilution technique uses results that are reported as positive or negative in one or more decimal dilutions of the sample to estimate the number of organisms present. Thus, unlike the aerobic plate count, the MPN does not provide a direct measure of the bacterial count. In addition, the MPN is more variable than the plate count.21 Although the MPN is not a precise measure, a specific value can be computed for a single dilution13 or for multiple dilutions,14 provided the results are not all positive or negative for all dilutions used and assuming that the organisms to be measured are distributed randomly throughout the sample and that growth will occur when one or more organisms are present in a tube.4 Halvorson and Ziegler14 demonstrated that, for a multiple tube MPN, accuracy depends only on the number of tubes per dilution; for single dilution tests,13 it depends on bacterial population and number of tubes. Eisenhart and Wilson10 and Oblinger and Koburger22 discuss the early history of dilution techniques. The latter article is useful for training students in understanding the test. The composition of many food products and ingredients makes it difficult to use standard plating procedures, particularly when the microbial concentration of the sample is less than 10 CFU per gram. Ziegler and Halvorson33 showed that, in these low-count situations, the MPN technique gave higher values for bacterial populations than did the plate count method. The direct microscopic count gave the same value as the plating and MPN method only when it was used on cultures that had not entered the death phase. McCarthy et al.21 also demonstrated a considerable positive mathematical bias in MPN values relative to plate counts. Applications of the MPN method are numerous. Use of the method is particularly important in the standard coliform procedure used for water and wastewater testing, and in testing foods in general. The method is also used in the isolation and enumeration of staphylococci, streptococci, Vibrio parahaemolyticus, and salmonellae when quantitative rather than qualitative analysis is necessary. The method also can be applied when a single sample dilution is used in several tubes (e.g., five 0.1 g samples for enumeration of very low numbers of organisms). For this type of application, special tables are required.1,22 Because the method uses liquid media, it offers the user considerable flexibility as to sample size. If allowances for appropriate dilutions of sample and ratios of medium to sample are made, sample volumes can be quite large. Increasing the number of tubes within each effective dilution improves precision. At low population levels, sensitivity is generally greater with the MPN than with the plate count31; however, this is not always the case. The ‘‘bathing’’ aspect of nutrient availability in a liquid medium may enhance recovery of organisms. Subsequent transfer of samples to a more inhibitory environment is possible after a period of resuscitation. Variation among replicate aliquots is by far the most important source of error.29 Extreme care is needed in preparing dilutions and transferring aliquots within the 82 |

same dilutions. Other important factors that contribute to spurious results include difficulty in obtaining truly representative samples from a given lot and the possibility of an uneven distribution of microorganisms within the sample units selected.26 If the sample contains inhibitory substances, or the product itself is inhibitory (e.g., sodium chloride), growth in the tubes with high concentrations of sample may be inhibited. The possibility of injured cells that cannot grow out should not be overlooked. Nutrient in the sample may also interfere with the selectivity of the medium. For example, sucrose in a food will lead to a false indication of the presence of coliforms. One set of tubes from each batch of medium prepared should be used as an uninoculated control. If, for example, the five-tube MPN method is being used, a set of five tubes should be incubated as uninoculated controls to ensure that the medium was properly sterilized.

6.532 1. 2. 3. 4.

5.

6.

Procedures

Preparation: Prepare sample (see chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis’’). Labeling: Label all tubes. Dilutions: Unless previous experience with a sample indicates the appropriate number of dilutions needed, a minimum of five 10-fold dilutions should be prepared. Inoculating tubes: Usually three or five tubes are inoculated for each dilution used within 20 min of preparation of the initial dilution. Typically, one part of sample to 10 parts of medium should be maintained; for example, 0.1 g sample should be dispersed in 1 mL of medium, or 1-mL aliquots into 10 mL of broth. The strength of the medium can be adjusted so that the concentration of medium after the sample is added equals single strength medium. For example, in water analysis, one frequently uses double-strength broth with an equal sample volume to avoid excessive nutrient and inhibitor dilution. Incubation: Incubation conditions for specific methods and commodities are presented in appropriate chapters for each analysis. An air incubator or a water bath may be used. Detection of positive tubes: a. Turbidity: When using samples that do not cloud the medium in the tubes, the development of turbidity after incubation indicates growth (positive tubes). When the sample causes turbidity, other methods must be used to determine positive tubes. b. Metabolic end products: i. Detection of gas production: Gases produced by developing microorganisms can be captured and observed with gas traps or inverted vials that are placed in the medium in the growth vessels before sterilization. A positive reaction is recorded when gas bubbles are observed in traps at the end of the incubation period. Other methods used to capture and observe the gases produced include overlay with vaspar or agar. These are useful only when the microorganisms to be enumerated

|

Table 6-2. Selected MPN Estimates and 95% Confidence Limits8 for Fermentation Tube Tests When Five Tubes With 0.1 g, 0.01 g, and 0.001 g Volumes Useda No. of Positive Tubes/5 Tubes

95% Confidence Limits b

0.1 g

0.01 g

0.001 g

MPN/g

0 0 0 1 1 1 1 2 2 2 2 2 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5

0 0 1 0 0 1 2 0 0 1 1 2 0 0 1 1 2 2 3 0 0 1 1 2 2 3 3 4 0 0 1 1 1 2 2 2 3 3 3 4 4 4 4 4 5 5 5 5 5 5

0 1 0 0 1 0 0 0 1 0 1 0 0 1 0 1 0 1 0 0 1 0 1 0 1 0 1 0 0 1 0 1 2 0 1 2 0 1 2 0 1 2 3 4 0 1 2 3 4 5

,2 2+ 2 2 4+ 4 6+ 4 7+ 7 9+ 9 8 11 11 14+ 14 17+ 17+ 13 17 17 21 22 26+ 27 33+ 34+ 23 31 33 46 63+ 49 70 94+ 79 110 140 130 170 220 280+ 350+ 240 350 540 920 1,600 .1,600

Lower

Upper

— ,1 ,1 ,1 1 1 2 1 2 2 3 3 3 4 4 6 6 7 7 5 7 7 9 9 12 12 15 16 9 13 14 20 22 21 30 40 30 40 60 50 70 100 120 160 100 100 220 300 600 —

— 10 10 11 15 15 18 17 20 21 25 25 24 29 30 35 35 40 41 38 45 46 55 56 65 67 77 80 68 110 120 150 180 170 210 250 250 300 360 390 480 580 690 820 940 1,300 2,000 2,900 5,300 —

Culture Methods for Enumeration of Microorganisms

Table 6-2. (continued ) a Normal results, obtained in 95% of tests, are not followed by a plus (+). Less likely results, obtained in only 4% of tests, are followed by a plus (+). Combinations of positive tubes not shown occur in less than 1% of tests, and their frequent occurrence indicates that technique is faulty or that assumptions underlying the MPN estimate are not being fulfilled. MPN estimates for combinations that are not shown may be obtained by extrapolation (or by Thomas’ formula, Section 6.534) to the next highest combination that is shown in the table. For example, a result of 40-2 has an MPN of approximately 21, which is the MPN for a more likely result of 4-1-1. b All values under MPN/g in this table may be multiplied by 100 for reporting MPN/100 g.

are known to produce gas under the conditions of the test. If tubed media are stored at low temperature, small bubbles may accumulate in the inverted fermentation tubes when media is brought to incubation temperatures due to air dissolving in the cold medium. Steaming or boiling the tubed media before use removes these bubbles; however, one must consider the possibility of denaturing sensitive medium components (depending on the medium being used). ii. Detection of acid or base: Acid or base production can be determined after incubation by measuring the pH or titratable acidity in each tube or by using a medium containing a pH-indicating dye. Detecting positive tubes by this method requires that the microorganisms being enumerated produce a pH change from a defined substrate. iii. Detection with reduction methods: Electron acceptors (e.g., resazurin, methylene blue, or 2, 3, 5-triphenyltetrazolium chloride) that change color upon reduction can be incorporated into the medium. Reduction of any of these compounds by microbial action also indicates growth. iv. Other: Specific media can be developed to assay for certain metabolic activities (e.g., NO3 reduction, indole production, starch hydrolysis, and H2S production, depending on the information desired). 7. Confirming inconclusive tests: a. Direct microscopic examination: Microscopic examination is done by placing a loopful of the medium from the tube on a slide, drying, heat fixing, and staining. The slide is examined for the specific microorganism using oil immersion at 1,000 6 magnification. Care should be taken if the original sample contains high numbers of killed or inactivated microorganisms to distinguish live from dead cells using procedures such as dye techniques. b. Subculturing: To confirm growth in questionable tubes, medium from the tube is transferred to a non-selective medium and incubated for an appropriate additional period of time. Growth in this medium confirms the presence of viable | 83

Compendium of Methods for the Microbiological Examination of Foods |

Table 6-3. Examples of Determining MPN Estimates: Three-Tube Series (1-mL Sample Aliquot/Tube) Sample Volume (g) Example

a b c d e a

0.1

3/3 3/3 0/3 3/3 3/3

a

Reported

MPN

0.01

0.001

0.0001

0.00001

Positive Values

Estimate/g

3/3 3/3 0/3 3/3 3/3

2/3 3/3 1/3 2/3 3/3

0/3 2/3 0/3 1/3 3/3

0/3 0/3 0/3 1/3 3/3

3-2-0 3-2-0 0-1-0 3-2-2 3-3-3

930 9,300 30 2,100 .110,000

Numerator/denominator 5 no. positive tubes/no. tubes inoculated.

per gram (or mL).’’ When specific groups of microorganisms have been estimated, results can be reported as a ‘‘presumptive MPN estimate’’ for that specific group until appropriate confirmatory tests have been completed. Tables 6-1 and 6-2 show the MPNs of microorganisms corresponding to the frequency of positive tubes obtained from three 1:10 dilution series beginning with 0.1 g test portions. Results for both three tubes and five tubes per dilution are given along with the 95% confidence limits. Tables 6-3 and 6-4 give examples for determining MPN estimates for three-tube and five-tube MPN series, respectively, when 1-mL sample aliquots from serial dilutions are planted. Note that the tabular values are treated in terms of the actual sample volumes planted in these dilutions. When more than three dilutions of a sample are prepared, the results from only three consecutive dilutions are used in determining the MPN. First, for all dilutions having all tubes positive, select the highest dilution (smallest sample size). Second, use the next two higher dilutions (smaller sample sizes), as shown in Examples a and b of Tables 6-3 and 6-4. When none of the tested dilutions yield all tubes positive, select the first three consecutive dilutions for which the middle dilution contains the positive result, as in Example c of Tables 6-3 and 6-4. If a positive result occurs in a higher dilution (smaller sample size) than the three selected, add the number of positive tubes in this dilution to the highest dilution (smallest sample size) of the three selected, as in Example d of Tables 6-3 and 6-4. When all dilutions tested have all tubes positive, select the three highest dilutions (smallest sample sizes), as in Example e of Tables 6-3 and 6-4. Often it is necessary to calculate the MPN from initial sample sizes other than those listed in Tables 6-1 and 6-2. If the largest (greatest) sample size used for the table reference is 0.01 g, multiply the MPN index listed in the table by 10. Thus, results of a five-tube MPN determination

microorganisms in the tube. To confirm a questionable growth reaction (such as acid) in tubes of selective media with heavy food turbidity, media from the incubated tube is transferred to a tube of the identical medium and similarly incubated. Growth reactions in such subcultures can be readily observed since they are free of the color and turbidity. Aliquots may also be streaked from MPN tubes onto either selective or non-selective media to ascertain appropriate growth.

6.533 Use of MPN Tables Because the MPN procedure provides an estimate of the count present, confidence intervals are used to indicate the precision of the MPN estimates. If we are considering a 95% confidence interval, then the true but unknown number of organisms in the sampled population lies within the limits 95% of the time. Tables in this book follow those shown in the Standard Methods for the Examination of Dairy Products.31 Halvorson and Ziegler14 presented the formulae for computing the probability of a combination (i.e., 3-2-0) for given dilutions and organism concentrations. Tables 6-2 and 6-3 provide combinations of positive results that occur frequently enough to be statistically significant. Combinations that occur less than 1% of the time are omitted. When compared to other references,1,22,29 reported confidence limits are slightly different, a situation attributed to assumptions made and computational methods used to derive the values. Review the original work if further insight into these tabular differences or the methods used to compute the values are needed.7,8 If the appropriate combination of positive results is not found, the analysis should be repeated or more complete tables should be consulted. When the multiple tube method is used, results are usually reported as ‘‘the most probable number of microorganisms

Table 6-4. Examples of Determining MPN Estimates: Five-Tube Series (1-mL Sample Aliquot/Tube) Sample Volume (g)

a

Reported

MPN

Example

0.1

0.01

0.001

0.0001

0.00001

positive values

estimate/g

a b c d e

5/5a 5/5 0/5 5/5 5/5

5/5 5/5 0/5 5/5 5/5

2/5 5/5 1/5 3/5 5/5

0/5 2/5 0/5 1/5 5/5

0/5 0/5 0/5 1/5 5/5

5-2-0 5-2-0 0-1-0 5-3-2 5-5-5

490 4,900 20 1,400 .160,000

Numerator/denominator 5 no. positive tubes/no. tubes inoculated.

84 |

|

showing three positive 0.01 g portions, two positive 0.001 g portions, and one positive 0.0001 g portion (3-2-1) are read from Table 6-2 as 17 and multiplied by 10 to arrive at 170 as the actual MPN/g for the sample. Similarly, if the greatest portion used for the table reference is 1 g rather than 0.1 g, divide the MPN derived from the table by 10. Thus, the result of a three-tube MPN determination for salmonellae showing three positive 1 g portions, one positive 0.1 g portion, and zero positive 0.01 g portions (3-1-0) is read from Table 6-1 as 43 and divided by 10, or 4.3 as the presumptive MPN per gram for the sample. An alternative approach to obtain the MPN per gram uses the following formula26: [(MPN/g from Table/100)] 6 dilution factor of the middle tube 5 MPN/g. To obtain an MPN/100 g, multiply by 100. MPN estimates are often credited with unfounded precision. The tabular MPN estimate represents a range and not an absolute value. Most MPN tables29,31 include 95% confidence limits for the tabular MPN estimates. The true number of organisms lies between these limits 95% of the time. One must be able to read each table properly and understand the significance of the results. For a three-tube test, the 95% confidence limits cover a 33-fold range from approximately 14% to 458% of the actual tabular MPN estimate and for a fivetube dilution test, the 95% confidence limits cover a 13-fold range from approximately 24% to 324% of the MPN.32 Results should be recorded as ‘‘Number of microorganisms per quantity (g or mL) of sample by the MPN method,’’ for example, coliform MPN/g 5 11. With the report of microbiological counts by the MPN method, the number of tubes used in each dilution is included, that is, five-tube MPN or three-tube MPN, and the particular method used.

6.534

Approximate MPN and 95% Confidence Limits Thomas27 published a simple formula to approximate MPNs. The results may not exactly agree with those in the MPN tables; however, deviations are usually slight and of no practical consequence. Additionally, the formula is not restricted to the number of tubes and dilutions used and may be applied to all types of tests. Thomas’ approximation is given by the following: pffiffiffiffiffiffiffiffi MPN/g 5 P/ NT where P 5 number of positive tubes N 5 total quantity of sample (g) in all negative tubes T 5 the total quantity of sample (g) in all tubes For example, consider the two-fold dilution series given below: Grams of Sample 8 4 2 1 0.5 0.25

No. of Tubes 5 5 5 5 5 5

No. of Positive Tubes 5 4 2 0 1 0

Culture Methods for Enumeration of Microorganisms

P 5 (5 + 4 + 2 + 1) 5 12 N 5 (8 6 0) + (4 6 1) + (2 6 3) + (1 6 5) + (0.5 6 4) + (0.25 6 5) 5 18.25 T 5 5(8 + 4 + p 2 +ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 + 0.5 + 0.25) 5 78.75 ffi MPN/g 512/ (18:25)(78:75)5 0.32/g Estimates of the 95% confidence limits can be obtained as follows:4 pffiffiffiffiffiffiffiffiffiffiffiffiffiffi log (MPN/g) ¡ (1.96)(0.55) ( log a)/n, where a 5 the dilution ratio n 5 the number of tubes per dilution For the above MPN example, the approximate 95% confidence limits are as follows: pffiffiffiffiffiffiffiffiffiffiffiffiffiffi log 0.32 ¡ (1.96)(0.55) ( log 2)/5 5 –0.495 ¡ 0.265 Then the lower limit is antilog (–0.76) 5 0.17/g and the upper limit is antilog (–0.23) 5 0.59/g.

6.54

Membrane Filtration Plate Count Method

For certain foods or food ingredients, the ability to test relatively large samples will improve the accuracy of quantitative microbiological analyses. Large volumes of liquid foods or solutions of dry foods that can be dissolved and passed through a bacteriological membrane filter (pore size 0.45 mm) may be analyzed for microbial content by the membrane filter method. The method is especially useful for samples that contain low numbers of bacteria. Additional details on rapid and commercial kits using membrane filtration methods are described in the chapter ‘‘Rapid Methods for the Detection and Identification of Foodborne Pathogens.’’

6.6

ALTERNATIVE METHODS

A number of innovative and convenient methods are commercially available for enumeration of microorganisms. These include Petrifilm Plate method, pectin gel method, calibrated loop method, drop plate method, Hydrophobic Grid-Membrane Filter method, Spiral Plate method, SimPlate, impedance, and luminescence. Refer to the chapter ‘‘Culture Methods for Enumeration of Microorganisms’’ for general procedures and information.

6.7

ANAEROBIC OR OTHER ATMOSPHERES

The choice of atmospheric conditions to which the plates will be exposed is vital to the successful enumeration of microorganisms. The three most commonly used atmospheres for the growth of microorganisms are aerobic, anaerobic, and microaerophilic. If the preferred atmosphere is aerobic, incubation is conducted in an incubator under conventional atmosphere. An atmosphere devoid of measurable oxygen is preferred for the manipulation and enumeration of anaerobic microorganisms such as Clostridium spp. There are a number of methods that enable a researcher to achieve an anaerobic atmosphere. For example this can be | 85

Compendium of Methods for the Microbiological Examination of Foods |

achieved with an incubator in a sealed hood with an atmosphere of approximately 85% nitrogen and 15% hydrogen. Use of anaerobe jars with a sachet system is the most commonly used method to achieve anaerobiosis for incubation of plates. Under conditions of anaerobiosis it is important to verify the absence of oxygen within the system. A number of testing methods are available, ranging from sophisticated electrical sensors to rudimentary indicator strips. Methods for determining total anaerobic colony counts are similar to those outlined in the chapter ‘‘Mesophilic Anaerobic Sporeformers,’’ except that steps to eliminate vegetative cells are omitted. If the analyst is concerned with enumerating strict or sensitive anaerobes, methods developed at the Anaerobe Laboratory, Virginia Polytechnic Institute and State University, may be of help.16 In particular, the Hungate method with pre-reduced, anaerobically sterilized media in tubes with butyl rubber stoppers is suggested for analysis of food samples for obligatory anaerobic bacteria. The specific procedure for enumerating anaerobic or microaerophilic organisms can follow individual steps of the pour plate or spread plate techniques (Sections 6.523 and 6.524). Anaerobic media may include plate count agar, Andersen’s pork pea agar, blood agar, or similar complex media and may be overlaid with thioglycollate agar. Incubation is under anaerobic conditions in an anaerobic culture jar, an anaerobic incubator, or some similar container that can contain and maintain an atmosphere free of oxygen. Such atmospheres can be achieved using an incubator connected to a free-flowing cylinder of an appropriate gas mixture. Similar techniques, such as using an anaerobe jar with appropriate sachet, may be used for creating microaerophilic conditions.

ACKNOWLEDGMENT Fourth edition authors: Katherine M. J. Swanson, Ruth L. Petran, and John H. Hanlin.

REFERENCES 1. American Public Health Association. 2012. Standard Methods for the Examination of Water and Wastewater. 22nd ed. American Public Health Association, Washington, D.C. 2. Berry, J. M., D. A. McNeill, and L. D. Witter. 1969. Effect of delays in pour plating on bacterial counts. J. Dairy Sci. 52:1456-1457. 3. Clark, D. S. 1967. Comparison of pour and surface plate methods for determination of bacterial counts. Can. J. Microbiol. 13:1409-1412. 4. Cochran, W. G. 1950. Estimation of bacterial densities by means of the ‘‘Most Probable Number.’’ Biometrics. 6:105116. 5. Courtney, J. L. 1956. The relationship of average standard plate count ratios to employee proficiency in plating dairy products. J. Milk Food Technol. 19:336-344. 6. Craig, D. L., H. J. Fallowfield, and N. J. Cromar. 2002. Enumeration of faecal coliforms from recreational coastal sites: evaluation of techniques for the separation of bacteria from sediments. J. Appl. Microbiol. 93:557-565. 7. deMan, J. C. 1975. The probability of most probable numbers. Eur. J. Appl. Microbiol. 1:67-78.

86 |

8. deMan, J. C. 1977. MPN tables for more than one test. Eur. J. Appl. Microbiol. 4:307-316. 9. Donnelly, C.B., E. K. Harris, L. A. Black, and K. H. Lewis. 1960. Statistical analysis of standard plate counts of milk samples split with state laboratories. J. Milk Food Technol. 23:315-319. 10. Eisenhart, C., and P. W. Wilson. 1943. Statistical methods and control in bacteriology. Bacteriol. Rev. 7:57-137. 11. Foladori, P., B. Laura, A. Gianni, and Z. Guiliano. 2007. Effects of sonication on bacteria viability in wastewater treatment plants, evaluated by flow cytometry—fecal indicators, wastewater, and activated sludge. Water Research. 41:235-243. 12. Fowler, J. L., W. S. Clark, Jr., J. F. Foster, and A. Hopkins. 1978. Analyst variation in doing the standard plate count as described in Standard Methods for the Examination of Dairy Products. J. Food Prot. 41:4-7. 13. Halvorson, H. O., and N. R. Ziegler. 1933a. Application of statistics to problems in bacteriology. II. A consideration of the accuracy of dilution data obtained by using a single dilution. J. Bacteriol. 26:331-339. 14. Halvorson, H. O., and N. R. Ziegler. 1933b. Application of statistics to problems in bacteriology. III. A consideration of the accuracy of dilution data obtained by using several dilutions. J. Bacteriol. 26:559-567. 15. Hartman, P. A., and D. V. Huntsberger. 1961. Influence of subtle differences in plating procedure on bacterial counts of prepared foods. Appl. Microbiol. 9:32-38. 16. Holdeman, L.V., and W. E. C. Moore (Editors). 1977. Anaerobe Laboratory Manual. 4th ed. Virginia Polytechnic Institute Anaerobe Laboratory, Virginia Polytechnic Institute and State University, Blacksburg, VA. 17. Huhtanen, C. N., A. R. Brazis, W. L. Arledge, E. W. Cook, C. B. Donnelly, R. E. Ginn, J. N. Murphy, H. E. Randolph, E. L. Sing, and D. I. Thompson. 1970. Effect of dilution bottle mixing methods on plate counts of raw-milk bacteria. J. Milk Food Technol. 33:269-273. 18. Huhtanen, C. N., A. R. Brazis, W. L. Arledge, E. W. Cook, C. B. Donnelly, R. E. Gin, J. J. Jezeski, D. Pusch, H. E. Randolph, and E. L. Sing. 1972. Effects of time and holding dilutions on counts of bacteria from raw milk. J. Milk Food Technol. 35:126-130. 19. International Commission on Microbiological Specifications for Foods (ICMSF). 1978. Microorganisms in foods: their significance and methods of enumeration. 2nd ed. Intern. Comm. on Microbiol. Spec. Foods. Univ. of Toronto, Canada. 20. Koburger, J. A. 1980. Stack-pouring of Petri plates: a potential source of error. J. Food Prot. 43:561-562. 21. McCarthy, J. A., H. A. Thomas, Jr., and J. D. Delaney. 1958. Evaluation of the reliability of coliform density tests. Am. J. Public Health. 48:1628-1635. 22. Oblinger, J. L., and J. A. Koburger. 1975. Understanding and teaching the most probable number technique. J. Milk Food Technol. 38:540-545. 23. Punch, J. D., and J. C. Olson, Jr. 1964. Comparison between standard methods procedure and a surface plate method for estimating psychrophilic bacteria in milk. J. Milk Food Technol. 27:43-47. 24. Ray, B., and M. L. Speck. 1973. Discrepancies in the enumeration of Escherichia coli. Appl. Microbiol. 25:494-498. 25. Richards, O. W., and P. C. Heijn. 1945. An improved darkfield Quebec colony counter. J. Milk Technol. 8:253-256. 26. Silliker, J. H., D. A. Gabis, and A. May. 1979. ICMSF methods studies XI. Collaborative/comparative studies on determination of coliforms using the most probable number procedure. J. Food Prot. 42:638-644. 27. Thomas, H. A. 1942. Bacterial densities from fermentation tube tests. J. Am. Water Work Assoc. 34:572-576.

|

28. Tomasiewicz, D. M., D. K. Hotchkiss, G. W. Reinbold, R. B. Read, Jr., and P. A. Hartman. 1980. The most suitable number of colonies on plates for counting. J. Food Prot. 43:282-286. 29. U.S. Food and Drug Association. 2001. Aerobic plate count. Bacteriological Analytical Manual. Available at: http://www. fda.gov/Food/FoodScienceResearch/LaboratoryMethods/ ucm063346.htm. Accessed July 30, 2013. 30. Vanderzant, C., and A. W. Matthys. 1965. Effect of temperature of the plating medium on the viable count of psychrophilic bacteria. J. Milk Food Technol. 28:383-388.

Culture Methods for Enumeration of Microorganisms

31. Wehr, H. M., and J. F. Frank (Editors). 2004. Standard Methods for the Examination of Dairy Products. 17th ed. American Public Health Association, Washington, D.C. 32. Woodward, R. L. 1957. How probable is the Most Probable Number? J. Am. Water Works Assoc. 49:1060-1068. 33. Ziegler, N. R., and H. O. Halvorson. 1935. Application of statistics to problems in bacteriology. IV. Experimental comparison of the dilution method, the plate count, and the direct count for the determination of bacterial populations. J. Bact. 29:609-634.

| 87

|

CHAPTER 7

|

Cell Injury and Methods of Analysis Alissa M. Wesche and Elliot T. Ryser

7.1

7.2

INTRODUCTION

Bacterial survival is dictated by the balance of stresses to which an organism is exposed in the internal (i.e., intrinsic) environment and surrounding (i.e., extrinsic) environment. These stresses have been categorized as chemical (e.g., acids, preservatives, sanitizers), physical (e.g., temperature, osmotic pressure, irradiation) or nutritional (e.g., starvation), and can occur at any stage within the farm-to-fork continuum.39 After exposure to one or more stresses, a portion of the bacterial population will become injured to varying degrees, based the type of stress, length of exposure, and physiological state of each cell in the population (Figure 7-1). Two types of injury are recognized after exposure to intrinsic and extrinsic stress: (1) metabolic injury, characterized by damage to various cellular components such as deoxyribonucleic acid (DNA), ribonucleic acid (RNA), and key enzymes; and (2) structural injury, characterized by damage to the cell wall and membrane. Regardless of the type of injury, detection of sublethally injured bacteria, including foodborne pathogens, remains challenging since, by definition, such cells do not grow in the presence of the various selective agents in most enrichment and plating media. Sublethally injured cells may adapt to their new surroundings, regain full cell function and pathogenicity by repairing the previous cellular damage, enter a viable-butnonculturable (VNC) state that is characterized by very low metabolic activity and an inability to divide, or succumb to the injury and die.39 After minor stress, most cells will adapt to the new environment and resume growing. However, prolonged exposure will result in a longer lag phase and will be accompanied by a series of temporary physiological changes that may lead to increased stress tolerance—a condition called ‘‘transient adaptation.’’49 When moderately stressed, the culture will include healthy cells, dead cells, and cells exhibiting various degrees of injury.15,27 Exposure to a lethal stress will typically kill most of the population; however, some survivors, including foodborne pathogens, may persist because of adaptive gene mutations.5,36 | 89 |

7.21

TYPES OF STRESS Chemical

Chemical stress results from exposure to acids and bases and a wide range of food preservatives and chemical sanitizers. Acute acid shock and stress leading to sublethal injury can occur at a low pH when hydrogen (H+) ions cross the bacterial cell membrane or when organic acids diffuse across the cell membrane and lower the internal pH of the cell on dissociation.1,10 The fermentation processes used to manufacture various cheeses and meat products can also lead to acid stress. Exposure to alkaline detergents and chemicals such as caustic soda (NaOH) and quaternary ammonium compounds commonly used to clean and sanitize food- and nonfood-contact surfaces can also induce sublethal injury with such cells becoming difficult to detect when using standard enrichment and plating protocols.37

7.22

Physical

Physical stress can result from exposure to high and low temperatures, drying, changes in osmotic pressure, high pressure processing, and radiation, among other stressors. With regard to temperature, three stages of cold shock— which involve the initial cessation of growth, resumption of growth after an adaptive period, and changes in protein synthesis—have been recognized. Low temperature sensitivity varies widely among organisms and is based on the cooling rate, population density, and growth range.10,25 Heat shock leading to enhanced thermal resistance can occur when organisms are exposed to temperatures above their normal growth range such as during low-temperature pasteurization of eggs, slow cooking of meats or sous-vide processing.40 Decreasing the water activity of a food through adding various water-binding agents such as sugars, salt (NaCl), and phosphates can lead to osmotic stress, which is also seen during freezing, drying, and rapid rehydration of foods.26

7.23

Nutritional

Nutritional stress can occur in the natural environment and in food processing facilities when nutrient levels become

Compendium of Methods for the Microbiological Examination of Foods |

Figure 7-1. The impact of increasing stress on bacteria viability. VBNC 5 viable-but-non-culturable. Courtesy of Wesche and Ryser.41

too low to support metabolic activity or microbial growth. Starvation stress has been reported on animal carcasses, in food, on equipment surfaces, on walls, and on floors. Microorganisms in nutritionally deficient environments typically undergo a series of cell surface modifications to use alternative energy sources. Transformations during starvation, particularly those related to cellular morphology and cell wall/cell membrane composition, also enhance bacterial adherence and may contribute to biofilm formation.20,31,41

7.3

CELLULAR REPAIR AND CROSS PROTECTION

Production of various stress-induced proteins has a critical role in adaptation to growth or survival limiting conditions and recovery from stress-induced damage due to changes in temperature, pH, osmolarity, and nutrient availability, among other stressors.23,38 Some stress-induced proteins have clear functions for managing a specific stress, whereas other stress-induced proteins are produced under multiplestress conditions. A range of heat-shock proteins are typically produced in response to moderate increases in temperature with some of these same proteins also produced after exposure to various types of nonthermal stress such as starvation and exposure to ethanol and other organic solvents, oxidative agents, and high salt concentrations.41 The phenomenon whereby one type of stress is protective against higher levels of the same stress (especially for heat) or a different stress is widely documented and is referred to as ‘‘cross-protection’’ or ‘‘stress hardening.’’24,25 Some important foodborne pathogens such as Escherichia coli O157:H7, Salmonella and Listeria monocytogenes exhibit increased thermotolerance when heat-shocked in laboratory media; various foods that have undergone nutrient deprivation and abrupt changes in pH also sometimes have enhanced thermal resistance.21,22,25 The extent of crossprotection after any given stress response will vary based on the specific bacterial species/strain and the magnitude and nature of the stress.39

90 |

7.4

VIRULENCE

A wide range of virulence genes can be induced in response to different stresses as pathogens move from soil, water, food processing environments or food into human hosts with these host-related stresses similar to stresses in the external environment such as changes in temperature, acidity, and oxygen availability. Development of acid tolerance from exposure to low pH foods (e.g., apple cider, cheese) or stomach acid, which can enhance survival, is a major contributing factor to the low infective doses occurring in many foodborne outbreaks.51,52 The stressful anaerobic environment of the small intestine can also enhance the virulence of pathogens. Therefore, the inability of many standard testing protocols to account for such sublethally injured cells in food and environmental samples raises important public health concerns.

7.5

RECOVERY AND DETECTION OF SUBLETHALLY INJURED BACTERIA

The two universally accepted standard methods for recovering foodborne pathogens—enrichment and direct plating— are used to determine the presence/absence and the number, respectively, of the target organism. Both methods generally rely on the use of various selective agents such as antibiotics, acids, dyes, and surface-active agents to suppress competing background microflora. Under such conditions, only healthy uninjured cells are likely to grow. In the absence of these inhibitors, injured cells can undergo cellular repair and regain their lost cell functions so that these resuscitated and now fully repaired cells are able to grow and divide normally. Because sublethally injured foodborne pathogens can regain their virulence after repair, additional steps must be performed when examining foods (e.g., acidic, fermented, low moisture or processed foods) that are likely to contain stressed or injured cells, or when taking environmental swab and sponge samples from food processing environments where

| Cell Injury and Methods of Analysis

organisms may have become sublethally injured from desiccation or exposure to cleaning and sanitizing agents. Resuscitation of sublethally injured cells in the laboratory is based on the use of various nonselective broth and agar media. However, these media will promote the growth of uninjured target and nontarget cells in the sample and are typically unable to differentiate the target organism from the background microflora. After the repair is complete, the resuscitated healthy cells can be recovered by using various selective/differential broth-based and agar-based media. By definition, sublethally injured cells can only be grown using nonselective enrichment or plating media; whereas after resuscitation, healthy cells will grow in the presence or the absence of selective agents. Hence, the percentage of injured cells in a population can be calculated as follows: population on nonselective media{ population on selective media |100 % injury~ population on nonselective media Achieving 90% or greater injury after exposing a bacterial population to a given stress in the laboratory is critical when assessing survival in foods that are subjected to thermal and nonthermal processing.

7.6

ENRICHMENT METHODS

Various nonselective broth media have been developed to facilitate the repair of injured bacteria before selective enrichment and/or selective plating with the choice of the medium dictated by the organism being targeted for recovery and the sample being analyzed. Some examples of media described in the U.S. Food and Drug Administration’s (FDA’s) Bacteriological Analytical Manual for the resuscitation of injured foodborne pathogens include brain-heart infusion broth for pathogenic strains

of E. coli,9 lactose broth, reconstituted nonfat dry milk, nutrient broth, buffered peptone water, or Universal preenrichment broth for Salmonella, depending on the food product and expected type of injury4 and alkaline peptone water for Vibrio spp.19 The temperature and length of incubation for the repair of injured cells is influenced by the target pathogen, the expected type of stress, and the method used with 3–5 hr of incubation at 25uC to 37uC generally considered optimal for the resuscitation of mesophilic foodborne pathogens (Figure 7-2). Most detection and recovery protocols include a separate secondary selective enrichment or plating step; however, for L. monocytogenes, a sample is pre-enriched in buffered Listeria enrichment broth containing the Listeria selective agents acriflavine HCl, nalidixic acid, and cycloheximide, which are added after 4 hr of initial incubation at 30uC.13 Alternatively, Kang and Siragusa18 diluted food samples 2-fold in buffered peptone water, and followed this by adding a double-strength selective broth 3 hr later for the resuscitation of sublethally injured coliform bacteria with this same approach likely amenable for other organisms. However, since injured cells will repair and uninjured cells will grow during this period of nonselective enrichment, the length of incubation must be kept to a minimum if the number of target organisms is to be determined by subsequent plating or most probable number (MPN) methods with the end result likely to overestimate the initial population. During resuscitation in nonselective media, hydrogen peroxide produced by other respiring microorganisms is highly toxic to sublethally injured cells because of decreased catalase and superoxide dismutase activity. Catalase,28 pyruvate,34 3,39-thiodipropionic acid,29 or Oxyrase (Mansfield, OH; a commercial preparation of partially purified membrane fragments from E. coli)2,14,50 can consequently be added to neutralize the toxicity of hydrogen peroxide. The addition of Tween 80 (a lipid and surfactant)

Figure 7-2. Impact of sublethal injury on the resuscitation and growth of bacteria on nonselective and selective media after repair. CFU 5 colony-forming unit. Courtesy of Wesche and Ryser.41

| 91

Compendium of Methods for the Microbiological Examination of Foods |

and magnesium chloride30 can enhance cell membrane and ribosome repair. Regardless of the approach, many foodborne pathogens such as Salmonella, Listeria, Campylobacter, E. coli, and Vibrio can enter the VBNC State. These difficultto-resuscitate, morphologically smaller, less metabolically active cells are able to persist long-term in this state. However, under the right conditions, these formerly semidormant cells will repair, become fully functional, and regain their pathogenicity; therefore, detecting them is of major importance.

7.7

PLATING METHODS

Bacterial populations are typically determined by surface plating, pour plating, filtration (see the chapter ‘‘Mesophilic Aerobic Plate Count’’) or an MPN method (see the chapter ‘‘Culture Methods for Enumeration of Microorganisms’’) with the types of organisms recovered determined by the growth media selected. However, by definition, sublethally injured cells will not grow on selective media. Therefore, when the presence of healthy and sublethally injured cells is suspected, a combination of selective and nonselective plating media are typically used in what is now broadly known as the agar overlay technique. Four different versions of the agar overlay technique are recognized, which are the pouroverlay plating method, 32,33 surface-overlay plating method,12,35 thin layer agar method,16,17,43,45,46 and multicompartmented thin layer agar method.47,48 As an alternative, several filtration-based methods also have been developed in which a standard 0.45 mm membrane6,11 or hydrophobic grid membrane filter7 with the target organism is transferred from a nonselective medium to a selective medium after a short incubation period. These methods for resuscitation and recovery of sublethally injured cells are summarized below with readers referred to a recent review article by Wu42 and to the chapter ‘‘Mesophilic Aerobic Plate Count’’ for specific details regarding sample preparation for pour plating (Section 8.72), spread or surface plating (Section 8.73), membrane filtration (Section 8.81), and the hydrophobic grid-membrane filter method.

7.71

Pour-Overlay Plating Method

In this method, an appropriately diluted sample is pourplated, as described in the chapter ‘‘Mesophilic Aerobic Plate Count’’ (Section 8.72) using 5 mL of a nonselective agar medium (e.g., trypyticase soy agar, brain-heart infusion agar, nutrient agar) with the optional addition of catalase, pyruvate, or other additives to enhance the rate of repair.32,33 After solidification of the medium followed by 3–5 hr of incubation at 25uC–37uC, the plate is overlayed with 10–12 mL of a selective medium appropriate for the target organism (e.g., modified Oxford agar for Listeria, violet red bile agar for coliforms), and then incubated accordingly (typically at 35uC–37uC for 24 hr) to recover the organism in question. This method provides fairly accurate counts since all cells are immobilized in the medium during repair and resuscitation. However, these embedded colonies may be difficult to pick for further characterization.

7.72

Surface-Overlay Plating Method

The first step of the surface-overlay plating method (also known as the overlay resuscitation method) is identical to 92 |

that for surface plating or spread plating (see the chapter ‘‘Mesophilic Aerobic Plate Count’’; Section 8.73) with an appropriately diluted sample surface, or spread plated using prepoured plates containing approximately 12 mL of a nonselective agar medium (e.g., trypyticase soy agar, brain heart infusion agar, nutrient agar with optional addition of catalase, pyruvate, or other additives to promote repair).12,35 After 3–5 hr for resuscitation at 25uC–37uC, the plate is overlayed with 10–12 mL of the selective medium that is appropriate for the target organism and then reincubated, as described for the aforementioned pour-overlay plating method. Two advantages of this method over the pouroverlay plating method are (1) the prepoured plates used in the first step simplify the procedure and (2) there is greater recovery since potentially injured cells are plated directly rather than being exposed to warm agar during pour-plating.

7.73

Thin Agar Layer Method

The thin agar layer method is a simple one-step procedure based on diffusion of selective agents from an underlying selective medium into the upper nonselective medium over a period of 3–6 hr with this protocol having now become the method of choice.16 In this procedure, 7 mL of a nonselective plating medium is overlayed on a prepoured plate containing 14 mL of the desired selective medium. When spread-plated or surface-plated immediately after preparation, the top nonselective layer provides a favorable environment for the resuscitation of injured cells. As incubation continues, the now resuscitated cells will interact with the selective and differential agents migrating from the bottom to the top layer to produce typical colony morphologies and color reactions for the target organism in question. Hence, the ratio of selective to nonselective media (approximately 2:1) is critical in maximizing the recovery of the target organism and minimizing the background microflora.17,43,45,46 In contrast to the pour-overlay and surface-overlay plating methods, the thin agar layer method eliminates the risk of further injury from exposure to molten agar and allows colonies to develop on the agar surface that can be easily picked for further characterization. This method has proven successful for the recovery of many heat-, acid-, and coldinjured foodborne pathogens including E. coli O157:H7, L. monocytogenes, S. Typhimurium, Staphylococcus aureus, Vibrio parahaemolyticus, and Yersinia enterocolitica.8,42,44,46,52,53

7.74

Multicompartment Thin Agar Layer Method

The multicompartment thin agar layer method is a modification of the thin agar layer method, which uses a two-, three- or four-compartmented Petri plate with each compartment containing a different selective medium, overlayed with the same nonselective medium to simultaneously resuscitate and recover different organisms (e.g., Listeria, Salmonella, E. coli, Yersinia).47,48 While more efficient in media, labor, and time, detecting and accurately quantifying low numbers of target organisms can be problematic because of the smaller surface area available for each organism.

7.75

Membrane Filter Method

Several membrane filter methods have also been developed. Using the original procedure reported by Goff et al.,11 an appropriately diluted food sample is filtered through a

| Cell Injury and Methods of Analysis

sterile 47-mm diameter membrane (pore size of 0.45 mm), as described in the chapter ‘‘Mesophilic Aerobic Plate Count’’ with the membrane first incubated on a nonselective medium for 3–6 hr, and then moved to a selective medium for an additional 18 hr of incubation. This method is best suited to resuscitate low numbers of the target organism in question, but food particulates may clog the membrane and make filtration difficult. Anderson and Baird-Parker3 introduced an alternative membrane filter-plating method that involved spread-plating a homogenized and appropriately diluted sample directly on the membrane, followed by similar incubation on a selective medium; this was then followed by a nonselective medium. Many years later, Blackburn and McCarthy6 used this same method with trypticase soy agar and sorbitol MacConkey agar to recover sublethally cells of E. coli O157:H7. However, any preservatives or inhibitors present in the food particulates may negatively impact the rate of repair on the filter membrane.

7.76

Hydrophobic Grid-Membrane Filter Methods

A hydrophobic membrane filtration as described in Section 8.83 of the chapter ‘‘Mesophilic Aerobic Plate Count’’ has also been adapted for the recovery of sublethally injured microorganisms. In this procedure, an appropriately diluted food sample is first passed through a sterile prefilter to remove larger food particulates and then through a membrane filter with a hydrophobic grid containing 1600 individual growth compartments that prevent organisms from spreading. This membrane can then be transferred to a selective medium, followed by a nonselective medium,7 or transferred to a thin agar layer plate44 for the repair and recovery of the target organism in question. Unlike the other plating procedures, the hydrophobic grid-membrane filter method yields an MPN rather than a direct plate count since multiple organisms may reside in the same growth compartment after filtration.

7.8

LIMITATIONS AND CONCLUSIONS

The physiological state of any microbial population will change after exposure to a chemical, physical, or nutritional stress with different portions being killed, injured to various degrees, or unaffected. Increased resistance of sublethally injured foodborne pathogens to subsequent stress—a phenomenon called ‘‘cross-protection’’—poses a major threat to public health since many such organisms can undergo repair and regain their virulence. Various combinations of selective and nonselective plating media can be used to resuscitate and recover sublethally injured bacteria, depending on the specific target organism in question, with the selectivity of the medium inversely related to the rate of repair. Currently recognized direct plating methods for quantitative recovery of sublethally injured bacteria include pour-overlay and surface-overlay plating and the thin layer agar and multicompartmented thin layer agar methods. Two alternative membrane-based approaches include membrane filtration (or direct membrane plating) and hydrophobic grid membrane filtration. Because of the likelihood for sublethally injured bacteria to survive in foods and persist in food processing environments, detecting them has important ramifications for food quality and safety.

REFERENCES 1. Abee, T., and J. A. Wouters. 1999. Microbial stress response in minimal processing. Int. J. Food Microbiol. 50:65-91. 2. Ali, M. S., and D. Y. C. Fung. 1991. Occurrence of Clostridium perfringens in ground beef and ground turkey evaluated by three methods. J. Food Safety. 11:197-203. 3. Anderson, J. M., and A. C. Baird-Parker. 1975. A rapid and direct plate method for enumerating Escherichia coli biotype I in food. J. Appl. Bacteriol. 38:390-394. 4. Andrews, W. H., A. Jacobson, and T. Hammack. 2011. Salmonella. Bacteriological Analytical Manual. Available at http://www.fda.gov/Food/FoodScienceResearch/Laboratory Methods/ucm070149.htm. Accessed February 4, 2014. 5. Archer, D. L. 1996. Preservation microbiology and safety: evidence that stress enhances virulence and triggers adaptive mutations. Trends Food Sci. Technol. 7:91-95. 6. Blackburn, C. W., and J. D. McCarthy. 2000. Modification to methods for the enumeration and detection of injured Escherichia coli O157:H7 in foods. Int. J. Food Microbiol. 55:285-290. 7. Brodsky, M. H., P. Bolesczcuk, and P. Entis. 1982. Effect of stress and resuscitation on recovery of indicator bacteria from foods using hydrophobic grid-membrane filtration. J. Food Prot. 45:1326-1331. 8. Duan, J., C. Liu, and Y.-C. Su. 2006. Evaluation of a double layer agar plate for direct enumeration of Vibrio parahaemolyticus. J. Food Sci. 71:M77-82. 9. Feng, P., S. D. Weagant, and K. Jinneman. 2013. Diarrheagenic Escherichia coli. FDA Bacteriological Analytical Manual. Available at http://www.fda.gov/Food/FoodScienceResearch/ LaboratoryMethods/ucm070080.htm. Accessed February 4, 2014. 10. Foster, J. W. 2000. Microbial responses to acid stress. In: Bacterial Stress Responses, (G. Storz and R. Hengge-Aronis, eds.), pp. 99-115, ASM Press, Washington, D.C. 11. Goff, J. H., T. J. Claydon, and J. J. Iandolo. 1972. Revival and subsequent isolation of heat-injured bacteria by a membrane filter technique. Appl. Microbiol. 23:857-862. 12. Hartman, P. A., P. S. Hartman, and W. W. Lanz. 1975. Violet red bile 2 agar for stressed coliforms. Appl. Microbiol. 29:537539. 13. Hitchins, A. D., and K. Jinneman. 2013. Detection and enumeration of Listeria monocytogenes. Bacteriological Analytical Manual. Available at http://www.fda.gov/ Food/FoodScienceResearch/LaboratoryMethods/ ucm071400.htm. Accessed February 4, 2014. 14. Hoskins, C. B., and P. M. Davidson. 1988. Recovery of Clostridium perfringens from food samples using an oxygenreducing membrane fraction. J. Food Prot. 51:187-191. 15. Hurst, A. 1984. Revival of vegetative bacteria after sublethal heating. In: The Revival of Injured Microbes, (M. H. E. Andrew and A. D. Russell, eds.), pp. 77-103, Academic Press, London, UK. 16. Kang, D. H., and D. Y. C. Fung. 1999. Thin agar layer method for recovery of heat-injured Listeria monocytogenes. J. Food Prot. 62:1346-1349. 17. Kang, D. H., and D. Y. C. Fung. 2000. Application of thin agar layer method for recovery of injured Salmonella Typhimurium. Int. J. Food Microbiol. 54:127-132. 18. Kang, D. H., and G. R. Siragusa. 2001. A rapid two-fold dilution method for microbial enumeration and resuscitation of uninjured and sublethally injured bacteria. Lett. Appl. Microbiol. 33:232-236. 19. Kaysner, C. A., and A. DePaola Jr. 2004. Vibrio. Bacteriological Analytical Manual. Available at http://www.fda.gov/Food/ FoodScienceResearch/LaboratoryMethods/ucm070830.htm. Accessed February 4, 2014.

| 93

Compendium of Methods for the Microbiological Examination of Foods |

20. Lazazzera, B. A. 2000. Quorum sensing and starvation: signals for entry into stationary phase. Curr. Opin. Microbiol. 3:177-182. 21. Leenanon, B., and M. A. Drake. 2001. Acid stress, starvation, and cold stress affect post-stress behavior of Escherichia coli O157:H7 and nonpathogenic Escherichia coli. J. Food Prot. 64:970-974. 22. Leyer, G. J., and E. A. Johnson. 1993. Acid adaptation induces cross-protection against environmental stresses in Salmonella Typhimurium. Appl. Environ. Microbiol. 59:1842-1847. 23. Lindquist, S. 1992. Heat-shock proteins and stress tolerance in microorganisms. Curr. Opin. Gen. Develop. 2:748-755. 24. Lou, Y., and A. E. Yousef. 1996. Resistance of Listeria monocytogenes to heat after adaptation to environmental stresses. J. Food Prot. 59:465-471. 25. Lou, Y., and A. E. Yousef. 1997. Adaptation to sublethal environmental stresses protects Listeria monocytogenes against lethal preservation factors. Appl. Environ. Microbiol. 63:12521255. 26. Mackey, B. M. 1984. Lethal and sublethal effects of refrigeration, freezing and freeze-drying on micro-organisms. In: The Revival of Injured Microbes, (M. H. E. Andrew and A. D. Russell, eds.), pp. 45-75, Academic Press, London, UK. 27. Mackey, B. M. 2000. Injured bacteria. In: The Microbiological Safety and Quality of Food, (B. Lund, T. C. Baird-Parker, and G. W. Gould, eds.), pp. 315-341, Aspen Publishers, Inc., Gaithersburg, MD. 28. Martin, S. E., R. S. Flowers, and J. J. Ordal. 1976. Catalase: the effect on microbial enumeration. Appl. Environ. Microbiol. 32:732-734. 29. McDonald, L. C., C. R. Hackney, and B. Ray. 1983. Enhanced recovery of injured Escherichia coli by compounds that degrade hydrogen peroxide or block its formation. Appl. Environ. Microbiol. 45:360-365. 30. Murthy. T. R. K., and R. Gaur. 1987. Effect of incorporation of Tween 80 and magnesium chloride on the recovery of coliforms in VRB medium from fresh, refrigerated and frozen minced buffalo meat. Int. J. Food Microbiol. 4:341-346. 31. Postgate, J. R., and J. R. Hunter. 1963. The survival of starved bacteria. J. Appl. Bacteriol. 26:295-306. 32. Ray, B. 1979. Methods to detect stressed microorganisms. J. Food Prot. 42:346-355. 33. Ray, B., and M. L. Speck. 1973. Enumeration of Escherichia coli in frozen samples after recovery from injury. Appl. Microbiol. 25:499-503. 34. Rayman, M. K., B. Aris, and H. B. El Derea. 1978. The effect of compounds which degrade hydrogen peroxide on the enumeration of heat stressed cells of Salmonella Senftenberg. Can. J. Microbiol. 24:883-885. 35. Speck M. L., B. Ray, and R. B. Read. 1975. Repair and enumeration of injured coliforms by a plating procedure. Appl. Microbiol. 29:549-550. 36. Storz, G., and R. Hengge-Aronis. 2000. Preface. In: Bacterial Stress Responses, (G. Storz and R. Hengge-Aronis, eds.), pp. 12-14, ASM Press, Washington D.C. 37. Taormina, P. J., and L. R. Beuchat. 2001. Survival and heat resistance of Listeria monocytogenes after exposure to alkali and chlorine. Appl. Environ. Microbiol. 67:2555-2563.

94 |

38. Vo€lker, U., H. Mach, R. Schmid, and M. Hecker. 1992. Stress proteins and cross-protection by heat shock and salt stress in Bacillus subtilis. J. Gen. Microbiol. 138:2125-2135. 39. Wesche, A. M., J. Gurtler, B. P. Marks, and E. T. Ryser. 2009. Stress, sublethal injury, resuscitation and virulence of foodborne pathogens—a review. J. Food Prot. 72:1121-1138. 40. Wesche, A. M., B. P. Marks, and E. T. Ryser. 2005. Thermal resistance of heat-, cold-, and starve-injured Salmonella in irradiated comminuted turkey. J. Food Prot. 68:942-948. 41. Wesche, A. M., and E. T. Ryser. 2013. Stress adaptation, survival and recovery of foodborne pathogens. In: Guide to Foodborne Pathogens, (R. G. Labbe´ and S. Garcı´a, eds.), John Wiley & Sons, Hoboken, NJ. 42. Wu, V. C. H. 2008. A review of microbial injury and recover methods in food. Food Microbiol. 736-744 43. Wu, V. C. H., and D. Y. C. Fung. 2001. Evaluation of thin agar layer method for recovery of heat-injured foodborne pathogens. J. Food Sci. 66:580-583. 44. Wu, V. C. H., and D. Y. C. Fung. 2004. An improved method for ISO-grid hydrophobic grid membrane filter (HGMF) system to detect heat-injured pathogens in ground beef. J. Food Sci. 69:85-89. 45. Wu, V. C. H, D. Y. C. Fung, and D. H. Kang. 2001. Evaluation of thin agar layer method for recovery of acid-injured foodborne pathogens. J. Food Prot. 64:1067-1071. 46. Wu, V. C. H., D. Y. C. Fung, and D. H. Kang. 2001. Evaluation of thin agar layer method for recovery of cold-injured foodborne pathogens. J. Rapid Methods Autom. Microbiol. 9:11-25. 47. Wu, V. C. H., D. Y. C. Fung, and D. H. Kang. 2003. Simultaneous recovery of four injured foodborne pathogens in the four-compartment thin agar layer plate. J. Food Sci. 68:646-648. 48. Wu, V. C. H., D. Y. C. Fung, and D. H. Kang. 2006. Simultaneous recovery and detection of four heat-injured foodborne pathogens in ground beef and milk by a four-compartment thin agar layer plate. J. Food Safety 26:126-136. 49. Yousef, A. E., and P. D. Courtney. 2003. Basics of stress adaptation and implications in new-generation foods. In: Microbial Stress Adaptation and Food Safety, (A. E. Yousef and V. K. Juneja, eds.), pp. 1-30, CRC Press, Boca Raton, FL. 50. Yu, L. S. L., and D. Y. C. Fung. 1991. Effect of Oxyrase enzyme on Listeria monocytogenes and other facultative anaerobes. J. Food Safety 11:163-175. 51. Yuk, H.-G., and D. L. Marshall. 2004. Adaptation of Escherichia coli O157:H7 to pH alters membrane lipid composition, verotoxin secretion, and resistance to simulated gastric fluid. Appl. Environ. Microbiol. 70:3500-3505. 52. Yuk, H.-G., and D. L. Marshall. 2005. Influence of acetic, citric, and lactic acids on Escherichia coli O157:H7 membrane lipid composition, verotoxin secretion, and acid resistance in simulated gastric fluid. J. Food Prot. 68:673-679. 53. Yuste, J., D. Y. C. Fung. 2003. Evaluation of Salmonella Typhimurium, Yersinia enterocolitica and Staphylococcus aureus counts in apple juice with cinnamon, by conventional media and thin agar layer method. Food Microbiol. 20:365-370.

|

CHAPTER 8

|

Mesophilic Aerobic Plate Count Elliot T. Ryser and James D. Schuman

8.1

INTRODUCTION

The mesophilic aerobic plate count (APC) has been successfully used for many years to gauge product shelflife, organoleptic acceptability, sanitary conditions, and adherence to good manufacturing practices. The APC has only been used marginally as an indicator of safety. The APC can provide food processors with valuable information regarding the bacteriological quality of raw materials and information on conditions surrounding food processing, handling, and storage. This test may also provide information regarding shelf-life or impending organoleptic changes in a food.26 Mesophilic APCs, sometimes referred to as aerobic colony counts, are generally poor indicators of safety since this test is not specific for the presence of pathogens or microbial toxins. A low APC does not indicate that the entire product or ingredient lot is pathogen-free because the lower limit of detection ranges between 1 and 100 colony-forming units per gram or per milliliter (CFU/g or CFU/mL, respectively), depending on the type of sample and the plating method used. Products or ingredients showing excessively or unusually high APCs are frequently organoleptically acceptable and free of foodborne pathogens. However, knowledge of the product and whether a high APC is expected must first be considered when interpreting any APC results. Depending on the situation, APC data can be valuable for evaluating food quality since high counts may indicate poor sanitation or problems with process control or with the ingredients. Certain products (e.g., products produced through fermentation) may have a higher APC because of the growth of lactic acid bacteria from a starter culture. The APC numbers, whether high or low, cannot be equated with the presence or absence, respectively, of pathogens. Foods need to be assayed for specific pathogens before ruling on food safety concerns. Quality and safety guidelines or specifications are often applied to raw materials and finished goods. Using the APC for ingredients may or may not be appropriate as a quality indicator. A food manufacturer’s decision to apply APC guidelines to ingredients must be based on the ingredient’s impact on the finished product. For instance, | 95 |

in dried foods the APC is useful for assessing the adequacy of moisture control during the drying process. For meat products, APC tests can be used to assess the condition of incoming carcasses to potentially identify suppliers who provide products with excessively high counts. The APC can also be used to evaluate sanitary conditions of equipment and utensils. This can be measured during processing to monitor buildup, and measured after sanitation to gauge its effectiveness. Tables 8-1 and 8-2 provide some general APC guidelines for various ingredients and finished products.2,24 Using narrowly defined APC specifications for a particular commodity is not always appropriate. For example, raw agricultural commodities can have widely fluctuating plate counts. In these situations, the APC can provide meaningful data to the processor who has a better understanding of factors that may influence the count; however, the APC data provide little value in relation to the acceptance criteria.

8.2 8.21

GENERAL CONSIDERATIONS Other Tests on the Same Sample

If additional analytical or sensory tests are to be performed on the sample, the portions for microbiological analysis must first be aseptically removed.

8.22

Preparation

Equipment and supplies should meet the specifications described in the chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis,’’ unless otherwise specified. All media, buffers and other materials to be sterilized should be autoclaved at 121uC for 15 minutes.

8.23

Controls

Sterility tests of the media, plates and diluents should be conducted for each lot. Laboratory quality assurance procedures should include periodic checks of old versus new lots of media with known cell numbers to verify that microbial recovery is optimum (see the chapters ‘‘Laboratory Quality Management Systems’’ and ‘‘Culture Methods for Enumeration of Microorganisms’’).

Compendium of Methods for the Microbiological Examination of Foods |

Table 8-1. Typical Mesophilic Aerobic Plate Counts of Selected Foods2,24 Food Commodity

APC (log CFU/g)

Raw ground beef Cooked, sliced luncheon meats Raw poultry (broilers) Pasteurized liquid egg products Dried egg products Pasteurized fluid milk products Dried milk products Finfish and shrimp (raw) Fruits and vegetables (frozen) Fruits and vegetables (dehydrated) Botanical gums (guar, carageenan, locust bean gum) Spices Sweeteners and starches Unprocessed Cereal Grains Breakfast cereals Confectionery products (chocolate and marshmallows) Tree nuts (fresh, in shell) Nuts/seeds/legume butters and pastes Bottled water (with a bactericidal or reverse osmosis treatment) Bottled water (still mineral waters without a bactericidal treatment)

4 2 3 3 4 2 3 4 2 2 4 4 2 4 2 3 4 3 2 4

to 7 to 3 to 4

to to to to to to to to to to to

4 4 6 7 5 5 5 4 5 3 4

to 4 (maximum) to 5

Note: APC 5 aerobic plate count.

8.3

PRINCIPLE

The APC (also known as the heterotrophic plate count and formerly as the standard plate count) is used to quantify the number of mesophilic aerobic bacteria in a sample. This test is based on the assumption that each viable cell, pair of cells, or small cluster of cells will form a visible colony, termed a colony-forming unit (CFU), when mixed with a growth medium containing the appropriate nutrients.

8.4

GENERAL DESCRIPTION OF METHOD

8.41

Equipment, Materials, and Reagents

N N

Work area: clean, sanitizable; level bench or table Refrigerator to cool and maintain samples at 0uC to 5uC

N N N N N N N N N N

Media: plate count agar (PCA) or equivalent (see the chapter ‘‘Microbiological Media, Reagents, and Stains’’) Balance scale (with a minimum sensitivity of ¡ 0.1 g) for weighing media and samples Hot plate with a magnetic stirrer for agar preparation Autoclave for sterilization of media and waste Water bath for tempering agar at 45 ¡ 1uC Homogenizer: mechanical stomacher with sterile stomacher bags or mechanical blender with sterile blender jars Sterile test tubes and dilution bottles (6 oz [160 mL]), glass, with plastic screw caps or rubber stoppers Vortex mixer Pipettes with pipette aid (no mouth pipetting) or mechanical pipettors: 1 mL, 5 mL, and 10 mL Petri plates, plastic or glass (at least 90 6 15 mm)

Table 8-2. FAO/WHO Microbiological Specifications for Foods20 Product Category

Maximum APC/g

Dried and frozen whole egg Dried instant products Dried products requiring heating before consumption Precooked frozen shrimp Ice mixes Edible ices Dried milk Caseins

50,000 1,000 10,000 100,000 25,000 50,000 50,000 30,000

Note: APC 5 aerobic plate count; FAO/WHO 5 Food and Agricultural Organization of the United Nation/World Health Organization.

96 |

| Mesophilic Aerobic Plate Count

Table 8-3. Effect of Delay in Pouring Plates on Total Counts5

N N N

Delay (minutes)

Sample 1 (CFU)

Sample 2 (CFU)

Sample 3 (CFU)

0 5 10 15 20 30 45 60

173 156

137 102 90 68 69

138 123 101 98 93 84

28

28

120 98 81 51

Sterile disposable ‘‘hockey sticks’’ or bent glass rods Incubator, 35 ¡ 1uC for most foods and beverages; 32 ¡ 1uC for dairy products Automated colony counter or dark-field Quebec colony counter with a grid plate

8.42

N

N

Additional Equipment for Alternative Methods

Membrane filtration method Sterile membrane filters, 0.45-mm pore diameter # Media: PCA or nutrient pads # Filter holder # Filtration flask # Vacuum tubing # Vacuum source # Sterile buffer # Sterile forceps Petrifilm method # Petrifilm Aerobic Count Plate # Petrifilm AC plates (3M Food Safety, St. Paul, MN) # Plastic spreader Hydrophobic grid–membrane filter (HGMF) method # NEO-GRID/ISO-GRID HGMF filters and filtration unit (Neogen Corp., Lansing, MI) # Vacuum manifold # Peptone/Tween 80 diluent # Tryptic soy-fast green agar (Neogen) # Tris buffer 1.0M # Enzyme reagents (e.g., amyloglucosidase, alkaline pro tease, cellulase, hemicellulase, papain) # HGMF Interpreter (Model MI-200; Richard Brancker, Research Ltd, Kanata, Ontario, Canada) or a Linecounter (Gelman Sciences, Ann Arbor, MI) Spiral plating method # Autoplate 4000 Automated Spiral Plater (Spiral Biotech) # Spiral plate reader (Spiral Biotech, Norwood, MA) # Vacuum trap # Disposable beakers, 5 mL # 5% Sodium hypochlorite (NaOCl) Calibrated loop method # Volumetrically calibrated loop Drop plate method # Calibrated pipette SimPlate method # SimPlate Total Plate Count Color Indicator (TPC-CI) # SimPlate pretreated dish (BioControl, Bellevue, WA)

8.5

TEMPO TVC (Total Viable Count MPN) method (bioMe´rieux, St. Louis, MO)8 # Vials of culture media # TEMPO filler device # 48-well TEMPO card # TEMPO reader

PRECAUTIONS

#

N N

N

N N N

For accurate results use the procedures described in the chapters ‘‘Laboratory Quality Management Systems.’’ ‘‘Culture Methods for Enumeration of Microorganisms,’’ and ‘‘Measurement of Water Activity, Acidity, and Brix’’. A delay between sample dispensing into Petri dishes and agar addition can result in lower counts for several reasons, including possible diluent toxicity or the adherence of the bacteria to the dish. Berry et al.5 demonstrated the risk of obtaining a significantly lower plate count if the plates were not poured within 10 minutes (Table 8-3). Check the pH of the food/buffer suspension. If the pH is outside the range of 5.5 to 7.6, adjust the pH to 7.0 ¡ 0.1 by using sterile sodium hydroxide or hydrochloric acid.19 Another potential source of error in plate counts can result from the stack-pouring of Petri plates. Koburger18 found that, in a stack of three plates, the middle and top plates took longer to cool, thereby resulting in lower counts. Huhtanen et al.17 demonstrated that, for raw milk, increasing the holding time of the dilutions in the buffer leads to higher counts. Holding dilutions for 20 minutes resulted in a count that was 22% higher than the control, possibly because of bacterial growth or the breaking up of clumps. A 5-minute holding time gave the closest approximation to the true counts.

8.6

LIMITATIONS

The APC is not a measure of the entire bacterial population, as is falsely suggested by other frequently used terms such as ‘‘total plate count’’ and ‘‘total viable count,’’ but as its name implies, it is a generic test for organisms that grow aerobically at mesophilic temperatures (approximately 20uC to 45uC). For information on the enumeration of psychrotrophic microorganisms, refer to the chapter ‘‘Psychrotrophic Microorganisms.’’ The APC does not differentiate between different types of bacteria. Alterations in agar nutrient | 97

Compendium of Methods for the Microbiological Examination of Foods |

content, incubation time and temperature, and the type of atmosphere will change the types of organisms that will grow and thus be counted.

8.7

PROCEDURES

8.71

Dilutions

Refer to the chapter ‘‘Culture Methods for Enumeration of Microorganisms’’ for complete details on the preparation and proper use of dilutions.

N

N

N

N N

Basic principles: Enumeration of microorganisms requires diluting samples to achieve a population that is countable by the chosen method. Generally decimal or ten-fold dilutions are used because of the ease in calculating the final results. Various diluents are available such as phosphate buffer and 0.1% peptone water. Distilled water should be used in the preparation of the diluents. However, using plain distilled or deionized water as a diluent is inappropriate because of the potential for osmotic stress on the diluted cells. Liquids: For nonviscous liquid samples (i.e., the viscosity is not greater than that of milk), aseptically pipette 11 mL of a well-mixed sample into a 99 mL dilution blank. For a viscous liquid sample, weigh 11 ¡ 0.1 g of a well-mixed sample into a 99 mL dilution blank to assure accuracy of the sample size. Solid sample homogenates: Refer to the chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis’’ for the initial preparation of sample homogenates. For fine granular or powdered samples, mix samples thoroughly, weigh 11¡ 0.1 g into a sterile sample container, and add 99 mL of diluent. If a larger sample is desired, other sample and diluent quantities can be used to arrive at a 1:10 dilution (e.g., a 50 ¡ 0.1 g sample can be added to 450 mL of diluent). For solid and particulate material samples, prepare a 1:10 dilution by selecting a sample size that assures that a representative sample is tested. The appropriate diluent volume for various sample sizes may be determined by multiplying the sample weight by nine. Dilutions may also be determined by weight. Homogenizing by blending or stomaching: For this procedure, refer to the chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis.’’ Swabs and sponges: For the use of these items, refer to the chapter on ‘‘Microbiological Monitoring of the Food Processing Environment.’’

8.72

Pour Plate Techniques

There are several inherent limitations in enumerating microorganisms by the colony count method, although many of the errors can be minimized if the analyst carefully follows directions and exercises extreme care in all measurements. Consistently accurate and meaningful results can be obtained from the routine examination of a food only if the same procedures are used to analyze each sample of that food. This includes sampling procedures, sample preparation, preparation of dilutions, plating media, incubation conditions (35uC for 48 ¡ 2 hr), and 98 |

counting procedures.1,11,14 Refer to the chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis’’ for sample preparation guidelines and to the chapter ‘‘Culture Methods for Enumeration of Microorganisms’’ for the pour plate method, including details on the correct way to compute colony counts.

8.73

Surface or Spread Plate Method

Plating methods designed to produce only surface colonies on agar plates have certain advantages over the pour plate method.7 Using translucent media is not essential with a surface or spread plate, but it is necessary with a pour plate to see colonies that are to be counted. The morphology of surface colonies is easily observed, thereby improving an analyst’s ability to distinguish different types of microorganisms. During surface plating, higher counts may be observed in some situations since microorganisms are not exposed to the heat of the melted agar medium. However, the inherently small sample volumes (0.1 to 0.5 mL) that must be used typically result in a higher minimum level of detection, compared to pour plating. Incubate the plates at 35uC for 48 ¡ 2 hr (as in the pour plate method of plating). Refer to the chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis’’ for sample preparation guidelines and to the chapter ‘‘Culture Methods for Enumeration of Microorganisms’’ for the spread plate method.

8.8 8.81

ALTERNATIVE APPROVED METHODS Membrane Filtration

For certain foods or food ingredients, the ability to test relatively large samples will improve the accuracy of quantitative microbiological analysis. Large volumes of liquid foods or solutions of dry foods that can be dissolved and passed through a bacteriological membrane filter (diameter of approximately 47 mm; pore size of 0.45 mm) may be analyzed for microbial levels by using the membrane filter method. This method is especially useful for samples that contain low numbers of bacteria. Additional details on rapid and commercial kits using membrane filtration methods are described in the chapter ‘‘Rapid Methods for the Detection and Identification of Foodborne Pathogens.’’ The procedure is as follows: 1.

2.

3.

Aseptically assemble the membrane filter apparatus, following the manufacturer’s instructions, and connect to the vacuum system. Prepare the sample as previously discussed in the chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis.’’ Introduce an appropriate amount of sample into the funnel with a sterile pipette or a graduated cylinder. For sample volumes of less than 10 mL, aseptically pour approximately 20 mL of sterile diluent in the funnel before adding the sample. If a graduated cylinder is used, rinse the cylinder with approximately 50 mL of sterile diluent and add this rinsate to the funnel.

| Mesophilic Aerobic Plate Count

4. 5.

6.

7. 8. 9.

8.82

Apply a vacuum to the filter apparatus and allow the liquid to pass completely through the filter into the flask. Do not turn off the vacuum. Rinse the inside of the funnel with sterile diluent by using a volume that is at least equal to the volume of liquid just filtered. After the rinse has passed completely through the filter, turn off the vacuum. Carefully and aseptically disassemble the portion of the apparatus containing the filter. By using alcohol flame-sterilized smooth-tipped forceps, remove the filter and carefully roll the filter, avoiding air bubbles on the surface of the chosen saturated nutrient pad or plate count agar medium in a 50-mm diameter Petri plate. Incubate at 35uC for 48 ¡ 2 hours. Count the colonies under low power magnification. The targeted range is 20 to 200 colonies per filter. Compute the counts and report as the membrane filter colony count per milliliter or per gram, based on the amount of sample filtered.

Dry Rehydratable Film Method: Petrifilm Plate Method

The dry rehydratable film method (3M Petrifilm; 3M Microbiology, St. Paul, MN) is a ready-made culture medium system that consists of two plastic films coated with standard methods nutrients, a cold-water-soluble gelling agent, and a tetrazolium indicator that facilitates colony enumeration. The Petrifilm AC plate—which has been collaboratively studied with milk, dairy products and other foods—statistically yields similar results as the APC method.4,6,9,13,23,25,28 In this method, the Petrifilm plate is inoculated with 1 mL of an undiluted or a diluted sample by using a pipette, pipettor, or plate loop continuous pipetting syringe. After lowering the top plastic film, the sample is evenly distributed with a special spreader over a growth area of approximately 20 cm2. The gelling agent is allowed to solidify. The colonies are then counted after the prescribed incubation period. 1. 2. 3. 4. 5.

6. 7. 8.

Prepare sample (see the chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis’’). Place the labeled Petrifilm plate on a flat surface. Prepare ten-fold decimal dilutions by following general procedures. Lift the top film and inoculate 1 mL of sample onto the center of the bottom film with a pipette, pipettor, or plate loop continuous pipetting syringe. Lower the top film onto the inoculum. Place the supplied plastic spreader on the top film over the inoculum in accordance with the manufacturer’s instructions. Distribute the sample by pressing downward on the center of the plastic spreader. Remove the spreader and leave the plate undisturbed for 1 minute to permit the gel to solidify. Incubate the plates (with the clear side facing up) in stacks, as in the pour plate method, but do not exceed 20 plates. Count all red colonies regardless of size or intensity and record the results. The circular growth area is

approximately 20 cm2. Estimates can be made on plates containing more than 250 colonies by counting the number of colonies in one or more representative squares and determining the average number per square. To determine the estimated count per plate, multiply the average number per square by 20.

8.83

Hydrophobic Grid–Membrane Filter (HGMF) Method

The hydrophobic grid–membrane filter (HGMF) method uses a specialized 0.45 mm pore size square membrane (Neogen Corp., Lansing, MI) that contains a hydrophobic grid comprised of 1600 squares. This method produces square colonies; their enumeration is based on the most probable number (MPN) technique. The aerobic count HGMF method uses a tryptic soy agar medium containing the stain Fast Green FCF, which eliminates the need for postincubation staining since the colonies develop varying intensities of green during incubation. At the recommended concentration the stain produces no evidence of toxicity. The method was accorded Official First Action status by the AOAC International.10 Refer to HGMF general procedures in the chapter ‘‘Rapid Methods for the Detection and Identification of Foodborne Pathogens’’ for general information on HGMF counts.

8.84

Spiral Plate Method

The spiral plate method (Spiral Systems Instruments, Inc., Bethesda, MD) for enumerating microorganisms has been collaboratively tested with milk and milk products, foods, and cosmetics; it has been found equivalent to the standard pour plating procedure.12 A known volume of sample is dispensed onto a rotating agar plate in an Archimedes spiral. The volume of sample decreases as the spiral moves out toward the edge of the plate. A modified counting grid, which relates the area of the plate to sample volume, is used to count the colonies in an appropriate area of the plate. With this information, the colony count for the sample can be computed. The procedure is as follows. 1.

2.

3.

4.

Prepare the sample (see the chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis’’). If necessary, let diluted solid samples settle a few minutes before removing test portions since particles may clog the tubing. Check the stylus tip angle daily by using the vacuum to hold a cover slip against the face of the stylus. The cover slip should be parallel to and 1 mm from the surface of the agar. Adjust, if necessary. Decontaminate the stylus tip and tubing by first pulling household bleach and then pulling sterile water through the system, before pulling the liquid sample into the stylus. Place a pre-poured plate count agar Petri plate on the turntable, and lower the stylus. The sample is differentially dispersed as the stylus tip rides on the surface of the rotating agar plate. Remove the inoculated plate while returning the stylus to the starting position. Decontaminate the stylus before processing the next sample. | 99

Compendium of Methods for the Microbiological Examination of Foods |

5. 6.

Incubate the plates, as in the pour plate method. Count the colonies and report the results, as described by the manufacturer.

8.85

Calibrated Loop Method

The plate loop method29 and the oval tube (or bottle culture) method15 use volumetrically calibrated loops (0.01 mL or 0.001 mL) for transferring samples in lieu of decimal dilutions. These methods are useful only for nonviscous liquids with counts greater than 2,500 CFU/mL or for viscous and solid foods with counts greater than 25,000 CFU/mL. For the plate loop method, fit a calibrated loop at the end of a Cornwall continuous pipetting device. After dipping the loop into the sample, rinse the measured volume in the charged loop into a Petri plate by depressing the Cornwall plunger. The oval tube method transfers a calibrated loopful of sample or diluted sample directly to agar in a tube. After solidification, incubate the tube as desired.

8.86

Drop Plate Method

The drop plate method of enumerating microorganisms is similar in principle to the spread plate method, except bent glass rods are not used to spread the diluted sample on the agar surface. The diluted samples are measured onto the surface of pre-poured plates by adding a predetermined number of drops from a specially calibrated pipette. The drops are allowed to spread and dry over an area of the agar surface; the area is usually 1.5 to 3 cm in diameter. The plates are incubated at the required temperature and time. After incubation, the colonies are counted and the computation of the colony count is based on the number of drops per plate, the number of drops per milliliter, and the dilution factor. The method is not recommended for food samples having counts of less than 3,000 CFU/g. For details, see the literature cited below.3,16

8.87

SimPlate Total Plate Count Color Indicator (TPC-CI)

The SimPlate Total Plate Count Color Indicator (BioControl, Bellevue, WA) for total plate count is used to quantify bacterial populations in food. This method— which has been collaboratively tested with various foods and compared to the conventional plate count7,21,22,27 and Petrifilm methods6—is based on the correlation between enzyme activity and the presence of viable bacteria. For growth and survival, the medium contains multiple enzyme substrates that are used by key enzymes in many foodborne bacteria. In this method, the food sample is mixed with the medium and distributed onto a SimPlate that contains a fixed number of wells. When the substrates are hydrolyzed by bacterial enzymes, a blue fluorescent molecule (i.e., 4-methylumbelliferone) is produced that is readily visible when exposed to ultraviolet light. The number of visible wells is then correlated to the number of bacteria by using an MPN table.

8.9

INTERPRETATION

Table 8-1 provides some general APC guidelines for various ingredients and finished products, drawn from 100 |

several expert sources.2,24 The range listed for certain commodities (e.g., raw ground beef, frozen fruits, and vegetables) is intentionally broad, reflecting varietal differences that are beyond the scope of this chapter. As stated previously, the APC values provide only a general indicator of the bacteriological profile of a given food or beverage sample. The use of narrowly defined APC specifications for a particular commodity or product is not always appropriate. The interpretive value of any given APC data point can only be determined in light of historical data and complementary counts for Enterobacteriaceae, coliforms, Escherichia coli (see the chapter ‘‘Enterobacteriaceae, Coliforms, and Escherichia coli as Quality and Safety Indicators’’), yeast and mold (see the chapter ‘‘Yeasts and Molds’’), or other organisms of concern.

ACKNOWLEDGMENT Fourth edition author: R. Dale Morton.

REFERENCES 1. Babel, F. J., E. B. Collins, J. C. Olson, and I. I. Peters. 1955. The standard plate count of milk as affected by the temperature of incubation. J. Dairy Sci. 38:499-503. 2. Banwart, G. J. Estimating the number of Microorganisms. 1981. In: Basic Food Microbiology, Abridged ed., pp. 10-33, AVI Publishing, Westport, CT. 3. Barbosa, H. R., M. F. A. Rodrigues, C. C. Campos, M. E. Chaves, I. Nunes, I. Juliano, and N. F. Novo. 1995. Counting of viable cluster-forming and non cluster-forming bacteria: a comparison between the drop and spread methods. J. Microbiol. Methods. 22:39-50. 4. Beloti, V., M. de Aguiar Ferreira Barros, L. A. Nero, J. A. S. Pachemshy, E. H. W. Santana, and B. D. G. M. Franco. 2002. Quality of pasteurized milk influences the performance of ready-to-use systems for enumeration of aerobic microorganisms. Int. Dairy J. 12:413-418. 5. Berry, J. M., D. A. McNeill, and L. D. Witter. 1969. Effect of delays in pour plating on bacterial counts. J. Dairy Sci. 52:1456-1457. 6. Beuchat, L. R., F. Copeland, M. S. Curiale, T. Danisavich, V. Gangar, B. W. King, T. L. Lawlis, R. L. King, J. Okwusoa, C. F. Smith, and D. E. Townsend. 1998. Comparison of the SimPlate Total Plate Count method with Petrifilm, Redigel, and conventional pour-plate methods for enumerating aerobic microorganisms in foods. J. Food Prot. 61:14-18. 7. Clark, D. S. 1967. Comparison of pour and surface plate methods for determination of bacterial counts. Can. J. Microbiol. 13:1409-1412. 8. Crowley, E. S., P. M. Bird, M. K. Torontali, J. R. Agin, D. G. Goins, and R. Johnson.2009. TEMPO TVC for the enumeration of aerobic mesophilic flora in foods: a collaborative study. J. AOAC Int. 92:165-174. 9. Curiale, M .S., P. Fahey, T. L. Fox, and J. S. McAllister. 1989. Dry rehydratable films for enumeration of coliforms and aerobic bacteria in dairy products: collaborative study. J. Assoc. Off. Anal. Chem. 72:312-318. 10. Entis, P. 1986. Hydrophobic grid membrane filter method for aerobic plate count in foods. J. Assoc. Off. Anal. Chem. 69:671676. 11. Fowler, J. L., W. S. Clark, J. F. Foster, and A. Hopkins. 1978. Analyst variation in doing the standard plate count as described in standard methods for the examination of dairy products. J. Food Prot. 41:4-7.

| Mesophilic Aerobic Plate Count

12. Gilchrist, J. E., C. B. Donelley, J. T. Peeler, and J. E. Campbell. 1977. Collaborative study comparing the spiral plate and aerobic plate count methods. J. Assoc. Off. Anal. Chem. 60:807-812. 13. Ginn, R. E., V. S. Packard, and T. L. Fox. 1986. Enumeration of total bacteria and coliforms in milk by dry rehydratable film methods: collaborative study. J. Assoc. Off. Anal. Chem. 69:527-531. 14. Hartman, P. A., and D. V. Huntsberger. 1961. Influence of subtle differences in plating procedure on bacterial counts of prepared frozen foods. Appl. Microbiol. 9:32-38. 15. Heinemann, B., and M. R. Rohr. 1953. A bottle agar method for bacterial estimates. J. Milk Food Technol. 16:133-135. 16. Herigstad, B., M. Hamilton, J. Heersink. 2001. How to optimize the drop plate method for enumerating bacteria. J. Microbiol. Methods. 44:121-129. 17. Huhtanen, C. N., A. R. Brazis, W. L. Arledge, E. W. Cook, C. B. Donnelly, R. E. Ginn, J. J. Jezeski, D. Pusch, H. E. Randolph, and E. L. Sing. 1972. Effects of time of holding dilutions on counts of bacteria from raw milk. J. Milk Food Technol. 35:126-130. 18. Koburger, J. A. 1980. Stack pouring of Petri plates: a potential source of error. J. Food Prot. 43:561-562. 19. Microbiology Evaluation Division, Bureau of Microbial Hazards, HPFB. 2001. Determination of the Aerobic Colony Count in Foods. MFHPB-18. Available at http://www.hc-sc. gc.ca/fn-an/res-rech/analy-meth/microbio/volume2/ mfhpb18-01-eng.php. Accessed May 22, 2013. 20. National Research Council, Subcommittee on Microbiological Criteria for Foods and Food Ingredients. 1985. An Evaluation of the Role of Microbiological Criteria for Foods and Food Ingredients. National Academy Press, Washington D.C. 21. Nero, L. A., V. Beloti, M. de Aguiar Ferreira Barros, E. H. W. de Santana, M. S. Pereira, V. V. Gusma˜o, and L. B. de Moraes.

22.

23.

24.

25.

26.

27.

28.

29.

2002. Assessment of the efficiency of SimPlate total plate count color indicator (TPC CI) to quantify mesophilic aerobic microorganisms in pasteurized milk. Braz. J. Microbiol. 33:4448. Pangloli, P., F. Jackson, H. A. Richards, J. R. Mount, and F. A. Draughon. 2006. Comparison of conventional plating and SimPlate methods for enumeration of aerobic microorganisms, coliform and Escherichia coli in farm environmental samples. J. Rapid Meth. Auto. Microbiol. 14:258-265. Piton, C., R. Grappin. 1991. A model for statistical evaluation of precision parameters of microbiological methods: application to dry rehydratable film methods and IDF reference methods for enumeration of total mesophilic flora and coliforms in raw milk. J. Assoc. Off. Anal. Chem. 74:92-103. Ray, B. 2004. Normal microbiological quality of foods and its significance. In: Fundamental Food Microbiology, 3rd ed., pp. 43-53. CRC Press, Boca Raton, FL. Rosmini, M. R., M. L. Signorini, R. Schneider, and R. Bonazza. 2004. Evaluation of two alternative techniques for counting mesophilic bacteria in raw milk. Food Control. 15:39-44. Silliker J. H. 1963. Total counts as indexes of food quality. In: NRC Subcommittee on Microbiological Criteria: Microbiological Quality of Foods, pp. 102-112. Academic Press, New York, NY. Smith, C. F., and D. E. Townsend. 1999. A new medium for determining the total plate count in food. J. Food Prot. 62:1404-1410. Smith, L. B., T. L. Fox, and F. F. Busta. 1985. Comparison of a dry medium culture plate (Petrifilm SM Plates) method to the aerobic plate count method for enumeration of mesophilic aerobic colony-forming units in fresh ground beef. J. Food Prot. 48:1044-1045. Thompson, D. I., C. B. Donnelly, and L. A. Black. 1960. A plate loop method for determining viable counts of raw milk. J. Milk Food Technol. 23:167-171.

| 101

|

CHAPTER 9

|

Enterobacteriaceae, Coliforms, and Escherichia coli as Quality and Safety Indicators Jeffrey L. Kornacki, Joshua B. Gurtler, and Bradley A. Stawick

9.1

INTRODUCTION

In 1887, Escherich observed the ubiquity of the organism, now known as Escherichia coli, in human stools. Schardinger (Australia) in 1892 and Smith (USA) in 1895 independently introduced the use of E. coli and related organisms to indicate the potential presence of enteric pathogens (e.g., Salmonella Typhi) in water.69 The premise was that E. coli in water indicated fecal contamination because E. coli and enteric pathogens were found in the feces of warm-blooded animals. Around 1915, the U.S. Public Health Service began using coliforms instead of E. coli as the standard for water, based on the false assumption that all coliforms possess equal value as sanitary indicators. The original practice of testing for E. coli and coliforms to assess water contamination expanded first to incorporate pasteurized milk and dairy products, and then to other foods. Little thought seems to have been given to the validity of using indicator organisms in widely disparate food matrices with divergent microbial ecologies. As early as 1924, papers cautioned that some of the analytical methods that worked well with water were not well suited to the analysis of milk.81 In 1937, Breed et al.11 stated that ‘‘the usefulness of tests of organisms of the Escherichia-Aerobacter types and so-called intermediates in dairy products has been complicated by a tendency on the part of many workers to carry over the interpretations of results from the water sanitation field into the dairy field.’’ These historical paradigms originating in the field of water hygiene continue to cause confusion in food testing. The assumption that the presence of or high numbers of E. coli, coliforms, fecal coliforms, or Enterobacteriaceae in a food indicates fecal contamination is invalid for the following reasons: (1) E. coli, coliforms and other Enterobacteriaceae are not obligate inhabitants of the intestinal tract of warm-blooded animals; (2) environmental reservoirs of these organisms exist16,17,51,58 and therefore the assumption is incorrect that Enterobacteriaceae are obligate

enteric dwellers; (3) these organisms are common in food manufacturing environments and may become part of the resident microflora of the facility, especially when sanitation is inadequate; (4) the growth of E. coli can occur on some foods when temperatures exceed approximately 7uC to 8uC; and (5) the growth of some coliforms, fecal coliforms, and other Enterobacteriaceae can occur on some foods during refrigeration.15,18,55 An attempt was made in the 1970s to differentiate between using E. coli, the Enterobacteriaceae, and coliforms as a marker or index of the potential presence of pathogens (for food safety purposes) and using these organisms as indicators of overall food quality.66 This differentiation in function is critical because it is rare that an organism or a group of organisms can be used to simultaneously address food safety and food quality. Index organisms are assumed to signal the increased likelihood of a pathogen originating from the same source as the index organism and thus serve a predictive function.13 For example, the original application of E. coli to assess water safety used E. coli as an index of Salmonella contamination from the feces of warm-blooded animals. As with most other historical uses of the Enterobacteriaceae family as index organisms, this application incorrectly assumed that the only source of these bacteria was the intestinal tract. However, there is no reason to assume that Enterobacteriaceae would have any value in predicting the presence of pathogens (such as Listeria monocytogenes) that typically originate from extraintestinal sources. Numerous studies have determined that E. coli, coliforms, fecal coliforms and Enterobacteriaceae are unreliable when used as an index of pathogen contamination of individual production lots of foods.64,84,87 Higher levels of index organisms may, over many lots, be correlated with a greater probability of one or more enteric pathogens being present (Table 9-1). However, the absence of the index organism does not indicate that the food is free from enteric

| 103 |

Compendium of Methods for the Microbiological Examination of Foods |

Table 9-1. Salmonella Incidence in Relationship to E. coli Most Probable Number (MPN) in Raw Preformed Meat Patties

,3 3-5 51–100 101–240 241–1,100 1,101–11,000 . 11,000

Samples

Samples Positive

Percent Positive

Within

for Salmonella

Within

E. coli MPN/g

Within MPN Range

MPN Range

270 406 54 96 65 56 25

2 20 3 4 3 9 5

0.7 4.9 5.6 4.1 4.6 16.1 20.0

pathogens. An unpublished study of raw hamburger patties from 16 producers indicated that as the level of E. coli increased in ground beef patties, the incidence of Salmonella also increased (Table 9-1). Siragusa et al.86 and Wyss and Hockenjos101 showed that aerobic plate count (APC) levels were generally correlated with the incidence of E. coli and Shiga toxin-producing E. coli (STEC), respectively, on beef carcasses. National and international advisory committees such as the Food and Agriculture Organization of the United Nations/World Health Organization and the U.S. National Research Council’s Subcommittee on Microbiological Criteria not surprisingly have concluded that it is invalid to predict the safety of food products based on levels of coliforms, fecal coliforms, Enterobacteriaceae or E. coli.34,71 Data from the 1994 outbreak of Salmonella in ice cream support this view. Sufficient Salmonella Enteritidis cells were present in the ice cream to cause nearly a quarter million human infections, despite very low coliform and E. coli counts (, 1 CFU/g).96 The greatest application of coliform, Enterobacteriaceae, and E. coli testing is in assessing the overall quality of a food and the hygienic conditions present during food processing. For example, enumeration of these organisms in heat-pasteurized foods can be used to assess the adequacy of a heating process that is designed to inactivate vegetative bacteria. As early as 1927, dairy microbiologists used E. coli as a true indicator organism to assess postpasteurization contamination of milk, especially contamination originating from improperly cleaned bottles.92 The process of milk pasteurization is known to destroy E. coli; thus, the presence of any E. coli in milk, after pasteurization may indicate inadequate pasteurization, poor hygienic conditions in the processing plant, and/or post-processing contamination. Several factors must be considered before testing for a particular indicator organism or group of organisms: (1) the physicochemical nature of the food; (2) the native microflora of the food; (3) the extent to which the food has been processed; (4) the effect that processing would be expected to have on one or more of the tested indicator organisms; (5) the physiology of the indicator 104 |

organisms; and (6) the accuracy with which the intended testing method can identify the indicator organisms. Improved hygienic indicator assays may be those which detect a wide variety of microbes, including potential pathogens of concern and nonpathogens. Thus, positive results would indicate the need to improve hygienic practices, whereas negative results suggest that pathogens are not present. This concept has been illustrated in case studies with an assay for hydrogen sulfide-producing thermophilic Enterobacteriaceae.54

9.2

DEFINITIONS

The term ‘‘Enterobacteriaceae group’’ will be used in this chapter to refer collectively to coliforms, fecal coliforms, E. coli and other bacteria in the taxonomic family Enterobacteriaceae. The definitions used for coliforms and fecal coliforms are best described as working concepts since these groups have no taxonomic status. ‘‘Coliforms’’ and ‘‘fecal coliforms,’’ practically speaking, are detected by the ‘‘coliform test’’ and the ‘‘fecal coliform test,’’ respectively.

9.21

Enterobacteriaceae Family

The taxonomically defined family Enterobacteriaceae includes fifty species of facultatively anaerobic Gramnegative straight bacilli (which ferment glucose to acid) that are oxidase-negative, usually catalase-positive, usually nitrate-reducing, motile by peritrichous flagella or nonmotile, and vary with respect to their virulence in humans. Common foodborne genera of the family Enterobacteriaceae include Citrobacter, Enterobacter, Erwinia, Escherichia, Hafnia, Klebsiella, Proteus, Providencia, Salmonella, Serratia, Shigella, and Yersinia. The family also includes several lesser known and rather obscure species (e.g., Aquamonas, Obesumbacterium, Wigglesworthia, Photorhabdus, Samsonia, Phlomobacter). Psychrotrophic members of this family are not uncommon,55,68 although the Enterobacteriaceae are widely regarded as being mesophilic. Psychrotrophic strains of Enterobacter, Hafnia, Yersinia and Serratia may grow at temperatures as low as 0uC.76 The Enterobacteriaceae have been used for years in Europe as indicators of food quality and as indices of food safety. In the United States, the use of coliforms as indicators of food quality or insanitation in food processing environments is based on tradition. This practice arbitrarily judges food quality or manufacturing plant insanitation based on recovering members of the Enterobacteriaceae group (i.e., coliforms) that ferment lactose, thus ignoring the presence of non–lactose-fermenting members. Mossel et al.70 recommend examining food products for the presence of Enterobacteriaceae, rather than coliforms, to better assess glucose-positive, lactose-negative members (e.g., Salmonella, Citrobacter) present in the food microflora. However, the practice of using coliforms as a measure of sanitation efficacy, hygienic process control, and air quality may continue in some sectors of the United States food industry because of tradition or liability-related concerns since Salmonella microorganisms are mainly lactose-negative and are included in the Enterobacteriaceae family, despite the lack of correlation between Enterobacteriaceae and Salmonella as an indicator.

| Enterobacteriaceae, Coliforms, and Escherichia coli as Quality and Safety Indicators

9.22

Coliforms

The term ‘‘coliform’’ appears to have been coined in 1893 by Blachstein8 to refer to bacilli that resembled E. coli and yielded similar colonies in culture (specifically on plates). The coliform group is defined on the basis of biochemical reactions, not genetic relationships. Thus, the term ‘‘coliform’’ has no taxonomic status. Coliforms are traditionally considered aerobic and facultatively anaerobic, Gramnegative, non-sporeforming rods that ferment lactose, forming acid and gas within 48 hr at 35uC.41 An incubation temperature of 32uC is usually used for dairy products.21,42 Test media include a variety of lactose-containing liquids or solid media supplemented with dyes, and/or surfaceactive agents. Some coliform tests only rely on gas and not on acid production from lactose. The coliform group may contain organisms (e.g., Aeromonas species) that are not included, or only provisionally included in the Enterobacteriaceae family. Representatives of 20 or more species may be classified as coliforms, on the basis of available evidence, in which lactose fermentation to gas is used as the defining criterion. By using synthetic substrates, it is also possible to detect coliforms by assaying cultures for their b-galactosidase activity, rather than by lactose fermentation. Detection may be based on the formation of colored end-products or on the formation of volatile o-nitrophenol.25,30,91 However, assays using synthetic substrates for b-galactosidase may detect Enterobacteriaceae that do not ferment lactose to gas in the traditional most probable number (MPN) assay and hence would not be classified as coliforms by traditional approaches.

9.23

Fecal Coliforms

‘‘Fecal coliforms’’ are coliforms that can ferment lactose to acid and gas within 48 hr at 44.5uC or 45.5uC, depending on the food matrix (usually E. coli [EC] broth).44 Strains of E. coli, Klebsiella pneumoniae, Enterobacter spp. (including agglomerans, aerogenes, and cloacae), and Citrobacter freundii may be recovered by a fecal coliform test, depending on the food product and incubation temperature.89 The term ‘‘fecal coliform,’’ as with coliforms, has no taxonomic status. The term ‘‘thermotolerant coliforms’’ is sometimes used to refer to these organisms and is, in our view, a more accurate description than is ‘‘fecal coliform.’’ Eijkman in 1904 began the practice of incubating coliform tests at elevated temperatures to separate organisms thought to be of fecal origin from other coliform organisms. This practice was originally used to assess fecal contamination in water. However, high-temperature incubation and gas production from lactose will not, be reliably selective for organisms originating in the intestinal tract or in feces.5,33 The assay has no value as a direct measurement of fecal contamination, and this misinterpretation of the assay has led to much confusion for many years.23 The fecal coliform test is essentially a truncated version of the E. coli MPN test. It does not involve isolation or the lengthy and laborious indole, methyl red, Voge-Proskauer, and citrate (IMViC) tests that are traditionally used to confirm the presence of E. coli. However, a 48-hr IMViC assay was developed that offers the advantage of contact with differential chemicals in solid media of the much

higher bacterial concentration in a colony, compared to the traditional assays using a broth culture that may take 120 hr. Reactions can be read on a single quad plate.77 A variety of incubation temperatures are used for detecting fecal coliforms. A temperature of 45.5uC ¡ 0.2uC is widely used for foods, whereas 44.5uC is recommended for shellfish, water, and shellfish harvest water.45 Data indicate that the incubation of EC broth at 45.5uC may be more specific for E. coli, whereas incubation at 44.5uC may yield slightly higher numbers of other fecal coliforms. Evidence indicates that an incubation temperature of 45.0uC ¡ 0.2uC for all tests would be a logical compromise.97 Well-regulated circulating water baths are preferred to air incubators, given the narrow temperature range (typically ¡ 0.2uC) allowed for incubation of fecal coliform and E. coli assay tubes. It should be noted that the elevated incubation temperatures used to recover fecal coliforms and E. coli from foods are unsuitable for the growth of some pathogenic E. coli. Strains of E. coli O157:H7, for example, grow poorly at 44uC and not at all at 45.5uC. 22 It is unclear whether this is true of other enterohemorrhagic strains of E. coli. Tuttle et al.93 found that the MPN of E. coli incubated at 45.5uC did not show any correlation with the presence of E. coli O157:H7 in ground beef. Enteropathogenic strains of E. coli will be described in the chapter ‘‘Pathogenic Escherichia coli.’’

9.24

Escherichia coli

The identification and enumeration of E. coli for sanitary significance relies on isolate conformance to the coliform and fecal coliform group definitions. E. coli isolates are traditionally identified by their IMViC pattern: + + – – (Type I) and – + – – (Type II). In this scheme, ‘‘I’’ refers to the ability of the organism to produce indole from the metabolism of tryptophan; ‘‘M’’ indicates the ability of the organism to ferment glucose to ‘‘high’’ acid, as detected by methyl red pH indicator dye in the medium; ‘‘Vi’’ stands for the production of the neutral products 2,3-butanediol and/or acetoin from glucose metabolism (otherwise known as the ‘‘Vogues-Proskauer’’ reaction); and ‘‘C’’ represents the ability of the bacterium to use citrate as the sole carbon source. Recent data indicate that defining E. coli by the IMViC profile is inadequate for identification of the species (see Section 9.37). For example, there are E. coli strains that do not give IMViC reactions that correspond to biotype I or biotype II.47 The relatively high incidence of Type II E. coli in some specimens is at least partly explained by the fact that many isolates require 48 hr to produce a detectable amount of indole, and additional tests are essential for speciation.

9.3 9.31

PRECAUTIONS Cultures

Stock cultures of E. coli (IMViC pattern + + – – ) and Enterobacter aerogenes (IMViC pattern – – + +) should be maintained to test the quality control of the IMViC media and reagents. Check the performance of all media. The control strain of E. coli should produce gas in the EC medium at 45.5uC within 24 hr. The control strain of E. aerogenes should produce a negative reaction. | 105

Compendium of Methods for the Microbiological Examination of Foods |

9.32

Dilutions

Prepare only as many test dilutions as can be inoculated within 15 min.

9.33

Media

Exhausting autoclaves too rapidly can result in air bubbles forming inside Durham tubes that are used in MPN assays for Enterobacteriaceae, coliforms, fecal coliforms or E. coli. This can confound the interpretation of the assays. Liquid media also absorb air during cold storage and should be allowed to reach laboratory temperature after their removal from refrigeration. Tubes should be inspected before use. Tubes with gas bubbles should be discarded. Include an uninoculated medium as a negative control.

9.34

General Limitations of the MPN Liquid Enrichment Test

Most probable number determinations potentially enhance recovery because of the enrichment of injured cells, absence of cells injured by heat from molten solid media (which can occur with direct plating), lack of confounding by particulates on solid media, and adjustment of the sensitivity of the assay by changing the quantity and/or dilution tested (e.g., enriching a quantity of 100 g or larger at the lowest dilutions for increased sensitivity). The most probable number (MPN) technique nevertheless only indicates the most likely number of organisms present in a sample. The convention among food microbiologists has been to state only the MPN value, rather than the MPN and its associated confidence interval. This reporting method unfortunately does not convey the idea that the MPN is purely a statistical approximation; the true number of organisms in the sample is an unknown value occurring within the MPN confidence interval. The confusion about the MPN value is regrettably common. A glance at a 3-dilution (i.e., 9-tube) MPN table indicates that the lower limit of detection is accompanied by quite a wide confidence interval.10 (See the chapter ‘‘Culture Methods for Enumeration of Microorganisms.’’) For example, a MPN of less than 3.0 occurs when quantities of 0.1 g, 0.01 g, and 0.001 g are tested in triplicate and none of the tubes is positive (i.e., 0-0-0 result). However, the upper limit of the 95% confidence interval is 9.5 MPN per gram or milliliter. The confidence intervals increase with greater numbers of positive tubes. A MPN value of 1100 (from a 3-3-2 result) consequently has a 95% confidence interval ranging from 180 MPN/g to 4100 MPN/g. Given this inexactitude, one may question whether the labor needed to perform the MPN is justified or whether simpler, less laborious plating methods may be used. It is possible to achieve a lower limit of 1 CFU per gram or milliliter by pour-plating 1 mL of a 1:10 dilution onto each of 10 plates. In practical terms, a result of less than 1 CFU per gram or milliliter by a direct count method may be just as useful as an MPN result of less than 0.30 MPN per gram or milliliter. Therefore, the decision to use a MPN versus a direct count approach should take into consideration the food matrix, the level of required sensitivity and expected accuracy, labor and time considerations and the analyst’s experience with the product.

106 |

9.35

Minimal Number of Fecal Coliforms

The fecal coliform assay is identical to the assay used for the detection of presumptive E. coli; however, the confirmation step to determine what proportion of the fecal coliform count (if any) are E. coli is not performed. Some food manufacturers stop at this point and do not undertake the time-consuming and laborious IMViC, testing that is typically performed to confirm E. coli. A fecal coliform count is occasionally wrongly interpreted as being equivalent to the E. coli count. All E. coli that are recovered from EC broth (at 44.5uC or 45.5uC) or that have a 4-methylumbelliferyl-b-D-glucuronic acid (MUG) positive reaction are fecal coliforms, but all fecal coliforms are not E. coli. It is impossible to extrapolate the E. coli population from a fecal coliform count, unless confirmatory testing is performed or unless a particular food manufacturer has a comprehensive collection of in-house data that indicates a preponderance of E. coli among fecal coliforms for a specific food product.

9.36

General Limitations of Eosin Methylene Blue Agar

Levine’s eosin methylene blue (L-EMB) agar is not especially selective for E. coli, and physical and subjective limitations are encountered in the use of this agar.90 The ability to discern one distinctive colony among many, and specifically to recognize an E. coli–like colony from among many coliform colonies, is critical to the success of the entire analytical procedure. Some biotypes of E. coli unfortunately do not produce a colony with the typical green sheen; slow or non– lactose-fermenters produce colorless colonies. Some nonE. coli colonies (e.g., Klebsiella pneumoniae) may also exhibit the typical E. coli–like morphology on L-EMB agar.95 Some authors have reported a 28% E. coli confirmation rate for ‘‘typical’’ colonies taken from L-EMB agar versus a confirmation rate of 86% for MUG-positive colonies taken from violet red bile agar (VRBA). Inoculum taken from a ‘‘typical’’ E. coli colony on an L-EMB plate may represent a mixed population and mixed cultures confound IMViC and other confirmatory tests.47 The selection of a pure culture is therefore essential.

9.37

Limitations of the IMViC Tests for E. coli Confirmation

The use of the biochemical tests, now referred to as the IMViC series (indole, methyl red, Voges Proskauer, and citrate), to characterize lactose-fermenting bacilli dates back to the 1920s. The IMViC tests were originally used as an attempt to differentiate coliforms believed to be of intestinal origin from those believed to be soil-borne. However, the IMViC tests were incapable of differentiating coliforms on the basis of habitat, even when supplemented with other biochemical characterization tests.59 Commercial identification systems can be used in place of the IMViC series. The AOAC Official Method 978.24 lists the following kits: API 20e, Remel MicroID, Microgen GN-ID, BD Enterotube II, and bioMe´rieux VITEK systems. These biochemical identification systems compare results to databases that provide the most likely identification of the isolate. After inoculation with a freshly isolated bacterial colony, results from these assays require approximately 1 day before reading.

| Enterobacteriaceae, Coliforms, and Escherichia coli as Quality and Safety Indicators

As early as 1934, the routine use of the IMViC tests for confirming E. coli was questioned.7 Advances in bacterial taxonomy and improvements in biochemical and genetic confirmation methods for E. coli also cast doubt on the wisdom of relying on IMViC tests for E. coli confirmation. Various references indicate that the following members of the Enterobacteriaceae may give IMViC reactions indicative of biotype I E. coli: Escherichia spp. (E. fergusonii, E. hermannii, E. blattae, and non-gas producing strains of E. coli); Enterobacter agglomerans, Edwardsiella spp. (viz., hoshinae, tarda); Leclercia (formerly Escherichia) adecarboxylata; Kluyvera cryocrescens; Morganella morganii; Proteus spp. (viz., mirabilis, morganii, vulgaris), Providencia rustigianii; Shigella spp. (viz., boydii, dysenteriae, flexneri); and Yersinia spp. (viz., enterocolitica, frederiksenii, intermedia, kristensenii). Biotype II E. coli reactions are also shared by some strains of the following Enterobacteriaceae: Budvicia aquatica; Edwardsiella hoshinae; Enterobacter spp. (viz., agglomerans, hafniae); Escherichia spp. (viz., blattae, vulneris); Klebsiella pneumoniae; Moellerella wisconsensis; Pragia fontium; Proteus penneri, Providencia heimbachae; Salmonella enterica subsp. indica; Shigella spp. (viz., boydii, dysenteriae, flexneri, sonnei); and Yersinia spp. (viz., aldovae, bercovieri, enterocolitica, kristensenii, mollaretii, pestis, pseudotuberculosis, ruckeri). 12,26,29,47 Silliker Laboratories of Wisconsin in the mid-1990s (unpublished results) examined 16 isolates from a freezedried meat extract recovered from a modified USP enrichment procedure94; the isolates displayed IMViC reactions typical of E. coli biotype I and II organisms. Further biochemical characterization (using the Biolog identification system) indicated that none of the isolates was E. coli. Isolates were instead identified as Escherichia spp. (vulneris, hermanii), Citrobacter freundii (lactose and non-lactose fermenting biotypes), Leclericia adecarboxylata, and Enterobacter spp. (agglomerans and cloacae, with isolates corresponding to biotype I and to biotype II). Selected strains did not produce gas after a 48-hr incubation in EC broth at 45.5uC. This indicates the importance of an elevated incubation temperature (e.g., 44.5uC to 45.5uC) to reduce the number of false positives in the ‘‘completed’’ MPN test for E. coli (see Section 9.91). It is important to note that most classical methods that test for for coliforms, fecal coliforms, Enterobacteriaceae, and E. coli are targeted at ‘‘typical’’ organisms. The MPN assay for E. coli, for example, quantifies only typical, gasproducing organisms. Anaerogenic strains that do not produce gas from lactose within the required 48 hr or strains that are incapable of growth at 44.5uC to 45.5uC would not be detected. Furthermore, classical detection methods for coliforms and fecal coliforms are based on biochemical and physiological responses to enrichment– the isolation of pure cultures and confirmatory testing is not performed. A similar situation exists with the enumeration of Enterobacteriaceae on violet red bile glucose agar (VRBGA). These tests all sacrifice some accuracy to obtain (relatively) rapid results. It remains to be seen whether these limitations will continue to be ignored in favor of expediency or whether more sophisticated testing procedures (e.g., genetic or extensive biochemical confirmation) will be utilized to detect E. coli from foods. These approaches may become more necessary in the face of an

increasing litigious climate in the food industry and at least two such AOAC-approved phenotypic-based tests are on the market, namely (1) the Solaris assay for E. coli (AOAC RI Certification No. 101101; AOAC Research Institute), which provides negative results in 24 hr and reportedly qualitatively detects E. coli in as little as 7 hr, and (2) the Tempo (AOAC OMA 2009.02) quantitative assay, which reportedly provides results in 22 to 27 hr.

9.38

Interferences

9.381 False-Negative Results The importance of false-negative reactions in the recovery of the Enterobacteriaceae group from water supplies has been described.28,57 False-negative results can similarly be obtained when analyzing food samples or food-plant environmental samples. Difficulty in recovering these organisms is partially attributable to sublethal cellular damage or stress introduced by food preservation practices or by environmental conditions such as drying, refrigeration and freezing, heating, acidification/fermentation, and/or the use of bactericidal or bacteriostatic agents.13,32,45,79 Sublethally damaged or stressed cells are unable to tolerate inhibitory agents (e.g., bile salts, sodium desoxycholate, crystal violet) that are present in selective media, especially if high incubation temperatures are used.90 Recovery of sublethally damaged or stressed members of the Enterobacteriaceae group can often be improved by resuscitation procedures.82,88 Resuscitation can occur in diluent or in nonselective agar or in broth media (e.g., before addition of selective agents). For the repair of injured cells, 2 to 6 hr at room temperature in nonselective liquid media are usually sufficient, but this may permit some growth of healthy and repaired cells.60 This type of approach has been used to improve the recovery of coliforms with VRB agar and on Petrifilm Coliform Count Plates72 and recovery of Enterobacteriaceae on VRBGA.49,68 Resuscitation in solid media (i.e., agar) appears to be as effective as resuscitation in liquid media and minimizes problems caused by growth during resuscitation.60,88 Speck et al.88 determined that 1 to 2 resuscitation periods in tryptic soy agar (TSA) was sufficient to allow the repair of injured coliforms in frozen foods. The value of spread-plating over pour-plating on the nonselective agar has been debated with some researchers advocating spread-plating88 and others finding no advantage in this procedure. Alternate overlay techniques have been proposed that involve adding nonselective agar on top of a pre-solidified selective medium.52,102 One technique is the thin agar layer (TAL) method.6,19,24,39,52,75,78,99,100,103 The TAL method requires that a nonselective layer be poured over a selective medium on the day of the experiment. Plates are incubated immediately for a maximum of 2 to 4 hr after the nonselective overlay. This is followed by spread plating. During subsequent incubation, selective agents from the bottom layer diffuse through the nonselective top layer to inhibit the growth of nontarget organisms. The benefits of the nonselective overlay method versus a selective overlay technique include the preclusion of further injuring cells by molten media at 45uC to 48uC and the elimination of a twostep process where nonselective plates are incubated for 1 to 4 hr, followed by adding the molten selective agar on | 107

Compendium of Methods for the Microbiological Examination of Foods |

top. In brief, pour 20 mL of VRBA tempered to 45uC to 50uC into petri dishes. Plates may be poured before the day of the experiment. Pour 10 mL of TSA (with 0.1% sodium pyruvate, if desired) to enhance the recovery of injured cells38 on top of the solidified VRBA and to allow the TSA to solidify and cool. Within 2 to 4 hr of pouring the nonselective top layer, samples may be spread or pourplated and incubated 18 hr to 24 hr at 32uC (for dairy products) or at 35uC (for other food products). Count the colonies, as described later in Section 9.73. A second type of false-negative result occurs when an analytical procedure fails to recover or enumerate strains that exhibit atypical behavior. The recovery method can be modified so as to include these organisms when past experience indicates that a particular food product commonly contains atypical strains. For example, anaerogenic strains of E. coli, which do not produce gas in lauryl sulfate tryptose (LST) broth, are present in foods. In muscle foods, anaerogenic strains of E. coli may constitute 3% to 74% of the E. coli population.2 Detection of these anaerogenic E. coli can be accomplished by streaking turbid, but not gassy LST tubes (see Section 9.71) onto a solid medium containing 4-methylumbelliferyl-b-D-glucuronic acid (MUG) and enumerating fluorescent (i.e., MUG-positive) colonies under ultraviolet light or by testing the colonies for indole production. A false-negative result may also be obtained because of microbial competition during enrichment. For example, the growth of Proteus vulgaris may suppress gas produced by E. coli in LST broth when both organisms are present.32 Slow growth and/or enzyme production is another cause of false-negative results. A 48-hr incubation period is usually recommended for MUG-containing liquid media to detect E. coli that are slow producers of b-glucuronidase (GUD). One study indicated that 34% of E. coli strains isolated from human fecal samples were MUG-negative in LST broth containing MUG.14 In some cases, a false-negative MUG test result occurs because the gene responsible for GUD production is present but unexpressed because of catabolite repression by lactose in the media.96 Transfer to a lactosefree minimal medium restores GUD activity. Indole-based E. coli detection tests (specific for biotype I only) may give false-negative results for E. coli (i.e., E. coli biotype II may be present) when dairy foods are tested directly since indole production is inhibited by high carbohydrate levels.45 False-negative results may also occur because of an inappropriate pH. For example, MUG hydrolysis is optimal under alkaline conditions.27 False-negative MUG results may occur in agars that contain high levels of fermentable carbohydrates, but lack good buffering capacity.

9.382 False-Positive Results Various false-positive reactions occur when analyzing foods for members of the Enterobacteriaceae group. Falsepositive reactions are especially common in MUG-based assays. Some batches of LST/MUG reportedly autofluoresce,27 as do glass test tubes containing cerium oxide.41 False-positive reactions and auto-fluorescence may occur when fish (e.g., salmon, tuna), Crustacea (e.g., shrimp), or shellfish (e.g., oysters, clams) homogenates are added to MUG-containing broths. The interference is 108 |

apparently caused by an enzyme native to the muscle tissue,43 and can usually be removed by centrifugation.95 Raw liver also contains endogenous b-D-glucuronidase, which may interfere with MUG-based assays.73 Falsepositive results because of food matrices can usually be ruled out by subculturing MUG-positive primary enrichment broths containing relatively high levels of the potentially interfering food into a secondary MUG-containing medium. Fluorescence in the secondary medium is nearly always representative of bacterial GUD activity since subculturing effectively dilutes the interfering food matrix. Non-Enterobacteriaceae present in a food sample may also produce b-D-glucuronidase and interfere with MUG-based E. coli tests. Strains of Staphylococcus warneri and Staphylococcus xylosus may exhibit fluorescence when grown in LST-MUG at 37uC, but no fluorescence is observed when the LST-MUG enrichments are subcultured to EC-MUG and incubated at 44.5uC.43 Incubation periods for MUG-containing agars are typically limited to 24 hr or less since MUG diffusion into the agar surrounding the GUD-positive colonies occurs during prolonged incubation periods and will give false-positive results.32 False-positive test reactions because of food matrices may also occur when the test relies on pH indicators. For example, tests that rely on the detection of acid production may yield false-positive results when an acidic food is tested without neutralization. Individual food particles may infrequently contain sufficient acid to change the color of the pH indicator in the agar immediately surrounding the food particle. False-positives (e.g., reddish particulates) have been observed with swine hair and flour products on VRB agar (Kornacki, unpublished observations).

9.4

EQUIPMENT, MATERIALS, AND REAGENTS

This section lists equipment, media, reagents and stains used in many common assays for the Enterobacteriaceae. Media and reagents specified for use with various commercially available assay kits are not included. The chapter ‘‘Microbiological Media, Reagents, and Stains’’ discusses the composition and preparation of media, reagents, and stains.

9.41

N N N N N N N N N N N N N

Equipment

Air incubators: 32uC, 35uC, 37uC, and 44uC Balance Refrigerator Stomacher/blender Glass-rod or plastic spreader, sterile Microscope with illumination, capable of 10006 magnification Petri dishes: 15 6 150 mm, glass or plastic, sterile Petrifilm spreader Hydrophobic grid membrane filter (HGMF) unit Pipette bulb or filler pipettes: 1 mL, 2 mL, 5 mL, 10 mL sterile serological; Pasteur Test tube racks (stainless steel, epoxy-coated or plastic) for various-sized tubes Total immersion thermometer: approximately 45 to 55 cm long, range of 1uC to 55uC, and standardized against a NIST-certified thermometer or equivalent Ultraviolet lamp: 365 nm wavelength (long wave): 250 to 400 nm ultraviolet lamp

| Enterobacteriaceae, Coliforms, and Escherichia coli as Quality and Safety Indicators

N N

Water-bath for tempering agar: 45uC ¡ 2uC Water baths with plastic or metal gable covers and mechanical circulation systems capable of maintaining temperatures of 44.5uC ¡ 0.2uC, 45uC ¡ 0.2uC, or 45.5uC ¡ 0.2uC

9.42

N N N N N N N N N N N N N N N N N N N N N N N N

9.43

N N N N N N N N N

Materials

Brilliant green bile (BGB) broth Buffered MUG agar (BMA) Chromocult Coliform agar E. coli (EC) broth E. coli broth with MUG (EC-MUG) Enterobacteriaceae enrichment (EE) broth Koser’s citrate medium or Simmons’ citrate slants Lactose monensin glucuronate (LMG) Lauryl sulfate tryptose (LST) broth: single-strength and double-strength Lauryl sulfate tryptose with MUG (LST-MUG) Levine’s eosin methylene blue (L-EMB) agar Methyl red-Voges Proskauer (MR-VP) broth m-FC Agar: without rosolic acid Petrifilm: coliform, E. coli, high-sensitivity, Enterobacteriaceae Rapid’E. coli 2 Agar Redigel ColiChrome Redigel violet red bile (VRB) Tryptic or trypticase soy agar (TSA) Tryptose bile agar (TBA) Tryptone (tryptophan) broth Violet red bile (VRB) agar Violet red bile Agar-2 (VRBA-2) Violet red bile (VRB) agar with MUG Violet red bile glucose agar (VRBGA)

Reagents

Butterfield’s phosphate-buffered dilution water ColiComplete reagent Gram stain reagents IDEXX Simplate Coliforms/E. coli Indole reagent, Kovac’s formulation Methyl red indicator 4-Methylumbelliferyl-b-D-glucuronic acid (MUG) 0.1% Peptone water diluent Voges-Proskauer reagents

9.5 9.51

SAMPLE PREPARATION Preparation of Food Test Samples

The chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis’’ provides information on sample collection and preparation before analysis, whereas the chapter ‘‘Culture Methods for Enumeration of Microorganisms’’ deals with the enumeration of foodborne microorganisms. If the food is frozen and must be thawed, refrigerate it at 2uC to 5uC for approximately 18 hr before analysis. Aseptically weigh 25 g of the unfrozen food into a sterile weighing jar, sterile blender jar, or sterile Stomacher bag. Add 225 mL of sterile Butterfield’s phosphate buffer diluent or sterile 0.1% peptone to the jar or Stomacher bag and homogenize for 2 min. If a larger

sample size is desired (e.g., 50 g), maintain a 1:10 ratio between the sample weight and diluent volume. A typical dilution series used for the analysis of the Enterobacteriaceae group is 1021 through 1023. However, additional dilutions may be required for analyzing samples suspected of having higher counts. The U.S. Food and Drug Administration (FDA) recommends in its Bacteriological Analytical Manual (BAM) that dilutions be prepared by adding 10 mL of the previous (i.e., lower) dilution to a fresh 90 mL diluent blank.3 Dilutions prepared in this manner are mixed by shaking 25 times in a 1-ft (30 cm) arc for 7 sec. Many food microbiology laboratories use 9 mL diluent blanks and prepare serial 1:10 dilutions by adding 1 mL of the previous dilution to a test tube containing 9 mL of sterile diluent. The contents of tubes may be mixed by using a vortex mixer or shaking, as described previously.

9.6

THE ENTEROBACTERIACEAE

Analyses for the Enterobacteriaceae have traditionally been conducted at 35uC to 37uC. Lower incubation temperatures (4uC, 10uC, 25uC, or 30uC) may be more effective when analyzing a refrigerated food that may contain psychrotrophic Enterobacteriaceae. Psychrotrophic organisms are often incapable of growing at 35uC but may grow at 30uC; they will grow at lower temperatures.68

9.61

Enrichment Method

Enrichment for Enterobacteriaceae is generally performed less frequently than is enumeration via plating with VRBGA (Section 9.62). Enrichment procedures for this family typically use Enterobacteriaceae enrichment (EE) broth, a modification of brilliant green bile (BGB) broth in which lactose has been replaced by glucose. The EE broth may be used in a straight enrichment procedure or in a MPN assay. For the isolation of Enterobacteriaceae, Mossel et al.70 originally described incubating this broth at 37uC for 20 to 24 hr, followed by streaking on VRBGA, and then incubating at 37uC. In North America, food microbiologists rarely use the 37uC incubation temperature, preferring 35uC instead. If desired, differentiation between psychrotrophic Enterobacteriaceae and ‘‘regular’’ mesophilic types can be achieved by incubation at 43uC for 18 hr; psychrotrophic types are unable to grow at the elevated temperature.70

9.62

Plating With VRBGA

Pour approximately 10 mL of VRBGA tempered to 48uC into plates containing 1.0 mL portions of the diluted sample. Swirl plates to mix well and allow the media to solidify. Overlay each plate with 5 to 8 mL VRBGA. After solidification, invert the plates and incubate them for 18 to 24 hr at 35uC. Examine the plates with illumination under a magnifying lens. Count purple-red colonies that are 0.5 mm or larger in diameter and are surrounded by a zone of precipitated bile acids. Plates should optimally have 15 to 150 colonies. Enterobacteriaceae colonies on more crowded plates may remain small and fail to reach 0.5 mm in diameter. Multiply the number of typical Enterobacteriaceae colonies by the reciprocal of the dilution used and report the results as the Enterobacteriaceae count | 109

Compendium of Methods for the Microbiological Examination of Foods |

(CFU/g or CFU/mL). If desired, Enterobacteriaceae colonies on VRBGA may be isolated and speciated by conventional or miniaturized biochemical tests.

9.63

Petrifilm Methods

All Petrifilm methods employ a dehydrated film medium that is applied to a card. Petrifilm plates used for the detection of Enterobacteriaceae, coliforms, and E. coli use a film containing selective and/or differential agents with a cold-water soluble gelling agent. The plating medium is hydrated when 1 mL of a diluted or undiluted sample is added to a 20 cm2 Petrifilm or when 5 mL of the sample is added to a high-sensitivity Petrifilm (60 cm2). The plastic overlay film is then carefully lowered onto the inoculated plate to prevent trapping small gas bubbles. Pressure applied to a plastic spreader placed on the overlay film distributes the test portions over 20 cm 2 (for the Enterobacteriaceae, coliform, and E. coli count plates) or over 60 cm2 (for the high-sensitivity coliform plate). The sample-containing Petrifilm plate is allowed to stand at room temperature for several minutes to allow gelation. Petrifilm plates are then stacked upright (in stacks of 20 or less) and placed in an incubator. Incorporation of triphenyltetrazolium dye in the Petrifilm facilitates colony counting. A magnified illuminator (e.g., Quebec colony counter) may be also be used when counting Petrifilm plates. The uptake of triphenyltetrazolium dye results in red colonies. However, red colonies are not specifically indicative of Enterobacteriaceae, coliforms, or E. coli, but only indicative of microbial metabolism in general. Gas production is another characteristic detected by Petrifilm plate methods since members of the Enterobacteriaceae group typically produce gas from the fermentation of carbohydrates (lactose in the case of coliforms and E. coli; glucose in the case of the family Enterobacteriaceae). These organisms typically produce colonies on Petrifilm plates that have a gas bubble adjacent to the colony or within 1 colony-diameter of the colony, or exhibit a ring of gas bubbles around the colony. The preferred counting range for standard, high-sensitivity, and Series 2000 Petrifilm plates is typically 15 to 150 colonies for coliforms and E. coli and 15 to 100 colonies for Enterobacteriaceae. Colonies on the white foam dam (i.e., outside the gridded well of the Petrifilm) should not be counted. If desired, colonies may be isolated from the Petrifilm gel and subjected to further culturing procedures. Difficulties with Petrifilm interpretation may be encountered when sample particulates are dark and present little contrast to the background medium. Low dilutions of chocolate milk, cocoa powder, and dried herbs have been reported to be problematic. Buffers containing citrate, bisulfite, or thiosulfate should not be used with Petrifilm, otherwise growth inhibition may result. Differential characteristics for colony counting are discussed below.

9.631

Petrifilm Enterobacteriaceae Method (AOAC 2003.01) The Petrifilm Enterobacteriaceae plate is one of the very few rapid methods directed at the enumeration of this family of bacteria. The Enterobacteriaceae Petrifilm method is an alternative to the standard Enterobacteriaceae count 110 |

method using VRBGA. First, ensure that the pH of the diluted sample is between 6.5 and 7.5; inoculate the Petrifilm plates (20 cm2) with 1 mL of sample, as described above (Section 9.63); and finally incubate the plates aerobically at 35uC for 24 ¡ 2 hr. All colonies visible on the plate will be red after incubation. Enterobacteriaceae colonies are associated with one or more gas bubbles (within one colony diameter of the colony) and/or are surrounded by a yellow zone, indicative of acid production.

9.7

COLIFORMS

It should not be assumed that the analysis of a single food sample by several coliform enumeration methods will recover the same types of organisms or yield the same quantitative results. For example, organisms classified as coliforms on solid media (e.g., VRBA or Petrifilm) may be incapable of producing gas when tested by the standard LST/BGB method and therefore may not be counted as coliforms by the MPN procedure.63,83 The proportion of true coliforms among VRBA isolates varies markedly with the food product under analysis.63 This section deals only with methods for the enumeration of coliforms. Methods for the simultaneous enumeration of coliforms and E. coli or of coliforms, fecal coliforms, and E. coli are discussed in Section 9.9.

9.71

Presumptive Test for Coliform MPN

A three-replicate, three-dilution tube MPN procedure is generally used for the analysis of foods. Certain procedures do, however, specify a five-replicate tube MPN format. The MPN for fecal coliforms in shellfish and shellfish meats, for instance, specifies that five replicate tubes be prepared at each dilution. Inoculate three replicate tubes of LST broth per dilution with 1 mL of the previously prepared 1:10, 1:100, and 1:1000 dilutions. Using the current three-tube MPN table (Appendix 2 of the BAM),31,10 this dilution range will cover an MPN range of less than 3.6 MPN/g or MPN/mL to more than 1,100 MPN/g or 1,100 MPN/mL. It is permissible to add 10 mL of the original 1:10 sample dilution to tubes containing 10 mL of double-strength LST (in addition to the aforementioned dilutions) if the food is expected to contain low levels of coliforms.50 This adaptation lowers the limit of detection of the three-tube MPN to less than 0.36 MPN/g or MPN/mL. Tubes are incubated at 35uC ¡ 0.5uC for 24–48 ¡ 2 hr after inoculation. Tubes are examined for evidence of gas production at the end of 24 hr incubation. Gas production is measured either by gas displacement in the inverted vial (e.g., Durham tube) or by effervescence produced when the tube is gently shaken. After recording the results, negative tubes are re-incubated for an additional 24 hr. Tubes are again examined for gas production (see the chapter ‘‘Culture Methods for Enumeration of Microorganisms’’). After reference to the MPN rules and using the appropriate MPN table (see the chapter ‘‘Culture Methods for Enumeration of Microorganisms’’), report results as the presumptive MPN of coliform bacteria per gram or per milliliter. Tubes giving presumptive-positive coliform results are confirmed, as described in Section 9.72.

| Enterobacteriaceae, Coliforms, and Escherichia coli as Quality and Safety Indicators

9.72

Confirmed Test for Coliforms

Gently agitate all LST tubes exhibiting gas production within 48 ¡ 2 hr (see Section 9.71), and then subculture each gassing tube into BGB broth by a 3-mm loop or other appropriate transfer device. Some laboratories utilize pre-sterilized wooden sticks (approximately 120 mm 6 3 mm) as a convenient and inexpensive transfer device. Avoid the pellicle (if present) when transferring. Incubate all BGB tubes at 35uC ¡ 0.5uC for 48 ¡ 2 hr. Examine the tubes for gas production. Gas production in BGB tubes at 35uC ¡ 2uC is considered confirmation of coliform presence. Record the results and refer to the appropriate MPN table (see the chapter ‘‘Culture Methods for Enumeration of Microorganisms’’). Report results as the confirmed MPN of coliform bacteria per gram (or per milliliter ).

9.73

VRBA Method for Coliforms Not Expected to Be Stressed or Damaged

Transfer 1-mL aliquots of each dilution to separate labeled petri dishes. Pour into the plates 10 mL of VRBA (boiled after hydration and not autoclaved, per manufacturer’s recommendations) tempered at less than 48uC and adjusted to a pH of 7.0 to 7.2. (VRBA at pH 6.9 or lower should not be used since these values are indicative of flaws in media preparation and/or storage).42 Swirl plates to mix well and allow to solidify. Overlay each plate with approximately 5 mL VRBA and allow it to solidify. Invert the plates after solidification and incubate 18 to 24 hr at 35uC. Incubate the plates at 32uC for dairy products.21 Examine the plates with illumination under a magnifying lens. Count the purplered colonies that are 0.5 mm or larger in diameter and are surrounded by a zone of precipitated bile acids. A plate should ideally have 15 to 150 colonies. Coliform colonies on more crowded plates may remain small and fail to reach 0.5 mm in diameter. Multiply the number of presumptivecoliform colonies by the reciprocal of the dilution used and report the results as it presumptive VRBA count (CFU/g or CFU/mL). The coliform count obtained on VRBA can be confirmed by selecting representative colonies and testing them for gas production in BGB (see Section 9.72). Colonies producing gas from lactose in BGB are confirmed as coliform organisms. Determine the confirmed number of coliforms per gram (or milliliter) by multiplying the percentage of BGB tubes confirmed as positive by the presumptive VRBA count. Report as ‘‘estimated’’ counts that are derived from plates outside the range of 15 to 150 colonies per plate. A modification of this method is the VRBA/MUG method for E. coli and coliforms, which is outlined in Section 9.933.

9.74

VRBA-Overlay Method for Stressed or Damaged Coliforms

Sublethally damaged or stressed coliforms may be unable to grow and form typical colonies on selective agars such as VRBA, as noted previously in section 9.381. This limitation is commonly overcome by plating the sample in a nonselective agar (e.g., tryptic soy agar [TSA]), allowing several hours of resuscitation at room temperature, and then overlaying the plate with VRBA.88 BAM45 describes an overlay of single-strength VRBA, whereas Standard Methods for the

Examination of Dairy Products21 specifies overlaying TSA with double-strength VRBA (VRBA-2). Studies have indicated that the VRBA-2 method consistently yields higher numbers of coliforms than nonresuscitative methods (e.g., VRBA or Petrifilm) when analyzing foods that are expected to contain sublethally damaged or stressed cells. Differences in coliform levels obtained with the VRBA-2 method versus nonresuscitative methods may exceed 1 log10 CFU/g.83 Transfer 1-mL aliquots of each dilution to separate labeled petri dishes. Pour approximately 10 mL of TSA tempered to 48uC into the plates. Swirl the plates to mix well, and allow them to solidify. Allow TSA plates to incubate at room temperature for 2.0 ¡ 0.5 hr. Overlay the plates with 8 to 10 mL of melted, cooled VRBA or VRBA-2 and allow them to solidify. Invert the plates after solidification and incubate 18 to 24 hr at 32uC (for dairy products) or at 35uC (for other food products). Count the colonies, as described previously in Section 9.73.

9.75

Petrifilm Methods for Coliforms

9.751

Petrifilm Coliform Count Plate [AOAC 991.14 (food), 986.33 (milk), 989.10 (other dairy)] This coliform method is an alternative to the coliform plate count method using VRBA. Petrifilm plates (20 cm2) are inoculated with 1 mL of sample, as described previously (Section 9.63), and then incubated aerobically at 32uC (for dairy samples) or at 35uC (for other food samples) for 24 ¡ 2 hr. After incubation, all colonies visible on the plate will be red. In this method, coliform colonies are associated with one or more gas bubbles. No additional confirmation is necessary because gas production from lactose fermentation by bile salt-resistant colonies is a characteristic of coliforms and is assumed to result from the fermentation of lactose in the medium (as opposed to fermentation of other carbohydrates in the food matrix). 9.752

Petrifilm High-Sensitivity Coliform Count Plate (AOAC 996.02 [dairy]) The high-sensitivity Petrifilm Coliform Plate was developed to analyze sample volumes of 5 mL, rather than the standard 1-mL sample analyzed with the Petrifilm Coliform Count Plate. Analysis of larger sample volumes on 60 cm2 Petrifilm plates improves sensitivity when recovering low levels of coliforms. Analysis of samples with high-sensitivity coliform plates is identical to the procedure described in Section 9.751, except the pH of the diluted sample must be adjusted (if necessary) to 6.5 to 7.5 and a 5-mL sample is used. Highsensitivity coliform count plates should be incubated at 32uC (for dairy products) or at 35uC (for other food products). The AOAC Official Method is for dairy products only. However, testing other foods with a plate incubation temperature of 35uC has been verified by other bodies (e.g., AFNOR 3M-1/ 7-3/99). 9.753

Petrifilm Series 2000 Rapid Coliform Count Plate (AOAC 2000.15) The Petrifilm Series 2000 Rapid Coliform Count Plate incorporates the usual features of the Petrifilm Coliform Count Plate plus a pH indicator that permits more rapid identification of presumptive coliform colonies. Analysis | 111

Compendium of Methods for the Microbiological Examination of Foods |

of samples with the Series 2000 Rapid Coliform Count plate is identical to the procedure described in Section 9.751, except the diluted sample must be adjusted (if necessary) to a pH of 6.5 to 7.5 and the counting procedure is different. Plates may be examined for yellow zones (indicative of acid production and presumptive coliform growth) as early as 6 to 14 hr after incubation. The final coliform count is usually read after 24 hr of incubation and is performed in a manner identical to that used for other Petrifilm coliform plates.

loop or other appropriate transfer device. Incubate EC tubes for 24 ¡ 2 hr at 44.5uC ¡ 0.2uC for water and shellfish or at 45.5uC ¡ 0.2uC for foods, preferably in a circulating water bath. Examine the tubes for gas, which indicates a positive result. Report results as fecal coliform MPN per gram or MPN per milliliter, after referring to an appropriate MPN table such as the table in BAM Appendix 2.10 Fecal coliforms are organisms giving positive results by this procedure.

9.82 9.76

Pectin Gel Method (Redigel Violet Red Bile Test) (AOAC 989.11)

Instead of using the standard VRB agar-based medium for preparing pour plates, the pectin gel method uses an agarfree medium that gels when poured into specially coated petri plates. The sample homogenate (or dilutions thereof) is typically added to a single-use bottle of Redigel VRB medium. The bottle is mixed gently by inversion, and the inoculated medium is added to a Redigel plate. The AOAC procedure calls for adding the Redigel medium to a Redigel plate, and then adding the inoculum to the medium in the plate. The plates are swirled to mix the medium and sample and then left at room temperature to solidify. Within 15 min of pouring the plate, the inoculated medium in the plate should be overlaid with sterile Redigel medium. The plates are allowed a solidification period of approximately 40 min, and then incubated aerobically for 24 ¡ 2 hr at 32uC (for dairy samples) or at 35uC (for other food samples). After incubation, all pink or red colonies are counted to obtain the coliform count. The pectin gel method (e.g., with Redigel violet red bile test) is an AOAC Official Method56 for the analysis of coliforms in dairy products (Method 989.11).

9.77

Hydrophobic Grid Membrane Filter Method for Coliforms (AOAC 983.25)

The hydrophobic grid membrane filter (HGMF) method involves filtering a diluted food sample through a membrane filter, placing the filter on a plate of selective/ differential agar, incubating the plate, and then counting colonies possessing certain color characteristics. To obtain a coliform count, the inoculated membrane filter is placed right-side up on the surface of a predried plate of m-FC agar (without rosolic acid). The plate is then incubated at 35uC for 24 ¡ 2 hr. Grid cells containing one or more colonies of any shade of blue are counted. This value is then plugged into a standard formula to calculate the MPN. The MPN is multiplied by the reciprocal of the dilution used. This number is reported as the MPN coliforms per gram or per milliliter of food. Additional confirmation is not required. The HGMF4 is an AOAC Official Method56 for the analysis of coliforms in foods.

9.8 9.81

FECAL COLIFORM GROUP

The official FDA procedure for the bacteriological analysis of domestic and imported shellfish is fully described elsewhere.1 The BAM method describes an approach for examining shellfish, fresh-shucked frozen shellfish, and shellfish frozen on the half-shell.31 This procedure does not apply to the examination of crustaceans (e.g., crabs, lobsters, shrimp) or to processed shellfish meats such as breaded, shucked, precooked, and heat-processed products. A 200-g quantity of shellfish liquor and meat is obtained from 10 to 12 shellfish and blended for 2 min with 200 mL presterilized phosphate buffered dilution water. A 5-tube MPN procedure is performed wherein 2 mL of blended homogenate (1 g) is added to each of 5 tubes containing 10 mL lactose broth or LST broth. This is repeated with a 1:10, 1:100, and 1:1000 dilution of the homogenate (5 tubes each). The tubes are incubated at 35uC and the assays are continued, as described previously. The coliform density is expressed as MPN per 100 g of sample. Gas production at 44.5uC in the EC broth is considered confirmatory for fecal coliforms. The MUG assay can also be used to confirm the presence of E. coli. However, some precautions must be taken since some shellfish contain GUD, which can interfere with the result. The MUG reagent consequently should not be added LST tubes to which an oyster homogenate is added since interference can occur. The MUG reagent is instead added to the EC broth in the confirmatory step of the assay for E. coli. Appropriate controls should include three EC-MUG medium tubes: one tube is inoculated with E. coli, one tube is inoculated with K. pneumonia, and one tube remains uninoculated.

9.9 9.91

ESCHERICHIA COLI Completed MPN Test for E. coli

The completed test for E. coli begins with LST cultures generated during the presumptive test for coliforms (Section 9.71). Subculture all positive LST tubes into EC broth and incubate the EC tubes in a circulating water bath at 45.5uC ¡ 0.2uC. All EC tubes that show gas within 48 ¡ 2 hr should be subcultured by streaking on L-EMB agar plates and incubating the plates aerobically for 18 to 24 hr at 35uC. Examine the plates for the typical nucleated, darkcentered colonies with or without a metallic sheen, which are indicative of E. coli.

E. coli Broth Method for Fecal Coliform MPN

The first stage of the E. coli (EC) broth MPN method is the presumptive coliform MPN test, as noted previously. Subculture all LST tubes exhibiting gas within 48 ¡ 2 hr (Section 9.71) to E. coli (EC) broth by using a standard 3-mm 112 |

Procedure for Shellfish and Shellfish Meats

9.92

Confirmation of E. coli (Including IMViC Tests)

If typical E. coli–like colonies are present on L-EMB plates, select two colonies from each L-EMB plate by touching an inoculating needle to the top center of the colony and

| Enterobacteriaceae, Coliforms, and Escherichia coli as Quality and Safety Indicators

transferring each isolate to a plate count agar (PCA) slant. If there are no typical colonies on the L-EMB plates, pick two or more atypical colonies and transfer them to PCA slants. Incubate the slants at 35uC for 18 to 24 hr. Transfer the growth from PCA slants into the following broths for identification by biochemical tests:

N N

N

N N

Tryptone broth: Incubate 24 ¡ 2 hr at 35uC and test for indole by using Kovac’s indole reagent. Methyl red-Voges Proskauer (MR-VP) medium: Incubate 48 ¡ 2 hr at 35uC. Remove 1 mL into a small glass test tube and test for acetylmethycarbinol by adding VogesProskauer reagents (a-napthol solution, 40% potassium hydroxide [KOH] solution). Incubate the remainder of the MR-VP culture an additional 48 hr. Test for methyl red reaction by adding methyl red indicator. Koser’s citrate broth: Incubate 96 hr at 35uC and record the growth (as evidenced by turbidity). As an alternative, Simmon’s citrate slants (a modification of Koser’s formulation) may be used. The advantage of Simmon’s slants is that citrate utilization is signaled by a green to bright blue color change, rather than by turbidity.85 Incubate Simmon’s slants at 35uC for 96 hr. Lauryl sulfate tryptose (LST) broth: Incubate 48 ¡ 2 hr at 35uC. Examine the tubes for gas formation from lactose. Gram stain: Perform Gram staining on a smear prepared from an 18-hr to 24-hr PCA slant. Coliforms are nonsporeforming bacilli that stain red (i.e., Gram-negative). Gram-positive organisms stain purple.

Compute the MPN of E. coli per gram (or per milliliter). E. coli are Gram-negative, non-sporeforming rods that produce gas in lactose and produce IMViC patterns of + + – – (biotype I) or – + – – (biotype II). Note that this procedure will not enumerate anaerogenic strains. If desired, a miniaturized biochemical identification system may be used to confirm the identity of isolates suspected of being E. coli. As an alternative, the 48-hr quad plate approach mentioned previously could also be employed for IMViC testing. The media are struck from colonies directly isolated from L-EMB agar.77 Other biochemical confirmation assays are discussed in Section 9.37.

9.93

b-Glucuronidase-Based Tests for E. coli

Presumptive E. coli colonies may be identified by their production of b-glucuronidase (GUD). Testing suspect colonies on L-EMB plates for GUD production has been proposed as an alternative to the IMViC tests.47 bGlucuronidase is commonly produced by E. coli and has been utilized as a differential characteristic in coliform recovery media containing various b-D-glucuronic acid substrates. For example, 4-methylumbelliferyl-b-D-glucuronic (MUG) acid is a fluorogenic substrate. A fluorescent product, 4-methylumbelliferone, is generated when nonfluorescent MUG is cleaved by GUD. The 4-methylumbelliferone exhibits a bluish fluorescence when exposed to longwave (365 nm) ultraviolet light in a darkened room. Fifty milligrams of MUG per 1 mL of broth is usually used for liquid media, whereas agar media is supplemented at 100-mg MUG per 1 mL agar. Chromogenic GUD substrates

such as 5-bromo-4-chloro-3-indolyl-b-D-glucuronide (BCIG, sometimes called X-GLUC) may also be incorporated into coliform-selective agars. Enzymatic cleavage of BCIG gives a dark blue color to E. coli colonies on the agar plate. Levels of 50- to 125-mg BCIG per liter of agar reportedly give optimal differentiation of E. coli colonies; the blue chromophore does not diffuse into the agar-like 4methylumbelliferone.35,74 Reports indicate that 92% to 99% of E. coli isolates, including many anaerogenic strains, produce GUD.20,32,40,41 However, complete reliance on GUD production to indicate E. coli is not recommended. Some pathogenic serotypes of E. coli (principally the Enterohemorrhagic E. coli O157:H7 strains) do not produce GUD22 and GUD production has been observed in various non-E. coli organisms. Enterobacteriaceae that are GUD-producing include strains of Shigella (especially S. sonnei), Salmonella (including S. enterica subsp. Indica), Escherichia vulneris, Citrobacter, Enterobacter, Proteus (including P. mirabilis), Klebsiella (including K. ozaenae), Serratia, and Yersinia enterocolitica.46,47 Non-Enterobacteriaceae known to produce GUD include Flavobacterium spp., Pseudomonas spp., Clostridium spp., Micrococcus, and Staphylococcus spp.46 Gram-positive organisms that produce GUD is especially problematic when minimally selective media (e.g., Peptone Turgitol Glucuronide agar) are incubated for 48 hr or longer.20

9.931 LST-MUG MPN for E. coli and Coliforms The procedure for the simultaneous presumptive-MPN determination of coliforms and E. coli is the same as that outlined in Section 9.71, with two exceptions.65 First, the LST broth contains 50 mg MUG per 1 mL of broth. Second, incubation is usually terminated after 24 ¡ 2 hr, which will identify 83% to 95% of E. coli–positive tubes, depending on the product. Incubation for 48 hr will identify 96% to 100% of E. coli–positive tubes.65 The LST-MUG tubes are examined for fluorescence under longwave (365 nm) ultraviolet light in a darkened area. A 6-watt, hand-held ultraviolet (UV) lamp is satisfactory for this purpose. More powerful UV sources (e.g., a 15-watt fluorescent tube type of lamp) may be used, but the user should be aware of the potential for false-positive fluorescence31 and the need for protective safety gear (e.g., protective glasses or goggles and gloves). Fluorescence results are used to obtain an MPN value by consulting a standard MPN table. Fluorescent-positive tubes are then streaked onto L-EMB plates, which are incubated at 35uC for 24 ¡ 2 hr. Confirmation of L-EMB isolates with E. coli–like morphology is performed by using IMViC tests (Section 9.92) or another confirmation method. The LST/MUG assay is an AOAC Official Method56 for the analysis of E. coli in chilled or frozen foods (AOAC Method 988.19). When analyzing foods known to have endogenous GUD activity (e.g., shellfish or fin fish), all growth-positive LST tubes should be transferred to EC broth containing 50-mg MUG per 1-mL broth. The EC/MUG tubes exhibiting fluorescence should then be struck to L-EMB and the presumptive colonies confirmed, as described immediately above. | 113

Compendium of Methods for the Microbiological Examination of Foods |

9.932

The ColiComplete MPN Method for E. coli and Coliforms (AOAC 992.30) The ColiComplete method combines the principles used in the LST-MUG test with an enzymatic assay for coliforms. ColiComplete uses a substrate supporting disc containing MUG and 5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside (a colorimetric indicator for b-galactosidase activity, which is a trait common to most coliforms). Discs are added to inoculated tubes of LST, and then the LST tubes are incubated at 35uC, as usual (Section 9.71). After 24 hr and after 48 hr, the tubes are examined for an insoluble blue precipitate, indicative of the presence of coliforms. After 30 ¡ 2 hr of incubation, the tubes are examined under longwave UV light for fluorescence resulting from MUG hydrolysis. Fluorescence indicates the presence of E. coli. The manufacturers of ColiComplete recommend that a known E. coli sample be used as a positive control and two samples (e.g., uninoculated media and media inoculated with a non-E. coli coliform such as Klebsiella or Enterobacter) be used as negative controls. The ColiComplete method yields confirmed results for E. coli and coliforms; no additional confirmation is necessary. The ColiComplete assay is an AOAC Official Method56 for the analysis of E. coli and coliforms in all food products. The ColiComplete system can also be used for citrus juices. The method should be run in duplicate. A 10-mL portion of juice is added to 90 mL of universal preenrichment broth (UPEB) and incubated for 24 hr at 35uC. After enrichment, transfer 1 mL from each UPEB to 9 mL of EC broth containing a ColiCcomplete disc. Incubate at 44.5uC for 24 hr. Include a tube with MUG (+) E. coli as a positive control and K. pneumoniae as a negative control. Examine tubes in the dark under longwave UV light. The presence of blue fluorescence is indicative of E. coli.31 9.933 VRBA/MUG Method for E. coli and Coliforms Depending on the nature of the sample and the expected physiological state of the target organisms, one of several VRBA/MUG plating methods may be used to determine the presence of and/or distinguish generic E. coli from other coliforms.21,31 Pour-plating typically uses 1.0-mL portions of sample dilutions plated in standard-sized 10-cm diameter/1.5-cm depth petri plates. For unprocessed foods expected to contain healthy cells, the VRBA/MUG method follows the VRBA method outlined in Section 9.73, except that MUG is added to the VRBA at a level of 100-mg MUG per 1-L agar. Commercially available agar plates or dehydrated VRBA media already containing MUG may be purchased. Colonies are coliform if they grow to a 0.5-mm diameter at 32uC, whereas E. coli are those colonies that fluoresce under longwave (365 nm) UV light. For processed foods or environmental samples that may contain sublethally damaged or stressed E. coli or coliforms, a procedure similar to that described in Section 9.74 is used (which includes a nonselective agar base [e.g., TSA]), except MUG is added at a level of 100-mg MUG per 1 L VRBA. The procedure for preparing the basal TSA plate is not well standardized and several alternative procedures exist:

N

Pour-plate 1 mL of the diluted sample with TSA, and then proceed with resuscitation.

114 |

N N N

Spread-plate 0.1 mL of the diluted sample onto a TSA plate, and then proceed with resuscitation. Pour-plate 1 mL of the diluted sample with TSA containing MUG at a level of 100-mg MUG per 1 L agar, and then proceed with resuscitation. Spread-plate 0.1 mL of the diluted sample onto a TSA plate containing MUG at a level of 100-mg MUG per 1 L agar, and then proceed with resuscitation.

After incubation, the coliform colonies are counted as usual (see Section 9.73). The plates are then exposed to longwave UV light and fluorescent colonies, indicative of E. coli, are counted. The coliform and E. coli counts are then multiplied by the reciprocal of the dilution used and results reported. Confirmation is typically not performed when using the VRBA/MUG method.

9.934

HGMF With MUG Method for E. coli and Coliforms The HGMF-MUG method involves filtering a diluted food sample through a membrane filter, and then placing the filter on a plate of selective/differential agar, incubating the plate, and counting colonies possessing certain characteristics (e.g., color or fluorescence) (see Section 9.93). To obtain a coliform count, the filter is placed on a predried plate of lactose monensin glucuronate (LMG) agar, and then incubated at 35uC for 24 ¡ 2 hr. Squares containing one or more blue colonies are counted. The number of positive (i.e., blue) squares is converted to a MPN by using a mathematical formula. The coliform MPN per gram or MPN per milliliter is calculated by multiplying the MPN by the inverse of the dilution factor of the filtered sample homogenate. If coliforms are present, an E. coli count can be obtained on the same filter by transferring it to the surface of a predried buffered MUG agar (BMA) plate. The filter is incubated at 35uC for 2 hr, and then examined under longwave UV light in a darkened room. Squares containing one or more large blue-white fluorescent colonies are positive. The number of positive squares is then converted to a MPN by using a mathematical formula and the E. coli MPN per gram or MPN per milliliter is calculated by multiplying the MPN by the inverse of the dilution factor. The HGMF-MUG method yields confirmed results for E. coli and coliforms. No additional confirmation is necessary. The HGMF/MUG ISO-GRID/Neo-Grid method is an AOAC Official Method56 for the analysis of E. coli and coliforms in foods (AOAC Method 990.11). 9.935 Petrifilm for E. coli and Coliforms The Petrifilm E. coli count plate allows simultaneous enumeration of coliforms and E. coli on a single Petrifilm. Differentiation of E. coli is accomplished by the addition of BCIG to a dehydrated film similar to the film used in the coliform count plate (Section 9.751). The plates are inoculated as described previously (Section 9.63), and then incubated aerobically at 32uC (for dairy samples) or at 35uC (for other food samples) for 24 ¡ 2 hr. Non-E. coli coliform colonies (i.e., red colonies associated with one or more gas bubbles within one colony diameter of the colony) and E. coli colonies (i.e., blue colonies associated with one or more gas bubbles within one colony diameter of the colony) are

| Enterobacteriaceae, Coliforms, and Escherichia coli as Quality and Safety Indicators

counted after 24 hr incubation in accordance with the Interpretation Guide provided with the kit. The total number of coliforms equals the number of red gasproducing colonies plus the number of blue gas-producing colonies. To recover slow GUD producers, Petrifilm E. coli count plates are re-incubated for an additional 24 ¡ 2 hr after counting. The E. coli colonies are again counted. The number of E. coli equals the number of blue gas-producing colonies. The Petrifilm E. coli Count Plate yields confirmed results for E. coli and coliforms. No additional confirmation is necessary. The Petrifilm E. coli Count Plate is an AOAC Official Method56 (AOAC Method 991.14) for the analysis of E. coli and coliforms in foods. The AOAC procedure specifies that only blue colonies with gas are counted as E. coli; however, this practice may result in the omission of anaerogenic E. coli. Some researchers9,62 have recommended that blue, non–gas-producing colonies be confirmed to determine whether they represent E. coli.

9.936

The Redigel ColiChrome 2 Test for E. coli and Coliforms The Redigel ColiChrome method follows the same procedure as the Redigel violet red bile test (Section 9.76), except a different gel formulation is used. Like the Petrifilm E. coli count plate, the ColiChrome medium incorporates indicator systems for b-galactosidase (to detect coliforms) and for b-glucuronidase (to detect E. coli). Redigel ColiChrome plates are incubated at 32uC (for dairy products) or at 35uC (for other foods) for 24 to 48 hr. Coliform colonies are pink to red, E. coli colonies are purple, and noncoliform colonies are white or cream. The countable range for the Redigel ColiChrome plate is 15 to 150 CFU per plate. No additional confirmation is necessary. 9.937 The IDEXX SimPlate Coliforms / E. coli test The general principle behind the IDEXX SimPlate Coliforms/E. coli test is very similar to the principle outlined in Section 9.76 and 9.936 in that the sample is added to a special selective or differential medium, and then plated onto a special plate. The medium enters the wells on the dimpled SimPlate or Bio-Plate. The remainder of the sample-containing medium is then poured off the plate and discarded. The plates are incubated at 35uC or 37uC for 24 hr. The number of colored wells is counted and this number is entered into a special MPN table. The MPN value is multiplied by the reciprocal of the dilution used to obtain the MPN coliforms per gram or per milliliter. The plate is next examined under longwave ultraviolet light and fluorescent wells are counted. The number of fluorescent wells is entered into the special MPN table and this value is multiplied by the reciprocal of the dilution factor to determine the MPN E. coli per gram or per milliliter. No confirmation is necessary for the IDEXX SimPlate Coliforms/E. coli test. 9.94

Indole-based Methods for E. coli and Coliforms

Since 1948, indole production at 44uC has been used for the detection and identification of E. coli.59 Type I (i.e., indolepositive) E. coli strains are regarded as typical E. coli.

Approximately 95% of E. coli strains recovered from foods are indole-positive.2 Bacteria other than E. coli that are indole-positive include Klebsiella spp. (pneumoniae and oxytoca), Citrobacter diversus, and Providencia spp. Indolepositive bacteria other than E. coli may comprise 3% to 5% of the indole-positive isolates on foods.80 Indole-based detection methods, such as MUG-based assays, offer a means of detecting anaerogenic strains of E. coli that may be missed by tests that are predicated on the production of gas from lactose. Combining indole and GUD assays improves the specificity of rapid E. coli determinations.

9.941

Anderson Baird-Parker Procedure for Presumptive Biotype / E. coli The Anderson Baird-Parker procedure is a rapid screening method for detecting E. coli biotype I and anaerogenic strains of E. coli. It involves inoculating a membrane flattened to a dried surface of a tryptone bile agar (TBA) plate with a diluted food sample. Indole production is subsequently used to identify E. coli colonies. Portions (0.5 to 1.0 mL) of a diluted food sample are inoculated onto a membrane placed and flattened on a plate of dried TBA. Plates are incubated (right-side up) overnight at 44uC ¡ 1uC. The membrane is removed from the surface of the TBA plate after incubation, and placed in an empty petri dish containing 1 to 2 mL of Vracko-Sherris indole reagent for 5 min. Pink-stained indole-positive colonies, which are representative of E. coli, are then counted. The stained filter may be fixed by drying in sunlight or under a UV lamp. E. coli CFU per gram or CFU per milliliter is then calculated by multiplying the number of indole-positive colonies by the reciprocal of the dilution used. Confirmation is not possible with this approach once the indole reagent is added since the indole reagent kills the E. coli during the staining process. If confirmation testing is desired, transfer the colonies to an appropriate medium before performing the indole test.2 A modified Anderson Baird-Parker method was developed to allow the resuscitation of injured cells. The modified method involves the preparation of the same membrane, which is placed first on TSA and incubated at 35uC to 37uC for 4 hr. The membrane is then transferred to TBA and incubated at 44.5uC ¡ 0.5uC until the entire incubation period is 24 hr. The membrane is then placed on filter paper saturated with indole reagent (0.5 g 4-dimethylaminobenzaldehyde in 100 mL 1 mol/l HCl). The membrane and filter are placed under a UV lamp (250 to 400 nm) for 10 to 15 min. E. coli CFU/g or CFU/mL is then calculated by multiplying the number of indolepositive colonies by the reciprocal of the dilution used.80 9.942

HGMF Method for E. coli and/or Fecal Coliforms The HGMF method is similar to the modified Anderson Baird-Parker procedure, except it uses filtration and an oxidant-accelerated indole reagent. The same filter can be used to obtain an E. coli count and a fecal coliform count. A diluted food sample is first filtered through a membrane filter, as with other HGMF procedures. The inoculated filter is then placed on the surface of a predried plate of either nonselective agar or a selective/differential agar, | 115

Compendium of Methods for the Microbiological Examination of Foods |

depending on the nature of the sample and the expected physiological state of the target organisms. A predried plate of TSA with magnesium sulfate (TSAM) is used if the food or environmental sample is expected to contain sublethally damaged or stressed organisms. The TSAM plate is then inverted and incubated at 25uC for 4 to 5 hr (if analyzing dry foods) or at 35uC for 4 to 5 hr (for all other foods) to allow resuscitation. The filter is transferred to the surface of a predried plate of TBA agar after filtration or resuscitation. The plate is then inverted and incubated at 44.5uC ¡ 0.5uC for 18 to 24 hr in an incubator with good temperature control. The number of grid squares containing one or more colonies of any shade of blue is counted after incubation. This count is used to calculate the MPN of fecal coliforms with a standard formula. The MPN value is multiplied by the reciprocal of the dilution used and the MPN fecal coliforms per gram or per milliliter of food is reported. Additional confirmation is not required. If an E. coli count is desired, the filter can then be stained with an oxidant-accelerated indole reagent to detect indole-positive colonies. The grids should be stained at room temperature for 10 to 15 min, and afterwards the HGMF grids containing one or more pink-red (indole-positive) colonies are counted as E. coli. This count is used to calculate the E. coli MPN with a standard formula. The MPN value is then multiplied by the reciprocal of the dilution used and the MPN E. coli per gram or per milliliter of food reported. Additional confirmation is not required. The HGMF4 is an AOAC Official Method56 for the analysis of E. coli and fecal coliforms in foods (Method 983.25).

9.95

Chromogenic and Fluorogenic Determination of Coliforms and E. coli

Another alternative for differentiating coliforms and E. coli is by fluorogenic-containing and chromogenic-containing media. These media allow the user to identify colonies as E. coli, coliforms, or other organisms. There are generally four groups of these compounds. They include fluorogenic dyes, pH-fluorescent indicators, Eh indicators, and fluorogenic and chromogenic enzyme substrates.61 These media may be used for direct plating for presumptive detection.31 These compounds may also be added to tubes for a modified MPN method.37 Examples of these media are RAPID’E. coli 2 Agar by Bio-Rad (Hercules, CA ) (AOAC RI #050601) and Chromocult Coliform Agar by Merck KGaA (Darmstadt, FRG) (AOAC RI #020902).

9.10

INTERPRETATION OF DATA: THE VALUE OF ENTEROBACTERIACEAE, COLIFORMS, AND E. COLI AS INDICATORS OF QUALITY AND INDEXES OF PATHOGENS

The enumeration of Enterobacteriaceae, coliforms (including fecal coliforms), and E. coli in foods is far from an exact science, as has already been discussed. Considerable effort has been expended over the past 100 years to improve the specificity of assays for Enterobacteriaceae, coliforms, and E. coli, but these efforts have achieved relatively little. Most food manufacturers seem willing to sacrifice some specificity in exchange for more rapid results. Results obtained 116 |

from assays for Enterobacteriaceae, coliforms, and E. coli (including those obtained by direct count methods) are best regarded as estimates and data interpretation must take this limitation into account. Variance in assays will also differ from laboratory to laboratory, technician to technician and method to method. An example of variance can readily be seen when looking at the published confidence intervals associated with MPN data (e.g., where many values differ by more than 1 log10 CFU).10

9.101

Use of Enterobacteriaceae, Coliform, or E. coli Counts in Microbiological Criteria

Regulatory agencies, for the most part, have recognized the futility of assuming that index counts reflect the safety of a particular food product. This is particularly true with processed (as opposed to raw) food products. The assumption that index counts reflect the presence or absence of a particular pathogen in a processed fod (as opposed to raw commodity) is not supported by the scientific literature (see Section 9.1). As a result, regulatory criteria for index or indicator organisms, when issued, typically take the form of guidelines (which are relatively lenient), rather than standards (which are more restrictive in that they are legally enforceable). The presence of high levels of coliforms, Enterobacteriaceae, or E. coli in foods processed for safety may indicate one of the following possibilities: (1) inadequate processing and/or postprocessing contamination and/or (2) microbial growth. The presence of high levels of any of these organisms is not de facto evidence that fecal contamination has occurred.48 E. coli has historically been considered an obligate enteric bacterium, requiring periodic passage through the gastrointestinal tract of animals for multiplication and dissemination. The presence of fecal coliforms, including generic biotype I E. coli, has furthermore been used to assess the level of fecal or sewage contamination in the water of food products. Recent studies, however, suggest that the highly phenotypically heterogeneous bacterial species contain strains that are known to have other ecologic reservoirs; this overturns the idea that E. coli that have been isolated must be of fecal origin.16,17,51,58 In raw foods, the presence of members of the Enterobacteriaceae group is to be expected since these organisms are widespread in the natural environment. End-product specifications describing acceptable levels of indicator organisms in various food products and food ingredients are commonly issued by purchasers, often to the consternation of the would-be supplier. Specifications are, too often, scientifically unsupportable and arbitrary. There is a common tendency among food manufacturers to assume that a specification is universal and can be applied to a variety of foods, despite their diverse origins, composition, physical/chemical characteristics, processing, and storage. To be effective, a microbiological specification must be (1) food-specific or ingredient-specific; (2) supported by organoleptic and/or microbiological data relating the specification to the quality and/or safety of the product in question; and (3) capable of being met when the particular product/ingredient is produced under optimal hygienic conditions. An end-product specification should also detail the sampling procedure and the microbiological

| Enterobacteriaceae, Coliforms, and Escherichia coli as Quality and Safety Indicators

method used to obtain the results. Analysis of a single food sample by two different microbiological assays, as discussed earlier in this chapter, often yields different numerical results that may reflect different sampled microbial populations. Confirmation tests such as the IMViC tests for E. coli may also be unreliable (see Section 9.37). The point at which a food is sampled in the production and distribution process has a tremendous impact on the recovery of coliforms, Enterobacteriaceae, or E. coli from that food. Levels of these organisms on raw foods (e.g., vegetables or raw meats) are not indicative of product quality unless processing procedures (e.g., washing) have been applied to reduce the levels of naturally occurring microorganisms. Analytical results obtained from products in refrigerated distribution may not be reflective of production conditions since some coliforms and Enterobacteriaceae are psychrotrophic55 and can therefore grow during refrigerated storage. Coliforms and E. coli levels may similarly decrease over time in dry foods or in cultured dairy products such as yogurt, buttermilk, and sour cream.

9.102

The Enterobacteriaceae

Despite their name, members of the family Enterobacteriaceae are not confined to the intestinal tract and may be isolated from a variety of nonintestinal sources. 18 Enterobacteriaceae are relatively heat-sensitive and easily killed during cooking or milk pasteurization. The resistance to freezing and other processing procedures varies among members of this family. Enterobacteriaceae are good indicators of environmental hygiene because they are readily inactivated by sanitizers and are capable of colonizing a variety of niches in the processing plant when sanitation is inadequate.18,36

9.103

The Coliform Group

Recovery of the coliform group from foods has less interpretive impact than the single index organism, E. coli, or the fecal coliform group because the coliform group may contain nonenteric members such as Serratia and Aeromonas. The specificity of the coliform group as an index of fecal contamination is diminished by the anonymity of its individual members and the diversity of this group.67 It is possible to use coliforms as indicators of inadequate sanitation on preoperational equipment contact surfaces since these organisms are not resistant to sanitizers. The presence of coliforms on ready-to-eat heatprocessed foods can serve as an indication of either inadequate heat-processing or postpasteurization contamination. The inability of many coliforms to survive freezing makes them of questionable use when analyzing frozen foods.98 Coliform counts can differ significantly, depending on the food tested, the medium used, and other testing conditions. Various conditions in food processing establishments (such as drying, acidification, heating, sanitation) may cause cell injury. Assays must be capable of recovering injured cells. Specification of the medium and the temperature used to obtain coliform counts is critical to the interpretation of data.

9.104

The Fecal Coliform Group

The term ‘‘fecal coliform’’ is a misnomer since organisms enumerated by a fecal coliform assay may or may not have originated in the intestinal tract. For example, organisms such as Klebsiella spp., Enterobacter spp., and Citrobacter freundii may grow outside the intestinal tract.18,53,89 The presence of these organisms within the fecal coliform group compromises the group’s specificity and represents a deficiency in methodology or nomenclature (see Section 9.23). Since the proportion of E. coli within the fecal coliform population varies between samples, there is little reason to stop at the fecal coliform test when E. coli is really the object of interest.

9.105

Escherichia coli

E. coli is regarded as the most valid indicator of fecal contamination of raw foods. This is not to say, however, that E. coli is a good indicator of fecal contamination of processed foods. Contemporary data indicate that E. coli can grow in a variety of extraintestinal niches, including in the processing plant environment and other ecological niches.16,17,18,51,58 Recovery of E. coli from heat-processed foods is indicative of either inadequate processing or subsequent contamination. Differences in sensitivity to various food processing technologies (e.g., drying) between E. coli and enteric pathogens also limits the usefulness of E. coli as an index organism.

ACKNOWLEDGMENT Fourth edition authors: Jeffery L. Kornacki and Jennifer L. Johnson.

REFERENCES 1. American Public Health Association. 1985. Laboratory Procedures for the Examination of Seawater and Shellfish, 5th ed. American Public Health Association, Washington, D.C. 2. Anderson, J. M., and A. C. Baird-Parker. 1975. A rapid and direct plate method for enumerating Escherichia coli biotype I in food. J. Appl. Bacteriol. 39:111-117. 3. Andrews, W. H., and T. S. Hammack. 2003. Food sampling/preparation of sample homogenate. Bacteriological Analytical Manual. Available at: http://www.fda.gov/Food/ FoodScienceResearch/LaboratoryMethods/ucm063335.htm. Accessed April 16, 2015. 4. AOAC International. 1985. Total coliform, fecal coliform and Escherichia coli in foods: hydrophobic grid membrane filter method. J. Assoc. Off. Anal. Chem. 68:404. 5. Bagley, S. T., and R. J. Seidler. 1977. Significance of fecal coliform-positive Klebsiella. Appl. Environ. Microbiol. 33:1141-1148. 6. Bansal, S., N. Malik, M. Ghosh, and A. Ganguli. 2005. Recovery of bacterial pathogens from Indian green chutneys by the thin agar layer method. J. Food Sci. Technol. 42:495497. 7. Bardsley, D. A. 1934. The distribution and sanitary significance of B. coli, B. lactis aerogenes, and intermediate types of coliform bacilli in water, soil, faeces, and ice-cream. J. Hyg. 34:38-68. 8. Blachstein J. 1893. Contribution a l’e´tude microbique de l’eau. Ann. Inst. Pasteur. 10:689-692. 9. Bloch, N., H. Sidjabat-Tambunan, T. Tratt, K. Lea, and A. J. Frost. 1996. The enumeration of coliforms and E. coli

| 117

Compendium of Methods for the Microbiological Examination of Foods |

10.

11. 12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23. 24.

25.

26.

27.

118 |

on naturally contaminated beef: a comparison of the Petrifilm method with the Australian standard. Meat Sci. 43:187-193. Blodgett, R. 2010. Most probable number from serial dilutions. Bacteriological Analytical Manual. Available at: http://www. fda.gov/Food/FoodScienceResearch/LaboratoryMethods/ ucm109656.htm. Accessed April 16, 2015. Breed, R. S., and J. F. Norton. 1937. Nomenclature for the colon group. Am. J. Public Health. 27:560-563. Brenner, J. D., N. R. Krieg, and J. T. Staley (eds.) 2005. Bergey’s Manual of Systematic Bacteriology, 2nd ed., vol. 2, The Proteobacteria, Part B: The Gammaproteobacteria, pp. 595-602. Springer, New York, NY. Brodsky, M. H., P. Boleszczuk, and P. Entis. 1982. Effect of stress and resuscitation on recovery of indicator bacteria from foods using hydrophobic grid-membrane filtration. J. Food Prot. 45:1326-1331. Chang, G. W., J. Brill, and R. Lum. 1989. Proportion of b-Dglucuronidase-negative Escherichia coli in human fecal samples. Appl. Environ. Microbiol. 55:335-339. Chordash, R. A., and N. F. Insalata. 1978. Incidence and pathological significance of Escherichia coli and other sanitary indicator organisms in food and water. Food Technol. 32:5458. Clermont, O., D. M. Gordon, S. Brisse, S. T. Walk, and E. Denamur. 2011. Characterization of the cryptic Escherichia lineages: rapid identification and prevalence. Environ. Microbiol. 13: 2468-2477. Cohan, F. M., and S. M. Kopac. 2011. Microbial genomics: E. coli relatives out of doors and out of body. Current Biology. 21, no. 15:R587-R589. Cox, L. J., N. Keller, and M. van Schothorst. 1988. The use and misuse of quantitative determinations of Enterobacteriaceae in food microbiology. J. Appl. Bacteriol. Symp. Suppl. 237S-249S. Crozier-Dodson, B. A., and D. Y. C. Fung. 2002. Comparison of recovery of airborne microorganisms in a dairy cattle facility using selective agar and thin agar layer resuscitation media. J. Food Prot. 65:1488-1492. ˜ , J. M., D. F. Campbell, and R. W. Johnston. 1985. DamarO Simplified direct plating method for enhanced recovery of Escherichia coli in food. J. Food Sci. 50:1736-1737. Davidson, P. M., L. A. Roth, and S. A. Gambrel-Lenarz. 2004. Coliform and other indicator bacteria. In: Standard Methods for the Examination of Dairy, 17th ed. American Public Health Association, Washington, D.C. Ser. 17:237S249S. Doyle, M. P., and J. L. Schoeni. 1984. Survival and growth characteristics of Escherichia coli associated with hemorrhagic colitis. Appl. Environ. Microbiol. 48:855-856. Doyle, M. P., and M. C. Erickson. 2006. Closing the door on the fecal coliform assay. Microbe. 1:162-163. Duan, J., C. Liu, and Y.-C. Su. 2006. Evaluation of a double layer agar plate for direct enumeration of Vibrio parahaemolyticus. J. Food Sci. 71:M77-M82. Edberg, S. C., M. J. Allen, and D. B. Smith. 1991. Defined substrate technology method for rapid and specific simultaneous enumeration of total coliforms and Escherichia coli from water: collaborative study. J. Assoc. Off. Anal. Chem. 74:526-529. Edwards, P. R., and W. H. Ewing. 1972. Identification of Enterobacteriaceae, 3rd ed. Burgess Publishing Company, Minneapolis, MN. Entis, P., and P. Boleszczuk. 1990. Direct enumeration of coliforms and Escherichia coli by hydrophobic grid membrane filter in 24 hr using MUG. J. Food Prot. 53:948-952.

28. Evans, T. M., R. J. Seidler, and M. W. LeChevallier. 1981. Impact of verification media and resuscitation on accuracy of the membrane filter total coliform enumeration technique. Appl. Environ. Microbiol. 41:1144-1151. 29. Ewing, W. H. 1986. Edwards and Ewing’s Identification of Enterobacteriaceae, 4th ed. Elsevier, New York, NY. 30. Feldsine, P. T., M. T. Falbo-Nelson, and D. L. Hustad. 1993. Substrate supporting disc method for confirmed detection of total coliforms and E. coli in all foods: collaborative study. J. AOAC Int. 76:988-1005. 31. Feng, P., S. D. Weagent, M. A. Grant, and W. Burkhardt. 2002. Enumeration of Escherichia coli and the coliform bacteria. Bacteriological Analytical Manual. Available at: http://www. fda.gov/Food/FoodScienceResearch/LaboratoryMethods/ ucm064948.htm. Accessed April 16, 2015. 32. Feng, P. C. S., and P. A. Hartman. 1982. Fluorogenic assays for immediate confirmation of Escherichia coli. Appl. Environ. Microbiol. 43:1320-1329. 33. Fishbein, M. 1962. The aerogenic response of Escherichia coli and strains of Aerobacter in EC broth and selected sugar broths at elevated temperatures. Appl. Microbiol. 10:79-85. 34. Food and Agricultural Organization of the United Nations and World Health Organization (FAO/WHO). Report of a joint FAO/WHO working group on microbiological criteria for foods. 20-26 February, 1979. WHO, Geneva, Switzerland. Document WG/Microbiol; 79. 35. Frampton, E. W., L. Restaino, and N. Blaszko. 1988. Evaluation of the b-glucuronidase substrate 5-bromo-4chloro-3-indolyl-b-D-glucuronide (X-GLUC) in a 24-hr direct plating method for Escherichia coli. J. Food Prot. 51:402-404. 36. Gabis, D. A., and R. E. Faust. 1988. Controlling microbial growth in food processing environments. Food Technol. 12:81-82, 89. 37. Geissler, K., M. Manafi, I. Amoros, and J. L. Alonso. 2000. Quantitative determination of total coliforms and Escherichia coli in marine waters with chromogenic and fluorogenic media. J. Appl. Microbiol. 88:280-285. 38. Gurtler, J. B., and J. L. Kornacki. 2009. Comparison of media supplements to enhance the recovery of Salmonella spp. from thermally-treated egg albumen. Lett. Appl. Microbiol. 49:503-509. 39. Hajmeer, M. N., D. Y. C. Fung, J. L. Marsden, and G. A. Milliken. 2001. Effects of preparation method, age, and plating technique of thin agar layer media on recovery of Escherichia coli O157:H7 injured by sodium chloride. J. Microb. Methods. 47:249-253. 40. Hansen, W., and E. Yourassowsky. 1984. Detection of bglucuronidase in lactose-fermenting members of the family Enterobacteriaceae and its presence in bacterial urine cultures. J. Clin. Microbiol. 20:1177-1179. 41. Hartman, P. A. 1989. MUG (b-glucuronidase) test for Escherichia coli in food and water. In: Rapid Methods and Automation in Microbiology and Immunology, (A. Ballows, R. C. Tilton, and A. Turano, eds.), p. 290. Brixia Academic Press, Brescia, Italy. 42. Hartman, P. A., and P. S. Hartman. 1976. Coliform analyses at 30uC. J. Milk Food Technol. 39:762-767. 43. Himelbloom, B. H., and R. C. Pfutzenreuter. 1998. Falsepositive fluorescence by pink Salmon tissue and Staphylococci in a rapid test for Escherichia coli. J. Food Prot. 61:1119. 44. Hitchins, A. D., P. Feng, W. D. Watkins, S. R. Rippey, and L. A. Chandler. 1998. Escherichia coli and the coliform bacteria. In: FDA Bacteriological Analytical Manual, rev. A, 8th ed., p. 401. AOAC International, Gaithersburg, MD. 45. Holbrook, R., J. M. Anderson, and A. C. Baird-Parker. 1980. Modified direct plate method for counting Escherichia coli in foods. Food Technol. Aust. 32:78-83.

| Enterobacteriaceae, Coliforms, and Escherichia coli as Quality and Safety Indicators

46. Holt, S. M., P. A. Hartman, and C. W. Kaspar. 1989. Enzymecapture assay for rapid detection of Escherichia coli in oysters. Appl. Environ. Microbiol. 55:229-232. 47. Huang, S. W., C. H. Chang, T. F. Tai, and T. C. Chang. 1997. Comparison of the b-glucuronidase assay and the conventional method for identification of Escherichia coli on eosin-methylene blue agar. J. Food Prot. 60:6-9. 48. International Commission on Microbiological Specifications for Foods of the International Union of Biological Societies. 1978. Microorganisms in Foods 1: Their Significance and Methods of Enumeration, p. 10. University of Toronto Press, Toronto, Canada. 49. International Standardization Organization. 2004. ISO 215281:2004, Microbiology of food and animal feeding stuffs— horizontal methods for the detection and enumeration of Enterobacteriaceae—part 1: detection and enumeration by MPN technique with pre-enrichment. ISO, Geneva, Switzerland. 50. International Standardization Organization. 2006. ISO 4831:2006, Microbiology of food and animal feeding stuffs—horizontal method for the detection and enumeration of coliforms—most probable number technique. ISO, Geneva, Switzerland. 51. Ishii, S., W. B. Ksoll, R. E. Hicks, and M. J. Sadowsky. 2006. Presence and growth of naturalized Escherichia coli in temperate soils from Lake Superior watersheds. Appl. Environ. Microbiol. 72:612-621. 52. Kang, D.-H., and D. Y. C. Fung. 1999. Thin agar layer method for recovery of heat-injured Listeria monocytogenes. J. Food Prot. 62:1346-1349. 53. Knittel, M. D., R. J. Seidler, C. Eby, and L. M. Cabe. 1977. Colonization of the botanical environment by Klebsiella isolates of pathogenic origin. Appl. Environ. Microbiol. 34:557-563. 54. Kornacki, J. L. 2012. Hygiene control in the dry food products industry: the roles of cleaning methods and hygienic indicators. In: Case Studies in Food Safety and Authenticity: Lessons From Real-life Situations, (J. Hoorfar, ed.). Woodhead Publishing, Cambridge, UK. 55. Kornacki, J. L., and D. A. Gabis. 1990. Microorganisms and refrigeration temperatures. Dairy Food Environ. Sanit. 10:192-195. 56. Latimer, G., Jr. (ed.), AOAC. 2012. Microbiological methods: E. coli. In: Official Methods of Analysis, 19th ed., 5th rev. AOAC International, Gaithersburg, MD. 57. LeChevallier, M. W., and G. A. McFeters. 1984. Recent advances in coliform methodology for water analysis. J. Environ. Health. 47:5-9. 58. Luo, C., S. T. Walk, D. M. Gordon, M. Feldgarden, J. M. Tiedje, and K. T. Konstantinidis. 2011. Genome sequencing of environmental Escherichia coli expands understanding of the ecology and speciation of the model bacterial species. Proc. Natl. Acad. Sci. U. S. A. 108:7200-7205. 59. Mackenzie, E. F. W., E. W. Taylor, and W. E. Gilbert. 1948. Recent experiences in the rapid identification of Bacterium coli type I. J. Gen. Microbiol. 2:197. 60. Mackey, B. M., C. M. Derrick, and J. A. Thomas. 1980. The recovery of sublethally injured Escherichia coli from frozen meat. J. Appl. Bacteriol. 48:315-324. 61. Manafi, M. 1996. Fluorogenic and chromogenic enzyme substrates in culture media and identification tests. Int. J. Food Microbiol. 31:45-58. 62. Matner, R. R., T. L. Fox, D. E. McIver, and M. S. Curiale. 1990. Efficacy of Petrifilm E. coli count plates for E. coli and coliform enumeration. J. Food Prot. 53:145-150. 63. Mercuri, A. J., and N. A. Cox. 1979. Coliforms and Enterobacteriaceae isolates from selected foods. J. Food Prot. 42:712-714.

64. Miskimin, D. K., K. A. Berkowitz, M. Solberg, W. E. Riha Jr., W. C. Franke, R. L. Buchanan, and V. O’Leary. 1976. Relationships between indicator organisms and specific pathogens in potentially hazardous foods. J. Food Sci. 41:1001-1006. 65. Moberg, L. J., M. K. Wagner, and L. A. Kellen. 1988. Fluorogenic assay for rapid detection of Escherichia coli in chilled and frozen foods: collaborative study. J. Assoc. Off. Anal. Chem. 71:589-602. 66. Mossel, D. A. A. 1978. Index and indicator organisms—a current assessment of their usefulness and significance. Food Technol. Aust. 30:212-219. 67. Mossel, D. A. A. 1985. Media for Enterobacteriaceae. Int. J. Food Microbiol. 2:27. 68. Mossel, D. A. A., I. Eelderink, M. Koopmans, and F. van Rossem. 1979. Influence of carbon source, bile salts, and incubation temperature on recovery of Enterobacteriaceae from foods using MacConkey-type agars. J. Food Prot. 42:470-475. 69. Mossel, D. A. A., and P. A. Van Netten. 1991. Microbiological reference values for foods: a European perspective. J. Assoc. Off. Anal. Chem. 74:420-432. 70. Mossel, D. A. A., M. Visser, and A. M. R. Cornelissen. 1963. The examination of foods for Enterobacteriaceae using a test of the type generally adopted for the detection of Salmonellae. J. Appl. Bacteriol. 3:444-452. 71. National Research Council (NRC), Food and Nutrition Board, Committee on Food Protection, and Subcommittee on Microbiological Criteria. 1985. An Evaluation of the Role of Microbiological Criteria for Foods and Food Ingredients, p. 436. National Academy Press, Washington, D.C. 72. Nelson, C. L., T. L. Fox, and F. F. Busta. 1984. Evaluation of dry medium film (Petrifilm VRB) for coliform enumeration. J. Food Prot. 47:520-525. 73. Ogden, I. D., and N. J. C. Strachan. 1993. Enumeration of Escherichia coli in cooked and raw meats by ion mobility spectrometry. J. Appl. Bacteriol. 74:402-405. 74. Ogden, I. D., and A. J. Watt. 1991. An evaluation of fluorogenic and chromogenic assays for the direct enumeration of Escherichia coli. Lett. Appl. Microbiol. 13:212-215. 75. Osaili, T. M., A. A. Al-Nabulsi, R. R. Shaker, M. M. Al-Holy, M. S. Al-Haddaq, A. N. Olaimat, M. M. Ayyash, M. K. Al Ta’ani, and S. J. Forsythe. 2010. Efficacy of the thin agar layer method for the recovery of stressed Cronobacter spp. (Enterobacter sakazakii). J. Food Prot. 73(10):1913-1918. 76. Patterson, J. T., and P. A. Gibbs. 1977. Incidence and spoilage potential of isolates from vacuum-packaged meat of high pH value. J. Appl. Microbiol. 43(1):25-38. 77. Powers, E. M., and T. G. Latt. 1977. Simplified 48-hr IMViC test: an agar plate method. Appl. Environ. Microbiol. 34:274-279. 78. Qiu, X., and V. C. H. Wu. 2007. Evaluation of Escherichia coli O157:H7, Listeria monocytogenes, Salmonella Typhimurium and Staphylococcus aureus in ground beef with cranberry concentrate by thin agar layer method. J. Rapid Methods Autom. Microbiol. 15:282-294. 79. Ray, B. 1986. Impact of bacterial injury and repair in food microbiology: its past, present and future. J. Food Prot. 49:651-655. 80. Rayman, M. K., G. A. Jarvis, C. M. Davidson, S. Long, J. M. Allen, T. Tong, P. Dodsworth, S. McLaughlin, S. Greenberg, B. G. Shaw, H. J. Beckers, S. Qvist, P. M. Nottingham, and B. J. Stewart. 1979. ICMSF methods studies. XIII. An international comparative study of the MPN procedure and the AndersonBaird-Parker direct plating method for the enumeration of Escherichia coli biotype I in raw meats. Can. J. Microbiol. 25:1321-1327. 81. Rochaix, A. 1924. La recherche du colibacille dans l’eau et dans le lait, au moyen des milieux a` l’esculine [Research

| 119

Compendium of Methods for the Microbiological Examination of Foods |

82.

83.

84.

85.

86.

87.

88.

89.

90. 91.

92.

120 |

examining E. coli in water and milk, using media containing esculin]. Lait. 4:541-544. Sharpe, A. N., and L. J. Parrington. 1998. Membrane filter method based on FC-5-bromo-4-chloro-3-indolyl-b- D glucuronide medium facilitates enumeration of Escherichia coli in foods and poultry carcass rinses. J. Food Prot. 61:360-364. Silk, T. M., E. T. Ryser, and C. W. Donnelly. 1997. Comparison of methods for determining coliform and Escherichia coli levels in apple cider. J. Food Prot. 60:13021305. Silliker, J. H., and D. A. Gabis. 1976. ICMSF methods studies. VII. Indicator tests as substitutes for direct testing of dried foods and feeds for Salmonella. Can. J. Microbiol. 22:971-974. Simmons, J. S. 1926. A culture medium for differentiating organisms of typhoid-colon aerogenes groups and for isolation of certain fungi. J. Bacteriol. 39:209-214. Siragusa, G. R., W. J. Dorsa, C. N. Cutter, G. L. Bennett, J. E. Keen, and M. Koohmare. 1998. The incidence of Escherichia coli on beef carcasses and its association with aerobic plate count categories during the slaughter process. J. Food Prot. 61(10):1269-1274. Solberg, M., D. K. Miskimin, B. A. Martin, G. Page, S. Goldner, and M. Libfeld. 1977. Indicator organisms, foodborne pathogens and food safety. Assoc. Food Drug Off. Quart. Bull. 41:9. Speck, M. L., B. Ray, and R. B. Read Jr. 1975. Repair and enumeration of injured coliforms by a plating procedure. Appl. Microbiol. 29:549-550. Splittstoesser, D. F., D. T. Queale, J. L. Bowers, and M. Wilkison. 1980. Coliform content of frozen blanched vegetables packed in the United States. J. Food Safety. 2:1-11. Stiles, M. E., and L.-K. Ng. 1980. Estimation of Escherichia coli in raw ground beef. Appl. Environ. Microbiol. 40:346-351. Strachan, N. J. C., and I. D. Ogden. 1993. A rapid method for the enumeration of coliforms in processed foods by ion mobility spectrometry. Lett. Appl. Microbiol. 17:228-230. Swenarton, J. C. 1927. Can B. coli be used as an index of the proper pasteurization of milk? J. Bacteriol. 13:419-429.

93. Tuttle, J., T. Gomez, M. P. Doyle, J. G. Wells, T. Zhao, R. V. Tauxe, and P. M. Griffin. 1999. Lessons from a large outbreak of Escherichia coli O157:H7 infections: insights into the infectious dose and method of widespread contamination of hamburger patties. Epidemiol. Infect. 122:185-192. 94. United States Pharmacopeial Convention. 1995. Microbial limits test. In: U.S. Pharmacopeia: National Formulary, 23rd ed., p. 1681. U.S. Pharmacopeial Convention, Rockville, MD. 95. Venkateswaran, K., A. Murakoshi, and M. Satake. 1996. Comparison of commercially available kits with standard methods for the detection of coliforms and Escherichia coli in foods. Appl. Environ. Microbiol. 62:2236-2243. 96. Vought, K. J., and S. R. Tatini. 1998. Salmonella enteritidis contamination of ice cream associated with a 1994 multistate outbreak. J. Food Prot. 61:5-10. 97. Weiss, K. F., N. Chopra, P. Stotland, G. W. Reidel, and S. Malcolm. 1983. Recovery of fecal coliforms and of Escherichia coli at 44.5, 45.0, 45.5uC. J. Food Prot. 46:172-177. 98. Wilderson, W. B., J. C. Ayres, and A. A. Kraft. 1961. Occurrence of enterococci and coliform organisms on fresh and stored poultry. Food Technol. 15(6):286-292. 99. Wu, V. C. H., and D. Y. C. Fung. 2001. Evaluation of thin agar layer method for recovery of heat-injured foodborne pathogens. J. Food Sci. 66:580-583. 100. Wu, V. C. H., D. Y. C. Fung, D. H. Kang, and L. K. Thompson. 2001. Evaluation of thin agar layer method for recovery of acid injured foodborne pathogens. J. Food Prot. 64:1067-1071. 101. Wyss, R. and P Hockenjos. 1999. Detection of enterohaemorrhagic Escherichia coli (EHEC) on beef carcasses. Fleischwirtschaft. 12/99:84-86. 102. Yan, Z., J. B. Gurtler, and J. L. Kornacki. 2006. A solid agar overlay method for recovery of heat-injured Listeria monocytogenes. J. Food Prot. 69:428-431. 103. Yuste, J., M. Capellas, R. Pla, S. Llorens, D. Y. C. Fung, and M. Mor-Mur. 2003. Use of conventional media and thin agar layer method for recovery of foodborne pathogens from pressuretreated poultry products. J. Food Sci. 68:2321-2324.

|

CHAPTER 10

|

Enterococci Katie Laird

10.1

INTRODUCTION

The classification of the enterococci has been reviewed on many occasions over the last 30 years; originally, all streptococci of fecal origin that produce group D antigen were considered enterococci. These included Streptococcus avium, S. bovis, S. faecalis (and its varieties liquefaciens and zymogenes), and S. faecium (and its varieties casseliflavus and durans). Due to advances in molecular characterization of the genus (including oligonucleotide cataloging of 16S rRNA, DNA-DNA and DNA-rRNA hybridization, and whole cell protein analysis) combined with physiological studies, there are now 23 distinct Enterococcus species.23 Members of this genus include Enterococcus avium, E. casseliflavus, E. durans, E. faecalis, E. faecium, E. gallinarum, E. hirae, E. malodoratus, and E. mundtii (Table 10-1). All these bacteria usually grow at 45uC, in 6.5% NaCl, and at pH 9.6; most grow at 10uC. S. bovis and S. equinus, which are negative in two or more of these properties, were assigned to a miscellaneous group of ‘‘Other Streptococci.’’42 Almost all (99%) are susceptible to vancomycin, and very few (less than 1%) produce gas from glycerol.1 The genera of lactic acid bacteria with which Enterococcus is grouped are also identified by a low G + C content of less than 50%.33 The enterococci have conventionally been identified by physiological as well as serological methods. When the former are employed, a spectrum of characteristics (Table 10-1) must be examined, because no single, two, or three traits will establish a definitive identification and are, therefore, often identified by the use of reverse identification (i.e., elimination of other species traits first).10 Thus the use of fermentation patterns, enzyme activities such as pyroglutamyl aminopeptidase (PYRase) activity,11 growth at defined temperatures, and physiological characteristics are essential in the identification of Enterococcus sp.43 Phenotypic, genotypic, and phylogenetic techniques for identification of enterococci and interpretation of these tests have been described by Domig et al.11 Generally, the Enterococcus sp. habitat is the intestinal contents of both warm and cold-blooded animals, including insects.19 Because enterococci are an essential part of the microflora of both humans and animals, their distribution is very similar in these sources. Some enterococci have

adapted to an epiphytic relationship with growing vegetation. None of the enterococci can be considered as absolutely host specific, although some species evidence a degree of host specificity.28 E. faecalis and E. faecium are relatively heat resistant and characteristically may survive traditional milk pasteurization procedures. E. faecium is markedly heat tolerant and is a spoilage agent in marginally processed canned hams. Most of the enterococci are relatively resistant to freezing, and, unlike Escherichia coli, they readily survive this treatment.28 Increasing antibiotic resistance and virulence factors in E. faecalis and E. faecium means that not only food poisoning outbreaks and food spoilage are being associated with these bacteria, but also nosocomial infections.20,22,29 Yet enterococci are also important probiotics, fermentors, and starter cultures, thus giving rise to concern for the use of enteroccci in the food industry.20

10.2

GENERAL CONSIDERATIONS

An abundance of media has been advocated for the selective isolation and/or quantification of enterococci.11,29 Many selective agents, incubation conditions, and combinations of these have been described, but all have one or more shortcomings. The media and methods that are available presently lack selectivity, differential ability, quantitative recovery, relative ease of use, or a combination of these to various degrees, with over 100 modifications of selective media for enterococci being described. Therefore, some considerations must be made before selection of media (e.g., specimen type, method of cultivation, and whether other contamination exists).8,41 In foodstuff, if only overall counts are required, two complex culture media can be used: Enterococcus selective Slantez and Bartley (SB) agar and kanamycin aesculin azide (KAA) agar; however, if enterococci are the only microbial component of the foodstuff, MRS or Rogosa agar can be used.11 In food microbiology, E. faecalis and E. faecium are the most common enterococci encountered. This undoubtedly influences the rationale of employing Kenner fecal (KF) streptococcal agar for the estimation of enterococci in

| 121 |

122 |

39–40 41–45 38–40 37–40 37–40 39–40 37–38 40–41 38–39 36–38 36–39

PYR (see text)

+ + + + + + + + + – –

Serological Group

D+Q D D D D D D D D D D

Acid from arabinose

Hydrol. of hippurate

Hydrol. of esculin

Hydrol. of arginine

Growth, pH 9.6

Growth, 6.5% NaCl

Growth, 45uC

Growth, 10uC

V V – (+) – – – – – – –

Acid from glycerol

+ + – – + + – – + V –

+ – – (+) – (–) V – – nd nd

Acid from melezitose

– V V V (–) (+) (–) – – – –

V + V – V + (+) + + + nd

Acid from melibiose

+ + + (+) + + + + + + +

+ V – (+) V – – + V V –

Acid from sorbitol

+ V + + + + + + +

+ – – – – – – + – – nd

Acid from sorbose

+ + + + + + + + + – –

+ (–) – + (–) (+) V – – nd nd

Acid from tagatose

+ + + + + + + + + – –

– + – – – + – – – – –

Motility

+ + + + + + + + + + +

– + – – – – – – + – –

Yellow pigment

+ + + + + + + + + – –

+ – – – – – nd + nd – –

H2S produced

Species

Note: + 5 positive; (+) 5 most strains positive; (–) 5 most strains negative; – 5 negative; nd 5 not determined (but probably would be hydrogen sulfide negative); v 5 variable Differential characteristics of Enterococcus spp. are listed in the references.3,5,6,14,15,17 See Facklam and Collins (1989)14 for abbreviated identification schemes, including that for E. pseudoavium, E. raffinosus, and E. solitarius. a All are Gram-positive, catalase-negative, facultatively anaerobic cocci or coccobacilli. Note that false-positive and false-negative test reactions can be obtained, depending on the sensitivity of the assay that is used.

G + C content (mole %)

E. avium E. casseliflavus E. durans E. faecalis E. faecium E. gallinarum E. hirae E. malodoratus E. mundtii S. bovis S. equinus

Table 10-1. Some Characteristics of the Enterococci and Group D Fecal Streptococcia

Compendium of Methods for the Microbiological Examination of Foods |

| Enterococci

foods.31 The selectivity of KF streptococcal agar is not absolute,11,14,15,48 quantitative recovery is less than ideal, and preparation of the medium necessitates an aseptic addition of an indicator—1% triphenyltetrazolium chloride (TTC) solution. Nevertheless, many industry and regulatory agencies have accepted KF agar for the quantitative estimation of enterococci in non-dairy foods. For dairy products, a more selective medium such as citrate azide tween carbonate (CATC) medium can be used; a higher incubation temperature (45uC) may be necessary to reduce background growth of lactobacilli and lactic streptococci. E. faecalis forms pronounced growth and formazan production on CATC medium, whereas E. faecium produces a weaker formazan reaction.8 KF streptococcal agar is a selective differential medium that employs sodium azide as the chief selective agent and TTC for differential purposes. The medium contains a relatively high concentration of maltose (2.0%) and a small amount of lactose (0.1%). Most, but not all, enterococci and streptococci ferment these sugars. The intensity of TTC reduction varies. E. faecalis reduces the compound to its formazan derivative, imparting a deep red color to the colony. Other group D enterococci and streptococci, if they grow on KF agar, are feebly reductive and the colonies appear light pink; however, tetrazolium reactions of E. hirae17 and E. malodoratus5 have not been described. Most other lactic acid bacteria are partially or completely inhibited; however, some strains of Pediococcus, Lactobacillus, and Aerococcus may grow, producing light pink colonies. A ‘‘repair-detection’’ procedure26 should be considered when the enterococcal population of a food may contain a large proportion of injured cells (see the chapter ‘‘Cultural Methods for the Enrichment and Isolation of Microorganisms’’). KF Streptococcal medium is available commercially with or without agar. A broth is available for the most probable number (MPN) procedure to detect low numbers of enterococci, but the MPN procedure is rarely used for foods. Other isolation media for Enterococcus spp. from food include KAA medium, which contains sodium azide and kanamycin as the selective agents and, upon hydrolysis of aesculin, a black halo is formed around the Enterococcus sp. colonies; increased incubation temperatures for shorter periods (42uC for 18 hr) can increase selectivity. Thallous acetate-tetrazolium-glucose (TITG) medium is also particularly good at distinguishing between E. faecalis and E. faecium; however, successful isolation of enterococci on this medium is often due to the method of preparation.35 KF streptococcal agar and KAA contain azide, which is inhibitory to many strains of S. bovis and S. equinus and possibly some of the newly named Enterococcus spp. Therefore, alternative procedures that permit the recovery of a wider variety of enterococci from foods are included in this section. Fluorogenic gentamicin-thallous-carbonate (fGTC) agar utilizes inhibitors other than azide.34 Dyed starch and a fluorogenic substrate are included to impart differential qualities to the medium. Enterococcal counts from foods may be two or more orders of magnitude higher on fGTC agar than on KF agar. Further, the incubation period for fGTC agar is only 18 to 24 hr, whereas it is 48 hr for KF agar.

Once Enterococcus sp. have been isolated from food samples, antibiotic resistance screening can be advisable due to increased epidemiological evidence of links between the use of antibiotics in humans and animal husbandry and the emergence of antibiotic-resistant strains in animal and dairy products.23 Enrichment is advised for the isolation and detection of low levels of vancomycin-resistant Enterococcus (VRE), and Enterococcosel broth (Becton Dickinson and Company, Sparks, MD) and KAA broth are particularly effective for recovery. As in other microbiological plating procedures, sample preparation is important. For example, dried foods are often reconstituted and immediately diluted and plated. In one study,45 however, the optimum procedure of sample preparation involved the addition of 25 mL of 0.1% peptone water diluent to 25 g of dry food in a sterile pint jar. The jar was swirled and allowed to remain at 4uC for 60 min. Then 200 mL of sterile peptone water were added to the jar and mixed to obtain a final 1:10 dilution. Enterococcal counts of dried soup mix were increased by 42% by using the ‘‘swirl-holddilute’’ method.45

10.3

N N N N N N N N N N N

EQUIPMENT, MATERIALS, AND REAGENTS

Bile-esculin agar15 Brain heart infusion (BHI) broth Filter-sterilized 1% aqueous triphenyltetrazolium chloride (TTC) Fluorogenic gentamicin-thallous-carbonate (fGTC) agar34 Hydrogen peroxide (3%) KF streptococcus (KF streptococcal) agar31 Long-wave (365 nm) ultraviolet light 5% Salt medium (BHI + 6.0% NaCl) Kanamycin aesculin azide (KAA) agar base Kanamycin sulphate supplement (10 mg/500 mL agar base) Tryptone water

10.4

PRECAUTIONS

Many foods contain from small (101) to large (107) numbers of enterococci, especially E. faecalis and E. faecium. Certain varieties of cheese and, occasionally, fermented sausage may contain more than 106 organisms per gram. Relatively low levels (101 to 103 per gram) are common in a wide variety of other foods. The shelf life of sliced, pre-packaged ham and sometimes other similarly prepared cured meats, may be dictated by controlling the initial numbers of contaminating enterococci. Many investigators have reported a lack of correlation between Enterococcus sp. and E. coli counts, and the unreliability of Enterococcus counts as a reflection that fecal contamination is established. The ability of enterococci to grow in food processing plants, and possibly other environments, long after their introduction, as well as the observation that enterococci can establish extraintestinal epiphytic relationships, reinforce those observations. No acceptable levels of enterococci can be stated because Enterococcus sp. counts vary with product, holding conditions, time of storage, and other factors. In general, enterococci serve as a good index of sanitation and proper holding conditions. However, the entire history of each product must be | 123

Compendium of Methods for the Microbiological Examination of Foods |

established and the culture medium and conditions must be standardized before setting specific criteria.

10.5 10.51

ENUMERATION OF ENTEROCOCCI KF Streptococcal Agar31

Prepare the sample for culturing by the pour-plate method as directed in the chapter ‘‘Culture Methods for Enumeration of Microorganisms.’’ Dispense 1 mL of decimal dilutions into duplicate petri plates. If a low count is expected, the accuracy and sensitivity may be increased by plating 1 mL of a 1:10 dilution into each of 10 Petri plates, in which case the total number of colonies on the 10 plates represents the count per gram of food. Add 12 to 15 mL of KF agar cooled to 45uC and allow to solidify. Incubate the plates for 48 ¡ 2 hr at 35 ¡ 1uC. Using a dissecting microscope with a magnification of 15 diameters or a colony counter, count all red and pink colonies. Report this number as the KF enterococcal count.

10.52

fGTC Agar

34

Prepare sample as directed in Section 10.51 above. Add 12 to 15 mL of fGTC agar34 cooled to 45uC, and allow to solidify. Incubate the plates for 18 to 24 hr at 35 ¡ 1uC. Observe for starch hydrolysis (a zone of clearing around a colony under visible light) and fluorescence (a zone of bright bluish fluorescence when the opened plate is held under a long-wave ultraviolet lamp). Three phenotypic groups are identifiable: (1) starch hydrolysis and fluorescence, indicative of S. bovis; (2) no starch hydrolysis but fluorescence, indicative of E. faecium and related biotypes; and (3) no starch hydrolysis or fluorescence, indicative of E. faecalis, E. avium, S. equinus, and other streptococci.34 Use all colonies to calculate the fGTC enterococcal count, which can be divided, if desired, into subgroups based on starch hydrolysis and fluorescence.

10.53

KAA Agar

Add 1 g or 1 mL of foodstuff to 9 mL of pre-chilled diluent (tryptone water), and prepare dilutions as outlined in Section 10.51. Streak the sample onto the KAA agar surface and incubate for 16 to 24 hr at 35 ¡ 1uC. Using a colony counter, count all the colonies surrounded by black haloes.

10.6 10.61

CONFIRMATION OF ENTEROCOCCI Conventional Procedures

If confirmation is desired, pick 5 to 10 typical colonies and transfer each into a separate tube of BHI broth. Incubate at 35uC for 18 to 24 hr. Prepare Gram-stained smears of the BHI cultures and observe for typical enterococcal morphology, Gram-positive cocci, elongated, in pairs, and occasionally short chains. Test for catalase activity by adding 1 mL of 3% hydrogen peroxide to a culture and observe for the generation of oxygen bubbles. Enterococci are catalase negative, and no reaction should occur. Caution: do not test for catalase activity directly on azide-containing media such as KF streptococcal agar. Observe for growth and black haloes on KAA agar after incubation for 24 hr at 35uC. Examine for growth in BHI broth containing 6.5% NaCl after incubation for 72 hr at 35uC. Test for growth at 45uC in 124 |

BHI broth that has been tempered to 45uC prior to incubation. Note: if growth in the salt-containing medium and growth at 45uC are to be determined, subcultures must be inoculated before testing for catalase. S. equinus and S. bovis are not enterococci, but they can be of fecal origin. Most do not grow at 10uC, in media containing 6.5% NaCl, or at pH 9.6, but all should grow at 45uC. An excellent confirmatory test for enterococci/fecal streptococci is the ability of an isolate to grow on bileesculin agar. Enterococci and group D streptococci tolerate bile (grow on bile-esculin agar) and hydrolyze esculin13 (6,7-dihydroxycoumarin-b-D-glucoside). Some bacteria produce an ‘‘esculinase’’ (b-D-glucosidase) that hydrolyzes esculin and releases esculetin (6,7-dihydroxycoumarin); the esculetin reacts with Fe+3 in the medium to form a dark brown or black complex.

10.62

Rapid Methods

A 15-min esculinase test was devised using p-nitrophenyl-b-D-glucopyranoside as the substrate for b -Dglucosidase (esculinase) determination,46 and a 4 hr combined NaCl tolerance-esculin hydrolysis test also has been described.39 Another rapid confirmatory test is the PYR test (Table 10-1), which detects the ability of a culture to hydrolyze pyrrolidonyl-b-naphthylamide (L-pyroglutamic acid-b-naphthylamide). Hydrolysis of this aminopeptidase substrate is detected by formation of a reddish color within 2 min of addition of PYR reagent. Of the streptococci, only S. pyogenes (group A) and the enterococci are positive; S. bovis and S. equinus are negative.15,24 Prepackaged PYR test reagents are available40 (Visi-Spot Kit, Thermo Scientific, Fremont, CA; Strep-A-Chek, E-Y Laboratories, Inc., San Mateo, CA; Identicult-AE, Scott Laboratories, Inc., West Warwick, RI; Roscoe Diagnostica, 2630 Taastrup, Denmark). Convenient tri-plates, quad-plates, and tubed21,27,32 media for key identification tests are available from many suppliers of prepared media. API Rapid ID 32 kit (bioMe´rieux, Durham, NC)2,49 may also be time-saving, with 67 of 71 Enterococcus sp. being accurately identified. Supplemental tests for motility and pigmentation can improve the accuracy of the identification, but one drawback of this test is that it does not accurately identify vancomycin-resistant E. faecium.27,37 Another similar kit is the API 20S (Analytab Products, Plainview, NY), which works well on E. faecalis with a greater than 95% accurate identification; however, identification of E. faecium, E. gallinarum, E. casseliflavus, and E. durans is often inaccurate.1,2,6,7,15,27,30,35,37,38 The efficacy of this kit and the data bank used with it is based on isolates from human clinical material; the efficiency may differ when isolates from other animals or food are studied.38

10.63

Automated Identification The VITEK 2 system1,16,47 (bioMe´rieux, Durham, NC) has been further developed since the original system in the 1990s, with more recent research reporting improved identification. Again, however, supplemental tests are required for motility and pigmentation. Another system is MicroScan Gram Positive Panels (Siemens Healthcare Diagnostics, Tarrytown, NY). This system has shown to

| Enterococci

accurately identify 99.6% E. faecalis and 78.3% E. faecium; however, the system is not as accurate when identifying other enterococcal strains (68.6%); therefore, additional tests are required for accurate identification of strains other than E. faecalis.9

10.64

Serological Tests

If serological confirmation is deemed necessary, commercial grouping sera are available from BBL Microbiology Systems (Becton Dickinson and Company, Sparks, MD). A variety of serological kits are available: BBL Streptocard (Becton, Dickinson and Company)25; Prolex StreptococcalSelect Grouping Latex Kit (Pro-Lab Diagnostics, Round Rock, TX)36; Streptex (Remel, Lenexa, KS)5,19,34,38,40,44; and Phadebact (MKL Diagnostics, Sollentuna, Sweden).4,18,24,40,44 These kits vary in efficacy, and false-negative group D reactions are common. Consult the references listed before using these kits. It is often difficult to demonstrate the presence of the group D antigen in some strains; only 77% of 188 Enterococcus strains tested were positive.14 The method of group antigen preparation is important.12,37

12.

13.

14.

15.

16.

17.

18.

ACKNOWLEDGMENT Fourth edition authors: Paul A. Hartman, Robert H. Deibel, and Linda M. Sieverding.

REFERENCES 1. Appelbaum, P. C., M. R. Jacobs, J. I. Heald, W. M. Palko, A. Duffett, R. Crist, and P. A. Naugle. 1984. Comparative evaluation of the API 20S system and the AutoMicrobic system gram-positive identification card for species identification of streptococci. J. Clin. Microbiol. 19:164-168. 2. Appelbaum, P. C., M. R. Jacobs, W. M. Palko, E. E. Frauenhoffer, and A. Duffett. 1986. Accuracy and reproducibility of the IDS RapID STR system for species identification of streptococci. J. Clin. Microbiol. 23:843-846. 3. Brown, L. H., E. M. Peterson, and L. M. de la Maza. 1983. Rapid identification of enterococci. J. Clin. Microbiol. 17:369370. 4. Chang, G. T., and P. D. Ellner. 1983. Evaluation of slide agglutination methods for identifying group D streptococci. J. Clin. Microbiol. 17:804-806. 5. Collins, M. D., D. Jones, J. A. E. Farrow, R. Kilpper-BSˇlz, and K. H. Schleifer. 1984. Enterococcus avium nom. rev., comb. nov.; E. casseliflavus nom. rev., comb. nov.; E. durans nom. rev., comb. nov.; E. gallinarium comb. nov.; and E. malodoratus sp. nov. Intl. J. Syst. Bacteriol. 34:220-223. 6. Collins, M. D., J. A. E. Farrow, and D. Jones. 1986. Enterococcus mundtii sp. nov. Intl. J. Syst. Bacteriol. 36:8-12. 7. Colman, G., and L. C. Ball. 1984. Identification of streptococci in a medical laboratory. J. Appl. Bacteriol. 57:1-14. 8. Corry, J. E. L, G. D. W. Curtis, and R. M. Baird. 2003. Handbook of Culture Media for Food Microbiology. Vol. 37. Elsevier Science, Netherlands. 9. d’Azevedo, P. A., C. A. Dias, A. L. Gonclaves, F. Rowe, and L. M. Teixeira. 2001. Evaluation of an automated system for the identification and antimicrobial susceptibility testing of Enterococci. Diagn. Microbiol. Infect. Dis. 42:157-161. 10. Deibel, R. H. 1964. The group D streptococci. Bacteriol. Rev. 28:330-366. 11. Domig, K. J., H. K. Mayer, and W. Kneifel. 2003. Methods used for the isolation, enumeration, characterisation and

19.

20.

21.

22.

23. 24.

25.

26.

27.

28.

29.

identification of Enterococcus spp. 1. Media for isolation and enumeration. Int. J. Food Microbiol. 88:147-164. Elliott, S. D., M. McCarty, and R. C. Lancefield. 1977. Teichoic acids of group D streptococci with special reference to strains from pig meningitis (Streptococcus suis). J. Exper. Med. 145:490-499. Facklam, R. R., and M. D. Moody. 1970. Presumptive identification of group D streptococci: the bile-esculin test. Appl. Microbiol. 20:245-250. Facklam, R. R., and M. D. Collins. 1989. Identification of Enterococcus species isolated from human infections by a conventional test scheme. J. Clin. Microbiol. 27:731-734. Facklam, R. R., L. G. Thacker, B. Fox, and L. Eriquez. 1982. Presumptive identification of streptococci with a new test system. J. Clin. Microbiol. 15:987-990. Facklam, R., G. S. Bosley, D. Rhoden, A. R. Franklin, N. Weaver, and R. Schulman. 1985. Comparative evaluation of the API 20S and AutoMicrobic gram-positive identification systems for non-beta-hemolytic streptococci and aerococci. J. Clin. Microbiol. 21:535-541. Farrow, J. A. E., and M. D. Collins. 1985. Enterococcus hirae, a new species that includes amino acid assay strain NCDO 1258 and strains causing growth depression in young chickens. Intl. J. Syst. Bacteriol. 35:73-75. Fertally, S. S., and R. Facklam. 1987. Comparison of physiologic tests used to identify non-beta-hemolytic aerococci, enterococci, and streptococci. J. Clin. Microbiol. 25:18451850. Fisher, K., and C. Phillips. 2009. The ecology, epidemiology & virulence of Enterococcus sp: a review. Microbiol. 155:17491757. Franz, C. M. A. P., W. H. Holzapfel, and M. E. Stiles. 1999. Enterococci at the crossroads of food safety? Int. J. Food Microbiol. 47:1-24. Freney, J., S. Bland, J. Etienne, M. Desmonceaux, J. M. Boeufgras, and J. Fleurette. 1992. Description and evaluation of semiautomated 4-hour rapid ID 32 Strep method for identification of streptococci and members of related genera. J. Clin. Microbiol. 30:2657-2661. Gilmore, M. S. 2002. The Enterococci, Pathogenesis, Molecular Biology and Antibiotic Resistance. American Society for Microbiology Press, Washington, D.C. Giraffa, G. 2002. Enterococci from foods. FEMS 26:163-171. Gordon, L. P., M. A. S. Damm, and J. D. Anderson. 1987. Rapid presumptive identification of streptococci directly from blood cultures by serologic tests and the L-pyrrolidonyl-bnaphthylamide reaction. J. Clin. Microbiol. 25:238-241. Green, N. M., S. B. Beres, E. A. Graviss, J. E. Allison, A. J. McGeer, et al. 2005. Genetic diversity among type emm28 group A Streptococcus strains causing invasive infections and pharyngitis. J. Clin. Microbiol. 43:4083-4091. Hackney, C. R., B. Ray, and M. L. Speck. 1979. Repair detection procedure for enumeration of fecal coliforms and enterococci from seafoods and marine environments. Appl. Environ. Microbiol. 37:947-953. Hamilton-Miller, J. M. T., and S. Shah. 1999. Identification of clinically isolated vancomycin-resistant enterococci: comparison of API and BBL crystal systems. J. Med. Microbiol. 48:695-696. Hartman, P. A., G. W. Reinbold, and D. S. Saraswat. 1966. Indicator organisms—a review. II. The role of enterococci in food poisoning. J. Milk Food Technol. 28:344-350. Hartman, P. A., J. P. Petzel, and C. W. Kaspar. 1986. New methods for indicator organisms. In: M. D. Pierson and N. J. Stern (Eds). Foodborne Microorganisms and Their Toxins: Developing Methodology. Marcel Dekker, Inc., New York, NY, p. 175.

| 125

Compendium of Methods for the Microbiological Examination of Foods |

30. Jorgensen, J. H., S. A. Crawford, and G. A. Alexander. 1983. Rapid identification of group D streptococci with the API 20S system. J. Clin. Microbiol. 17:1096-1098. 31. Kenner, B. A., H. F. Clark, and P. W. Kabler. 1961. Fecal streptococci. I. Cultivation and enumeration of streptococci in surface waters. Appl. Microbiol. 9:15-20. 32. Kim, M. J., M. Weiser, S. Gottschall, and E. L. Randall. 1987. Identification of Streptococcus faecalis and Streptococcus faecium and susceptibility studies with newly developed antimicrobial agents. J. Clin. Microbiol. 25:787-790. 33. Klein, G. 2003. Taxonomy, ecology and antibiotic resistance of Enterococci from food and the gastro-intestinal tract. Int. J. Food Microbiol. 88: 123-131. 34. Littel, K. J., and P. A. Hartman. 1983. Fluorogenic selective and differential medium for isolation of fecal streptococci. Appl. Environ. Microbiol. 45:622-627. 35. Mead, G. C. 1985. Isolation media for group D streptococci: comments. Int. J. Food Microbiol. 2:115-117. 36. Petts, D. N. 1995. Evaluation of a modified nitrous acid extraction latex agglutination kit for grouping beta-hemolytic streptococci and enterococci. J. Clin. Microbiol. 33:1016-1018. 37. Poutrel, B. 1983. Comparative evaluation of commercial latex agglutination and coagglutination reagents for groups B, C, and D mastitis streptococci. Amer. J. Vet. Res. 44:490-492. 38. Poutrel, B., and H. Z. Ryniewicz. 1984. Evaluation of the API 20 Strep system for species identification of streptococci isolated from bovine mastitis. J. Clin. Microbiol. 19:213-214. 39. Qadri, S. M. H., D. J. Flournoy, and S. G. M. Qadri. 1987. Sodium chloride-esculin hydrolysis test for rapid identification of enterococci. J. Clin. Microbiol. 25:1107-1108. 40. Rappaport, T., K. P. Sawyer, and I. Nachamkin. 1988. Evaluation of several commercial biochemical and immunologic methods

126 |

41. 42.

43.

44.

45.

46.

47.

48.

49.

for rapid identification of gram-positive cocci directly from blood cultures. J. Clin. Microbiol. 26:1335-1138. Reuter, G. 1992. Culture media for enterococci and group Dstreptococci. Int. J. Food Microbiol. 17:101-111. Schleifer, K. H., and R. Kilpper-BSˇlz. 1987. Molecular and chemotaxonomic approaches to the classification of streptococci, enterococci and lactococci: a review. Syst. Appl. Microbiol. 10:1-19. Shanks, O. C., J. W. Santo Domingo, and J. E. Graham. 2006. Use of competitive DNA hybridization to identify differences in the genomes of bacteria. J. Microbiol. Methods 66:321-330. Shlaes, D. M., Z. Toossi, and A. Patel. 1984. Comparison of latex agglutination and immunofluorescence for direct Lancefield grouping of streptococci from blood cultures. J. Clin. Microbiol. 20:195-198. Ting, W.-T., and G. J. Banwart. 1985. Enumeration of enterococci and aerobic mesophilic plate count in dried soup using three reconstitution methods. J. Food Prot. 48:770-771. Trepeta, R. W., and S. C. Edberg. 1987. Esculinase (bglucosidase) for the rapid estimation of activity in bacteria utilizing a hydrolysable substrate, p-nitrophenyl-b-D-glucopyranoside. Antonie van Leeuwenhoek 53:273-277. van Den Braak, N., W. Goessens, A. van Belkum, H. A. Verbrugh, and H. P. Endtz. 2001. Accuracy of the VITEK 2 system to detect glycopeptides resistance in enterococci. J. Clin. Microbiol. 34:924-927. Versalovic, J. 2011. Streptococcus. In: B. Spellerberg and C. Brandt (Eds). Manual of Clinical Microbiology. 10th ed. American Society for Microbiology, Washington, D.C. You, M. S., and R. R. Facklam. 1986. New test system for identification of Aerococcus, Enterococcus, and Streptococcus species. J. Clin. Microbiol. 24:607-611.

|

CHAPTER 11

|

Rapid Methods for the Detection and Identification of Foodborne Pathogens Hari P. Dwivedi, Ronald D. Smiley, and David H. Pincus

11.1

INTRODUCTION

Traditional methods for the detection and identification of pathogens in food mostly rely on the sequential steps of cultural enrichment, selective and differential plating, confirmation, and strain typing. Advances in the areas of immunology, molecular biology, and the biochemical sciences have resulted in the advancement and availability of diagnostic technologies that can be exploited for the detection of foodborne pathogens. These improved diagnostic technologies reduce the total time needed to detect and identify pathogens by reducing culture enrichment time and eliminating biochemical identification steps, which also require lengthy incubation periods. Many technologies provide similar levels of sensitivity and specificity to culture-based methods but with faster sample turnaround times and a substantial reduction in manual labor. This chapter covers some of the types and availability of rapid methods for the detection and identification of foodborne pathogens. The chapter is divided into three sections: antibody-based methods, nucleic acid amplification methods, and matrix-assisted laser desorption/ionization time-of-flight mass spectrometer (MALDI-TOF MS)–based methods.

11.2 11.21

ANTIBODY-BASED METHODS Introduction

Immunoassays are a broad class of technologies and methods that exploit the high affinity of antibodies for their specific antigens. Improvements in assay design and detection chemistries and the increased availability of monoclonal antibodies have significantly reduced the specificity and sensitivity problems that plagued some first-generation antibody-based technologies. Antibodybased detection technologies have been widely adapted to the field of food safety and are used for presumptive screening of selective enrichments, species-level confirmation of purified isolates, and subspecies-level testing such as serotyping. Some antibody-based detection platforms

can even be used quantitatively. Because of the large number of commercially available antibody-based methods, full descriptions of them will not be attempted here. Many of the subsequent chapters in this book provide methodspecific details of various immunoassays as they pertain to particular foodborne microorganisms. Additional information can also be found at the end of this section.7,31,38

11.22

Enzyme-Linked Immunosorbent Assay

The enzyme-linked immunosorbent assay (ELISA) is a highly popular and versatile immunoassay platform. It has been successfully employed at all stages of foodborne pathogen recovery/detection, including for the screening of selective enrichments to identify those most likely to contain the target pathogen,22,36,37,58,59,132 the screening of individual colonies following selective/differential plating, and the obtaining of subspecies-level information such as serotype determination111 or toxin-producing capability.150 The antibodies used in ELISAs are covalently linked to an enzyme; it is the enzymatic action upon the addition of substrate that yields a visible colored signal. Since the intensity of the signal is proportional to the concentration of the target antigen, this type of immunoassay can be used quantitatively. Two of the most popular enzymes used in ELISAs are alkaline phosphatase and horseradish peroxidase, which typically use the substrates p-nitrophenyl phosphate (PNPP) and 2, 29-azinobis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS), respectively; other chromogenic substrates are also available. In addition to chromogenic substrates, chemiluminescence signals can also be produced using substrates such as luminol for horseradish peroxidaselabeled antibodies. The chemiluminescence substrate 3-(29-spiroadamantane)-4-methyl-4-(39-b -D-galactopyranosyloxyphenyl-1, 2-dioxetane) (AMPGD) is popular for alkaline phosphatase-labeled antibodies. The use of fluorogenic substrates is a third option for ELISA signal production; this type of assay is sometimes referred to as enzyme-linked fluorescence assay (ELFA). Two frequently used fluorogenic substrates are 4-methylumbelliferyl phosphate (4-MUP) and

| 127 |

Compendium of Methods for the Microbiological Examination of Foods |

hydroxyphenylacetic acid (HPA) for alkaline phosphataselabeled antibodies and horseradish peroxidase-labeled antibodies, respectively. In some ELISA systems, an antibody that recognizes the target (e.g., bacterial cell or microbial toxin) is first bound to the wells of a microtiter plate and then used to capture the target following the addition of the test sample. This format is referred to as an antibody-capture ELISA (Figure 11-1). This type of assay is also referred to as a ‘‘sandwich ELISA’’ because the target is sandwiched between the capture antibody and a second detection antibody. In other ELISA formats, the test sample is applied directly to the wells of a microtiter plate and the target antigen (if present) binds non-specifically to the plate. The presence of the antigen is then determined by the addition of a labeled detection antibody. This assay format is referred to as antigen-capture ELISA. ELISAs are also classified based on the use of either enzyme-labeled primary antibodies (direct detection) or enzyme-labeled secondary antibodies (indirect detection) for detection (Figure 11-1). For indirect detection, an unlabeled primary antibody is first bound to the captured target antigen (i.e., foodborne pathogen or microbial toxin), and then the labeled secondary antibody, which recognizes the bound primary antibody, is added. Both types of detection formats (direct or indirect) can be used with both types of capture formats (antibody-capture or antigencapture), resulting in multiple ELISA formats. Regardless of which format is used, the incorporation of appropriate washing and blocking steps during the performance of the assay is necessary to prevent false positive readings.

11.23

Immunochromatographic Assays

Immunochromatographic assays for the detection of foodborne pathogens have become more popular in recent

Figure 11-1. Simple illustration contrasting direct and indirect detection ELISA formats. In both cases an antibody-capture ELISA is shown. C 5 colored product resulting from the action of the enzyme on the substrate; E 5 enzyme; S 5 substrate.

128 |

years; these assays are sometimes referred to as immunoprecipitate assays, lateral flow devices, or simply ‘‘dipsticks.’’ Lateral flow devices (LFDs) have three distinct zones (Figure 11-2). The test sample is added at the application point. If the target antigen is present in the test sample, it will react with label antibodies that are tethered to colored latex or nanogold particles; the label antibodies are located near the application point in zone 1. The sample flows (capillary action) along the solid support until it reaches the second zone containing the immobilized capture antibodies that recognize the antigen. The antigen (now labeled with either colored latex or nanogold particles) is held in place, forming a visible line (i.e., immunoprecipitate). The remaining sample continues moving downstream until it reaches the third zone, where remaining free colored latex or nanogold particles react with immobilized control antibodies, forming a second visible line (control line). LFDs are commercially available and have been described for many foodborne pathogens.1,2,54,55 These assays are typically easy to perform, are easy to interpret, and do not require specialized training or equipment. Because of their lower sensitivity, however, these assays usually require enrichment of the test sample in order to achieve minimum detection threshold levels. A small amount of sample preparation may also be required, such as centrifugation, cell-pellet washing, and cell lysis by boiling, before the actual assay is performed. LFDs have been reported for the detection of many foodborne pathogens, including Escherichia coli O157:H7,55 Salmonella,54,136 and Listeria.2,106 The use of LFDs to detect Clostridium botulinum neurotoxin has also been reported.126

11.24

Immunofluorescence Assays

Immunofluorescence assays use fluorescently labeled antibodies for signal production or use enzyme labeled antibodies that produce fluorescent products upon the addition of substrate. One very simple and popular immunofluorescence assay is the 96-well microplate. The wells of the plate are first coated with capture antibody, followed by the addition of the test sample. The presence of the target antigen (i.e., microbial cell or toxin) is revealed by the

Figure 11-2. Diagram of a ‘‘dipstick’’-style lateral flow device.

|

addition of a second fluorescently labeled detection antibody, which can be measured using a fluorescence microplate reader (Figure 11-3); a labeled primary or secondary detection antibody can be used. Because the signal is proportional to the amount of target present, this assay can also be used quantitatively. Another type of assay that can be performed using the fluorescence microplate reader is the ELFA (Figure 11-3), which was mentioned in section 11.22. This assay can be considered both a type of ELISA because the antibody is labeled with an enzyme and a type of immunofluorescence assay because the product formed by the enzyme is fluorescent. Similar to the ELISA, the ELFA can be either direct if the primary detection antibody is labeled or indirect if a labeled secondary antibody is required for detection. Automated immunofluorescence assay systems based on the ELFA principle are commercially available and widely popular; the manufacturers of these instruments also usually market a wide range of foodborne pathogen test kits designed to work on their detection systems. Other platforms that can be used for immunofluorescence detection include the fluorescence microscope and the flow cytometer.20,29 The recent development of fluorescently dyed microspheres has expanded the utility of flow cytometry for foodborne pathogen detection. The ability to generate multiple fluorescently colored beads makes it possible to create a multiplex immunoassay with different target-specific antibodies attached to different colored beads. During analysis, the color of the bead is determined and serves to identify the specific target (i.e., foodborne pathogen), and a fluorescently labeled secondary antibody determines whether the target–antibody–bead complex has been formed (i.e., the foodborne pathogen was present).

Figure 11-3. Illustration contrasting simple immunofluorescence detection and enzyme-linked immunofluorescence detection assays. E 5 enzyme; P 5 fluorescent product resulting from the action of the enzyme on the substrate; S 5 substrate.

Rapid Methods for the Detection and Identification of Foodborne Pathogens

11.25

Latex Agglutination

Latex agglutination (LA) is a relatively simple immunoassay format. LA assays use small (,1 mm) latex beads that have target-specific monoclonal or polyclonal antibodies attached to the surface. LA assays are relatively simple to perform, are highly specific, and are commercially available for most foodborne pathogens. These assays do not require any specialized equipment and can be performed using a clean glass microscope slide or Petri dish. Binding between the antibody-labeled latex beads and the target cells results in the formation of a visible precipitate (i.e., agglutination) (Figure 11-4); the typical ‘‘clumping’’ reaction occurs because each latex bead is coated in antibody and can bind multiple target cells and because each cell typically has multiple surface antigens that can be bound by more than one latex bead, resulting in complex crossbridging between cells and beads. Commercially available LA test kits come with their own specific set of instructions, but they each follow the same general format. Briefly, bacterial cells ($107) are suspended in a small pool (,50 mL) of the antibody-coated latex beads on the surface of a glass slide or Petri dish. The slide is gently rocked from side to side while observing for the formation of granular clumping. Both positive and negative control pools should be included when performing the assay; commercial LA test kits typically provide a positive control (inactivated target-specific antigen), negative control beads, and latex bead dilution buffer. LA assays have several limitations. Because they rely on visible turbidity, food particulates can be confused with cell-bead clumping, thus preventing them from being used for the detection of foodborne pathogens directly in foods or directly from selective enrichments. Additionally, commercial LA test kits use specific growth conditions in order to maximize cell-surface antigen expression. Before using the assay on cells that have been grown under different conditions (e.g., using colonies lifted directly from selective agar), the assay performance should be verified. Finally, although LA assays are highly specific, they display low sensitivity and usually require overnight culture of the test organism to achieve reliable results. Despite these limitations LA assays are well suited for

Figure 11-4. Simple illustration of an agglutination reaction which results from the interactions between antigen and antibody-labeled latex microspheres.

| 129

Compendium of Methods for the Microbiological Examination of Foods |

confirmatory testing of presumptive target microorganisms and for obtaining subspecies-level information about foodborne pathogens, such as serotype.

11.26

Immunomagnetic Cell Capture

The ability to tether an antibody to a solid support while still maintaining a high affinity for its intended target has expanded the range of antibody-based detection platforms available for foodborne pathogen detection. This has also given rise to the development of technologies for capturing and concentrating target foodborne pathogens from complex food matrices or food processing/production environments3,35,43,69,130; this is known as immunomagnetic cell capture or immunomagnetic separation (IMS). There are several advantages to including an IMS-based capturing step in any foodborne pathogen surveillance protocol, including (1) removing matrix-specific interferences, (2) concentrating the target, (3) shortening the sample analysis time by reducing or eliminating the selective enrichment period, and (4) increasing the specificity of the detection step by removing similar nontarget competing microorganisms. Cell capture using IMS is relatively straightforward. Antibody-labeled paramagnetic microspheres (beads) are mixed with the analytical sample; complex food matrices may need to be first blended with an appropriate volume of buffer or microbial growth medium (e.g., selective enrichment broth) in order to ensure good dispersion of the IMS beads and to facilitate interaction between the target cells and the IMS beads. The blended test sample and the labeled IMS beads are mixed in a test tube; the antibodies bind the target cell (i.e., capture) and then a magnet is applied to the side of the tube, drawing the IMS-captured cells out of the matrix (Figure 11-5). Commercial IMS-based sample recirculation systems are also available. The sample matrix is then removed (e.g., decanted or pipetted), leaving the captured cells fixed to the side of the tube by the magnetic field. The cells can be washed, removing nontarget cells and sample debris, and prepared for downstream detection. Immunomagnetic cell capture is independent of detection and can be used for preanalytical concentration

and purification with numerous detection-based technologies, such as polymerase chain reaction (PCR),3,35,43,101 flow cytometry, 52 simple immunofluorescence, 153 or ELISA.97 The use of IMS can also aid in the recovery and isolation of foodborne pathogens from a wide variety of food matrices.144,145,146

11.27

In addition to paramagnetic particles, antibodies can also be coupled to other support matrices, such as agarose, for use in purifying and concentrating target antigens by chromatography. Although not frequently used for the isolation of foodborne pathogens, immunoaffinity chromatography is useful in purifying and concentrating a wide range of toxins for subsequent detection, such as C. botulinum neurotoxin, 46 Bacillus cereus enterotoxin, 128 Staphylococcus enterotoxin,127 and mycotoxins. A wide range of commercially available support matrices with specialized chemistries for attaching antibodies are available; available chemistries include streptavidin linkages for attaching biotinylated antibodies and N-hydroxysuccinamide (NHS) linkages which react with primary amine groups (i.e., lysine residues) on the antibody. Immunoaffinity chromatography can be done in both column and batch formats. In batch chromatography, the antibody-labeled matrix and the sample are gently mixed in an appropriate container, such as a beaker. The matrix is allowed to settle by gravity and the supernatant carefully decanted, leaving behind the target antigen still attached to the antibody matrix. In the column format the antibody-coupled matrix is packed in a flow column. The sample is applied to the top of the matrix and allowed to flow (gravity or pump) through the column. The target antigen is retained in the column and the supernatant is discarded. In either format, the captured antigen (e.g., bacterial toxin) can be washed while still bound to the antibody, and subsequently recovered.

11.3 11.31

Figure 11-5. Depiction of single-tube immunomagnetic separation assay.

130 |

Immunoaffinity Chromatographic Purification

NUCLEIC ACID AMPLIFICATION METHODS Introduction

The specificity of the nucleic acid sequence of a microorganism provides an extremely accurate target for the development of in vitro diagnostics for food safety applications. Foodborne pathogen-specific nucleic acid sequences can be amplified in vitro using a variety of techniques, which can be selected based on the goals of the end user. Nucleic acid amplification methods can be used in food microbial diagnostics to (1) determine the presence/absence of a specific organism; (2) determine the levels of a specific organism; (3) establish the genus-, species-, or subspecieslevel identification of an organism; (4) determine virulence genes; and (5) distinguish between highly similar organisms. in vitro nucleic acid amplification assays typically follow the same basic procedure. First, the region of the doublestranded deoxyribonucleic acid (dsDNA) that is to be amplified is separated into single strands, or for full-length genes into single-stranded regions. In PCR, this is accomplished by heating the sample (see section 11.33). In the case of isothermal amplification methods, this is accomplished by the use of strand-displacing DNA polymerases or other

|

Rapid Methods for the Detection and Identification of Foodborne Pathogens

strand-displacing enzymes known as helicases (see section 11.39). Next, the DNA synthesis origin is established by the binding of a short synthetic oligonucleotide known as a primer. Finally, the ‘‘primed’’ strand serves as the template and the primer itself is extended with the addition of nucleotides to the growing strand, via the action of DNA polymerase, being determined by the sequence of the template strand. Amplification is achieved by using two primers that bind opposite strands of the dsDNA and flank the region to be amplified. Both primers are then extended in opposite directions. This entire cycle of strand separation, template priming, and primer extension is repeated for a predefined number of cycles, resulting in the accumulation of amplification products.

11.32

Types of Nucleic Acid for Amplification Assays

Genomic DNA is the most commonly used sample type in PCRs for food microbiology applications. Genomic DNA has several advantages over other nucleic acid targets, such as higher specificity and sensitivity, lower susceptibility to enzymatic degradation, and higher stability. Ribonucleic acid (RNA) is also frequently used for nucleic acid amplification reactions. RNA is mostly used for gene expression analysis rather than the simple detection (presence/absence) of foodborne pathogens. However, RNA is the only sample type available for the detection of RNA viruses, such as human noroviruses. There are various types of RNA that can be applied in the detection assays. Ribosomal RNA (rRNA), which is present in large amounts (100–10,000 copies/bacterial cell) compared to genomic DNA, also used for detection assays.48 Although rRNAs have highly conserved sequences that allow them to fold and function as ribozymes in all bacteria, hypervariable regions in rRNA sequences are unique to each bacterium, which is also helpful in the identification of microbes at genus, species, and even subspecies levels. Messenger RNA (mRNA) is primarily used for quantitative gene expression analysis. In addition to genomic DNA and RNA, there are other targets, such as plasmids, which can be used in nucleic acid amplification assays but are not as common for foodborne pathogen detection and identification.

11.33

Polymerase Chain Reaction

PCR has become an important diagnostic tool in the area of food microbiology since its invention in 1983. PCR is based on enzymatic amplification of a target DNA region (called template) to generate millions of copies within 1–2 hours (Figure 11-6). Double-stranded genomic or plasmid DNA generally serves as the template, although single-stranded DNA (ssDNA) can also be amplified by PCR. The PCR procedure consists of repeated thermal cycling to achieve single-strand DNA formation (referred to as denaturation), establishment of the DNA synthesis origin (referred to as primer annealing), and finally DNA synthesis (referred to as extension/elongation). Denaturation is usually performed at temperatures ranging from 90uC–98uC and separates the double-stranded target DNA into single strands. The annealing step is usually performed at temperatures ranging from 45uC–65uC to facilitate the association of the primers with their complementary template sequences, thus forming

Figure 11-6. Steps in the polymerase chain reaction. dNTP 5 deoxynucleotide.

the initiation site for DNA synthesis. Two primers are needed that bind on opposite strands of the dsDNA and that flank the region to be amplified, so that DNA synthesis will occur in both directions. DNA synthesis (also referred to as polymerization) is performed by an enzyme called DNA polymerase. During the extension step (68uC–75uC), DNA polymerase adds nucleotide bases to the annealed primer and continues to synthesize this growing DNA strand until it reaches the end or the temperature is increased, resulting in strand denaturation. During PCR, a thermostable DNA polymerase enzyme is used owing to its ability to withstand the high temperature used during the denaturation step. Newly constructed dsDNA molecules then serve as the template for additional rounds of amplification, and the original target is exponentially amplified. Traditionally, gel electrophoresis followed by ethidium bromide staining has been used to detect the PCR amplification products. The movement of the negatively charged DNA towards the positive electrode during electrophoresis separates the DNA fragments based on their size. Ethidium bromide intercalates with DNA and fluoresces upon exposure to UV light. In gel electrophoresis, the band size of an amplified PCR product can be compared to a known DNA size standard known as a reference DNA ladder, which is run along with unknown sample(s) and a positive and negative control. Besides ethidium bromide, alternate DNA staining dyes such as Fast Blast DNA Stain (Bio-Rad Laboratories, Inc., Hercules, CA) and EZ-Vision DNA dye as loading buffer (Amresco LLC, Solon, OH) can be used. The detection of a band sequence can also be performed by blotting the gel followed by DNA probe hybridization. More details on DNA hybridization technology can be found in the previously published literature.118,151 In general, PCR assays are rapid, sensitive, and highly specific, which provides a high level of end-user confidence for pathogen detection and identification. It should be noted that PCR techniques cannot distinguish between the | 131

Compendium of Methods for the Microbiological Examination of Foods |

nucleic acid of viable and non-viable cells when using DNA as the target, as DNA is abundantly present in both dead and live cells. However, mRNA-based targets have been reported to correlate better with the viability of target cells.68

11.34

Real-Time Polymerase Chain Reaction

Real-time PCR (rtPCR) is a variation of PCR that uses a measurable fluorescence signal to monitor the level of amplification at each cycle throughout the entire PCR analysis. rtPCR circumvents the need for gel electrophoresis, which adds additional time to the analysis and can generate large volumes of hazardous waste (e.g., ethidium bromide). The success of rtPCR has been driven by the development and refinement of optics that could be incorporated into automated thermal cyclers, and by the development and refinement of fluorescence chemistries that could be incorporated into the reaction to a produce a measurable signal. In addition to diagnostic applications, rtPCR can be used for quantitative assessments such as gene expression analyses or bacterial population determination. Although there are different fluorescence chemistries available, there are some fundamental principles that are constant for all rtPCR protocols. During each cycle of the PCR analysis the fluorescence intensity is measured. There is typically some slight level of background fluorescence, which changes very little during the initial cycles; this determines the baseline for the amplification plot. The increase in the fluorescence signal above the baseline serves as a measurement of the accumulation of target amplicons (an amplicon is the fragment of DNA generated during the PCR). The threshold level is an arbitrary value determined by the end user that is set above the baseline, typically within the region of exponential increase in the accumulation of amplicons. Qualitative assessment (i.e., positive or negative) of a sample is determined by the ability of the fluorescence signal to cross the threshold within a set number of amplification cycles (frequently 40 cycles for most food pathogen detection protocols, although this number can vary); the cycle at which the fluorescence signal crosses the threshold is called the ‘‘come-up time’’ or Ct value. Automated rtPCR thermal cyclers automatically collect the data, determine the Ct value at the threshold level specified by the user, and designate the sample as positive or negative. rtPCR also affords the ability to analyze samples in a quantitative manner. By including standards (i.e., samples of known cell populations) in the rtPCR analysis, the correlation between target cell population and Ct value can be determined (i.e., the generation of a standard curve) and the Ct values from unknown samples can be converted to cell populations by comparison to that standard curve. Many modern instruments will determine the Ct value, generate the standard curve, and calculate the quantity of target(s) in unknown samples automatically. In addition to the qualitative assessment of the presence/absence of a target foodborne pathogen and the quantitative determination of the levels of target foodborne pathogens, rtPCR can also be used to determine the effects of food matrix composition (pH, salinity, water activity, etc.) and environmental stresses (e.g., temperature) on the physiological response of 132 |

foodborne pathogens. This is accomplished by measuring the levels of specific mRNA transcripts using a specialized rtPCR technique called reverse transcriptase PCR (RTPCR), which is covered in more detail in section 11.3512. rtPCR has become one of the primary techniques for the detection and quantification of foodborne pathogens in diagnostic laboratories. Some of the reasons for the success of rtPCR are as follows: (1) the assays are reproducible; (2) there is low propensity for non-specific amplification; (3) post-amplification analysis is not needed; and (4) userfriendly testing kits and supplies are commercially available.

11.341 Real-Time PCR Chemistries The various chemistries available for rtPCR can be classified as either dye binding–based or probe-based, and both types are available commercially from several different manufacturers. Dye-binding rtPCR relies on the non-sequence-specific binding of dyes to the minor groove or intercalation between the base pairs of dsDNA, resulting in increased fluorescence over the non-bound dye. Probebased rtPCR requires, in addition to a pair of primers, a third oligonucleotide (also known as a probe) that has been labeled with a fluorescent molecule and a quencher molecule, in such a way that the fluorescence is quenched until the probe is bound to the amplicon. The amplified product is detected by a fluorescence signal generated by one of these chemistries, which is measured at each cycle during the rtPCR analysis. There are obvious benefits and shortcomings to both methods. rtPCR based on dye binding can be used to perform both qualitative and quantitative sample analysis. Dye-binding assays are considerably less expensive than probe-based assays owing to the cost of labeling the probe. Frequently, DNA intercalating dyes can be added to existing PCR protocols, making them rtPCR assays; however, these assays will require further optimization to minimize the accumulation of non-specific products, which frequently appear late in the assay. A major disadvantage to dye-binding fluorescence rtPCR is that the DNA intercalating dyes bind non-specifically, so that both target amplicons and unwanted PCR products (e.g., primer dimers or non-specific amplicons resulting from mispriming) are measured, thereby affecting the results of the assay. Like dye-binding rtPCR assays, probebased assays can be used either qualitatively or quantitatively. The primary disadvantage to probe-based rtPCR is the initial cost of the fluorescently labeled probe. 11.3411 Dye-Binding rtPCR Assays. Intercalating and minor groove dye-binding chemistries produce higher fluorescence signals upon binding dsDNA than when free in solution. For example, SYBR Green, the most commonly used dye, fluoresces several times brighter when it binds dsDNA than as a free dye. With each successive PCR cycle, the fluorescence signal increases in proportion to the increase in accumulated dsDNA product (amplicon). SYBR Green can also bind to non-specific PCR products, hence this method may lack the desired degree of specificity. This can often result in overestimation of target. Highresolution melting curve analysis can help in differentiating the non-specific products that typically generate melting curve peaks different than those from target amplicons.

|

Optimal primer design can help prevent the formation of these non-specific products. SYBR Green can be employed in the multiplex rtPCR by using primers specific to each target. These primers allow the amplification of PCR products having distinct melting temperature values, resulting in the formation of distinct peaks representing the different targets. The use and application of dye binding–based rtPCR is widely reported for the detection of foodborne pathogens.24,30 Besides SYBR Green, dyes such as EvaGreen, SYBR GreenER, SYTO 9, and LCGreen have been reported to generate higher fluorescence readouts and minimize the detection of primerdimers.21 Dye binding-based rtPCR assays display a high level of sensitivity, as multiple intercalating dye molecules can bind to a single amplified DNA amplicon. Interestingly, the signal resulting from dye binding is proportional to the mass of the product. Thus, a longer PCR product will generate higher readout signal than a shorter product, assuming similar PCR amplification efficiencies for both sizes of target.

11.3412 TaqMan Probes. TaqMan assay chemistry is based on a specifically designed probe that has a fluorescent dye (reporter fluorophore) at one end (59) and a quencher moiety at the other end (39). The TaqMan probe is designed to bind within the target gene region being amplified. In its unbound state, fluorescence emission from the reporter fluorophore is absorbed by the quencher moiety in a process known as fluorescence resonance energy transfer (FRET). During the annealing step of the PCR amplification, both the primers and the probe bind to the target region. As the DNA polymerase extends the primer and approaches the annealed probe, it cleaves off the reporter fluorophore from the probe, which dissociates into solution via the 59-39 exonuclease activity of the polymerase. Because the reporter fluorophore and the quencher are no longer in close proximity, FRET ceases and the result is increased fluorescence. The fluorescence signal increases exponentially with each cycle owing to the exponential accumulation of amplified product. TaqMan probes are also referred to as hydrolysis probes, and the TaqMan-based rtPCR is sometime referred to as a 59 nuclease assay. TaqMan has the ability to detect multiple targets within the same reaction simultaneously by designing separate probes with spectrally unique fluorophores and quenchers for each target. TaqMan assays are considered to have a high degree of specificity, but it can be complicated to design and expensive to synthesize multiple target-specific probes. TaqMan assays are widely applied for pathogen detection in food and environmental samples.30,135 Modified probes with a conjugated minor groove-binding protein (MGB) have been reported to further enhance the specificity of TaqMan assays. TaqMan MGB probes typically form extremely stable duplexes with ssDNA targets; these complexes have a higher melting temperature (Tm), resulting in increased specificity.77 11.3413 Molecular Beacons. Molecular beacon probes consist of a target-recognition region (,15–25 bases) that is flanked on each side by complementary DNA sequences. The 59 end of the probe is labeled with a fluorescence reporter moiety and the 39 end with a fluorescence

Rapid Methods for the Detection and Identification of Foodborne Pathogens

quencher. The association of the complementary DNA sequences at the terminal ends of the probe result in a hairpin loop structure, bringing the fluorophore reporter and quencher moieties into close proximity and resulting in FRET. When the molecular beacon probe binds to a specific amplicon sequence the hairpin loop dissociates, increasing the distance between the reporter and quencher and resulting in increased fluorescence in the absence of FRET. Molecular beacons remain intact during the amplification reaction and rebind to the target in every cycle for signal measurement. Molecular beacons are widely used in the detection of pathogens in food.30,82 Similar to TaqMan assays, molecular beacon assays can be designed to detect more than one target within the same assay and can be used either qualitatively or quantitatively.

11.3414 Scorpions. Scorpion primers combine the primer and the probe into a single molecule so that the resulting fluorescence signal is unimolecular. Scorpion primers can be thought of as having both a primer element and a reporter element, despite being single molecules. The primer element is complementary to the target DNA being amplified. On the 59 end of the primer element there is a polymerase blocker which prevents DNA polymerization through the reporter region, which would result in a doublestrand probe with the loss of the hairpin loop and an increase in fluorescence intensity. Immediately upstream from the blocking group is the probe element, which begins with a 39 end fluorescence quencher. Adjacent to the fluorescence quencher in the 59 direction is a stem sequence that hybridizes internally with its complementary stem sequence located further upstream. Immediately following the stem sequence is a loop sequence, which is the reverse complement to the target DNA being amplified and is important for the mechanism of action of the assay. Immediately upstream of the loop region is the second stem region, followed by the fluorescent reporter molecule located at the 59 end of the probe region. During primer extension, the probe becomes part of the strand being synthesized. During the next PCR cycle, the loop region denatures and forms an intramolecular hybridization with the previously synthesized target DNA via the loop region sequence, resulting in a separation of reporter and quencher and an overall increase in fluorescence intensity. Scorpion probe technology has been reported for the detection of pathogens in food.129 11.3415 Hybridization Probes. Hybridization probes comprise a pair of DNA probes designed to hybridize adjacent to each other on a target sequence. Both probes have a fluorophore dye at the end; the dye ends of the probes face each other and interact with one another via FRET. Upon excitation the shorter wavelength–absorbing dye transfers its energy to the longer wavelength–absorbing dye of the adjacent probe, which then emits a detectable signal. The amount of signal produced is proportional to the level of accumulated amplicon. The design of hybridization probes is critical to make sure they hybridize adjacent to each other at the appropriate distance for efficient resonance energy transfer to produce a detectable signal. Hybridization probes are used in commercially available assays for the | 133

Compendium of Methods for the Microbiological Examination of Foods |

detection of pathogens in food.11 Some of the hybridization probe-based rtPCR assays combine the Ct value and melting curve analysis for the amplification reactions, thereby providing additional specificity to the assay. A melting curve helps to accurately determine the presence of target in a sample. A high fluorescence is observed in a positively amplified sample, which drops when the probe dissociates from the amplicon at the Tm of a probe set. This variation in fluorescence at different temperatures is recorded as a melting curve. A derivation of this curve appears as a melting peak during the melt analysis.

11.3416 Emerging rtPCR Chemistries. There are many emerging rtPCR chemistries that do not rely on a separate probe and are based on primer sets with fluorogenic chemistries. For example, the Plexor Primers utilize two nucleotide modifications called isoguanine (iso-dG) and 5methylisocytosine (iso-dC). These modified bases pair only with each other to form a unique base pair upon incorporation into dsDNA. Plexor primer pairs consist of one PCR primer having an iso-dC residue and a fluorescence label at the 59 end. The second PCR primer is an unlabeled standard oligonucleotide. Unlike the previously mentioned rtPCR assays, the amplification progress is measured not by the increase in fluorescence but rather by the decrease in fluorescence. In addition to the standard nucleotides, the PCR mix also contains a dabcyl-modified 29-deoxyisoguanosine (dabcyldiGTP) that acts as a fluorescence quencher. During the amplification cycle, the fluorescent label and the iso-dC are incorporated into the amplicon. Strand synthesis in the reverse reaction (from the unlabeled primer) results in the incorporation on the dabcyl-diGTP at the site containing the iso-dC nucleotide, placing the fluorescence quencher (dabcyl) in close proximity to the reporter fluorophore and resulting in a decrease in the fluorescence signal. Another technology, the Light Upon Excitation (LUX) primer pairs, consists of a primer that has a reporter fluorophore at its 39 end and a second corresponding unlabeled primer. The 59 end of the labeled primer includes a 4–6 nucleotide sequence tail that is complementary to the 39 end of the primer. This forms a hairpin structure, resulting in quenching of the fluorophore; LUX primers do not require an additional quencher moiety. During PCR, the incorporation of the primer into dsDNA leads to loss of quenching, resulting in a several-fold increase in fluorescence signal. The application of these emerging PCR chemistries has been reported in food safety applications.16 Overall, probe-based chemistries are more specific than DNA-binding dyes because a prerequisite for the signal production in probe-based chemistries is binding of the probe with target-specific complementary sequences. This minimizes the chances of non-specific amplification products contributing to the signal, as is the case with intercalating dyes. It is also well established that probes function better in quantitative assays than intercalating dyes do, owing to the fact that probe-based chemistries have less risk of producing signal from amplification byproducts or non-specific products. However, probes are expensive to synthesize and require expensive hardware and software to perform the assay and interpret the data. 134 |

11.35 11.351

Different Types of Polymerase Chain Reaction Approaches Polymerase Chain Reaction Approaches for Varying Numbers and Types of Targets

11.3511 Multiplex Polymerase Chain Reaction. Multiplex PCR refers to the detection of multiple targets in a single reaction. Multiple target sequences are amplified using different sets of primers specific to each target. In multiplex rtPCR, primers and probes specific to each target are used to amplify multiple nucleic acid targets within a single reaction. Multiplex rtPCR is widely applied in food applications.30 Detection of non-O157 Shiga toxin-producing E. coli (STEC) targeting Shiga toxin types 1 and/or 2 (stx 1 and/or 2), intimin (eae), and serogroups-specific O genes is a good example of a multiplex reaction currently used by the food industry (USDA-MLG 5B.04).138 The appropriate designing of primers and probes is critical for the success of multiplex rtPCR. The annealing temperatures of the primers and the probes should be close. In the case of traditional PCR, the primers should be designed so that the resulting amplicons are of sufficiently distinct sizes to allow identification by agarose gel electrophoresis. If multiple variants of a gene are reported, then degenerative primers can be used to amplify the variant gene targets in multiplex reactions.116 Degenerative primers can be easily designed by aligning the gene sequences found in GenBank. These primers are a mixture of primers corresponding to all permutations of variant types for a gene target. The primers in a multiplex reaction compete for the amplification, hence appropriate reaction optimization is essential to avoid the competition among the primers that might result in poor amplification of some targets. 11.3512 Reverse Transcription PCR. RNA can be used as the initial template in place of DNA in PCRs. In RTPCR, RNA is initially reverse transcribed to its complementary DNA (cDNA) using an enzyme called reverse transcriptase, and the cDNA is then amplified using a DNA polymerase enzyme in a standard or real-time PCR. This two-step process can be combined into one step using the enzyme Tth polymerase, which can perform both reverse transcription and DNA polymerization reactions. RT-PCR is commonly used for gene expression analysis. Other applications include determination of transcription start and termination sites and the location of exons and introns in a gene sequence. The use of RT-PCR is reported in the field of food microbiology for pathogen detection152 and gene expression analysis. RT-PCR is also used for detection of RNA viruses such as human noroviruses in food.48,142 11.3513 Colony PCR. In colony PCR a small quantity of a microbial colony is directly transferred to a PCR tube containing a reaction mix. Intact colonies are then lyzed to release the DNA. Approaches used to lyze the cells include extended denaturation time at 95uC or a short denaturation step at 100uC, depending on the thermal stability of the polymerase used in the reaction. Although colony PCR is mostly used in molecular biology for the screening of transformants for the presence of target plasmids or

|

vectors, the application of colony PCR to the detection of pathogens such as Salmonella has been reported.107

11.352

PCR Approaches to Enhance the Reaction Specificity

11.3521 Nested PCR. Nested PCR employs two sets of PCR primers that are used sequentially in the amplification process. This imparts better specificity and sensitivity to the detection assay than traditional PCR. The first primer set amplifies a target sequence that is subsequently used as the template for the second set of primers; the second primer set amplifies a region within the first amplicon. The second amplification can only occur if the first amplification was successful. A nested PCR requires the opening of the first reaction tube to add its contents to another reaction tube containing the reaction mix for the second reaction. This may result in a higher risk of contamination of PCRs. However, some commercial assays (e.g., FilmArray, BioFire Diagnostics, Inc., Salt Lake City, UT) based on nested PCR are fully automated and do not require manual opening of the reaction tube. Nested PCR has been reported for the detection of pathogens in food.48,141

Rapid Methods for the Detection and Identification of Foodborne Pathogens

11.3532 Droplet Digital PCR. Droplet digital PCR (ddPCR) works on the principle of separating the sample into several thousand tiny droplets within a dispersed phase of emulsion in microwell plates, capillaries, or other droplet generator chambers. Thus, the PCR is carried out both individually and simultaneously in a large number of partitions in a pool of tiny droplets. This allows more sensitive and accurate measurement of the amount of target nucleic acid in a sample. The partitioning of sample into a droplet emulsion follows the Poisson distribution, designating each drop as either a negative (0) or a positive (1) reaction based on the absence or presence of signals, respectively. Thus, each droplet provides an independent digital measurement. The fraction of positive droplets provides the measure of initial amount of target. As ddPCR does not rely on the number of amplification cycles to determine the initial sample amount, it provides absolute quantification with no further need for a standard curve. Owing to its high accuracy, ddPCR is ideal for applications such as determining copy number variants, point mutations, rare sequence detection, gene expression analysis, and clonal amplification of samples for next-generation sequencing.104,114 11.36

11.3522 Touch-Down PCR. In touch-down PCR, the initial higher annealing temperature (usually 3uC–5uC above the standard Tm of primers) is used in the early cycles to impart greater specificity for primer binding. This is followed by a lower annealing temperature (usually 3uC–5uC below the standard Tm of primers) to allow more efficient amplification towards the end of the PCR. Touch-down PCR ensures specific primer annealing and reduced PCR artifacts.28,83 11.3523 Hot-Start PCR. The reaction components in hot-start PCR are heated to 95uC before the polymerase is added, to avoid the formation of non-specific amplification products that can occur at lower temperatures. Hot-start PCR has been recommended to reduce non-specific priming. 11.353

Quantitative PCR Approaches

11.3531 Quantitative rtPCR. qPCR is used to measure the amount of target nucleic acid present in a sample. In food safety applications, qPCR can be used to measure the levels of a target pathogen present in a sample, based on the levels of DNA that are determined. The target nucleic acid amount in a PCR is determined by the fluorescence signal (probe fluorophore or fluorescent dye) resulting from the accumulation of amplified target. Target quantification using qPCR is discussed in section 11.372. If the PCR efficiency is within acceptable ranges, then Ct values will accurately reflect the initial amount of target. Ct numbers are inversely related to the amount of target nucleic acid in the sample. Thus the greater the initial amount of target nucleic acid in a sample, the quicker it will reach the cycle threshold and the lower the Ct value. Thermal cyclers capable of conducting rtPCR can be used to measure the amount of amplified product as the amplification progresses. qPCR has been extensively described in several reviews.51,115

Considerations for the Development of Nucleic Acid Amplification Assays

11.361 Sample Preparation Reliable sample preparation is important in maximizing the sensitivity of any PCR assay in order to detect foodborne pathogens, which are typically present at levels of ,1 CFU/g in large volumes of food samples (10–375 g).30 To achieve this, an appropriate cultural enrichment of the food sample is required. Factors such as lag phase and the growth rate of target microorganisms, the presence of stressed/injured target cells, the physicochemical properties of the food matrix, and levels and types of background microflora play a significant role in the cultural enrichment of food samples to achieve the required target cell biomass. Selective enrichment of the food sample is typically needed to ensure that sufficient levels of DNA/RNA are recovered for PCRs in pathogen detection. The quality of the nucleic acid sample is also critical for the success of the PCR, specifically for the sensitivity and reproducibility of the assay. There are several components in food matrices, including gelatin, fats, proteins, divalent cations, and phenolic compounds such as humic acid in soil, and hemoglobin and lactoferrin in blood, which can adversely affect the quality of extracted nucleic acid. The inclusion of wash steps following the recovery of cells from the selective enrichment can help reduce the effects of these PCR inhibitors. The carryover of inhibitory substances from the sample can inhibit the amplification reaction, either by inhibiting the activity of DNA polymerase or by interfering with the annealing of primers/probes. To achieve high-quality target DNA/RNA, preanalytical sample preparation strategies such as immunomagnetic separation can be applied to selectively capture and concentrate target microbial cells from selective enrichments, or in some cases directly from food products prior to nucleic acid extraction. There are various non-commercial/commercial methods for the extraction of nucleic acids. Lysis strategies | 135

Compendium of Methods for the Microbiological Examination of Foods |

including enzymatic lysis using a combination of lysozyme and proteinase K; detergent-based lysis using CHAPS, SDS, or Triton X; freezing and thawing; physical disruption, such as sonication or bead milling; phenol–chloroform extraction; and heating/boiling have all been successfully used to facilitate cell lysis and release DNA. It should be noted, however, that some of the extraction methods might not be efficient enough to recover the required amount and quality of nucleic acids, particularly when the target cells are Gram-positive bacteria and fungi. Thus the inclusion of controls such as a DNA extraction control (unrelated purified DNA oligonucleotide) inoculated into the sample before DNA extraction could be a good indicator of the quality and quantity of the extracted DNA. If nucleic acid extraction is performed using chemical solutions residual chemicals can affect the quality of the extracted nucleic acid. It must be noted that the extraction method should avoid the shearing of target nucleic acid. RNA extraction methods are more tedious in general than DNA extraction methods. RNA samples are less stable and can degrade quickly, and so they should be promptly analyzed. An RNA integrity number (RIN) can be calculated and used as an indicator to determine the suitability of RNA samples for further analysis.124

11.362 Controls for PCRs For analytical sample analysis, the set-up of PCRs should be performed in a designated isolated area dedicated to DNA extraction, reaction set-up, and thermal cycling in a unidirectional set-up. This should be done away from sample preparation areas (e.g., sample weighing area, enrichment set-up, and incubation areas). Cross-contamination of DNA is a major challenge for laboratories involved in PCR analysis, and appropriate measures should be taken to prevent its effects. Cross-contamination can often lead to false positive results, and so a negative DNA extraction control should be included in the PCR. In addition to a negative control, a positive control, non-template control, and control for environmental contamination should also be considered during the development and set-up of PCR assays.115 A PCR is susceptible to sample matrix-associated inhibitory compounds and other factors, such as thermal cycler malfunction, incorrect reaction mixture, and poor DNA polymerase activity. Any of these can result in amplification failure or reduced sensitivity and specificity. Internal amplification controls (IACs) can be used in PCRs to detect amplification failures resulting in false negative results.94,95 An IAC is usually a non-target DNA sequence that can be co-amplified simultaneously with the target.57 The IAC is usually amplified in PCRs irrespective of the presence or absence of the target. The absence of IAC amplification is indicative of improper reactions in cases where the target amplification is also negative. However, the absence of IAC amplification signal in a situation where target amplification is positive is not necessarily indicative of an improper PCR. The approaches for IAC design include (1) heterologous IAC that is co-amplified with the target using its own set of primers or (2) homologous IAC that is amplified with the same primers used for the target amplification. Several 136 |

practical considerations in designing an IAC have been recommended, including the following30,57: (1) the IAC and the target nucleic acid should share the common primerbinding sites; (2) IAC amplicons should be distinguished from target amplicons, using a separate sequence-dependent probe; and (3) the source of the IAC should be highly purified nucleic acid. The homologous IAC design is a preferred choice. This approach uses probe-based fluorescence chemistries such that the IAC probe and the target probe fluoresce at different wavelengths.93 It is important to optimize the IAC co-amplification reaction with the target to avoid the loss of detection sensitivity, as inherent competition between the target and the IAC for the amplification is obvious. Several homologous IACs have been designed for inclusion in probe-based rtPCR assays for the detection of foodborne pathogens.30 SYBR Green chemistry can also be employed for the PCRs with IAC by optimizing the melting curve analysis to discriminate between target and IAC sequences. Another approach to detect the inhibition due to matrixassociated PCR inhibitors is the use of a ‘‘spike’’ control. DNA of known quality and quantity can be added to the food matrix being tested, which can be used as a control to test for any potential amplification inhibition during PCR owing to the presence of matrix-associated PCR inhibitors. These so called external amplification controls (EACs) are also helpful in determining the malfunction of thermal cyclers and reaction mixture ingredients.

11.363 Instrument Specifications Temperature uniformity of the thermal cycling block is a key factor for the precision of the temperature cycle, which eventually affects the rate and efficiency of PCRs. The ramp rate (average ramp rate) and settling time for the temperature uniformity of a cycler are also important, as these are factors in determining the total time to finish a PCR run, besides other factors such as the quality of reaction components and thermal cycling protocol. The number of optical channels needed to read different wavelengths is also critical for multiplexing capability. Various features, such as temperature gradient option, broader linear dynamic range for the analysis, open system to perform customized reactions, high-throughput sample analysis capability ($96 samples), ability to connect to laboratory information management systems (LIMS)— preferably wirelessly—and connectivity to cloud data storage, and ease of performing the instrument quality check, are other key factors to be considered for quality instrument performance. 11.364 Reaction Performance The most important factors in PCRs are specificity and sensitivity. PCRs are susceptible to food matrix-based inhibitors, so a robust assay is always a need. Appropriate sample preparation should be adopted to remove matrixspecific effects and reach the desired level of assay sensitivity. The assay specificity is dependent on gene target selection.121 Highly specific candidate gene(s) should be used as assay target(s). Abundant microbial genome sequence information is now publicly available and can be utilized to develop genus-, species-, subspecies-, and serogroup-specific assays.

|

Other factors, including low background fluorescence, a steep increase in fluorescence, high amplification efficiency, and a high-level plateau, are indicators of the fidelity of PCRs.21 Standard curve analysis using the pure target should result in a high coefficient of determination between the Ct values and the concentration of target DNA. The correlation linearity (ideally 6–8 logs) between Ct values and DNA concentrations demonstrates a better quantitative performance of an assay. The amplification specificity of the reaction should be validated by running a dissociation curve analysis.115 The detection chemistry used is also critical, as it can affect the ability to perform multiplex reactions. For example, TaqMan and hybridization probes are more accurate, specific, and easier for the development of multiplex reactions than dye-based amplification assays.30

11.37

Application of Nucleic Acid Amplification Methods

11.371 Detection of Microorganisms A common use for nucleic acid amplification technologies is the detection of pathogenic and non-pathogenic microorganisms in food, feed, cosmetics, and environmental samples. Common gene targets for the detection of foodborne pathogens are listed in Table 11-1. In addition to 16S rDNA, 23S rDNA, and 16/23S rDNA intergenic regions, other gene targets have been reported and used for the nucleic acidbased detection of foodborne pathogens, including Listeria spp., 84,147 pathogenic E. coli, 9 Salmonella spp., 30 and Campylobacter spp.30 Selected examples of validated nucleic acid amplification-based methods for the detection of foodborne pathogens are given in Tables 11-2 to 11-5. 11.372 Quantification of Target rtPCR is currently the most popular technique for quantifying foodborne pathogens.51a,140 The two most common approaches for qPCR include (1) absolute quantification (i.e., the standard curve method); and (2) relative quantification (i.e., the comparative Ct method). 11.3721 Absolute Quantification. In the absolute quantification approach a standard curve is constructed using known concentrations of DNA/RNA and their corresponding Ct values resulting from rtPCR. The standard curve is then used to estimate the concentrations of DNA/RNA based on the resulting Ct values of the unknown samples.94,148 Further, the concentration of DNA/RNA quantified using the standard curve can be correlated with approximate CFU equivalents.63 The CFU equivalents obtained using this approach can provide an indirect count of the target microorganism. It must be noted that compared to DNA, the construction of an RNA standard curve is tedious and time-consuming as it involves the construction of cDNA that has to be transcribed in vitro into the RNA standard and accurately quantified. The quality and stability of RNA also play a major role in the construction of an RNA standard curve. RNA standards can be used to generate the absolute copy number data from unknown samples. Other nucleic acids that are used to construct the standard curve include purified plasmid dsDNA, in vitro–generated ssDNA, and

Rapid Methods for the Detection and Identification of Foodborne Pathogens

cDNA. Because absolute quantification requires a similar amplification efficiency for both the sample and the standard, the standard must be selected with care.

11.3722 Relative Quantification. In relative quantification (also referred to as the comparative Ct method), quantitative Ct value comparisons are performed between the target samples and a calibration control, for example RNA from an untreated sample and a treated sample. The Ct values of both the target samples and control are normalized to an appropriate endogenous housekeeping gene. The comparative Ct method is also known as the DDCt method. Amplification efficiency is a critical factor to be considered when performing comparative quantification. Various methods for the relative quantification of mRNA using RT-PCR can be found in a published review article by Cikos et al.18 11.373 Identification of Microorganisms Selective amplification of rDNA is widely used for sequencing for subsequent taxonomy and phylogenetic applications.73 Ribosomal targets are widely used, as they are abundantly present in microbial cells. There are several PCR assays that use 16S rDNA, 23S rDNA, and internal transcribed spacer (ITS) region-specific primers for genusand species-specific identification. In many cases where the 16S rDNA sequence is highly conserved and cannot discriminate closely related bacterial populations, the 23S rDNA sequence is used as an alternate target. The ITS region is widely used for sequencing fungi for species- and subspecies-level discrimination. The ITS region has a higher degree of variations than other generic regions of rDNA. ITS1 and ITS4 primers, along with several taxonspecific primers, are widely applied for the specific amplification of fungal sequences. There are many other applications of nucleic acid amplification technology, such as subtyping of pathogenic strains, sample preparation for sequencing, etc., which are outside of the scope of this chapter. 11.38

Selected Relevant Standards for Nucleic Acid Amplification Assays

ISO 22174:2005. Microbiology of food and animal feeding stuffs—Polymerase chain reaction (PCR) for the detection of food-borne pathogens—General requirements and definitions. This ISO standard was established for foodborne pathogens isolated from food and feed matrices, but is also applicable to other matrices (e.g., environmental samples) and for the detection of non-pathogenic microorganisms. ISO/TS 20836:2005. Microbiology of food and animal feeding stuffs—Polymerase chain reaction (PCR) for the detection of food-borne pathogens—Performance testing for thermocyclers. This standard provides basic requirements for the installation, performance, and maintenance of thermal cyclers. ISO 20837:2006. Microbiology of food and animal feeding stuffs—Polymerase chain reaction (PCR) for the detection of food-borne pathogens—Requirements for sample preparation for qualitative detection. This standard was established for food matrices but can be also applied to feed and agriculture/environmental matrices with some adaptations, if necessary. | 137

Compendium of Methods for the Microbiological Examination of Foods |

Table 11-1. Selected Gene Targets Used as Markers in the Nucleic Acid Amplification-Based Assays for Foodborne Pathogens Markers Including Virulence Factors

Pathogenic E. coli stx1 stx2 eae Wzx EHEC-hlyA (ehxA) espP katP cdt efa cnf1 and cnf2 fliC iha nleA-F irp-2 fyuA Salmonella fimC ttrRSBCA locus invA stn HindIII oriC ompC gene sipB/sipC hilA Campylobacter spp. including C. jejuni hipO ORF-C sequence cadF glyA ceuE ccoN lpxA flaA and flaB aspA gene L. monocytogenes ssrA gene hly iap plcA and plcB inlA and inlB actA prfA clpE lma/dth18 Listeria spp. liv22-228 lse24-315 lin0464 lin2483

Description

Reference

stx1 and its variants stx2 and its variant Intimin O-antigen-flippase EHEC hemolysin/enterohemolysin Serine protease Catalase Cytolethal distending toxin EHEC factor of adherence Cytotoxic necrotizing factor type 1 or type 2 Flagellar H7 gene IrgA homologue adhesin Non-LEE-encoded (Nle) effector proteins Iron-repressible protein 2 Yersiniabactin receptor

113 113 113 23 113 113 14 9 42 110 44 61 120 6 6

Type 1 fimbriae Tetrathionate respiration Invasion A Salmonella enterotoxin gene Fragments within a 1.8 kb Hindiii DNA sequence Origin of replication of Salmonella chromosome Outer membrane protein osmoporin C Encoding Salmonella invasion proteins Salmonella pathogenicity island 1 (SPI 1)

125 94 51 103 149 40 78 32 112

Hippuricase C. jejuni-specific region of ORF-C sequence Fibronectin-binding protein Serine hydroxymethyltransferase Siderophore transport Cytochrome c oxidase Lipid A gene Flagellin Aspartate ammonia-lyase

56 123 74 60 47 133 70 71,90 75

ssrA gene that codes for tmRNA Hemolysin listeriolysin O Invasion-associated surface protein p60 (common and variable regions between Listeria spp.) Encoding proteins PI-PLC and PC-PLC, respectively Internalin A and internalin B, respectively Encodes ActA protein Encodes transcriptional regulator PrfA protein Clp ATPase LmA antigen/delayed-type hypersensitivity protein

109 101 15,72

Putative Putative Putative Putative

N-acetylmuramidase (L. ivanovii) internalin (L. seeligeri) transcriptional regulator (L. innocua) transporter (L. innocua)

143 64 92 147 143 62 87 89 85 119 (continued on next page)

138 |

|

Rapid Methods for the Detection and Identification of Foodborne Pathogens

Table 11-1. (continued ) Markers Including Virulence Factors

Description

Reference

lwe7-571 lgr20-246

Putative phosphotransferase system enzyme IIBC (L. welshimeri) Putative oxidoreductase (L. grayi)

86 88

ISO 20838:2006. Microbiology of food and animal feeding stuffs—Polymerase chain reaction (PCR) for the detection of foodborne pathogens—Requirements for amplification and detection for qualitative methods. This standard was established for foodborne pathogens in or isolated from food and feed matrices but can also be applied to other matrices, for example environmental samples, or for the detection of other microorganisms under investigation. ISO 22118:2011. Microbiology of food and animal feeding stuffs—Polymerase chain reaction (PCR) for the detection and quantification of food-borne pathogens—Performance characteristics. This standard provides minimum requirements of performance characteristics for the detection of nucleic acid sequences (DNA or RNA) by molecular methods. Besides foodstuff, this standard can be also helpful for environmental and feed stuff samples. ISO 22119:2011. Microbiology of food and animal feeding stuffs—Real-time polymerase chain reaction (PCR) for the detection of food-borne pathogens—General requirements and definitions. Besides food stuffs, this standard can also be applicable to environmental and feed stuffs. AFNOR XP V03 044:2008. Intralaboratory validation criteria for the methods of detection and quantification of specific nucleic acid sequences. This standard provides intralaboratory validation criteria for the methods of detection and quantification of specific nucleic acid sequences.

11.39

Isothermal Amplification Technologies

Isothermal amplification is fundamentally different from PCR in that amplification is performed at a single temperature. Instead of requiring high heat (typically 95uC–98uC) to achieve single-stranded DNA, isothermal amplification relies on strand-displacing enzymes such as Bst DNA polymerase or other accessory enzymes and proteins, such as helicases and single-strand DNA-binding proteins. The three most popular methods of isothermal amplification used in the detection of foodborne pathogens are loop-mediated isothermal amplification (LAMP), strand-displacement amplification (SDA), and transcription-mediated amplification (TMA), and these are discussed below.

11.391 Loop-Mediated Isothermal Amplification LAMP employs two to three sets of primers that recognize six to eight distinct regions of target DNA and a polymerase having both strand displacement activity and replication activity. A strand-displacing DNA polymerase initiates synthesis and two of the primers form loop structures, facilitating subsequent rounds of amplification. Magnesium pyrophosphate is abundantly produced, forming a precipitate that can be visualized by eye. Alternatively, intercalating fluorescent dyes can be used to generate a fluorescence signal from the amplified target DNA. LAMP does not require a thermocycler, as amplification is performed at a constant

temperature (60uC–65uC) that can be achieved using a heating block or a water bath. LAMP can also be performed quantitatively. LAMP can be combined with a reversetranscription step to amplify RNA samples. The LAMP technology has been well reviewed by Niessen et al.108 and is widely reported for the detection of pathogens in food.49,108,152

11.392 Strand Displacement Amplification SDA employs a strand-displacing polymerase and a recognition sequence-specific nicking endonuclease to amplify a DNA target at a constant temperature of approximately 55uC–59uC. The SDA technique has essentially two segments: the first is the generation of a modified template that will be used in the second segment, which is the actual amplification. The modification of the template that occurs in the first segment is the incorporation of an endonuclease restriction site that will be used by the nicking enzyme in subsequent rounds of amplification. During the amplification segment, the strand-displacing DNA polymerase initiates synthesis at a nick created by the activity of the nicking endonuclease on the modified template. As the new strand is being synthesized, the old strand is being displaced. Amplification primers are also included in the reaction that binds to the displaced strands, initiating polymerization and rendering double-stranded products. The restriction site is regenerated during each cycle and is repeatedly nicked by the endonuclease and restored to the newly synthesized strand by the polymerase. SDA is rapid and does not require a thermocycler. Nicking enzyme amplification reaction (NEAR) is another isothermal amplification which utilizes technology similar to SDA. The assays based on NEAR technology are commercially available and have been reported for the detection of foodborne pathogens.105 11.393 Transcription-Mediated Amplification TMA is a transcription-based amplification that employs the two enzymes reverse transcriptase and RNA polymerase. In the first enzymatic reaction, reverse transcriptase creates a dsDNA copy from an RNA sample. In the second enzymatic reaction, an RNA polymerase makes several copies of the complementary RNA sequence from the dsDNA template.53,91 Assays using TMA are commercially available for the detection of foodborne pathogens.53 11.4

11.41

MATRIX-ASSISTED LASER DESORPTION/ IONIZATION—TIME-OF-FLIGHT MASS SPECTROMETRY Introduction

Matrix-assisted laser desorption/ionization—time-of-flight mass spectrometry (MALDI-TOF MS) is one of the newest methods used in diagnostic microbiology laboratories to | 139

Compendium of Methods for the Microbiological Examination of Foods |

Table 11-2. Selected Nucleic Acid Amplification-Based Methods for Detection of Listeria spp. in Food Method Name

AOAC Validation

ANSR for Listeria spp. (Neogen Corporation) Assurance GDS for Listeria monocytogenes (BioControl Systems, Inc.)

AOAC-RI # 101202 AOAC-RI # 070702

Assurance GDS for Listeria species (BioControl Systems, Inc.)

AOAC-RI # 070701

BAX system L. monocytogenes (DuPont Qualicon) BAX System PCR assay for genus Listeria 24E (DuPont Qualicon) BAX System PCR assay for Listeria monocytogenes 24E (DuPont Qualicon) BAX System PCR Assay for Listeria monocytogenes (DuPont Qualicon) BAX System PCR assay for screening genus Listeria and the BAX System media for Listeria (DuPont Qualicon) GeneDisc Plate Listeria monocytogenes detection kit (Pall GeneDisc Technologies, Inc.) GeneDisc Plate Listeria spp. detection kit (Pall GeneDisc Technologies, Inc.)

AOAC-OMA # 2003.12 AOAC-RI # 050903 AOAC-RI # 080901 AOAC-RI # 070202 AOAC-RI # 030502

GeneDisc Plate Listeria identification kit (Pall GeneDisc Technologies, Inc.) IEH Listeria spp. and Listeria monocytogenes test system (IEH Laboratories and Consulting Group) iQ-Check Listeria monocytogenes II real-time PCR (Bio-Rad Laboratories) iQ-Check Listeria spp. real-time PCR (Bio-Rad Laboratories) Listeria LT (Idaho Technology, Inc)

AOAC-RI # 031207 AOAC-RI # 021201b

MicroSEQ Listeria monocytogenes detection kit (Applied Biosystems) MicroSEQ Listeria spp. detection kit (Applied Biosystems) Molecular detection assay Listeria (3M) Roka Listeria detection assay (Roka Bioscience, Inc) InstantLabs Listeria species food safety kit (InstantLabs Medical Diagnostics Corporation) InstantLabs Listeria monocytogenes food safety kit (InstantLabs Medical Diagnostics Corporation) Thermo Scientific SureTect Listeria monocytogenes PCR assay (Thermo Fisher Scientific)

Selected Validated Matricesa

AOAC-RI # 031204 AOAC-RI # 031205

AOAC-RI 010802 AOAC-RI 090701 AOAC-RI 010901 AOAC-RI 011002 AOAC-RI 021108

Environmental surfaces (stainless steel, plastic, sealed concrete, ceramic tile, rubber) Liquid pasteurized milk, Mexican soft cheese, frankfurter, deli turkey, raw fish, raw green beans, environmental surfaces (stainless steel, rubber, concrete, plastic) Liquid pasteurized milk, Mexican soft cheese, frankfurter, deli turkey, raw fish, raw green beans, environmental surfaces (stainless steel, rubber, concrete, plastic) Dairy products, fruits and vegetables (except radishes), seafoods, raw and processed meats, poultry Bagged spinach, processed cheese, frankfurters, cooked shrimp, environmental surfaces (stainless steel) Bagged spinach, processed cheese, frankfurters, cooked shrimp, environmental surfaces (stainless steel) Raw meats, fresh produce/vegetables, processed meats, seafood, dairy cultured/non-cultured, egg and egg products, fruit juices Processed cheese, frankfurters, smoked salmon, spinach, environmental surfaces (plastic ceramic, tile, rubber, painted wood surfaces, unpainted wood, sealed concrete, cast iron, air filter material, drain swabs) Deli roast beef, hot dogs, deli turkey, raw shrimp, cold smoked salmon, romaine lettuce, pasteurized whole milk, vanilla ice cream, Brie cheese, liquid eggs, environmental surfaces (stainless steel, sealed concrete) Deli roast beef, hot dogs, deli turkey, raw shrimp, cold smoked salmon, romaine lettuce, pasteurized whole milk, vanilla ice cream, Brie cheese, liquid eggs, environmental surfaces (stainless steel, sealed concrete) Pure colonies from OXA, MOX, OAA, PALCAM, Rapid’L.mono agar, nutrient agar Raw beef trim, ready-to-eat turkey, environmental surfaces (stainless steel, plastic)

#

Smoked salmon, cottage cheese, hot dogs, deli turkey

#

Environmental surfaces (stainless steel, plastic, ceramic, sealed concrete)

Turkey deli meat, Mexican soft cheese, environmental surfaces (plastic, stainless steel, ceramic) # Pasteurized whole milk, dry infant formula, ice cream, roast beef, cured bacon, lox (cold-smoked salmon), lettuce, salad dressing and mayonnaise # Pasteurized whole cow’s milk, dry infant formula, hot dogs, roast beef, lox (smoked salmon), environmental surfaces (stainless steel, plastic cutting board, ceramic tile, rubber sheets, concrete sealed with Seal Hard) AOAC-RI # 081203 Environmental surfaces (stainless steel, concrete, plastic) AOAC-RI # Pasteurized whole milk, ice cream, Brie cheese, hot dogs, cured ham, 011201 deli chicken, chicken salad, cold-smoked salmon, romaine lettuce, environmental surfaces (stainless steel, sealed concrete, plastic) AOAC-RI # Hot dogs, raw shrimp, cheddar cheese, stainless steel, sealed concrete 041304 #

AOAC-RI # 051302

Hot dogs, deli turkey, romaine lettuce, raw shrimp, cheddar cheese, vanilla ice cream, pasteurized whole milk, stainless steel, sealed concrete

AOAC-RI # 061302

Raw ground beef, pork frankfurters, salami, cooked sliced turkey, fresh bagged spinach, cantaloupe, processed cheese, ice cream, smoked salmon, cooked prawns, stainless steel and plastic

Source: Table includes the modified information provided from the USDA-FSIS list of Foodborne Pathogen Test Kits Validated by Independent Organizations.139 a 25 g samples. b Validated for 25–375 g. 140 |

|

Rapid Methods for the Detection and Identification of Foodborne Pathogens

Table 11-3. Selected Nucleic Acid Amplification-Based Methods for Detection of Pathogenic E. coli in Food Method Name

Target Organism(s)

AOAC Validation

Selected Validated Matrices

Assurance GDS for E. coli O157: H7 (BioControl Systems)

E. coli O157:H7

AOAC-OMA # 2005.04

Assurance GDS Shigatoxin genes (BioControl Systems)

E. coli O157:H7 and E. coli O157:H7 nonmotile (NM)

AOAC-OMA # 2005.05

BAX real-time PCR assay E. coli O157:H7 (DuPont Qualicon) BAX System E. coli O157:H7 MP (DuPont Qualicon) E. coli O157:H7 test kit (Idaho Technology, Inc.) GeneDisc pathogenic E. coli O157 (Pall GeneSystems) iQ-Check E. coli O157:H7 Kit (Bio-Rad) MicroSEQ E. coli O157:H7 detection kit (Applied Biosystems)

E. coli O157:H7

AOAC-RI # 031002

E. coli O157:H7

AOAC-RI # 050501

E. coli O157:H7

AOAC-RI # 100901

Raw ground beef, beef trim, orange juice, apple juice, fresh vegetables, sprout process water Raw ground beef, beef trim, orange juice, fresh vegetables, sprout process water Lettuce and spinach, beef trim (375 g) and ground beef (65 g) Raw ground beef, beef trim, spinach and lettuce Raw ground beef, uncooked spinach

E. coli O157

AOAC-RI # 021102

E. coli O157:H7

AOAC-RI # 020801

E. coli O157:H7

AOAC-RI # 071001

Molecular detection assay E. coli O157 (including H7) (3M)

E. coli O157 (including H7)

AOAC-RI # 071202

SAS molecular tests Escherichia coli O157 detection kit (SA Scientific, Ltd.) GeneDisc STEC (Pall GeneDisc Technologies) iQ-Check STEC VirX and iQ-Check STEC SerO (Bio-Rad Laboratories)

E. coli O157

AOAC-RI # 031203

Shigatoxigenic E. coli

AOAC-RI # 021103

stx1, stx2, eae, O26, O45, O103, O11, O121, O145, O157:H7 STECs including O26, O103, O111, O145, 045, 0121 stx1, stx2, eae, O26, O45, O103, O11, O121, O145, O157:H7

AOAC-RI # 121203

GeneDisc Plate STEC Top 6 (Pall GeneDisc Technologies) iQ-Check STEC VirX and iQ-Check STEC SerO (Bio-Rad Laboratories)

AOAC-RI # 021106 AOAC-RI # 121203

Raw ground beef and raw beef trim (25 g, 375 g) Raw ground beef, fresh spinach, apple cider Raw ground beef and raw beef trim (375 g); ground beef and raw beef trim (25 g); spinach, orange juice, apple juice (25 g) Raw ground beef (325 g, 375 g), fresh bagged spinach (200 g), alfalfa sprouts (25 g) Raw ground beef (25 g, 375 g), raw beef trim (375 g), bagged mixed lettuce (200 g), fresh spinach (200 g) Fresh raw ground beef (25 g, 375 g), fresh raw beef trim (25 g, 375 g) Raw beef trim

Fresh raw ground beef (25 g, 375 g), fresh raw beef trim (25 g, 375 g) Raw beef trim

Source: Table includes the modified information provided from the USDA-FSIS list of Foodborne Pathogen Test Kits Validated by Independent Organizations.139

characterize microorganisms at the species level. Although it only gained a foothold for routine use in microbiology laboratories over the last decade, the method and wavelength

dependence were described by Karas et al. over 25 years ago.65 In 1988, two important studies showed that MALDITOF MS could be used to detect proteins with masses

Table 11-4. Selected Nucleic Acid Amplification-Based Methods for Detection of Campylobacter spp. in Food Method Name

Target Organism(s)

AOAC Validation

Selected Validated Matrices

ADIAFOOD rapid pathogen detection system for Campylobacter quantification (AOAC-RI # 050603) BAX system real-time PCR assay for Campylobacter jejuni/coli/lari (DuPont Qualicon) iQ-Check Campylobacter (Bio-Rad Laboratories)

Campylobacter jejuni, C. coli, C. lari

AOAC-RI # 050603

Poultry rinses

C. jejuni, C. coli, C. lari

AOAC-RI # 040702

Ready-to-eat turkey product (25 g) and chicken carcass rinses (30 mL)

C. jejuni, C. coli, C. lari

AOAC-RI # 031209

Chicken carcass rinse (30 mL), turkey carcass sponge, raw ground chicken (25 g)

Source: Table includes the modified information provided from the USDA-FSIS list of Foodborne Pathogen Test Kits Validated by Independent Organizations.139

| 141

Compendium of Methods for the Microbiological Examination of Foods |

Table 11-5. Selected Nucleic Acid Amplification-Based Methods for Detection of Salmonella spp. in Food AOAC Validation

Selected Validated Matricesa

ADIAFOOD detection system: Salmonella species (AES Chemunex Canada) ANSR for Salmonella (Neogen Corporation)

AOAC-RI # 070402

Cottage cheese, boneless pork, ground beef, chicken breast, cooked ham, chicken wings, clam chowder, apple juice, cauliflower, tortellini

AOAC-RI # 061203

Assurance GDS Salmonella (BioControl Systems, Inc.) Assurance GDS Salmonella (BioControl Systems, Inc.) Atlas Salmonella detection assay (Roka Biosciences, Inc.)

AOAC-OMA # 2009.03 AOAC-RI # 050602 AOAC-RI # 031201

BAX system PCR assay for Salmonella (DuPont Qualicon)

AOAC-RI # 100201

foodproof Salmonella detection kit (BIOTECON Diagnostics GmbH)

AOAC-RI # 120301

GeneDisc Salmonella spp. (Pall GeneDisc Technologies) GeneQuence Salmonella (Neogen Corporation) GeneQuence Salmonella (Neogen Corporation)

AOAC-RI # 021101 AOAC-OMA # 2007.02 AOAC-RI # 030201

Raw ground beef, hot dogs (25 g and 325 g), chicken carcass rinse (30 mL), raw ground turkey, oat cereal, surfaces (stainless steel, plastic, sealed concrete, ceramic tile, rubber) Meats, poultry, poultry rinse, seafood, dairy products, fruits and vegetables, egg, pasta, peanut butter, environmental surfaces Nonfat dry milk, liquid milk, egg, raw beef, raw pork, ground turkey, chicken rinse, raw shrimp, stainless steel, rubber concrete Fresh raw ground beef (375 g), frozen raw ground beef (375 g), raw ground chicken, cooked deli turkey (325 g), cooked deli chicken (325 g), pasteurized dried whole egg, raw cod, creamy non-organic peanut butter, romaine lettuce (375 g), tomatoes, instant nonfat dry milk, string cheese (mozzarella), milk chocolate, cocoa powder (375 g), raw cookie dough, dry pet food, dry pasta, shell eggs, nacho cheese seasoning, black pepper, soy flour, environmental surfaces (stainless steel, plastic, sealed concrete) Milk (2%), custard, nonfat dry milk, liquid egg, chipped ham, cooked chicken, hot dogs, ground beef, cooked fish, prawns, frozen peas, orange juice, peanut butter, alfalfa sprouts, black pepper, dry pet food, chilled ready-meal, chocolate, elbow macaroni, pizza dough, isolated soy protein Milk powder, ice cream, egg powder, chicken breast, minced meat, sliced sausage, sausage, smoked fish, watermelon, sliced cabbage, coconut, white pepper, cumin, wet pet food, dry pet food, dough, food dye, milk chocolate, cocoa powder, pasta Raw ground beef and raw beef trim

InstantLabs Salmonella species food safety kit (InstantLabs Medical Diagnostics Corporation) iQ-Check Salmonella II kit (Bio-Rad Laboratories) MicroSEQ Salmonella spp. detection kit (Life Technologies)

AOAC-RI # 031202

Molecular detection assay Salmonella (3M) SAS molecular tests Salmonella spp. detection kit (SA Scientific, Ltd.) TaqMan Salmonella enterica detection kit (Applied Biosystems) Atlas Salmonella G2 Detection Assay (Roka Biosciences, Inc.)

AOAC-RI 031208 AOAC-RI 021202 AOAC-RI 020803 AOAC-RI 041303

Thermo Scientific Sure Tect Salmonella spp. PCR assay (Thermo Fisher Scientific) BAX system real-time PCR Assay for Salmonella (DuPont Qualicon)

AOAC-RI # 051303

Method Name

AOAC-RI # 010803 AOAC-RI # 031001 # # # #

AOAC-RI # 081201

Raw turkey, dried, liquid and liquid frozen pasteurized eggs, milk chocolate, dry pet food Dried whole egg, nonfat dry milk, cheese powder, raw pooled shell egg, raw ground pork, beef franks, raw ground turkey, raw ground chicken, raw fish fillet, surimi, dried fruit, fresh mushrooms, frozen fruit, black pepper, dry pet food, dry cake mix, shelled walnuts, semi-sweet chocolate, refrigerated cookie dough, soy flour, egg noodles, food dye Raw ground beef (375 g), raw chicken breast, raw ground chicken, lettuce, rolled oats (750 g), oat flour (750 g), wheat flour (750 g) Eggs, raw beef, raw chicken, cantaloupe, environmental surfaces (ceramic, concrete, plastic, stainless steel), dry dog food, wet cat food Brie Cheese, shell eggs, raw ground beef, raw chicken wings, raw shrimp, cantaloupe, black pepper, dry pet food, chocolate, peanut butter, dry infant formula Pasteurized liquid whole egg, raw ground beef, cooked breaded chicken, raw shrimp, bagged spinach, wet pet food (375 g) Raw ground beef (25 g and 375 g), raw beef trim (375 g), raw ground turkey, chicken carcass rinses, bagged mixed lettuce (200 g), fresh spinach (200 g) Ground beef, chicken wings, cheddar cheese, dry pet food Fresh raw ground beef (375 g), fresh raw ground turkey (375 g), cooked deli turkey (325 g), romaine lettuce (375 g), oat cereal, stainless steel, plastic, sealed concrete Raw ground beef, raw chicken breast, chilled ready-to-eat dinner, pork frankfurters, raw ground pork, cooked shrimp, non-fat dried milk powder, fresh bagged lettuce, pasteurized liquid whole egg and stainless steel Raw ground beef (375 g), chicken carcass rinse (30 mL), cream cheese, bagged lettuce, dry pet food (375 g), stainless steel

Source: Table includes the modified information provided from the USDA-FSIS list of Foodborne Pathogen Test Kits Validated by Independent Organizations.139 a 25 g unless specified in parenthesis.

142 |

|

greater than 10,000 Da66 and up to 100,000 Da134 with the aid of various matrix compounds. At that time, the former study used a chemical matrix (nicotinic acid) method similar to that which was adopted later for microbial identification, whereas the latter study used a matrix of fine cobalt powder combined with glycerol. These matrix compounds allow for what is referred to as soft ionization or soft laser desorption, which assists in the ionization of proteins with minimal fragmentation of the molecules. Thanks to the ability to interrogate the intact proteins and related peptides, one can map their origin using a protein database (e.g., TagIdent; http://web.expasy.org/tagident).

11.42

MALDI-TOF MS Principle

In contrast to other mass spectrometry methods that require volatilization of the sample prior to introduction into the mass spectrometer, MALDI-TOF MS allows for the use of whole cell preparations (i.e., direct deposit, or an extract from microbial colonies) that are combined with a chemical matrix, typically about 3% a-cyano-4-hydroxycinnamic acid dissolved in ethanol and acetonitrile (CHCA), which forms a crystalline lattice once dried. A nitrogen laser (wavelength 337 nm) is pulsed on the sample and the sample is ionized (Figure 11-7). An electrostatic field and a high-voltage supply are applied to uniformly accelerate the ions into the flight tube, which is maintained under high vacuum. Once in the flight tube, the ions travel toward the detector based on their mass-to-charge ratio (m/z), and the time-offlight data are recorded at the detector (Figure 11-8). Timeof-flight data are then converted into a mass spectrum comprised of mass peaks and associated intensities (Figure 11-9).

11.43

Microbial Identification With MALDI-TOF MS

For use in microbial identification, the mass range of interest is 2,000–20,000 Da, where one can typically detect 100–200 mass peaks. Many of these mass peaks have been identified as small and large subunit ribosomal proteins and nucleic acid-binding proteins. The others are probably

Rapid Methods for the Detection and Identification of Foodborne Pathogens

representative of other housekeeping proteins found in abundance in the microbial cell.25

11.44

Commercial MALDI-TOF MS Systems

There are two systems used today in the microbiology laboratory, including VITEK MS (bioMe´rieux, Marcy l’Etoile, France) and Bruker Biotyper (Bruker Daltonics, Bremen, Germany). Various data analysis methods can be used to arrive at a species-level identification of the microbe of interest. These include (1) fingerprint (peak to peak) matching (e.g., VITEK MS Research Use Only [RUO] Saramis reference spectra); (2) population spectral matching (e.g., VITEK MS RUO Saramis SuperSpectra); and (3) population analysis based on an entirely numerical approach (e.g., Advanced Spectra Classifier (ASC) on the VITEK MS in vitro diagnostic (IVD) proprietary algorithm).

11.45

Specificity of MALDI-TOF MS

MALDI-TOF MS is a highly specific method used in the identification of microorganisms from cultures. In fact, MALDI-TOF MS provides the same level of resolution as sequencing ribosomal RNA genes (16S and 26S rDNA for bacteria and fungi, respectively) plus other housekeeping gene targets (e.g., RNA polymerase b subunit [encoded by rpoB] for coryneforms, superoxide dismutase [encoded by sodA] for Gram-positive cocci, DNA recombination protein [encoded by recA] for non-fermentative Gram-negative bacilli). This was demonstrated recently in a multicenter study conducted in the microbiology laboratories of five large medical centers to validate the clinical performance of the VITEK MS System. The performance data of several bacterial species known to be food pathogens were extracted from those publications13,45,96,117,122 and are shown in Table 11-6.

11.46

MALDI-TOF MS Performance

11.461 Campylobacter and Related Genera Aside from the data13 shown in Table 11-6, other studies have shown the reliable performance of MALDI-TOF MS

Figure 11-7. Matrix-assisted laser desorption/ionization.

| 143

Compendium of Methods for the Microbiological Examination of Foods |

A separate study by Martiny et al.100 showed the influence of isolation media and incubation conditions (i.e., time and temperature) on the performance of MALDITOF MS. Authors compared five media, two temperatures, and daily incubation for five days. At 24- to 48-h incubation, performance was optimal, with little or no effect from other variables. At 3- to 5- day incubation times, performance degradation was observed, with the most notable impact from the use of Butzler medium and 42uC cultures.

Figure 11-8. Time-of-flight (TOF) and mass spectrum acquisition.

with isolates of Campylobacter spp. A large study of 999 isolates was performed by Besse`de et al.10 using the Bruker Biotyper (version 2.0 software). Compared to real-time PCR as the reference standard, MALDI-TOF MS proved 99.6% overall correct to species level, with 99.5% correct for C. jejuni (n5785) and 100% correct for the following species: C. coli (n5149), C. fetus (n540), C. lari (n58), C. upsaliensis (n52), C. sputorum (n51), and Arcobacter butzleri (n514). Another study done by Martiny et al.98 showed a similar performance, where 230/234 isolates (98.3%) were correctly identified to species level. A compilation of data from these studies on Campylobacter is shown in Table 11-7.

Figure 11-9. Mass spectra of some notable food pathogens.

144 |

11.462 Listeria Barbuddhe et al.8 showed the potential of MALDI-TOF MS to differentiate several species of Listeria as well as determining clonal relatedness compared to pulsed field gel electrophoresis results. This earlier work employed an inactivation and tube extraction process and was prior to the establishment of a usable database. Interestingly, a later evaluation by Farfour et al.34 carried out with the Microflex MALDI-TOF MS spectrometer (Bruker Daltonics) in conjunction with the Andromas analysis software (Andromas SAS, Paris, France) showed that 32 isolates of Listeria monocytogenes and 24 isolates of five other Listeria spp. could only be identified to the genus level, owing to the high similarity of spectra between the six species. This lack of discriminatory power appears to be unique to direct colony preparation in conjunction with the Andromas software, as another study by Carbonnelle et al.17 using direct colony preparation in conjunction with the Bruker Biotyper software confirmed the findings of Barbuddhe et al.8 when L. monocytogenes isolates were tested, albeit with a small sample number. A later and more comprehensive study of L. monocytogenes122 demonstrated identification at the species level for 34/45 (76%) isolates using direct colony preparation in conjunction with the Vitek MS IVD database and ASC algorithm. 11.463 Salmonella One of the earliest studies reported on the use of MALDITOF MS for the characterization of serovars of Salmonella enterica ssp. enterica.81 Although this precluded the introduction of commercial MALDI-TOF MS systems and their associated databases, it showed the potential of this technology that would be exploited in the coming years. In 2008, Dieckmann et al.,25 using an Ultraflex II MALDI-TOF/TOF MS (Bruker Daltonics) in combination with Saramis software (Anagnostec GmbH, PotsdamGolm, Germany; software/database and expertise acquired in 2010 by bioMe´rieux), showed the ability to identify S. enterica subspecies and the similarity to DNA homology and sequencing of housekeeping genes for differentiation of S. enterica from S. bongori and the relatedness of various subspecies of S. enterica. Dieckmann and Malorny27 later showed the potential of MALDI-TOF MS for differentiating the main serovars of S. enterica ssp. enterica. By studying 913 isolates representing 89 serovars, they were able to identify specific biomarker peaks for the most prevalent five serovars (Enteritidis, Typhimurium, Virchow, Infantis, and Hadar). Although serovar identification is an application of great interest to both food and clinical laboratories for the surveillance and epidemiology of Salmonella outbreaks,

|

Rapid Methods for the Detection and Identification of Foodborne Pathogens

Table 11-6. Performance of Food Pathogens Tested on VITEK MS13,45,96,117,122 Species

No. of Isolates

% Correct to Species

% Correct to Genus

% Total Correct

% Total MisID

% Total NoID

Campylobacter jejunia Campylobacter colia Clostridium perfringens Escherichia coli Listeria monocytogenes Salmonella enterica Staphylococcus aureus Vibrio parahaemolyticusb Vibrio vulnificusb Yersinia enterocolitica

33 2 61 65 45 35 61 16 11 14

93.9 100.0 98.4 100.0 75.6 94.3 98.4 87.5 90.9 100.0

3.0 0.0 0.0 0.0 8.9 5.7 0.0 6.3 0.0 0.0

96.9 100.0 98.4 100.0 84.4 100.0 98.4 93.8 90.9 100.0

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

3.0 0.0 1.6 0.0 15.6 0.0 1.6 6.3 9.1 0.0

Note: MisID 5 misidentification; NoID 5 no identification. a Also shown in data compilation Table 11-7. b Also shown in Table 11-8 with other related species performances.

there is not yet an easy-to-use commercial tool for their rapid identification using MALDI-TOF MS. The Bruker Biotyper version 3.0 and security-relevant (SR) databases (containing 29 spectra from multiple serovars) were tested by Kuhns et al.76 who found that S. enterica serovar Typhi could not be reliably differentiated from the other S. enterica serovars by the routine software. However, a proteomic analysis revealed several biomarker ions that could be used to separate Typhi from non-Typhi isolates. Martiny et al.99 found the same issue with the Biotyper software, whereas the VITEK MS ASC was able to correctly identify 5/5 (100%) of Typhi serovar isolates. As demonstrated, one can collect data using the research modes of the commercial systems that allow for construction of dendrograms and more detailed manual analysis of specific biomarkers. Alternatively, one can export spectral data to one of various external software tools in order to perform other data analyses and reveal

serovar-specific biomarker ions. Hopefully, the commercial systems will address this need for easier applications in future software improvements. A review by Sandrin et al.123a raised the possibility that MALDI-TOF MS today (proteins and peptides in the 2,000– 20,000 Da range) may not be optimal for strain (e.g., serovar) typing and that perhaps one better solution might be to look at outer membrane molecules (e.g., lipids or lipopeptides).

11.464

Shiga Toxin–Producing E. coli and other pathogenic E. coli Currently, commercial MALDI-TOF MS systems are unable to distinguish less virulent strains of E. coli from their phylogenetically equivalent and toxigenic relatives (e.g., Shigella spp., STEC, enterohemorrhagic E. coli [EHEC]). There have been a few studies showing the potential to identify serotype-specific biomarker ions.

Table 11-7. Compilation of Performance on Different MALDI-TOF MS Methods With Species of Campylobacter and Related Genera10,13,98 Species

No. of Isolates

% Correct to Species

% Correct to Genusa

% Total MisIDb

% Total NoIDc

Campylobacter jejuni Campylobacter coli Campylobacter fetus Campylobacter lari Campylobacter upsaliensis Campylobacter curvus Campylobacter hyointestinalis Campylobacter sputorum Campylobacter peloridisd Arcobacter butzleri Arcobacter cryaerophilus Helicobacter pullorum

948 216 47 13 10 6 3 1 1 20 2 1

99.4 100.0 100.0 100.0 100.0 83.3 100.0 100.0 0.0 100.0 0.0 100.0

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.1 0.0 0.0 0.0 0.0 16.7 0.0 0.0 100.0 0.0 100.0 0.0

a

Genus-level identification with correct species listed in choices. All misidentifications (MisID) in correct genus. c All no identifications (noID) in correct genus. d Not in database. b

| 145

Compendium of Methods for the Microbiological Examination of Foods |

Fagerquist et al.33 used MALDI-TOF-TOF MS/MS (reflectron mode) to distinguish E. coli non-O157:H7 from strains of E. coli O157:H7 and identified a biomarker (YahO— protein of unknown function) that could be useful in their differentiation. Since the difference resulted in an amino acid substitution and a subsequent mass shift of only 1 Da, authors concluded that MALDI-TOF MS (linear mode) had limited utility for its recognition. However, authors also identified a biomarker at approximately 9060 Da (identified as an HdeB acid stress chaperone-like protein) that could be visualized by MALDI-TOF MS (linear mode) and was present in E. coli non-O157:H7 but absent in strains of E. coli O157:H7. Karger et al.67 were able to export peak lists collected with an Ultraflex I MALDI-TOF MS (Bruker) and convert them for use with a web-based external software, Rpackage caMassclass (available at http://cran.r-project. org/src/contrib/Archive/caMassClass) to differentiate STEC serotypes O26:H11, O156:H25, and O165:H25 with a high degree of specificity (99.3%). A recent study19 used the Bruker Biotyper to collect spectral data and exported these data to an in-house developed software, GenomeFisher, and through an iterative approach, investigators were able to utilize the presence and absence of specific mass peaks to reliably differentiate six pathotypes, EHEC, STEC, enteropathogenic E. coli (EPEC), enterotoxigenic E. coli (ETEC), enteroaggregative E. coli (EAEC), and urinary tract pathogenic E. coli (UPEC), although different serotypes within pathotypes were not so easily distinguished. This study also confirmed the earlier finding that the absence of a peak at approximately 9060 Da is characteristic of most strains of EHEC E. coli O157:H7 and could be useful as a screening tool for that pathotype. Shiga toxin is not yet detected by routine MALDI-TOF MS, but Antoine et al.4 described an assay using MALDITOF MS (Microflex) whereby incubation of the toxin with 12- or 14-mer oligonucleotides resulted in cleavage of an adenine residue, respective mass shifts of 135 Da, and appearance of novel mass peaks.

11.465 Vibrio The phylogenetic similarity of Vibrio and Aeromonas species often make it difficult to differentiate based solely on 16S rDNA sequencing. To address this issue, sequencing of other housekeeping genes has been used for higher levels of specificity. One of these targets, rpoB, is particularly useful. One study of various Vibrio and Aeromonas species

using the Ultraflex II MALDI-TOF MS coupled with external software (Gene Cluster; version 3.0, Human Genome Center, University of Tokyo, Japan) was able to delineate these species with similar specificity to rpoB sequencing.26 Other early studies using the Voyager DE STR MALDITOF MS (Applied Biosystems, Foster City, CA) showed the power of proteomics and the potential ability to differentiate V. parahaemolyticus from nine other Vibrio species using 30 unique mass peaks present only in strains of V. parahaemolyticus,50 as well as in differentiation of several Gram-negative species found in seafood spoilage.12 A recent study93 showed correct identification of 22 isolates of V. parahaemolyticus compared with the VITEK 2 GN card (bioMe´rieux), although the Biotyper data were only presented as a dendrogram for analysis of clonal relationships, rather than showing actual score matches to the Biotyper database. The use of a 10% dissimilarity cutoff allowed the segregation of clones that resembled those generated through pulsed-field gel electrophoresis, which is a standard method for strain typing. A multicenter evaluation of the VITEK MS IVD system and its associated ASC algorithm compared to a reference standard of 16S rDNA sequencing in combination with other phenotypic and/or genotypic testing (when necessary) showed excellent species-level performance of V. parahaemolyticus, two other Vibrio species, and three Aeromonas species. Data from that study are summarized in Table 11-8.

11.466 Yersinia Testing of 146 strains of 13 different Yersinia species, including 57 strains of Y. enterocolitica, on an Autoflex I MALDI-TOF MS (Bruker Daltonics) in conjunction with data analysis using Matlab (The Mathworks, Inc., Natick, MA) revealed several species-specific biomarker ions useful in their differentiation from one another, as well as from over 20 other species of Enterobacteriaceae. Respective dendrograms of mass spectra showed clusters bearing close similarity to known taxonomic relationships.79 The Autoflex II with Biotyper version 2.0 software (Bruker Daltonics) was used to create new database entries for 12 Yersinia species. Subsequent blind analysis showed that 11/11 (100%) isolates of Y. enterocolitica gave speciesspecific scores $ 2.0. Interestingly, before the database was extended, use of the original Bruker database showed a Y. pestis strain misidentified as Y. pseudotuberculosis with a score . 2.0.5

Table 11-8. Performance of Vibrio parahaemolyticus and Related Species on VITEK MS96 Species

No. of Isolates

% Correct to Species

% Correct to Genusa

% Total Correct

% Total MisIDb

% Total NoIDc

Vibrio parahaemolyticus Vibrio cholerae Vibrio vulnificus Aeromonas hydrophila/caviae Aeromonas sobria

16 11 11 25 10

87.5 90.9 90.9 64.0 40.0

6.3 0.0 0.0 24.0 50.0

93.8 90.9 90.9 88.0 90.0

0.0 0.0 0.0 8.0 10.0

6.3 9.1 9.1 4.0 0.0

a

Genus-level identification with correct species listed in choices. All misidentifications (MisID) in correct genus. c All no identifications (noID) in correct genus. b

146 |

|

Stephan et al.131 used the Axima Confidence (ShimadzuBiotech Corp., Kyoto, Japan) in conjunction with Saramis software to create their own library of Superspectra using 19 strains of Y. enterocolitica and 24 strains of 11 other Yersinia species. The library was then tested with a blind challenge of 117 Y. enterocolitica strains, and 100% were identified correctly.

11.467

Other Enterotoxigenic Genera (Staphylococcus, Bacillus, Clostridium) Although many other genera and species can be involved in forming enterotoxins and associated foodborne disease, the main additional species of concern include Staphylococcus aureus, Bacillus cereus, Clostridium perfringens, and C. botulinum.41 The performance of MALDI-TOF MS for S. aureus can be found in several global evaluations, although most (if not all) of these involved testing of clinical isolates. In general, MALDI-TOF MS is highly specific for S. aureus. Martiny et al.99 found 183/183 (100%) and 180/183 (98.4%) correct to species level for VITEK MS version 1.1 and Bruker Biotyper version 2.0, respectively. Dubois et al.28a found 35/36 (97.2%) correct to species level and 1/36 (2.8%) unidentified with the VITEK MS version 1.1. Rychert et al.122 found 60/61 (98.4%) correct to species level and 1/61 (1.6%) unidentified with the VITEK MS version 2.0 software. There are no clear-cut performance evaluations for B. cereus, although the literature suggests that MALDI-TOF MS does not give very good resolution among members of the B. cereus group, with the exception of B. anthracis, which appears to have some unique biomarkers that may be useful in its separation from other members of the complex.39,80 The other members of the B. cereus group (e.g., B. cereus, B. thuringiensis, and B. mycoides), require other (e.g., phenotypic) tests for their differentiation. A publication reported on the indirect detection of enterotoxins using MALDI-TOF MS in conjunction with tryptic digests of electrophoretic gel bands.137 Similarly, one can only find MALDI-TOF MS performance data for clostridia in clinical evaluations of anaerobic culture isolates. Garner et al.45 tested 61 isolates of C. perfringens with the VITEK MS with version 2.0 software and showed that 60/61 (98.4%) were correctly identified to species level and 1/61 (1.6%) was unidentified. Although with much smaller data sets for this species, Martiny et al.99 showed that both VITEK MS and Bruker Biotyper correctly identified 2/2 isolates (100%) to species level, and Dubois et al. also showed 2/2 (100%) correct to species level with VITEK MS. 11.468 Conclusion Conventional biochemical and morphologic testing that required hours to days in order to achieve a microbial identification from cultured colonies is no longer necessary in the microbiology laboratory. Today, MALDI-TOF MS is a revolutionary advance that allows for highly specific results within minutes of preparation. Testing is both rapid and cost-effective compared to the previous conventional approach. Additionally, the amount of microorganism needed to perform MALDI-TOF is much less than that required for biochemical methods. With MALDI-TOF MS

Rapid Methods for the Detection and Identification of Foodborne Pathogens

direct colony application on the target (as is done with the VITEK MS system), one colony or a portion of one colony is typically sufficient to meet the limit of detection, which is approximately 104–105 CFU on the target. Diagnostic microbiology had a slow evolution over the last century, but we are finally moving rapidly into the future.

ACKNOWLEDGMENTS Fourth edition authors: Phyllis Entis, Daniel Y. C. Fung, Mansel W. Griffiths, Lynn McIntyre, Scott Russell, Anthony N. Sharpe, and Mary Lou Tortorello. The authors acknowledge Dr. Gregory Devulder for critically reviewing this section. The authors thank Dr. Martin Welker for his kind assistance in supplying the relevant examples of MALDITOF MS spectra used in Figure 11-9.

REFERENCES 1. Aldus, C. F., A. van Amerongen, R. M. C. Ariens, M. W. Peck, J. H. Wichers, and G. M. Wyatt. 2003. Principles of some novel rapid dipstick methods for detection and characterization of verotoxigenic Escherichia coli. J. Appl. Microbiol. 95:380-389. 2. Alles, S., S. Curry, D. Almy, B. Jagadeesan, J. Rice, and M. Mozola. 2012. Reveal Listeria 2.0 test for detection of Listeria spp. in foods and environmental samples. J. AOAC Int. 95:424-434. 3. Amagliani, G., E. Omiccioli, A. del Campo, I. J. Bruce, G. Brandi, and M. Magnani. 2005. Development of a magnetic capture hybridization-PCR assay for Listeria monocytogenes direct detection in milk samples. J. Appl. Microbiol. 100:375-383. 4. Antoine, M. D., N. A. Hagan, J. S. Lin, A. B. Feldman, and P. A. Demirev. 2012. Rapid detection of ribosome inactivating protein toxins by mass-spectrometry-based functional assays. Int. J. Mass. Spectrom. 312:41-44. 5. Ayyadurai, S., C. Flaudrops, D. Raoult, and M. Drancourt. 2010. Rapid identification and typing of Yersinia pestis and other Yersinia species by matrix-assisted laser desorption/ ionization time-of-flight (MALDI-TOF) mass spectrometry. BMC Microbiol. 10:285. 6. Bach, S., A. de Almeida, and E. Carniel. 2000. The Yersinia high-pathogenicity island is present in different members of the family Enterobacteriaceae. FEMS Microbiol. Lett. 183:289294. 7. Banada, P. P., and A. K. Bhunia. 2008. Antibodies and immunoassays for detection of bacterial pathogens, 567-602. In Zourob, M. et al. (eds.), Principles of bacterial detection: biosensors, recognition receptors and microsystems. Springer Science and Media, New York, NY. 8. Barbuddhe, S. B., T. Maier, G. Schwarz, M. Kostrzewa, H. Hof, E. Domann, T. Chakraborty, and T. Hain. 2008. Rapid identification and typing of Listeria species by matrixassisted laser desorption ionization-time of flight mass spectrometry. Appl. Environ. Microbiol. 74:5402-5407. 9. Bekal, S., R. Brousseau, L. Masson, G. Prefontaine, J. Fairbrother, and J. Harel. 2003. Rapid identification of Escherichia coli pathotypes by virulence gene detection with DNA microarrays. J. Clin. Microbiol. 41:2113-2125. 10. Besse`de, E., O. Solecki, E. Sifre´, L. Labadi, and F. Me´graud. 2011. Identification of Campylobacter species and related organisms by matrix assisted laser desorption ionizationtime of flight (MALDI-TOF) mass spectrometry. Clin. Microbiol. Infect. 17:1735-1739.

| 147

Compendium of Methods for the Microbiological Examination of Foods |

11. Bohaychuk, V. M., G. E. Gensler, M. E. McFall, R. K. King, and D. G. Renter. 2007. A real-time PCR assay for the detection of Salmonella in a wide variety of food and foodanimal matrices. J. Food Prot. 70:1080-1087. 12. Bo¨hme, K., I. C. Ferna´ndez-No, J. Barros-Vela´zquez, J. M. Gallardo, P. Calo-Mata, and B. Can˜ as. 2010. Species differentiation of seafood spoilage and pathogenic Gramnegative bacteria by MALDI-TOF mass fingerprinting. J. Proteome Res. 9:3169-3183. 13. Branda, J. A., J. Rychert, C.-A. D. Burnham, M. Bythrow, O. B. Garner, C. C. Ginocchio, R. Jennemann, M. A. Lewinski, R. Manji, A. B. Mochon, G. W. Procop, S. S. Richter, L. F. Sercia, L. F. Westblade, and M. J. Ferraro. 2014. Multi-center validation of the VITEK MS v2.0 MALDI-TOF mass spectrometry system for the identification of fastidious Gramnegative bacteria. Diagn. Microbiol. Infect. Dis. 78:129-131. 14. Brunder, W., H. Schmidt, and H. Karch. 1996. KatP, a novel catalase-peroxidase encoded by the large plasmid of enterohaemorrhagic Escherichia coli O157:H7. Microbiology. 142:3305-3315. 15. Bubert, A., S. Kohler, and W. Goebel. 1992. The homologous and heterologous regions within the iap gene allow genusand species specific identification of Listeria spp. by polymerase chain reaction. Appl. Environ. Microbiol. 58:26252632. 16. Buh Gasparic, M., T. Tengs, J. L. La Paz, A. Holst-Jensen, M. Pla, T. Esteve, J. Zel, and K. Gruden. 2010. Comparison of nine different real-time PCR chemistries for qualitative and quantitative applications in GMO detection. Anal. Bioanal. Chem. 396:2023-2039. 17. Carbonnelle, E., P. Grohs, H. Jacquier, N. Day, S. Tenza, A. Dewailly, O. Vissouarn, M. Rottman, J.-L. Herrmann, I. Podglajen, and L. Raskine. 2012. Robustness of two MALDI-TOF mass spectrometry systems for bacterial identification. J. Microbiol. Meth. 89:133-136. 18. Cikos, S., A. Bukovska, and J. Koppel. 2007. Relative quantification of mRNA: comparison of methods currently used for real-time PCR data analysis. BMC Mol. Biol. 8:113. 19. Clark, C. G., P. Kruczkiewicz, C. Guan, S. J. McCorrister, P. Chong, J. Wylie, P. van Caeseele, H. A. Tabor, P. Snarr, M. W. Gilmour, E. N. Taboada, and G. R. Westmacott. 2013. Evaluation of MALDI-TOF mass spectroscopy methods for determination of Escherichia coli pathotypes. J. Microbiol. Meth. 94:180-191. 20. Clotilde, L. M., C. Bernard, A. Salvador, A. Lin, C. R. Lauzon, M. Muldoon, Y. Xu, K. Lindpaintner, and J. M. Carter. 2013. A 7-plex microbead-based immunoassay for serotyping Shiga toxin-producing Escherichia coli. J. Microbiol. Meth. 92:226-230. 21. Cocolin, L., A. Rajkovic, K. Rantsiou, and M. Uyttendaele. 2011. The challenge of merging food safety diagnostic needs with quantitative PCR platforms. Trends Food Sci. Technol. 22:S30-S38. 22. Curiale, M. S., and W. Lepper. 1994. Enzyme-linked immunoassay for detection of Listeria monocytogenes in dairy products, seafoods, and meats: collaborative study. J. AOAC Int. 77:1472-1489. 23. DebRoy, C., E. Roberts, A. M. Valadez, E. G. Dudley, and C. N. Cutter. 2011. Detection of Shiga toxin-producing Escherichia coli O26, O45, O103, O111, O113, O121, O145, and O157 serogroups by multiplex polymerase chain reaction of the wzx gene of the O-antigen gene cluster. Foodborne Pathog. Dis. 8:651-652. 24. Delibato, E., A. Fiore, F. Anniballi, B. Auricchio, E. Filetici, L. Orefice, M. N. Losio, and D. De Medici. 2011. Comparison between two standardized cultural methods and 24 hour

148 |

25.

26.

27.

28.

28a.

29.

30.

31.

32.

33.

34.

35.

36.

duplex SYBR green real-time PCR assay for Salmonella detection in meat samples. New Microbiol. 34:299-306. Dieckmann, R., R. Helmuth, M. Erhard, and B. Malorny. 2008. Rapid classification and identification of Salmonellae at the species and subspecies levels by whole-cell matrixassisted laser desorption ionization-time of flight mass spectrometry. Appl. Environ. Microbiol. 74:7767-7778. Dieckmann, R., E. Strauch, and T. Alter. 2010. Rapid identification and characterization of Vibrio species using whole-cell MALDI-TOF mass spectrometry. J. Appl. Microbiol. 109:199-211. Dieckmann, R., and B. Malorny. 2011. Rapid screening of epidemiologically important Salmonella enterica subsp. enterica serovars by whole-cell matrix-assisted laser desorption ionization-time of flight mass spectrometry. Appl. Environ. Microbiol. 77:4136-4146. Don, R. H., P. T. Cox, B. J. Wainwright, K. Baker, and J. S. Mattick. 1991. ‘Touchdown’ PCR to circumvent spurious priming during gene amplification. Nucleic Acids Res. 19:4008. Dubois, D., M. Grare, M. F. Prere, C. Segonda, N. Marty, E. Oswald. 2012. Performances of the VITEK MS matrixassisted laser desorption ionization-time of flight mass spectrometry system for rapid identification of bacteria in routine clinical microbiology. J. Clin. Microbiol. 50:2568-2576. Dunbar, S. A., C. A. Vander Zee, K. G. Oliver, K. L. Karem, and J. W. Jacobson. 2003. Quantitative, multiplexed detection of bacterial pathogens: DNA and protein applications of the Luminex LabMAP system. J. Microbiol. Meth. 53:245-252. Dwivedi, H. P., and L. A. Jaykus. 2011. Detection of pathogens in foods: the current state-of-the-art and future directions. Crit. Rev. Microbiol. 37:40-63. Dwivedi, H. P., P. Rule, and J. C. Mills. 2012. Detection and identification of bacterial pathogens in food using biochemical and immunological assays, 229-268. In Taormina, P.T. (ed.), Microbiological research and development for the food industry. CRC Press, Boca Raton, FL. Ellingson, J. L., J. L. Anderson, S. A. Carlson, and V. K. Sharma. 2004. Twelve-hour real-time PCR technique for the sensitive and specific detection of Salmonella in raw and ready-to-eat meat products. Mol. Cell. Probes 18:51-57. Fagerquist, C. K., B. R. Garbus, W. G. Miller, K. E. Williams, E. Yee, A. H. Bates, S. Boyle, L. A. Harden, M. B. Cooley, and R. E. Mandrell. 2010. Rapid identification of protein biomarkers of Escherichia coli O157:H7 by matrix-assisted laser desorption ionization-time-of-flight-time-of-flight mass spectrometry and top-down proteomics. Anal. Chem. 82:2717-2725. Farfour, E., J. Leto, M. Barritault, C. Barberis, J. Meyer, B. Dauphin, A.-S. Le Guern, A. Lefle`che, E. Badell, N. Guiso, A. Leclercq, A. Le Monnier, M. Lecuit, V. Rodriguez-Nava, E. Bergeron, J. Raymond, S. Vimont, E. Bille, E. Carbonnelle, H. Guet-Revillet, H. Le´cuyer, J.-L. Beretti, C. Vay, P. Berche, A. Ferroni, X. Nassif, and O. Join-Lambert. 2012. Evaluation of the Andromas matrix-assisted laser desorption ionizationtime of flight mass spectrometry system for identification of aerobically growing Gram-positive bacilli. J. Clin. Microbiol. 50:2702-2707. Fedio, W. M., K. G. Jinneman, K. J. Yoshitomi, R. Zapata, C. N. Wendakoon, P. Browning, and S. D. Weagant. 2011. Detection of E. coli O157:H7 in raw ground beef by Pathatrix immunomagnetic-separation, real-time PCR and cultural methods. Int. J. Food Microbiol. 148:87-92. Feldsine, P. T., M. T. Falbo-Nelson, S. L. Brunelle, and R. L. Forgey. 1997. Assurance enzyme immunoassay for detection of enterohemorrhagic Escherichia coli O157: H7 in selected foods: a collaborative study. J. AOAC Int. 80:530-543.

|

37. Feldsine, P. T., A. H. Lienau, R. L. Forgey, and R. D. Calhoon. 1997. Assurance polyclonal enzyme immunoassay for detection of Listeria monocytogenes and related Listeria species in selected foods: a collaborative study. J. AOAC Int. 80:775-790. 38. Feng, P. 1997. Impact of molecular biology on the detection of foodborne pathogens. Mol. Biotechnol. 7:267-278. 39. Ferna´ndez-No, I. C., K. Bo¨hme, M. Dı´az-Bao, A. Cepeda, J. Barros-Vela´zquez, and P. Calo-Mata. 2013. Characterisation and profiling of Bacillus subtilis, Bacillus cereus and Bacillus licheniformis by MALDI-TOF mass fingerprinting. Food Microbiol. 33:235-242. 40. Fluit, A. C., M. N. Widjojoatmodjo, A. T. A. Box, R. Torensma, and J. Verhoef. 1993. Rapid detection of Salmonellae in poultry with the magnetic im munopolymerase chain reaction assay. Appl. Environ. Microbiol. 59:1342-1346. 41. Food and Drug Administration. 2012. Bad Bug Book, Foodborne Pathogenic Microorganisms and Natural Toxins, 2nd ed. Center for Food Safety and Applied Nutrition, U.S. Department of Health and Human Services, Washington, DC. 42. Franke, J., S. Franke, H. Schmidt, A. Schwarzkopf, L. H. Wieler, G. Baljer, L. Beutin, and H. Karch. 1994. Nucleotide sequence analysis of enteropathogenic Escherichia coli (EPEC) adherence factor probe and development of PCR for rapid detection of EPEC harboring virulence plasmids. J. Clin. Microbiol. 32:2460-2463. 43. Fu, Z., S. Rogelj, and T. L. Kieft. 2005. Rapid detection of Escherichia coli O157:H7 by immunomagnetic separation and real-time PCR. Int. J. Food Microbiol. 99:47-57. 44. Gannon, V. P., S. D’Souza, T. Graham, R. K. King, K. Rahn, and S. Read. 1997. Use of the flagellar H7 gene as a target in multiplex PCR assays and improved specificity in identification of enterohemorrhagic Escherichia coli strains. J. Clin. Microbiol. 35:656-662. 45. Garner, O., A. Mochon, J. Branda, C.-A. Burnham, M. Bythrow, M. Ferraro, C. Ginocchio, R. Jennemann, R. Manji, G. W. Procop, S. Richter, J. Rychert, L. Sercia, L. Westblade, and M. Lewinski. 2014. Multi-centre evaluation of mass spectrometric identification of anaerobic bacteria using the VITEK MS system. Clin. Microbiol. Infect. 20:335-339. 46. Gessler, F., K. Hampe, and H. Bohnel. 2005. Sensitive detection of botulinum neurotoxin types C and D with an immunoaffinity chromatographic column test. Appl. Environ. Microbiol. 71:7897-7903. 47. Gonzalez, I., K. A. Grant, P. T. Richardson, S. F. Park, and M. D. Collins. 1997. Specific identification of the enteropathogens Campylobacter jejuni and Campylobacter coli by using a PCR test based on the ceuE gene encoding a putative virulence determinant. J. Clin. Microbiol. 35:759-763. 48. Green, J., K. Henshilwood, C. I. Gallimore, D. W. G. Brown, and D. N. Lees. 1998. A nested reverse transcriptase PCR assay for detection of small round-structured viruses in environmentally contaminated molluscan shellfish. Appl. Environ. Microbiol. 64:858-863. 49. Han, F., and B. Ge. 2008. Evaluation of a loop-mediated isothermal amplification assay for detecting Vibrio vulnificus in raw oysters. Foodborne Pathog. Dis. 5:311-320. 50. Hazen, T. H., R. J. Martinez, Y. Chen, P. C. Lafon, N. M. Garrett, M. B. Parsons, C. A. Bopp, M. C. Sullards, and P. A. Sobecky. 2009. Rapid identification of f by whole-cell matrixassisted laser desorption ionization-time of flight mass spectrometry. Appl. Environ. Microbiol. 75:6745-6756. 51. Heid, C. A., J. Stevens, K. J. Livak, and P. M. Williams. 1996. Real time quantitative PCR. Genome Res. 6:986-994. 51a. Hein, I., G. Flekna, M. Krassnig, and M. Wagner. 2006. Realtime PCR for the detection of Salmonella spp. in food: An

Rapid Methods for the Detection and Identification of Foodborne Pathogens

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

alternative approach to a conventional PCR system suggested by the FOOD-PCR project. J. Microbiol. Methods. 66:538-547. Hibi, K., A. Abe, E. Ohashi, K. Mitsubayashi, H. Ushio, T. Hayashi, H. Ren, and H. Endo. 2006. Combination of immunomagnetic separation with flow cytometry for detection of Listeria monocytogenes. Anal. Chim. Acta. 573–574:158163. Hill, C. S. 2001. Molecular diagnostic testing for infectious diseases using TMA technology. Expert Rev. Mol. Diagn. 1:445-455 Hoerner, R., J. Feldpausch, R. L. Gray, S. Curry, Z. Islam, T. Goldy, F. Klein, T. Tadese, J. Rice, and M. Mozola. Reveal Salmonella 2.0 test for detection of Salmonella spp. in foods and environmental samples. J. AOAC Int. 94:1467-1480. Hoerner, R., J. Feldpausch, R. L. Gray, S. Curry, P. Lewis, J. Tolan, T. Goldy, F. Klein, B. Neiditch, E. Hosking, P. Norton, J. Rice, and M. Mozola. 2011. Reveal E. coli 2.0 method for detection of Escherichia coli O157:H7 in raw beef. J. AOAC Int. 94:1835-1845. Hong, J., W. K. Jung, J. M. Kim, S. H. Kim, H. C. Koo, J. Ser J, and Y. H. Park. 2007. Quantification and differentiation of Campylobacter jejuni and Campylobacter coli in raw chicken meats using a real-time PCR method. J. Food Prot. 70:20152022. Hoorfar, J., N. Cook, B. Malorny, M. Wagner, D. De Medici, A. Abdulmawjood, and P. Fach. 2004. Diagnostic PCR, making internal amplification control mandatory. Lett. Appl. Microbiol. 38:79-80. Hughes, D., A. E. Dailianis, and L. Hill. 2001. TECRA Unique test for rapid detection of Salmonella in food: a collaborative study. J. AOAC Int. 84:416-429. Hughes, D., A. E. Dailianis, and L. Hill. 2003. Salmonella in foods: new enrichment procedure for TECRA Salmonella visual immunoassay using a single RV(10) only, TT only, or dual RV(10) and TT selective enrichment broths: a collaborative study. J. AOAC Int. 86:775-790. Jensen, A. N., M. T. Andersen, A. Dalsgaard, D. L. Baggesen, and E. M. Nielsen. 2005. Development of real-time PCR and hybridization methods for detection and identification of thermophilic Campylobacter spp in pig faecal samples. J. Appl. Microbiol. 99:292-300. Johnson, J. R., T. A. Russo, P. I. Tarr, U. Carlino, S. S. Bilge, J. C. Vary Jr., and A. L. Stell. 2000. Molecular epidemiological and phylogenetic associations of two novel putative virulence genes, iha and iroNE. coli among Escherichia coli isolates from patients with urosepsis. Infect. Immun. 68:3040-3047. Johnson, W., S. Tyler, E. Ewan, F. Ashton, G. Wang, and K. Rozee. 1992. Detection of genes coding for listeriolysin and Listeria monocytogenes antigen A (lmA) in Listeria spp. by the polymerase chain reaction. Microb. Pathog. 12:79-86. Joshi, R., H. Janagama, H. P. Dwivedi, T. M. Senthil Kumar, L. A. Jaykus, J. Schefers, and S. Sreevatsan. 2009. Selection, characterization, and application of DNA aptamers for the capture and detection of Salmonella enterica serovars. Mol. Cell. Probes. 23:20-28. Jung, Y. S., J. F. Frank, R. E. Brackett, and J. Chen. 2003. Polymerase chain reaction detection of Listeria monocytogenes on frankfurters using oligonucleotide primers targeting the genes encoding internalin AB. J. Food Prot. 66:237-241. Karas, M., D. Bachmann, and F. Hillenkamp. 1985. Influence of the wavelength in high-irradiance ultraviolet laser desorption mass spectrometry of organic molecules. Anal. Chem. 57:2935-2939. Karas, M., and F. Hillenkamp. 1988. Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal. Chem. 60:2299-2301.

| 149

Compendium of Methods for the Microbiological Examination of Foods |

67. Karger, A., M. Ziller, B. Bettin, B. Mintel, S. Schares, and L. Geue. 2011. Determination of serotypes of Shiga toxinproducing Escherichia coli isolates by intact cell matrix-assisted laser desorption ionization-time of flight mass spectrometry. Appl. Environ. Microbiol. 77:896-905. 68. Keer, J. T., and L. Birch. 2003. Molecular methods for the assessment of bacterial viability. J. Microbiol. Meth. 53:175-183. 69. Kim, J. S., C. R. Taitt, F. S. Ligler, and G. P. Anderson. 2010. Multiplexed magnetic microsphere immunoassays for detection of pathgogens in foods. Sens. Instrum. Food Qual. Saf. 4:73-81. 70. Klena, J. D., C. T. Parker, K. Knibb, J. C. Ibbitt, P. M. L. Devane, S. T. Horn, W. G. Miller, and M. E. Konkel. 2004. Differentiation of Campylobacter coli, Campylobacter jejuni, Campylobacter lari, and Campylobacter upsaliensis by a multiplex PCR developed from the nucleotide sequence of the lipid A gene lpxA. J. Clin. Microbiol. 42:5549-5557. 71. Klimuszko, A., and D. Krutkiewitz. 2010. Genotyping and PCR detection of potential virulence genes in Campylobacter jejuni and Campylobacter coli isolates from different sources in Poland. Folia Microbiol. 55:167-175. 72. Kohler, S., M. Leimeister-Wachter, T. Chakraborty, F. Lottspeich, and W. Goebel. 1990. The gene coding for protein p60 of Listeria monocytogenes and its use as a specific probe for Listeria monocytogenes. Infect. Immun. 58:1943-1950. 73. Kolbert, C. P., and D. H. Persing. 1999. Ribosomal DNA sequencing as a tool for identification of bacterial pathogens. Curr. Opin. Microbiol. 2:299-305. 74. Konkel, M. E., S. A. Gray, B. J. Kim, S. G. Garvis, and J. Yoon. 1999. Identification of the enteropathogens Campylobacter jejuni and Campylobacter coli based on the cadF virulence gene and its product. J. Clin. Microbiol. 37:510-517. 75. Kos, V. N., A. Gibreel, M. Keelan, and D. E. Taylor. 2006. Species identification of erythromycin-resistant Campylobacter isolates and optimization of a duplex PCR for rapid detection. Res. Microbiol. 157:503-507. 76. Kuhns, M., A. E. Zautner, W. Rabsch, O. Zimmermann, M. Weig, O. Bader, and U. Groß. 2012. Rapid discrimination of Salmonella enterica serovar Typhi from other serovars by MALDI-TOF mass spectrometry. PLoS One. 7:e40004. 77. Kutyavin, I. V., I. A. Afonina, A. Mills, V. V. Gorn, E. A. Lukhtanov, E. S. Belousov, M. J. Singer, D. K. Walburger, S. G. Lokhov, A. A. Gall, R. Dempcy, M. W. Reed, R. B. Meyer, and J. Hedgpeth. 2000. 39-minor groove binder-DNA probes increase sequence specificity at PCR extension temperatures. Nucleic Acids Res. 28:655-661. 78. Kwang, J., E. T. Littledike, and J. E. Keen. 1996. Use of the polymerase chain reaction for Salmonella detection. Lett. Appl. Microbiol. 22:46-51. 79. Lasch, P., M. Drevinek, H. Nattermann, R. Grunow, M. Sta¨mmler, R. Dieckmann, T. Schwecke, and D. Naumann. 2010. Characterization of Yersinia using MALDI-TOF mass spectrometry and chemometrics. Anal. Chem. 82:8464-8475. 80. Lasch, P., W. Beyer, H. Nattermann, M. Sta¨mmler, E. Siegbrecht, R. Grunow, and D. Naumann. 2009. Identification of Bacillus anthracis by using matrix-assisted laser desorption ionization-time of flight mass spectrometry and artificial neural networks. Appl. Environ. Microbiol. 75:72297242. 81. Leuschner, R. G. K., N. Beresford-Jones, and C. Robinson. 2003. Difference and consensus of whole cell Salmonella enterica subsp. enterica serovars matrix-assisted laser desorption/ionization time-of-flight mass spectrometry spectra. Lett. Appl. Microbiol. 38:24-31. 82. Liming, S.H., and A.A. Bhagwat. 2004. Application of a molecular beacon-real-time PCR technology to detect

150 |

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

Salmonella species contaminating fruits and vegetables. Int. J. Food Microbiol. 95:177-187. Lindstedt, B. A., E. Heir, T. Vardund, and G. Kapperud. 2000. Fluorescent amplified-fragment length polymorphism genotyping of Salmonella enterica subsp. enterica serovars and comparison with pulsed-field gel electrophoresis typing. J. Clin. Microbiol. 38:1623-1627. Liu, D. 2006. Identification, subtyping and virulence determination of Listeria monocytogenes, an important foodborne pathogen. J. Med. Microbiol. 55:645-659. Liu, D., A. J. Ainsworth, F. W. Austin, and M. L. Lawrence. 2003a. Characterization of virulent and avirulent Listeria monocytogenes strains by PCR amplification of putative transcriptional regulator and internalin genes. J. Med. Microbiol. 52:1066-1070. Liu, D., A. J. Ainsworth, F. W. Austin, and M. L. Lawrence. 2004b. Identification of a gene encoding a putative phosphotransferase system enzyme IIBC in Listeria welshimeri and its application for diagnostic PCR. Lett. Appl. Microbiol. 38:151157. Liu, D., A. J. Ainsworth, F. W. Austin, and M. L. Lawrence. 2004c. PCR detection of a putative N-acetylmuramidase gene from Listeria ivanovii facilitates its rapid identification. Vet. Microbiol. 101:83-89. Liu, D., M. Lawrence, F. W. Austin, and A. J. Ainsworth. 2005b. Isolation and PCR amplification of a species-specific, oxidoreductase coding gene region in Listeria grayi. Can. J. Microbiol. 51:95-98. Liu, D., M. L. Lawrence, A. J. Ainsworth, and F. W. Austin. 2004d. Species-specific PCR determination of Listeria seeligeri. Res. Microbiol. 155:741-746. Liu, G., Y. Han, X. Li, and S. Song. 2006. Applicability of a rapid method based on immunomagnetic capture-fluorescent PCR assay for Campylobacter jejuni. Food Control. 17:527-532. Livezey, K., S. Kaplan, M. Wisniewski, and M. M. Becker. 2013. A new generation of food-borne pathogen detection based on ribosomal RNA. Ann. Rev. Food Sci. Technol. 4:313-325. Longhi, C., A. Maffeo, M. Penta, G. Petrone, L. Seganti, and M. P. Conte. 2003. Detection of Listeria monocytogenes in Italian-style soft cheeses. J. Appl. Microbiol. 94:879-885. Malainine, S. M., W. Moussaoui, G. Pre´vost, J.-M. Scheftel, and R. Mimouni. 2013. Rapid identification of Vibrio parahaemolyticus isolated from shellfish, sea water and sediments of the Khnifiss lagoon, Morocco, by MALDITOF mass spectrometry. Lett. Appl. Microbiol. 56:379-386. Malorny, B., C. Lofstrom, M. Wagner, N. Kramer, and J. Hoorfar. 2008. Enumeration of Salmonella bacteria in food and feed samples by real-time PCR for quantitative microbial risk assessment. Appl. Environ. Microbiol. 74:1299-1304. Malorny, B., J. Hoorfar, M. Hugas, A. Heuvelink, P. Fach, L. Ellerbroek, C. Bunge, C. Dorn, and R. Helmuth R. 2003. Interlaboratory diagnostic accuracy of a Salmonella specific PCR-based method. Int. J. Food Microbiol. 89:241-249. Manji, R., M. Bythrow, J. A. Branda, C.-A. D. Burnham, M. J. Ferraro, O. B. Garner, R. Jennemann, M. A. Lewinski, A. B. Mochon, G. W. Procop, S. S. Richter, J. A. Rychert, L. Sercia, L. F. Westblade, and C. C. Ginocchio. 2014. Multi-center evaluation of the VITEK MS system for mass spectrometric identification of non-Enterobacteriaceae Gram-negative bacilli. Eur. J. Clin. Microbiol. Infect. Dis. 33:337-346. Mansfield, L. P., and S. J. Forsythe. 2000. The detection of Salmonella using a combined immunomagnetic separation and ELISA end-detection procedure. Lett. Appl. Microbiol. 31:279-283.

|

98. Martiny, D., A. Dediste, L. Debruyne, L. Vlaes, N. B. Haddou, P. Vandamme, and O. Vandenberg. 2011. Accuracy of the API Campy system, the Vitek 2 Neisseria-Haemophilus card and matrix-assisted laser desorption ionization timeof-flight mass spectrometry for the identification of Campylobacter and related organisms. Clin. Microbiol. Infect. 17:1001-1006. 99. Martiny, D., L. Busson, I. Wybo, R. A. El Haj, A. Dediste, and O. Vandenberg. 2012. Comparison of the Microflex LT and Vitek MS systems for routine identification of bacteria by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J. Clin. Microbiol. 50(4):1313-25. 100. Martiny, D., A. Visscher, B. Catry, S. Chatellier, and O. Vandenberg. 2013. Optimization of Campylobacter growth conditions for further identification by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). J. Microbiol. Meth. 94:221-223. 101. Mengaud, J., M. F. Vicente, J. Chenevert, J. M. Pereira, C. Geoffroy, B. Gicquel-Sanzey, F. Baquero, J. C. Perez-Diaz, and P. Cossart. 1988. Expression in Escherichia coli and sequence analysis of the listeriolysin O determinant of Listeria monocytogenes. Infect. Immun. 56:766-772. 102. Mercanoglu, B., and M. W. Griffiths. 2005. Combination of immunomagnetic separation with real-time PCR for rapid detection of Salmonella in milk, ground beef, and alfalfa sprouts. J. Food Prot. 68:557-561. 103. Moore, M. M., and M. D. Feist. 2007. Real-time PCR method for Salmonella spp targeting the stn gene. J. Appl. Microbiol. 102:516-530. 104. Morisset, D., D. Sˇtebih, M. Milavec, K. Gruden, and J. Zˇel. 2013. Quantitative analysis of food and feed samples with droplet digital PCR. PLoS. 8:e62583. 105. Mozola, M., P. Norton, S. Alles, R.L. Gray, J. Tolan, O. Caballero, L. Pinkava, E. Hosking, K. Luplow, and J. Rice. 2013. Validation of the ANSR Salmonella method for detection of Salmonella spp. in selected foods and environmental samples. J. AOAC Int. 96:842-853. 106. Muldoon, M. T., A. C. O. Allen, V. Gonzalez, M. Sutzko, and K. Lindpaintner. 2012. SDIX RapidChek Listeria F.A.S.T.TM environmental test system for the detection of Listeria species on environmental surfaces. J. AOAC Int. 95:850-859. 107. Nagarajan, A. G., G. Karnam, A. Lahiri, U. S. Allam, and D. Chakravortty. 2009. Reliable means of diagnosis and serovar determination of blood-borne Salmonella strains: quick PCR amplification of unique genomic loci by novel primer sets. J. Clin. Microbiol. 47:2435-2441. 108. Niessen, L., J. Luo, C. Denschlag, and R. F. Vogel. 2013. The application of loop-mediated isothermal amplification (LAMP) in food testing for bacterial pathogens and fungal contaminants. Food Microbiol. 36:191-206. 109. O’Grady, J., S. Sedano-Balba´s, M. Maher, T. Smith, and T. Barry. 2008. Rapid real-time PCR detection of Listeria monocytogenes in enriched food samples based on the ssrA gene, a novel diagnostic target. Food Microbiol. 25:75-84. 110. Oswald, E., P. Pohl, E. Jacquemin, P. Lintermans, K. Van Muylem, A. D. O’Brien, and J. Mainil. 1994. Specific DNA probes to detect Escherichia coli strains producing cytotoxic necrotising factor type 1 or type 2. J. Med. Microbiol. 40:428434. 111. Palumbo, J. D., M. K. Borucki, R. E. Mandrell, and L. Gorski. 2003. Serotyping of Listeria monocytogenes by enzyme-linked immunosorbent assay and identification of mixed-serotype cultures by colony immunoblotting. J. Clin. Microbiol. 41:564-571. 112. Pathmanathan, S. G., N. Cardona-Castro, M. M. Sa´nchezJime´nez, M. M. Correa-Ochoa, S. D. Puthucheary, and K. L. Thong. 2003. Simple and rapid detection of Salmonella strains

Rapid Methods for the Detection and Identification of Foodborne Pathogens

113.

114.

115.

116.

117.

118.

119.

120.

121.

122.

123.

123a.

124.

125.

by direct PCR amplification of the hilA gene. J. Med. Microbiol. 52:773-776. Paton, A. W., and J. C. Paton. 1998. Detection and characterization of Shiga toxigenic Escherichia coli by using multiplex PCR assays for stx1, stx2, eaeA, enterohemorrhagic E. coli hlyA, rfbO111, and rfbO157. J. Clin. Microbiol. 36:598-602. Pinheiro, L. B., V. A. Coleman, C. M. Hindson, J. Herrmann, B. J. Hindson, S. Bhat, and K. R. Emslie. 2012. Evaluation of a droplet digital polymerase chain reaction format for DNA copy number quantification. Anal. Chem. 83:1003-1011. Postollec, F., H. Falentin, S. Pavan, J. Combrisson, and D. Sohier. 2011. Recent advances in quantitative PCR (qPCR) applications in food microbiology. Food Microbiol. 28:848861. Reischl, U., M. T. Youssef, J. Kilwinski, N. Lehn, W. L. Zhang, H. Karch, and N.A. Strockbine. 2002. Real-time fluorescence PCR assays for detection and characterization of Shiga toxin, intimin, and enterohemolysin genes from Shiga toxin-producing Escherichia coli. J. Clin. Microbiol. 40:2555-2565. Richter, S. S., L. Sercia, J. A. Branda, C.-A. D. Burnham, M. Bythrow, M. J. Ferraro, O. B. Garner, C. C. Ginocchio, R. Jennemann, M. A. Lewinski, R. Manji, A. B. Mochon, J. A. Rychert, L. F. Westblade, and G. W. Procop. 2013. Identification of Enterobacteriaceae by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry using the VITEK MS system. Eur. J. Clin. Microbiol. Infect. Dis. 32:1571-1578. Rijpens, N. P., and L. M. Herman. 2002. Molecular methods for identification and detection of bacterial food pathogens. J. AOAC Int. 85:984-995. Rodrıguez-Lazaro, D., M. Hernandez, M. Scortti, T. Esteve, J. A. Vazquez-Boland, and M. Pla. 2004. Quantitative detection of Listeria monocytogenes and Listeria innocua by real-time PCR: assessment of hly, iap, and lin02483 targets and AmpliFluor technology. Appl. Environ. Microbiol. 70:1366-1377. Roe, A. J., L. Tysall, T. Dransfield, D. Wang, D. Fraser-Pitt, A. Mahajan, C. Constandinou, N. Inglis, A. Downing, R. Talbot, D. G. Smith, and D. L. Gally. 2007. Analysis of the expression, regulation and export of NleA-E in Escherichia coli O157:H7. Microbiol. 153:1350-1360. Roux, K. H. 1995. Optimization and troubleshooting in PCR. In Dieffenbach, C. W., and G. S. Dveksler (eds.), PCR Primers: a Laboratory Manual, 53-62. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Rychert, J., C.-A. D. Burnham, M. Bythrow, O. B. Garner, C. C. Ginocchio, R. Jennemann, M. A. Lewinski, R. Manji, A. B. Mochon, G. W. Procop, S. S. Richter, L. Sercia, L. F. Westblade, M. J. Ferraro, and J. A. Branda. 2013. Multicenter evaluation of the Vitek MS matrix-assisted laser desorption ionization-time of flight mass spectrometry system for identification of Gram-positive aerobic bacteria. J. Clin. Microbiol. 51:2225-2231. Sails, A. D., A. J. Fox, F. J. Bolton, D. R. Wareing, and D. L. Greenway. 2003. A real-time PCR assay for the detection of Campylobacter jejuni in foods after enrichment culture. Appl. Environ. Microbiol. 69:1383-1390. Sandrin, T. R., J. E. Goldstein, S. Schumaker. 2013. MALDI TOF MS profiling of bacteria at the strain level: a review. Mass Spectrom. Rev. 32:188-217. Schroeder, A., O. Mueller, S. Stocker, et al. 2006. The RIN: an RNA integrity number for assigning integrity values to RNA measurements. BMC Mol. Biol. 7:3. Seo, K. H., I. E. Valentin-Bon, and R. E. Brackett. 2006. Detection and enumeration of Salmonella Enteritidis in

| 151

Compendium of Methods for the Microbiological Examination of Foods |

126.

127.

128.

129.

130.

131.

132.

133.

134.

135.

136. 137.

138.

139.

152 |

homemade ice cream associated with an outbreak, comparison of conventional and real-time PCR methods. J. Food Prot. 69:639-43. Sharma, S. K., B. S. Eblen, R. L. Bull, D. H. Burr, and R. C. Whiting. 2005. Evaluation of lateral-flow Clostridium botulinum neurotoxin detection kits for food analysis. Appl. Environ. Microbiol. 71:3935-3941. Shinagawa, K., M. Mitsumori, N. Matsusaka, and S. Sugii. 1991. Purification of staphylococcal enterotoxins A and E by immunoaffinity chromatography using murine monoclonal antibody with dual specificity for both of these toxins. J. Immunol. Meth. 139:49-53. Shinagawa, K., T. Takechi, and N. Matsusaka. 1991. Purification of an enterotoxin produced by Bacillus cereus by immunoaffinity chromatography using a monoclonal antibody. Can. J. Microbiol. 38:153-156. Singh, J., V. K. Batish, and S. A. Grover. 2009b. Scorpion probe-based realtime PCR assay for detection of E. coli O157, H7 in dairy products. Foodborne Pathog. Dis. 6:395-400. Skjerve, E., L. M. Rorvik, and O. Olsvik. 1990. Detection of Listeria monocytogenes in foods by immunomagnetic separation. Appl. Environ. Microbiol. 56:3478-3481. Stephan, R., N. Cernela, D. Ziegler, V. Pflu¨ger, M. Tonolla, D. Ravasi, M. Fredriksson-Ahomaa, and H. Ha¨chler. 2011. Rapid species specific identification and subtyping of Yersinia enterocolitica by MALDI-TOF mass spectrometry. J. Microbiol. Meth. 87:150-153. Stewart, D. S., K. F. Reineke, and M. L. Tortorello. 2002. Comparison of Assurance Gold Salmonella EIA, BAX, for screening/Salmonella, and GENE-TRAK Salmonella DLP rapid assays for detection of Salmonella in alfalfa sprouts and sprout irrigation water. J. AOAC Int. 85:395-403. Stintzi, A. 2003. Gene expression profile of Campylobacter jejuni in response to growth temperature variation. J. Bacteriol. 185:2009-2016. Tanaka, K., H. Waki, Y. Ido, S. Akita, Y. Yoshida, T. Yoshida, and T. Matsuo. 1988. Protein and polymer analyses up to m/ z 100 000 by laser ionization time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 2:151-153. Tebbs, R. S., P. M. Brzoska, M. R. Furtado, and O. V. Petrauskene. 2011. Design and validation of a novel multiplex real-time PCR assay for Vibrio pathogen detection. J. Food Prot. 74:939-48. Thompson, L., and C. Lindhardt. 2006. Singlepath Salmonella. J. AOAC Int. 89:417-432. Tsilia, V., B. Devreese, I. de Baenst, B. Mesuere, A. Rajkovic, M. Uyttendaele, T. Van de Wiele, and M. Heyndrickx. 2012. Application of MALDI-TOF mass spectrometry for the detection of enterotoxins produced by pathogenic strains of the Bacillus cereus group. Anal. Bioanal. Chem. 404:1691-1702. US Department of Agriculture-Food Safety and Inspection Service. 2013. Detection and Isolation of non-O157 Shiga Toxin-Producing Escherichia coli (STEC) from Meat Products and Carcass and Environmental Sponges; Microbiology Laboratory Guidebook 5B.04. US Department of Agriculture-Food Safety and Inspection Service (USDA-FSIS). 2013. USDA-FSIS list of Foodborne Pathogen Test Kits Validated by Independent Organizations.

140.

141.

142.

143.

144.

145.

146.

147.

148.

149.

150.

151. 152.

153.

Available at http://www.fsis.usda.gov/wps/wcm/connect/ 909c8279-6865-424d-ab7a-e1f165646c63/Validated-Test-KitSpreadsheet.xls?MOD5AJPERES. Accessed July 14, 2014. VanGuilder, H. D., K. E. Vrana, and W. M. Freeman. 2008. Twenty-five years of quantitative PCR for gene expression analysis. BioTechniques. 44:619-626. Va´zquez-Novelle, M. D., A. J. Pazos, M. Abad, J. L. Sa´nchez, and M. L. Pe´rez-Paralle´. 2005. Eight-hour PCR-based procedure for the detection of Salmonella in raw oysters. FEMS Microbiol. Lett. 243:279-283. Vinje´, J., H. Vennema, L. Maunula, C. H. von Bonsdorff, M. Hoehne, E. Schreier, A. Richards, J. Green, D. Brown, S. S. Beard, S. S. Monroe, E. de Bruin, L. Svensson, and M. P. Koopmans. 2003. International collaborative study to compare reverse transcriptase PCR assays for detection and genotyping of noroviruses. J. Clin. Microbiol. 41:1423-33. Volokhov, D., A. Rasooly, K. Chumakov, and V. Chizhikov. 2002. Identification of Listeria species by microarray-based assay. J. Clin. Microbiol. 40:4720-4728. Wadud, S., C. G. Leon-Velarde, N. Larson, and J. A. Odumeru. 2010. Evaluation of immunomagnetic separation in combination with ALOA Listeria chromogenic agar for the isolation and identification of Listeria monocytogenes in readyto-eat foods. J. Microbiol. Meth. 81:153-159. Weagant, S. D., J. L. Bryant, and K. G. Jinneman. 1995. An improved rapid technique for isolation of Escherichia coli O157:H7 from foods. J. Food Prot. 58:7-12. Weagant, S. D., and A. J. Bound. 2001. Evaluation of techniques for enrichment and isolation of Escherichia coli O157:H7 from artificially contaminated sprouts. Int. J. Food Microbiol. 71:87-92. Wernar, K., K. Heuvelman, S. Notermans, E. Domann, E. M. Leimeister-Wachter, and T. Chakraborty. 1992. Suitability of the prfA gene, which encodes a regulator of virulence genes in Listeria monocytogenes in the identification of pathogenic Listeria spp. Appl. Environ. Microbiol. 58:765-768. Whelan, J. A., N. B. Russell, and M. A. Whelan. 2003. A method for the absolute quantification of cDNA using realtime PCR. J. Immunol. Meth. 278:261-269. Whyte, P., K. Mc Gill, J. D. Collins, and E. Gormley. 2002. The prevalence and PCR detection of Salmonella contamination in raw poultry. Vet. Microbiol. 89:53-60. Willford, J., K. Mills, and L. D. Goodridge. 2009. Evaluation of three commercially available enzyme-linked immunosorbent assay kits for detection of shiga toxin. J. Food. Prot. 72:741-747. Wulff, K. (1986) Nucleic acids as analytes in laboratory diagnosis. Arzneimittelforschung. 36:157-61. Zhang, G., E. W. Brown, and N. Gonza´lez-Escalona. 2011. Comparison of real-time PCR, reverse transcriptase real-time PCR, loop-mediated isothermal amplification, and the FDA conventional microbiological method for the detection of Salmonella spp. in produce. Appl. Environ. Microbiol. 77:6495-501. Zhu, P., D. R. Shelton, S. Li, D. L. Adams, J. S. Karns, P. Amstutz, and C. M. Tang. 2011. Detection of E. coli O157: H7 by immunomagnetic separation coupled with fluorescence immunoassay. Biosens. Bioelectron. 30:337-341.

|

CHAPTER 12

|

Molecular Typing and Differentiation Brian B. Oakley, Narjol Gonzalez-Escalona, and Marirosa Molina

12.1

INTRODUCTION

In this chapter, general background and bench protocols are provided for a number of molecular typing techniques in common use today. Methods for the molecular typing and differentiation of microorganisms began to be widely adopted following the development of the polymerase chain reaction (PCR) in the 1980s. For the foreseeable future, rapid advances in DNA sequencing will continue to drive the evolution of new molecular typing approaches. In particular, we foresee the increased adoption of wholegenome based approaches, as described below. Molecular typing methods generally have important advantages of speed and discriminatory power compared to phenotypic characterization methods, but can also change rapidly, given the rapid rate of technological advances. In this chapter, techniques that have either already been superseded by others or are expected to fall out of common usage in the near future are given brief mention for historical purposes, but the more modern technique is given preference. For example, multilocus enzyme electrophoresis (MLEE) has been largely replaced by multilocus sequence typing (MLST),1,92,93 a development which has led to the creation of large databases of MLST profiles—currently for nearly 100 bacterial species and four Candida species. See the MLST section below for more information. For each method, background information and selected noteworthy applications to food microbiology are given. Additional background information and recommendations for the molecular typing of foodborne pathogens can be found elsewhere.40,62,150 For many of the methods discussed here, data analysis can be an important task. In modern biology data analysis—that is, bioinformatics broadly defined—has become its own field and as such is outside the scope of this chapter. For the methods below which can generally be considered ‘‘fingerprinting-type‘‘ approaches (restriction fragment length polymorphism [RFLP], automated ribosomal intergenic spacer analysis [ARISA], repetitive extragenic palindromic-PCR [REP-PCR], pulsed-field gel electrophoresis [PFGE], etc.), the forms of the data derived are fairly simple, representing the number, types, and sometimes intensities of bands or peaks for each sample. These types of data are typically transformed to a

(dis)similarity matrix and can be represented as dendrograms to illustrate relatedness among samples. For the most data-intensive method covered here, whole-genome sequencing, readers are referred to several excellent recent reviews.27,87 The material presented here provides a broad overview of methods for molecular typing and differentiation with the inclusion of generalized laboratory protocols.

12.2

SEROTYPING

12.21

Background

Serotyping is the identification of strains of the same species of microorganism based on the differential expression of antigens on the cell surface. Antigenic determinants can be found in a variety of cell surface components, such as membrane proteins, flagella, fimbriae, lipopolysaccharides, or polysaccharide capsules. Organisms characterized by serotyping are called serotypes. The methodology is often used as a first-line typing method because most procedures are easy to implement, reproducible, and results are easy to interpret.156 Serologic tests may include fluorescent and enzyme-labeling assays, latex agglutination, or coagglutination. Some disadvantages of the procedure include a large antigen cross-reactivity among certain strains reducing the discriminatory power of the assay,84 and the high cost associated with or lack of availability of certain agglutination antisera.84,24

12.211

Noteworthy Applications to Foodborne Pathogens Serotyping has been applied to food outbreak investigations mainly because groups of strains can be rapidly screened and it is possible to quickly discriminate strains related to an outbreak from those that do not belong to the outbreak. This first line of identification allows for better resource utilization, as strains identified by serotyping can be further screened with other molecular techniques that allow for more detailed identification.156 Screening by serotyping can avoid other more expensive or time-consuming approaches. In the food industry, Listeria serotyping has been applied to track contaminant strains through the food chain. Usually, in outbreak investigations where

| 153 |

Compendium of Methods for the Microbiological Examination of Foods |

L. monocytogenes is involved, serotyping is followed by PFGE.30,156,157 More recently, a multiplex PCR serotyping assay was designed to identify the four major L. monocytogenes serotypes,30 and along with random amplification of polymorphic (RPD) DNA this allowed the genotypic analysis of 18 L. monocytogenes strains isolated from milk and ready-to-eat milk products.4 Serotyping has also been applied to characterize Salmonella isolates in a variety of retail foods, along with PFGE and antimicrobial resistance analysis.162,163

12.3 12.31

PCR-RFLP and AFLP Background

The digestion of PCR products with restriction enzymes creates fragments that vary in number and length depending on the sequence of the PCR product. Thus, RFLP can be used as a method of molecular typing and differentiation. Early examples of the use of RFLP for typing and epidemiology can be found in the Mycobacterium literature.22,140 RFLP is typically performed on PCR products obtained from a single isolate, and for the technique to be useful, one needs to find a target containing suitable polymorphisms to discriminate among strains of interest. For foodborne pathogens, genes involved in virulence are frequently a target for PCR-RFLP. The amplified DNA fragment is digested by a specific restriction endonuclease or combination of nucleases, and run on an agarose gel to visualize the resulting polymorphism. The digestion should result in a banding pattern containing at least 7–10 fragments. Amplified fragment length polymorphism (AFLP) is based on digestion of genomic DNA with restriction enzymes followed by ligation with primers complementary to the cut sites and subsequent PCR.154 Typically 40–200 bands per strain are produced, to give a fingerprint reflecting mutations in the restriction sites and size variation of amplified fragments.125

12.311

Noteworthy Applications to Foodborne Pathogens RFLP of PCR products derived from broad-range or universal primers has advantages of wide applicability with little prior knowledge of gene sequence; ribosomal ribonucleic acid (rRNA) genes are the most commonly used for such an approach. For example, 16S-RFLP has been used effectively to discriminate among Campylobacter, Helicobacter, Arcobacter, and Wolinella isolates13,95 and among Listeria isolates.152 RFLP patterns of 23S rRNA genes cut with up to four restriction enzymes were used to unambiguously classify Listeria isolates into one of six species, and also identify mixed cultures of L. monocytogenes and L. innocua.109 In a study of lactic acid bacteria70 active in the production of sorghum beer, RFLP was applied to the 16S– 23S intergenic spacer region to identify L. fermenterum as the dominant in the malt during mashing and acidification.126 RFLP applied to the listeriolysin O virulence gene was able to show that subtypes of Listeria isolated from various foods were different from clinical outbreak strains isolated from humans.50 For Campylobacter, flaA-RFLP has proved to be a useful tool for detection, speciation, and genotyping of strains.83,136 154 |

For Staphylococcus, RFLP of coagulase genes has been used as an effective epidemiological tool,47 but unrelated strains of S. aureus have also been found to share identical coagulase RFLP patterns in AluI digests of hypervariable regions.131 In a comparison of random amplified polymorphic DNA (RAPD), PFGE, and several other singlegene PCR approaches, Schmitz et al. found coa-RFLP to be the least discriminatory method.127 For any single gene approach such as RFLP, the gene must be chosen carefully for adequate discriminatory power. Modifications of RFLP include T-RFLP, in which one of the primers used for PCR is conjugated with a fluorescent molecule and thus the sizes of the terminal restriction fragment (T-RF) can be determined by an automated sequencer.85 T-RFLP is generally used to characterize the microbial community in a complex sample (by using broad-range primers to target 16S rRNA genes, for example) rather than to characterize axenic cultures. AFLP has been widely used for the typing of medically important and foodborne pathogens. In a classic study, Jansen et al. validated the use of AFLP for a variety of taxa, including Bacillus, Acinetobacter, Clostridium, Pseudomonas, and Vibrio.67

12.4 12.41

ARISA Background

ARISA, automated ribosomal intergenic spacer analysis,38,118 exploits length variation in the intergenic region between the 16S and 23S rRNA genes encoding for the small and large subunits of the ribosome, respectively. PCR products obtained from primers targeting this region can be individually electrophoresed in a manual approach, or if fluorescently labeled, detected with a capillary sequencer in relatively high-throughput fashion. Like T-RFLP, ARISA is most commonly used to characterize mixed populations in complex samples, but can also be used to type and discriminate individual strains.

12.411

Noteworthy Applications to Foodborne Pathogens One of the first examples of molecular typing and differentiation using length and copy number variations in the ribosomal spacer regions was by Jensen et al., who validated the method by discriminating among strains of Citrobacter, Enterobacter, Escherichia, Listeria, Proteus, Salmonella, Staphylococcus, and Yersinia.69 This approach has since been used for numerous applications, including distinguishing lactic acid bacteria and Staphylococcus in sausage,5 identifying bacteria involved in flavor and aroma production in cheeses,19,94 and clinical isolates of Staphylococcus.20,53 12.5 12.51

SSCP Background

Single-strand conformation polymorphism (SSCP) is based on the principle that nucleotide sequence determines the folding conformation of single-stranded DNAs and thus affects mobility when electrophoresed. Differential mobility can be used in SSCP to discriminate two sequences on the basis of a single base pair, either with electrophoresis in

| Molecular Typing and Differentiation

a non-denaturing polyacrylamide gel107,108 or with capillary electrophoresis.76 SSCP and the related heteroduplex analysis (HA) that detects changes in double-stranded DNA71,101 are both sensitive to mutations in any region of a gene. This is in contrast to the action of restriction enzymes used for RFLP, whereby discriminatory power relies on recognition of a particular sequence motif.

12.511

Noteworthy Applications to Foodborne Pathogens Oh et al.106 validated capillary SSCP of eight 16S rRNA assays as specific and sensitive for common foodborne pathogens, including E. coli, Campylobacter jejuni, Salmonella enterica, L. monocytogenes, Vibrio parahaemolyticus, S. aureus, and Bacillus cereus. In a study of the utility of SSCP for distinguishing Salmonella serovars, Nair et al.102 found that SSCP with a 1.6 kb fragment of the groEL gene could discriminate among 10 different serovars and also distinguish strains within a particular serovar. In this same study, RFLP grouped all the strains into one of only three profiles, indicating greater resolution of SSCP versus RFLP. For Listeria, Vaneechoutte et al.152 found generally equivalent discriminatory power of RFLP and SSCP of 16S rRNA genes applied to a collection of isolates belonging to six species of Listeria. 12.6 12.61

PHAGE TYPING Background

Phage typing is a phenotypic method used for detecting single strains of bacteria through the use of bacteriophages (phages) or viruses that infect bacteria. Phages are usually extremely host specific, and tend to infect only specific species and even specific strains of bacteria. The method takes advantage of the variable sensitivity of specific bacterial strains to a set of bacteriophages. Phage typing remains a useful, cost-effective method that can be used to augment other more sophisticated approaches available for the identification of specific bacterial strains. The specificity of the majority of the phages described to date derives from the recognition of surface molecules in susceptible bacteria by tail-associated phage proteins.33,54,130,137 The best candidates for detection purposes are virulent phages which are unable to integrate into their host genome, resulting in the death of their host after successful infection. For this reason, these agents are good candidates for biocontrol approaches.54 Phage typing is a fast, cost-effective, reproducible method that requires no specialized equipment. Briefly, a culture of the bacterial strain to be typed is grown in agar, usually for 18–24 hours. Then, the bacterial lawn is inoculated with a scheme of bacteriophages, excess inoculum is dried, and plates are incubated for plaque development.23 To facilitate reading, a grid is usually drawn on the base of the Petri dish to mark out different regions. Each square of the grid is inoculated with a different phage. The susceptible phage regions will show a circular clearing where the bacteria have been lyzed. Lytic patterns are used for differentiation.

12.611

Noteworthy Applications to Foodborne Pathogens Phage typing provides a way to compare international surveillance data for pathogen outbreaks when there is no molecular information available owing to the pathogen’s recent introduction or description in a new environment.23,6 Because of its specificity, this methodology has been used to track the source of outbreaks of infection in a number of epidemiological studies.2,6,51,59,117 Phage-based typing schemes have been described for the most common foodborne pathogens: Salmonella, Campylobacter, E. coli, and Listeria.41,51,59,86,128 12.7 12.71

SEQUENCING OF INDIVIDUAL GENES OR INTERGENIC REGIONS Background

Determining the exact sequence of a gene or intergenic region has several advantages over simply determining the length of a fragment or deriving a banding pattern. Sequence data generally provide greater discriminatory power and more robust taxonomic classifications than fingerprinting methods, and curated collections of sequence data become valuable for designing primers and assessing the specificity and sensitivity of novel assays. Moreover, sequence data, in contrast to banding patterns, are highly archivable and easily comparable among investigators. Although gene sequencing for typing and differentiation has become a standard technique in nearly every microbiology laboratory, historically, the relatively high cost of sequencing was sometimes a deterrent to its use and provided an incentive for researchers and clinicians to use less expensive approaches, such as RFLP. The increasingly widespread adoption of so-called next-generation sequencing in the last several years has greatly reduced the cost of sequencing, a trend expected to continue as technologies such as nanopore sequencing reach commercial maturity.87

12.8 12.81

MULTILOCUS SEQUENCE TYPING Background

Multilocus sequence typing (MLST) or multilocus sequence analysis (MLSA) was introduced in 1998 by Maiden and coworkers as a portable and universal method for characterizing bacteria.93 MLST is similar to multilocus enzyme electrophoresis, but instead of enzyme mobility on a gel, it is based on sequence analysis of chosen housekeeping genes.80 As a sequence-based approach, MLST is more reproducible and better-suited for archival retrieval than fingerprinting methods such as RFLP, which represent a derivation of primary sequence data. MLST is based on the sequencing of housekeeping gene fragments of approximately 400–600 bp in length. Most published MLST schemes vary between six and 10 loci.93 The number of loci depends on the application: for subtyping at least seven are recommended. If population genetic studies are intended, a larger number of loci will be preferable. By sequencing multiple loci, MLST provides two main advantages. First, much greater discriminatory power is possible than with sequencing of an individual gene or intergenic region, simply because of the greater amount of | 155

Compendium of Methods for the Microbiological Examination of Foods |

data and their information content. Second, MLST allows powerful inferences regarding the true evolutionary history of a strain that are not possible with single-locus approaches. In the 1990s, accumulating evidence of microbes exchanging genetic material (horizontal gene transfer) highlighted the value of interrogating multiple loci to determine the true evolutionary relationships among strains.133 In the simplest case, when a single genetic locus is relied on for typing one can easily imagine a scenario in which a recombination event between two strains would lead to either over- or underestimation of strain relatedness, depending on the choice of locus. MLST data are highly amenable to reconstructing evolutionary events and inferring parental and ancestral genotypes; specialized software developed for this use has been widely adopted.37 Reconstructing evolutionary events with MLST can be particularly valuable for understanding the epidemiology of outbreaks of foodborne or medically important pathogens.148 The availability of numerous bacterial genomes allows for the design of any new MLST scheme specific for each microorganism. After sequencing each locus, an allele number is given to each different sequence for that locus. After collecting all locus–allele combinations (or MLST profile), a sequence type (ST) number is assigned to each allele combination. For example, the Vibrio parahaemolyticus MLST scheme uses internal fragments of seven housekeeping genes (Table 12-1). For chromosome I, recA (RecA protein), dnaE (DNA polymerase III, alpha subunit) and gyrB (DNA gyrase, subunit B) and for chromosome II, dtdS (threonine 3-dehydrogenase), pntA (transhydrogenase alpha subunit), pyrC (dihydroorotase) and tnaA (tryptophanase), were chosen.48 A database (http://pubmlst.org/vparahaemolyticus) stores all the data from each strain and allows for searching individual locus sequences or ST profiles; new allele numbers or STs are assigned for new sequence types. MLST has become a preferred method for determining the global epidemiology of bacterial pathogens. Examples include Neisseria meningitidis, S. aureus, and Vibrio parahaemolyticus.48,92,93 Sequence-based MLST provides definitive characterization of bacterial isolates that is consistent from one laboratory to the next. This overcomes the main disadvantage of MLEE, which, because it is gel based, is not highly reproducible between laboratories. In the case of MLST, the sequences are typically stored in public databases that can be readily accessed (http://www.mlst.net or http://pubmlst.org). MLST studies have led to better understanding of the genetic relatedness of strains within a species and have identified the relative evolutionary importance of mutations and lateral transfer events in the evolution of different bacterial species.36,48,65,91,99 This tech-

nique has been used extensively during the last decade to study the epidemiology of numerous human and animal bacterial pathogens as well as fungi (http://pubmlst.org/ databases.shtml). MLST schemes are available at the time of this writing (Table 12-2).

12.82

General Description of Data Analysis and Interpretation

With sequence data obtained after following the wet-bench procedures described below, MLST data can be analyzed in two general ways: (1) methods that determine relationships among organisms on the basis of allelic designations and STs, and (2) methods that analyze nucleotide sequences directly, either by individual genes or by concatenating all loci into a single sequence for each isolate. For the former, which is handled similarly to MLEE data, Unweighted Pair Group Method with Arithmetic Mean (UPGMA) analyses, split decomposition, and/or eBURST (Figure 12-1A) can be employed. This kind of analysis allows for fast identification of clonal complexes for use in epidemiological studies and for studies of population genetics and evolution.36,48,92,93,99 MLST schemes provide a mechanism for timely recognition of evolutionary trends and the emergence of different human or animal pathogens, thus providing an early warning system for potential emerging pathogens.

12.821

Noteworthy Applications to Foodborne Pathogens Particularly notable applications of MLST to foodborne pathogens include characterizations of greater than 800 Campylobacter isolates to identify niche adaptations of particular sequence types and probable examples of horizontal gene transfer among Campylobacter species.28 In a comparison of MLST, PFGE, and serotyping of 175 L. monocytogenes isolates, Revazishvili et al. found MLST more discriminatory than PFGE, which in turn had better resolution than serotyping.120 In several cases, MLST has been found to be less discriminatory than PFGE for Salmonella35,55,147 and E. coli.104,142 12.9 12.91

MULTIPLE-LOCUS VARIABLE NUMBER TANDEM REPEATS Background

Multilocus variable number tandem repeat (VNTR) analysis (MLVA) is based on end-to-end duplication of a specific DNA sequence that repeats in tandem arrays within a locus. VNTRs are DNA sequences of varying copy numbers that are widely dispersed throughout the bacterial genome.81,149 MLVA has been developed for many bacterial species and is particularly useful in distinguishing serovars

Table 12-1. Example of MLST Data Available for Vibrio parahaemolyticus Isolate

Year

Country

Source

dnaE

gyrB

recA

dtdS

pntA

pyrC

tnaA

ST

428/00 30824 9808/1 906–97 357–99

1998 1999 2004 1997 1999

Spain Spain Spain Peru Peru

clinical clinical clinical clinical clinical

13 13 3 3 15

10 10 4 4 11

19 19 19 19 30

27 27 4 4 10

28 28 29 29 1

27 27 4 4 3

21 21 22 22 9

17 17 3 3 19

156 |

| Molecular Typing and Differentiation

Table 12-2. List of MLST Schemes Available as of March 2014 Organism

Database Website

Achromobacter spp. Acinetobacter baumannii#1 Acinetobacter baumannii#2 Aeromonas spp. Arcobacter spp. Aspergillus fumigatus Bacillus cereus Bacillus licheniformis Bifidobacterium Bordetella spp. Borrelia burgdorferi Brachyspira hyodysenteriae Brachyspira intermedia Brachyspira spp. Burkholderia cepacia complex Burkholderia pseudomallei Campylobacter concisus/curvus Campylobacter fetus Campylobacter helveticus Campylobacter hyointestinalis Campylobacter insulaenigrae Campylobacter jejuni Campylobacter lanienae Campylobacter lari Campylobacter sputorum Campylobacter upsaliensis Candida albicans Candida glabrata Candida krusei Candida tropicalis Chlamydiales spp. Clostridium botulinum Clostridium difficile Clostridium difficile#2 Clostridium septicum Corynebacterium diphtheriae Cronobacter spp. Cryptococcus neoformans Enterococcus faecalis Enterococcus faecium Escherichia coli#1 Escherichia coli#2 Flavobacterium psychrophilum Haemophilus influenzae Haemophilus parasuis Helicobacter cinaedi Helicobacter pylori Klebsiella pneumoniae Lactobacillus casei Lactobacillus salivarius Leptospira spp. Listeria monocytogenes

http://pubmlst.org/achromobacter http://pubmlst.org/abaumannii http://www.pasteur.fr/recherche/genopole/PF8/mlst/Abaumannii.html http://pubmlst.org/aeromonas http://pubmlst.org/arcobacter http://pubmlst.org/afumigatus http://pubmlst.org/bcereus http://pubmlst.org/blicheniformis http://www.pasteur.fr/recherche/genopole/PF8/mlst/Bifidobacterium.html http://pubmlst.org/bordetella http://borrelia.mlst.net http://pubmlst.org/brachyspira http://pubmlst.org/brachyspira http://pubmlst.org/brachyspira http://pubmlst.org/bcc http://bpseudomallei.mlst.net http://pubmlst.org/campylobacter http://pubmlst.org/cfetus http://pubmlst.org/campylobacter http://pubmlst.org/campylobacter http://pubmlst.org/campylobacter http://pubmlst.org/campylobacter http://pubmlst.org/campylobacter http://pubmlst.org/campylobacter http://pubmlst.org/campylobacter http://pubmlst.org/campylobacter http://calbicans.mlst.net http://cglabrata.mlst.net http://pubmlst.org/ckrusei http://pubmlst.org/ctropicalis http://pubmlst.org/chlamydiales http://pubmlst.org/cbotulinum http://pubmlst.org/cdifficile http://www.pasteur.fr/recherche/genopole/PF8/mlst/Cdifficile2.html http://pubmlst.org/csepticum http://pubmlst.org/cdiphtheriae http://pubmlst.org/cronobacter http://mlst.mycologylab.org/DefaultInfo.aspx?Page5Cneoformans http://efaecalis.mlst.net http://efaecium.mlst.net http://mlst.ucc.ie/mlst/dbs/Ecoli http://www.pasteur.fr/recherche/genopole/PF8/mlst/EColi.html http://pubmlst.org/fpsychrophilum http://haemophilus.mlst.net http://pubmlst.org/hparasuis http://pubmlst.org/hcinaedi http://pubmlst.org/helicobacter http://www.pasteur.fr/recherche/genopole/PF8/mlst/Kpneumoniae.html http://www.pasteur.fr/recherche/genopole/PF8/mlst/Lcasei.html http://pubmlst.org/lsalivarius http://leptospira.mlst.net http://www.pasteur.fr/recherche/genopole/PF8/mlst/Lmono.html (continued on next page)

| 157

Compendium of Methods for the Microbiological Examination of Foods |

Table 12-2. (continued ) Organism

Database Website

Mannheimia haemolytica Moraxella catarrhalis Mycobacterium abscessus Mycobacterium massiliense Mycoplasma agalactiae Neisseria spp. Pantoea agglomerans Pasteurella multocida#1 Pasteurella multocida#2 Pediococcus pentosaceus Plesiomonas shigelloides Porphyromonas gingivalis Propionibacterium acnes Propionibacterium freudenreichii Pseudomonas aeruginosa Salmonella enterica Sinorhizobium spp. Staphylococcus aureus Staphylococcus epidermidis Staphylococcus pseudintermedius Stenotrophomonas maltophilia Streptococcus agalactiae Streptococcus canis Streptococcus dysgalactiae equisimilis Streptococcus oralis Streptococcus pneumoniae Streptococcus pyogenes Streptococcus suis Streptococcus thermophilus Streptococcus uberis Streptococcus zooepidemicus Streptomyces spp. Vibrio parahaemolyticus Vibrio tapetis Vibrio vulnificus Wolbachia Xylella fastidiosa Yersinia pseudotuberculosis Yersinia ruckeri Yersinia spp.

http://pubmlst.org/mhaemolytica http://mlst.ucc.ie/mlst/dbs/Mcatarrhalis http://pubmlst.org/mabscessus http://pubmlst.org/mabscessus http://pubmlst.org/magalactiae http://pubmlst.org/neisseria http://www.pasteur.fr/recherche/genopole/PF8/mlst/Pantoea.html http://pubmlst.org/pmultocida http://pubmlst.org/pmultocida http://pubmlst.org/ppentosaceus http://www.pasteur.fr/recherche/genopole/PF8/mlst/references_Plesio.html http://pubmlst.org/pgingivalis http://pubmlst.org/pacnes http://www.pasteur.fr/recherche/genopole/PF8/mlst/Propio-freudenreichii.html http://pubmlst.org/paeruginosa http://mlst.ucc.ie/mlst/dbs/Senterica http://pubmlst.org/sinorhizobium http://saureus.mlst.net http://sepidermidis.mlst.net http://pubmlst.org/spseudintermedius http://pubmlst.org/smaltophilia http://pubmlst.org/sagalactiae http://pubmlst.org/scanis http://sdse.mlst.net http://pubmlst.org/soralis http://spneumoniae.mlst.net http://spyogenes.mlst.net http://ssuis.mlst.net http://www.pasteur.fr/recherche/genopole/PF8/mlst/strepto-thermophilus.html http://pubmlst.org/suberis http://pubmlst.org/szooepidemicus http://pubmlst.org/streptomyces http://pubmlst.org/vparahaemolyticus http://pubmlst.org/vtapetis http://pubmlst.org/vvulnificus http://pubmlst.org/wolbachia http://pubmlst.org/xfastidiosa http://mlst.ucc.ie/mlst/dbs/Ypseudotuberculosis http://pubmlst.org/yruckeri http://pubmlst.org/yersinia

within closely related species.15,79 Some researchers designate the DNA repeats found in prokaryotic organisms as short tandem repeats (STRs), although a prokaryotic repeat can be identified as a VNTR if the repeat number variation is associated with a single genetic locus.149 The technique can be performed through DNA digestion with restriction endonuclease enzymes, and sequences identified using gene probes with Southern blotting. More recently, a PCRbased method has been performed using primers complementary to the conserved sequences flanking the tandem repeats.81,151 The resultant band pattern can be compared by gel electrophoresis without the need for gene probes. PCR 158 |

amplicons can also be fluorescently labeled with different dye colors and separated by capillary electrophoresis in an automated DNA fragment sequencer.81,82 A variety of software programs are available to analyze typing information and even to infer phylogenetic relationships.72,77,143

12.911

Noteworthy Applications to Foodborne Pathogens VNTR can be extremely useful when investigating disease outbreaks believed to be caused by closely related organisms146 or when looking at international transfer of organisms causing disease. MLVA has been used to develop typing panels for a

| Molecular Typing and Differentiation

Figure 12-1. Example of MLST data analysis: (A) V. parahaemolyticus population snapshot obtained using eBURST v3; CC = clonal complex; three clonal complexes were identified; adapted from Gonzalez-Escalona et al.48 (B) Minimum evolution tree analysis using concatenated sequences of the seven loci using MEGA5.141 (C) Split tree decomposition of the concatenated sequences of the seven loci.

variety of foodborne pathogens, such as L. monocytogenes, E. coli O157:H7, Clostridium difficile, and Salmonella enterica.40,146,68,66,16 Because MLVA typing schemes developed for Salmonella typhi have demonstrated high discriminatory power in distinguishing closely related strains, they have often been used in combination with other techniques considered gold standards, such as PFGE.16 It has been demonstrated, however, that for certain bacterial genera MLVA has higher discriminatory power. For example, MLVA outperformed PFGE in a study determining the strains of E. coli O157:H7 responsible for five foodborne outbreaks in Pennsylvania and Minnesota. In addition, the technique identified sporadic E. coli O157:H7 isolates, which validated the accuracy of the methodology for outbreak applications.105

12.10 12.101

PCR-BASED GENOMIC FINGERPRINTING TECHNIQUES (REP, ERIC, BOX) Background

Genomic DNA of microorganisms contains a variety of repetitive DNA sequences interspersed throughout the genome.90

Repetitive elements in bacterial genomes were first reported in the early 1980s,18,58,135 and by the early 1990s had been recognized for their ubiquity and usefulness in molecular typing and differentiation.90,153 These elements are highly variable by location and copy number within the genome and are thus an attractive target for DNA typing, as banding patterns can be used to distinguish even closely related strains. Three general classes of conserved repetitive sequence motifs, repetitive extragenic palindromic (REP), enterobacterial repetitive intergenic consensus (ERIC) and BOX elements can be targeted by PCR primers and used to create fingerprints or banding patterns useful for strain typing. REP sequences are 35–40 bp,135,46 ERIC elements are 124–127 bp, and BOX sequences are found in three subunits of approximately 50 bp each.73 The BOX element was first discovered in the genome of Streptococcus pneumoniae and is the first example of repetitive elements to be described in Gram-positive organisms. REP and ERIC sequences were originally described in Gram-negative bacteria, but have also been successfully used in the typing of several Gram-positive species. ERIC elements were first described in 1990 and 1991 in E. coli, | 159

Compendium of Methods for the Microbiological Examination of Foods |

S. typhimurium, Y. pseudotuberculosis, Klebsiella pneumonia and Vibrio cholerae.60,132 BOX, REP, and ERIC sequence motifs are genetically stable and differ between species only in their copy number and chromosomal locations, making them a desirable target for strain differentiation using a variety of conserved primers.

12.1011

Noteworthy Applications to Foodborne Pathogens Fingerprinting approaches with various REP, ERIC, and BOX elements have been widely applied to foodborne pathogens, but it is important to note that results can be influenced by PCR conditions,70 and some have suggested that optimal comparisons of strains may only be possible for patterns produced in the same PCR run.119 An interesting study compared four methods (RAPD, ERIC, SSCP, and ribotyping) to differentiate among 57 Salmonella isolates and concluded that a combination of the ERIC and RAPD approaches provided the most discriminatory power and was able to distinguish all 57 isolates.80 In similar studies of Campylobacter isolates, Wilson et al.160 found that combined REP, ERIC, and BOX-A1R-based repetitive extragenic palindromic (BOX-PCR) fingerprints had greater discriminatory power than PFGE or MLST. Behringer et al.10 were not able to unequivocally distinguish C. coli from C. jejuni using REP-PCR or flaA-RFLP, but were able to distinguish isolates of these two species with MLST. 12.11

RIBOTYPING

12.111

Background

Ribotyping, as the name suggests, uses banding patterns of ribosomal RNA (rRNA) for strain typing. In this method, genomic DNA is digested, transferred to a membrane, and probed with a conserved rRNA probe to provide a pattern of rRNA genes. Each strain produces a unique pattern, which can be used for typing. A commercially available system by Dupont (Wilmington, DE) automates the bench procedures and types strains using a database of banding patterns.

12.1111

Noteworthy Applications to Foodborne Pathogens Ribotyping applied to outbreaks of foodborne diarrheal illness has been successfully used to distinguish individual epidemic strains of Campylobacter and E. coli.155 Ribotyping of L. monocytogenes helped to identify three distinct lineages differing in their virulence potential in humans.159 In a comparison of four phenotypic and six genotypic methods to discriminate among C. jejuni strains, Patton et al.114 found ribotyping, along with MLEE and whole-genome restriction enzyme analysis to be the most discriminatory. 12.12 12.121

PULSED-FIELD GEL ELECTROPHORESIS (PFGE) Background

The introduction of pulse-field gel electrophoresis (PFGE) as a method to separate genomic DNA fragments129,11 and its early adoption as a molecular typing methodology63,7 led to a long period when it was the preferred method of whole-genome based typing. PFGE has served as a de facto gold standard for the subtyping and source tracking of foodborne pathogens8,44,61,113,121,122 and was adopted as the pri160 |

mary genotyping methodology for the PulseNet database139 described below. Only recently has the adoption of rapid and cost-effective whole-genome sequencing begun to supplant PFGE. PFGE relies on restriction enzymes which cut genomic DNA relatively infrequently, then separates the resulting large fragments by pulsed-field gel electrophoresis.17,138 In many cases PFGE has proved more sensitive than other subtyping methods.9,97,124,138,145 Selection of a universal size standard for PFGE analysis has made comparisons of PFGE patterns easier, more reproducible, and comparable across laboratories (for example strains of Salmonella serotype Braenderup H9812),61 enabling the creation in 1996 of PulseNet,44 a national database and tracking system for major foodborne bacterial pathogens.

12.1211

Noteworthy Applications to Foodborne Pathogens PulseNet (http://www.pulsenetinternational.org/protocols/ Pages/default.aspx) has collected extensive PFGE data and established foodborne PFGE databases for most recognizable foodborne pathogens, such as C. jejuni, Clostridium botulinum, L. monocytogenes, V. cholerae, V. parahaemolyticus, Y. pestis, Salmonella, Shigella sonnei, S. flexneri, and E. coli (O157:H7 and non-O157).8,57 Although subtyping alone cannot prove or disprove a connection between two isolates in the absence of epidemiologic and environmental evidence, PulseNet has become a powerful tool for the detection, investigation, source tracking, and subsequent control of outbreaks of foodborne infections in the United States.44,89,100,113,121,122,139 12.13 12.131

OPTICAL MAPPING Background

Optical mapping is an aptly named technique in which strands of genomic DNA are digested with one or more restriction enzymes, and a physical map of the genome is constructed from the resulting image.12,14,164,165 In outline, genomic DNA is extracted from a pure culture, linearly arrayed on a specialized glass slide, cut with a restriction enzyme, and the resulting set of fragments imaged at high resolution. Gaps at the restriction sites provide reference marks to determine the size of fragments in comparison to standards. Finally, fragments from multiple strands are assembled by aligning restriction sites to build a contiguous map of the genome. In essence, an optical map is truly a linear barcode—a pattern comprised of particular fragment sizes in a particular order.75,164 Optical maps are based on whole genomes similar to other so-called genomic barcoding techniques such as PFGE, but contain much more information, as the order of fragments is preserved corresponding to the physical location of restriction sites on the genome. Although smaller fragments (,1 kb) can be missed by optical mapping, errors in the estimation of fragment sizes is typically low (,3%) and proportionally less for larger fragment sizes.

12.1311

Noteworthy Applications to Foodborne Pathogens Optical mapping has been used for strain-level discrimination of strains of the foodborne enteric pathogen E. coli

| Molecular Typing and Differentiation

O157:H7 using digestion with BamHI.75 Comparisons of optical maps to strains with fully sequenced genomes revealed multiple genomic rearrangements, many associated with prophages.75 Optical mapping was used with microarray-genotyping to rapidly characterize outbreak strains of E. coli O104:H4,64 and also used to identify a putative prophage in the genome of virulent S. typhimurium strains linked to a cluster of salmonellosis in Denmark.116

12.14 12.141

WHOLE-GENOME SEQUENCING Background

Whole-genome sequencing (WGS) is the term used to describe the sequencing of a full genome of an organism. The first bacterial genome sequenced was that of Haemophilus influenzae Rd KW20 in 1995 by Fleischmann and colleagues.39 The first decade (1995–2005) of bacterial WGS was mainly accomplished using Sanger sequencing.124 This method, although highly improved with time, was still labor-intensive, cumbersome, and expensive; the sequencing of a single bacterial genome could take years, and was generally restricted to human pathogens.98 Heavy public and private investment led to the creation of new sequencing technologies, referred to generally as second-generation sequencing technologies or next-generation sequencing (NGS).78,98,110 NGS is cheaper and faster: bacterial genomes can now be sequenced in a single day, depending on the NGS platform (Table 12-3). More detailed information about these second-generation techniques is provided elsewhere.87,88,98,110,111

Sequencing technologies are changing quickly and several new approaches based on single DNA molecule sequencing have been developed.25,111 Among these socalled third-generation sequencing (3GS) technologies are the Pacific BioSciences platform (the only sequencer today able to read through small genomes such as bacteria, plasmids, and phages), and the Oxford Nanopore DNA sequencer, among others. The new generation of sequencers is moving away from the use of PCR to amplify genomic libraries owing to errors in PCR amplification and PCR bias for certain fragments.111 Because of its low cost and high speed, WGS by NGS is becoming the method of choice for applications such as vaccine development, antibiotic resistance studies, finding new markers for pathogen detection, pathogen identification, and epidemiological pathogen tracking, among others.26,27,56,96,111 Because each sequencing platform has unique methods of library preparation and sequencing which change rapidly, a specific protocol is not provided here.

12.1411

Noteworthy Applications to Foodborne Pathogens The ability of NGS technologies to generate whole-genome information for pathogens has led to an explosion of comparative genomics and genomic epidemiology. A dramatic recent example123 involved the rapid (,1 wk) sequencing and open-source analysis of E. coli isolates from a foodborne outbreak in Germany.42 Subsequent whole-genome NGS of 17 isolates identified several Single-nucleotide

Table 12-3. Generalized Overview of Commercial Sequencing Technologies as of March 2013

Generation

Run Time (Hours)

Typical Read Lengths155

Chemistry

Platform Cost ($1,000’s)a

0.40–0.80

Sangerb

,300

emPCR

0.40–0.70

Pyro

,500

0.4–0.7

24

99.5

Template Prep

Max Output (Gb)

Accuracy (%)

Company

Platform

ABI

3700/3730

Roche

454-Flx Titanium 454 GS junior HiSeq 2000

emPCR

0.40

Pyro

,100

0.035

10

99.5

CBA

2 6 0.10 (PE)

RDT

,690

600

48–264

–

MiSeq SOLid

CBA emPCR

2 6 0.25 (PE) 2 x 0.05

RDT OPL

,125 ,525

8.5 60

4–40 144

.98.5 99

Ion Torrent Ion Proton

emPCR emPCR

0.4 0.2

Pyro (pH) Pyro (pH)

,50 ,150

1 10

1.5–4 2–4

99 99

5 (mean)

Single molecule

,700

0.25/ cell

1–2

86

1st ,100

2nd

IlluminaSolexa Life Technologies

3rd Pacific BioSciences

Note: CBA 5 clonal bridge amplification; emPCR 5 Emulsion PCR; OPL 5 oligonucleotide probe ligation; PE 5 paired-end; Pyro 5 pyrosequencing; RDT 5 reverse dye terminators. Pricing may vary between countries and/or sales territories. Platform costs do not include service contracts. Source: Behringer et al. and Chu et al.10,17; http://www.illumina.com/systems/miseq.ilmn; http://www.invitrogen.com/site/us/en/home/ brands/Ion-Torrent.html?cid5fl-iontorrent; http://www.invitrogen.com/site/us/en/home/brands/Ion-Torrent.html?cid5fl-iontorrent. a Estimated cost in thousands of dollars. b Fluorescent dideoxy-terminator.

| 161

Compendium of Methods for the Microbiological Examination of Foods |

polymorphisms (SNPs) correlated with geography and provided additional insights into the origin and epidemiology of the outbreak.49 Additional examples of the extent of the application of whole-genome sequencing include comparisons of pathogen-specific genes responsible for food outbreaks in E. coli O157:H7 isolates158 as well as horizontal transmission of genes and niche adaptations.31,32 NGS applied to WGS has also opened the possibility of determining how many genomes of an individual bacterial strain should be sequenced to determine the entire range of genes.144 Description of the extensive genetic diversity in C. jejuni and the discovery of a large number of hypervariable regions apparently important in its survival have been possible.112,29 Genomic comparisons of L. monocytogenes serotypes have identified core genes important for survival and growth in different environments,103 and separately, a large and novel genomic island encoding translocation and efflux functions.45

12.15 12.151

GENERALIZED PROTOCOLS Preparation of Genomic DNA From Bacteria for Use With PCR-Based Methods

The first step common to most molecular approaches to bacterial typing is to obtain high-quality purified genomic DNA from the organism of interest. Although protocols vary, the generic process is to lyze the cells by some combination of physical and/or chemical steps, solubilize the DNA, and remove contaminating proteins, RNA, and other macromolecules.

12.152

Preparation of Bacterial Lysates

Pure cultures of the organism are grown under optimal conditions to late log phase and harvested by centrifugation or scraping. Depending on the organism and laboratory settings, 1 mL of broth culture (usually Luria-Bertani; LB) or 1 colony from tryptic soy agar16 or LB plates may be needed per strain. Cells are resuspended in 50 mL of 1X Tris-EDTA (TE) (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). From this point on, DNA extraction will depend on Gram type. If the bacteria are Gram-positive then an initial lysozyme (50 mg/mL) treatment will be necessary in order to disrupt the cell wall. If Gram-negative, DNA extraction can proceed directly. A variety of commercially available kits, such as those available from MoBio or Qiagen, can be used to rapidly and efficiently obtain high-quality nucleic acids from microbial cells. Such kits are widely used and offer many advantages, such as speed, ease of use, and consistency. Disadvantages of commercial kits include a relatively high unit cost and the proprietary nature of reagents, which can make troubleshooting difficult. For some purposes, DNA extraction and purification steps can be greatly simplified or eliminated altogether. What follows are three protocols using basic laboratory supplies and reagents, ranging from the simplest to the most complex: (1) cell suspension used directly in a PCR, (2) a simple boiling lysis protocol, and (3) a basic protocol to easily and efficiently obtain genomic DNA from microbial cells and/or complex samples such as food matrices. 162 |

12.153

N N N N N N

General Equipment, Consumables, and Reagents

0.5–1.7 mL microcentrifuge tubes CTAB extraction buffer (2% cetyl trimethylammonium bromide, 100 mM Tris pH 8.0, 20 mM EDTA, 1.4 M sodium chloride [NaCl]) Phenol:chloroform:isoamyl alcohol (25:24:1) PEG.NaCl buffer (20% Polyethylene glycol MW8000, 2.5M NaCl) 70% ethyl alcohol (EtOH) Microcentrifuge

A. Simplest Cell Suspension PCR (or Colony PCR) 1. Lightly touch single colony with pipette tip and resuspend in 10 mL sterile distilled (sd) H2O. 2. Dilute this suspension 1:10 by transferring 1 mL to 9 mL sd H2O. 3. Add 1 mL each of 1:1 and 1:10 suspensions as template directly to PCR reactions prepared as below with 10 min hold at 95uC added to PCR thermocycling protocol. 4. Note that the most important factor for the success of cell suspension PCR is obtaining the optimal number of cells. Too many cells and/or introducing agar from the plate into the master mix will inhibit the PCR. Picking from a single colony on solid media is inherently variable; broth cultures standardized by optical density can reduce this variability. However, the time associated with measuring and normalizing optical density can defeat the purpose of cell suspension PCR as a rapid method. If broth culture is used, media should be removed before PCR. With some experience, picking single colonies as above can be regularly successful. B. Boiling Lysis Prep 1. Pick 1–5 colonies and resuspend in 100 mL of TE (10 mM Tris, 1 mM EDTA) or sdH2O in 1.7 mL Eppendorf tube. 2. Place tube in 97uC water bath for 5–10 min and/or bead beat for 5 min with 0.5 g sterilized glass beads (0.17–0.18 mm) at 2000 oscillations per min. 3. Centrifuge at 15 000 g for 5–15 min. 4. Remove supernatant for storage and use 1–5 mL for PCR. 5. Additional steps, including incubation with lysozyme, proteinase K, and/or heat shock treatments, have been shown to be helpful for Listeria,21 Lactococcus lactis, 3 4 Staphylococcus aureus, 4 7 Pseudomonas cepacia,74 Bacillus, Klebsiella, and Streptococcus, among other taxa.161 6. For any application in which a relatively pure (high-quality DNA, free of contaminants) preparation is required, the use of cell suspension PCR or crude lysis preparations is discouraged. For such cases, commercial kits may give the most reproducible results, particularly in the hands of different personnel. C. CTAB Protocol for Complex Samples 1. Resuspend bacterial pellet (up to 0.5 g wet weight) in 0.5 mL CTAB buffer and 0.5 mL phenol:chloroform:isoamyl alcohol (25:24:1).

| Molecular Typing and Differentiation

2.

Transfer suspension to tube containing 0.2 g sterile 100 mm beads. 3. Lyse cells for 30 s with bead beater or vortex. 4. Centrifuge at 16,000 g for 5 min, transfer aqueous phase to new 1.7 mL tube. 5. Add 1 volume chloroform:isoamyl alcohol, invert tube 5–10 times. 6. Transfer aqueous phase to new tube, centrifuge at 16,000 g for 5 min. 7. Add 2 volumes PEG/NaCl, incubate for ,2 h at room temperature (RT) to precipitate nucleic acids. 8. Centrifuge at maximum speed at 4uC for 10 min, immediately remove supernatant and resuspend pellet in 0.5 mL ice-cold 70% EtOH. 9. Centrifuge at maximum speed for 10 min, remove supernatant. 10. Air dry on benchtop or with vacuum centrifuge, resuspend pellet in 30–50 mL TE or 10:0.1 TE. 11. Repeat steps 7–10 for dirty samples. 12. Note that this protocol largely follows that of Griffiths et al.,52 which has been cited more than 400 times and compared favorably to a variety of other methods.3,115,134

general rule, template DNA should range from 10 pg/mL to 10 ng/mL. The following are typical concentrations of each reagent (volumes/25 mL rxn):

N N N N N N N

Always include a no-template control (NTC), a negative control, and a positive control for each run.

12.15313 1. 2. 3. 4.

12.1531 12.15311 Reagents

N N N N N N N N N N N N N N N N

Preparation, Amplification, Detection, and Analysis of PCR for Rapid Methods General Equipment, Consumables, and

Thermocycler 0.5–1.7 mL microcentrifuge tubes 0.2 mL thin-walled PCR tubes or plates sdH2O 2X Master Mix, 10X PCR buffer without MgCl2, MgCl2 (25 mM stock), or dNTPs (10 mM stock of 2.5 mM each dNTP) Polymerase Appropriate oligonucleotide primers Agarose or pre-cast gels 10X TBE or TAE buffer Gel loading buffer Molecular weight standards Gel-casting and electrophoresis tanks, power supply Ethidium bromide or alternative stain such as Sybr Green/Gold UV transilluminator Photo-documentation system; digital systems preferred to allow subsequent analysis of DNA profiles with computer software Computer and DNA analysis software (optional)

12.15312 Typical PCR Reaction Mix. Prepare reagents and reaction tubes and keep on ice until the vials are placed in the thermal cycler. Generally a master mix is prepared and aliquoted into each PCR tube or plate well. Most PCR amplifications can tolerate a wide range of reagent and template concentrations and still achieve adequate amplification. The values given below are typical recommended ranges. For the template, standardizing the amount of DNA used will improve reproducibility. As a

1X PCR buffer without MgCl2 (2.5 mL 10X buffer) 1.5–2.5 mM MgCl2 (1.5–2.5 mL of 25 mM stock) 200 mM dNTPs (0.5 mL of 10mM stock) 0.5 mM each primer (0.25–1.25 mL 10 mM stock) 0.25–1 unit DNA polymerase (0.25–1 mL 1 U/mL stock) sdH2O to 23 mL Template DNA (2 mL); 1 pg–1 ng plasmids or virus, 1 ng– 1 mg for genomic templates

5. 6. 7.

Typical Thermal Cycling Program

Initial denaturation is 94uC for 2–5 min. Denaturation is 94uC for 15–30 s. Annealing is 5uC below lowest T m of primers (typically 50–60uC) for 15–30 s. Extension is 72uC for 1 min per kb depending on processivity of polymerase. Go to Step 2 (25–40X). Final extension is 72uC for 10 min. Hold at 4uC.

Note that if using whole cells for cell-suspension PCR, an initial lysis step (which can extend initial denaturation step to 10 min) will usually be required.

12.15314 Typical Gel Electrophoresis. After the PCR run is complete, mix each amplicon with 1–2 mL loading buffer. Using parafilm for this purpose is a rapid and economical alternative to mixing in tubes. Load each sample on a horizontal 1–2% agarose gel made with 1X TBE or TAE buffer and submerged in the same buffer. The gel may also contain ethidium bromide stain or other non-toxic dye (e.g. SYBR Gold). Alternatively, poststaining can be done subsequent to the run. Load between 5 and 20 mL per well, depending on the depth and the yield of PCR product. Load a molecular weight marker (e.g., Lambda DNA/ HindIII, Promega, WI) in the first and last lanes of the gel. The marker chosen should have adequate range to size all expected fragments. Electrophorese the samples at 5 v/cm (between 80 and 150 V depending on the system used) until the tracking dye has migrated approximately 1 cm from the bottom of the gel for mini gels and 10 cm from the top for larger format gels. The time required will therefore depend on the size of the gel being used. Place the gel on a transilluminator (302 nm) and photograph. Saving images is required for subsequent analysis. A variety of software programs and analytical packages are available beyond the scope of this chapter. For many of the so-called fingerprinting methods (e.g., rep-PCR) reproducibility can be an issue, and so replication of experiments is considered best practice. | 163

Compendium of Methods for the Microbiological Examination of Foods |

12.1532 PFGE Protocol For PFGE results to be comparable across time and laboratories, protocols must be standardized. The specific protocol below has proved to provide reliable, readily comparable results122 and has been used by the Centers for Disease Control and Prevention (CDC), state health department laboratories in the United States, and partner laboratories in PulseNet International (http://www.pulse netinternational.org/publications/Pages/default.aspx). If rapid turnaround is needed, the shorter protocol of Gautom43 may be preferred for DNA preparation, but in order to obtain results comparable to those in the PulseNet database, researchers should use the running conditions described below. Our example uses the Salmonella serotype Braenderup strain H9812 (available from the Foodborne and Diarrheal Diseases Laboratory Section, CDC). Other standard PFGE protocols for the molecular subtyping of pathogens under surveillance by PulseNet laboratories can be found at the PulseNet International webpage (http://www. pulsenetinternational.org/protocols/Pages/default.aspx). The following protocol was developed for PFGE typing of E. coli O157:H7, Salmonella, and S. flexneri using the restriction enzyme XbaI. These same conditions can be used with the restriction enzymes BlnI/AvrII and SpeI when additional testing is warranted. 12.15321 Reagents

N N N N N N N N N N N N N N N N

General Equipment, Consumables, and

Trypticase soy agar with 5% defibrinated sheep blood (TSA-SB) Water baths (shaking and stationary) 12 mm x 75 mm Falcon tubes (BD Biosciences, San Jose, CA or equivalent) TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0) 1% SeaKem Gold (Lonza, Basel, Switzerland) (SKG) agarose Cell suspension buffer: 100 mM Tris, 100 mM EDTA, pH 8.0 50 mL polypropylene screw-cap tubes Cell lysis buffer (CLB) [50 mM Tris, 50 mM EDTA, pH 8.0 + 1% Sarcosyl (N-lauroylsarcosine, sodium salt)]. Proteinase K (20 mg/mL) CHEF Mapper (Bio-Rad Life Sciences Division, Hercules, CA) 10X TBE buffer Gel-casting and electrophoresis tanks, power supply Ethidium bromide or alternative stain such as Sybr Green/Gold UV transilluminator Photo-documentation system; digital systems preferred to allow subsequent analysis of DNA profiles with computer software Computer and DNA analysis software (optional)

A. Preparation of Bacterial Cultures 1. Inoculate a single colony from test cultures onto trypticase soy agar with 5% defibrinated sheep blood (TSA-SB) plates. 2. Incubate at 37uC for 14–18 h. 164 |

B.

Preparation of Gel Plugs Containing Bacterial DNA 1. Turn on shaker water bath (54–55uC) and stationary water bath (55–60uC), and spectrophotometer. 2. Label two sets of small tubes (12 mm x 75 mm Falcon tubes or equivalent) with culture numbers. 3. Prepare TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). Add 10 mL of 1M Tris, 2 mL of 0.5 M EDTA, and dilute to 1000 mL with sterile ultrapure water (Clinical Laboratory Reagent Water, CLRW). This TE buffer is used to make the plugs and to wash lyzed PFGE plugs. Prepare 1% SeaKem Gold (Lonza, Basel, Switzerland) (SKG) agarose in TE buffer for PFGE plugs. Weigh 0.5 g of SKG agarose into a 250 mL screw-cap flask and add 50 mL TE buffer, swirl gently to disperse the agarose. Loosen or remove cap and cover loosely with clear film, and microwave for 30 s; mix gently and repeat at 10-s intervals until the agarose is completely dissolved. Recap flask and return to 55–60uC water bath and equilibrate the agarose in the water bath for 15 min or until ready to use. 4. Prepare cell suspension buffer (CSB) (100 mM Tris, 100 mM EDTA, pH 8.0). Add 10 mL 1M Tris, 20 mL of 0.5 M EDTA and dilute to 100 mL with CLRW. 5. Add 2 mL of CSB to the first set of labeled tubes. Suspend growth (using a polystyrene fiber or cotton swab, previously moistened with sterile CSB) from overnight culture on TSA-SB plate in CBS. 6. Adjust OD610 of cell suspension to 1.0 (range 0.8– 1.00) by diluting with sterile CSB solution or by increasing amount of cells. Transfer 400 mL of adjusted cell suspension to labeled 1.5 mL microcentrifuge tube and discard the rest into disinfectant. 7. Label wells of PFGE plug molds with culture number. If using reusable plug molds, put a strip of transparent tape on lower part of reusable plug mold before labeling wells. 8. Add 20 mL of proteinase K (20 mg/mL stock solution) to each tube and mix gently with pipette tip. 9. Add 400 mL (0.4 mL) melted 1% SKG agarose to the 0.4-mL cell suspension; mix by gently pipetting mixture up and down a few times. Maintain temperature of melted agarose by keeping flask in beaker of warm water (55–60uC). 10. Dispense part of mixture into appropriate well(s) in plug mold. Do not allow bubbles to form. Allow plugs to solidify at least 15 min. They can be placed on ice or in the refrigerator to harden faster for 5 min. C. Lysis of Cells in Gel Plugs 1. Label 50 mL polypropylene screw-cap (or equivalent) tubes with culture numbers and date. 2. Prepare cell lysis buffer (CLB) [50 mM Tris, 50 mM EDTA, pH 8.0 + 1% Sarcosyl (N-lauroylsarcosine, sodium salt)]. Add 25 mL of 1M Tris, 50 mL of 0.5 M EDTA, 50 mL of 10% sarcosyl and dilute to 500 mL with CLRW. 3. Accurately measure 5 mL of CLB times the number of plugs (10 plugs 6 2 mL 5 total 20 mL) into the appropriate size test tube or flask.

| Molecular Typing and Differentiation

Add 25 mL of proteinase K (20 mg/mL) times the number of plugs of proteinase K final concentration of 0.1 mg/mL. Mix well. 4. Add 5 mL of cell lysis buffer with proteinase K to each labeled 50 mL polypropylene screw-cap tube. 5. Trim excess agarose from top of plug with scalpel. Open mold and transfer plugs from mold with a 5–6 mm-wide spatula to the appropriately labeled tube. Be sure plug is submerged under the buffer and not on the side of the tube. 6. Incubate plugs for 1.5–2 h in a 54–55uC water bath with constant and vigorous agitation (150–175 rpm). 7. Pre-heat enough sterile CLRW to 54–55uC so that plugs can be washed twice with 10–15 mL water. D. Washing of Gel Plugs After Cell Lysis 1. Remove tubes with plugs from water bath. Carefully pour off lysis buffer into discard; plug can be held in tube with a spatula or Pasteur pipet. 2. Add 10–15 mL sterile CRLW, mix, and shake the tubes in a 54–55uC water bath incubator for 10–15 min. 3. Pour off water from the plugs and repeat previous wash step with pre-heated water one more time. 4. Pre-heat enough sterile TE Buffer in a 54–55uC water bath to wash plugs four times with 10–15 mL TE after beginning last water wash. Pour off water from the plugs and add 10–15 mL preheated TE buffer and shake the tubes in a 54–55uC water bath incubator for 10–15 min. 5. Pour off TE buffer and repeat previous TE wash step three more times. If washing cannot be completed on the same day, store plugs in 5–10 mL TE buffer at 4uC overnight. 6. After last rinse, store plugs in 5–10 mL sterile TE buffer at 4uC until used. E. Restriction Digestion of DNA in Lysed Gel Plugs 1. Label 1.5 mL microcentrifuge tubes with culture numbers; label 3 (10-well gel) or 4 (15-well gel) tubes. 2. Dilute the appropriate 10X restriction buffer (Roche Applied Science, Indianapolis, IN) or equivalent, 1:10 with sterile CLRW. 3. Add 200 mL appropriate diluted restriction buffer to labeled 1.5 mL microcentrifuge tubes. 4. Carefully remove plug from TE buffer with narrow spatula and place in a sterile disposable Petri dish. 5. Cut a 2.0- to 2.5-mm-wide slice from test samples and the appropriate number of S. ser. Braenderup H9812 standards with a single-edge razor blade and transfer to tube containing diluted restriction buffer. Be sure plug slice is under buffer. Replace rest of plug in original tube that contains TE buffer. Store at 4uC. 6. Incubate sample and control plug slices in a 37uC water bath for 5–10 min or at room temperature. 7. After incubation, remove buffer from plug slice, being careful not to cut plug slice with pipette tip.

8.

F.

Prepare the restriction enzyme master mix by diluting 10X restriction buffer 1:10 with sterile CLRW and adding restriction enzyme (50 U/ sample) according to the table above. 9. Add 200 mL restriction enzyme master mix to each tube. Close tube and mix by tapping gently; be sure the plug is submerged in enzyme mixture. 10. Incubate sample and control plug slices at 37uC for 1.5–2 h in a water bath. 11. Approximately 1 h before restriction digest reaction is finished, pour the electrophoresis gel so it has time to harden. Preparation of Gel and Electrophoresis Unit for PFGE of Restriction Digested DNA 1. Turn on 55–60uC water bath. 2. Make 0.5X tris-borate EDTA buffer (TBE) by diluting: 105 mL 10X TBE to 2100 mL with reagent grade H2O (14-cm-wide gel) or 110 mL 10X TBE to 2200 mL with reagent grade H2O (21-cm-wide gel). 3. Make 1% SKG agarose in 0.5X TBE as follows: for 14-cm-wide gel form (10 or 15 wells): 1.0 g agarose/100 mL 0.5X TBE; for 21-cm-wide gel form (15 or more wells): 1.5 g agarose/150 mL 0.5X TBE. Note: , 4 mL melted 1% SKG agarose will be needed to fill wells after plugs are loaded. Place in 55–60uC water bath until ready to use. 4. Remove restricted plug slices from 37uC water bath. Remove enzyme/buffer mixture and add 200 ml 0.5X TBE. Incubate at room temperature for 5 min. 5. Remove plug slices from tubes; put comb on bench top and load plug slices on the bottom of the comb teeth as follows: load S. ser. Braenderup H9812 standards in lanes 1, 5, 10 (10-well gel) or lanes 1, 5, 10, 15 (15-well gel), and load samples in remaining lanes. 6. Remove excess buffer with tissue. Allow plug slices to air dry on the comb for approximately 3–5 min or seal them to the comb with 1% SKG agarose (55–60uC). 7. Carefully pour cooled melted SKG agarose into gel form. Be sure there are no bubbles. 8. Put black gel frame in electrophoresis chamber. Add 2–2.2 L freshly prepared 0.5X TBE. Close cover of unit. 9. Turn on cooling module (14uC), power supply, and pump (setting of 60–70, to achieve a flow rate of 1 L/min) approximately 30 min before gel is to be run. 10. Remove comb after gel solidifies for 30–45 min. 11. Fill in wells of gel with melted and cooled (55– 60uC) 1% SKG agarose. Unscrew and remove end gates from gel form; remove excess agarose from sides and bottom of casting platform with a tissue. Allow to harden for at least 5 min. Keep gel on the casting platform and carefully place the gel inside black gel frame in electrophoresis chamber. Close cover of chamber.

| 165

Compendium of Methods for the Microbiological Examination of Foods |

12. Select following conditions for E. coli O157:H7 and S. sonnei strains restricted with XbaI or AvrII (BlnI): a. Select following on CHEF Mapper (Bio-Rad Life Sciences Division, Hercules, CA): i. Select auto algorithm. ii. Select 30 kb (low MW). iii. Select 600 kb (high MW). iv. Select default values except where noted by pressing ‘‘enter.’’ v. Change run time to 18–19 h. vi. Initial switch time should be 2.16 s (default value). vii. Final switch time should be 54.17 s (default value). b. Set Chef DR II or III electrophoresis unit as follows: i. Initial A time is 2.2 s. ii. Final A time is 54.2 s. iii. Start ratio is 1.0. iv. Run time is 19–20 h (DR II); 18–19 h (DR III). v. Voltage is 200 V (DRII), 6V (DRIII) with included angle 120u. 13. Select following conditions for Salmonella strains restricted with XbaI or AvrII (BlnI): a. Select following on CHEF Mapper: i. Select auto algorithm. ii. Select 30 kb (low MW). iii. Select 700 kb (high MW). iv. Select default values except where noted by pressing ‘‘enter.’’ v. Change run time to 18–19 h. vi. Initial switch time should be 2.16 s (default value). vii. Final switch time should be 63.8 s (default value). b. Set Chef DR II or III electrophoresis unit as follows: i. Initial A time is 2.2 s. ii. Final A time is 54.2 s. iii. Start ratio is 1.0. iv. Run time is 19–20 h (DR II); 18–19 h (DR III). v. Voltage is 200 V (DRII), 6V (DRIII) with included angle 120u. G. Documentation of Gel 1. Stain gel by adding 40 mL of ethidium bromide stock solution (10 mg/mL) to 400 mL reagent grade water. Stain gel for 30 min in covered container. 2. Destain gel in 500 mL reagent grade water 60–90 min; change water every 20 min. Capture digital image of gel for subsequent analysis. BioNumerics (Applied Maths, Austin, TX) is commonly used for analysis. Additional instructions are provided in PNL07 of the PulseNet QA/ QC manual (http://www.pulsenetinternational. org/protocols/qualityassurance).

ACKNOWLEDGMENTS Fourth edition authors: Jeffrey M. Farber, Steven M. Gendel, Keven D. Tyler, Patrick Boerlin, Warren L. Landry, Scott J. Fritchel, and Timothy J. Barrett. 166 |

The chapter was partially supported by the Federal Drug Administration Foods Program Intramural Funds. Dr. Lili Fox Ve´lez is acknowledged for editorial assistance. Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the US Department of Agriculture, an equal opportunity provider and employer. The views expressed in this chapter are those of the authors and do not necessarily represent the views or policies of the US Department of Agriculture, the US Environmental Protection Agency, or the US Food and Drug Administration.

REFERENCES 1. Achtman, M., J. Hale, R. A. Murphy, E. F. Boyd, and S. Porwollik. 2013. Population structures in the SARA and SARB reference collections of Salmonella enterica according to MLST, MLEE and microarray hybridization. Infection, Genetics, and Evolution, 16:314-325. 2. Ahmed, R., G. Soule, W. H. Demczuk, C. Clark, R. Khakhria, S. Ratnam, S. Marshall, L.-K. Ng, D. L. Woodward, W. M. Johnson, and F. G. Rodgers. 2000. Epidemiologic Typing of Salmonella enterica Serotype Enteritidis in a Canada-Wide Outbreak of Gastroenteritis due to Contaminated Cheese. Journal of Clinical Microbiology, 38(6):2403-2406. 3. Anderson, K. L., and S. Lebepe-Mazur. 2003. Comparison of rapid methods for the extraction of bacterial DNA from colonic and caecal lumen contents of the pig. Journal of Applied Microbiology, 94(6):988-993. 4. Aurora, R., A. Prakash, and S. Prakash. 2009. Genotypic characterization of Listeria monocytogenes isolated from milk and ready-to-eat indigenous milk products. Food Control, 20(9):835-839. 5. Aymerich, T., B. Martin, M. Garriga, and M. Hugas. 2003. Microbial quality and direct PCR identification of lactic acid bacteria and nonpathogenic staphylococci from artisanal lowacid sausages. Applied and Environmental Microbiology, 69(8):4583-4594. 6. Baggesen, D. L., G. Sørensen, E. M. Nielsen, and W. HC. 2010. Phage typing of Salmonella Typhimurium - is it still a useful tool for surveillance and outbreak investigation? Eurosurveillance, 15(4):19471. 7. Bannerman, T. L., G. A. Hancock, F. C. Tenover, and J. M. Miller. 1995. Pulsed-field gel electrophoresis as a replacement for bacteriophage typing of Staphylococcus aureus. Journal of Clinical Microbiology, 33(3):551-555. 8. Barrett, T. J., P. Gerner-Smidt, and B. Swaminathan. 2006. Interpretation of pulsed-field gel electrophoresis patterns in foodborne disease investigations and surveillance. Foodborne Pathogens and Disease, 3(1):20-31. 9. Barrett, T. J., H. Lior, J. H. Green, R. Khakhria, J. G. Wells, B. P. Bell, K. D. Greene, J. Lewis, and P. M. Griffin. 1994. Laboratory investigation of a multistate food-borne outbreak of Escherichia coli O157: H7 by using pulsed-field gel electrophoresis and phage typing. Journal of Clinical Microbiology, 32(12):3013-3017. 10. Behringer, M., W. G. Miller, and O. A. Oyarzabal. 2011. Typing of Campylobacter jejuni and Campylobacter coli isolated from live broilers and retail broiler meat by flaA-RFLP, MLST, PFGE and REP-PCR. Journal of Microbiological Methods, 84(2):194-201. 11. Birren, B. W., E. Lai, S. M. Clark, L. Hood, and M. I. Simon. 1988. Optimized conditions for pulsed field gel electrophoretic separations of DNA. Nucleic Acids Research, 16(15):7563-7582.

| Molecular Typing and Differentiation

12. Cai, W., J. Jing, B. Irvin, L. Ohler, E. Rose, H. Shizuya, U. J. Kim, M. Simon, T. Anantharaman, B. Mishra, and D. C. Schwartz. 1998. High-resolution restriction maps of bacterial artificial chromosomes constructed by optical mapping. Proceedings of the National Academy of Sciences of the United States of America, 95(7):3390-3395. 13. Cardarelli-Leite, P., K. Blom, C. M. Patton, M. A. Nicholson, A. G. Steigerwalt, S. B. Hunter, D. J. Brenner, T. J. Barrett, and B. Swaminathan. 1996. Rapid identification of Campylobacter species by restriction fragment length polymorphism analysis of a PCR-amplified fragment of the gene coding for 16S rRNA. Journal of Clinical Microbiology, 34(1):62-67. 14. Cebula, T. A., E. W. Brown, S. A. Jackson, M. K. Mammel, A. Mukherjee, and J. E. LeClerc. 2005. Molecular applications for identifying microbial pathogens in the post-9/11 era. Expert Review of Molecular Diagnostics, 5(3):431-445. 15. Chiou, C., J. Liao, T. Liao, C. Li, C. Chou, H. Chang, S. Yao, and Y. Lee. 2006. Molecular epidemiology and emergence of worldwide epidemic clones of Neisseria meningitidis in Taiwan. BMC Infectious Diseases, 6(1):25. 16. Chiou, C.-S., H.-L. Wei, J.-J. Mu, Y.-S. Liao, S.-Y. Liang, C.-H. Liao, C.-S. Tsao, and S.-C. Wang. 2013. Salmonella enterica Serovar Typhi Variants in Long-Term Carriers. Journal of Clinical Microbiology, 51(2):669-672. 17. Chu, G., D. Vollrath, and R. W. Davis. 1986. Separation of large DNA molecules by contour-clamped homogeneous electric fields. Science, 234(4783):1582-1585. 18. Clement, J. M., and M. Hofnung. 1981. Gene sequence of the lambda receptor, an outer-membrane protein of Escherichia coli k12. Cell, 27(3):507-514. 19. Coppola, S., G. Blaiotta, D. Ercolini, and G. Moschetti. 2001. Molecular evaluation of microbial diversity occurring in different types of Mozzarella cheese. Journal of Applied Microbiology, 90(3):414-420. 20. Couto, I., S. Pereira, M. Miragaia, I. S. Sanches, and H. de Lencastre. 2001. Identification of clinical staphylococcal isolates from humans by internal transcribed spacer PCR. Journal of Clinical Microbiology, 39(9):3099-3103. 21. Czajka, J., and C. A. Batt. 1994. Verification of causal relationships between Listeria monocytogenes isolates implicated in food-borne outbreaks of listeriosis by randomly amplified polymorphic DNA patterns. Journal of Clinical Microbiology, 32(5):1280-1287. 22. Daley, C. L., P. M. Small, G. F. Schecter, G. K. Schoolnik, R. A. McAdam, W. R. Jacobs, and P. C. Hopewell. 1992. An outbreak of tuberculosis with accelerated progression among persons infected with the human-immunodeficiency-virus an analysis using restriction-fragment-length-polymorphisms. New England Journal of Medicine, 326(4):231-235. 23. Demczuk, W., G. Soule, C. Clark, H.-W. Ackermann, R. Easy, R. Khakhria, F. Rodgers, and R. Ahmed. 2003. Phage-Based Typing Scheme for Salmonella enterica Serovar Heidelberg, a Causative Agent of Food Poisonings in Canada. Journal of Clinical Microbiology, 41(9):4279-4284. 24. Dhanashree, B., S. K. Otta, I. Karunasagar, and I. Karunasagar. 2003. Typing of Listeria monocytogenes isolates by random amplification of polymorphic DNA. Indian Journal of Medical Research, 117:19-24. 25. Diaz-Sanchez, S., I. Hanning, S. Pendleton, and D. D’Souza. 2013. Next-generation sequencing: the future of molecular genetics in poultry production and food safety. Poultry Science, 92(2):562-572. 26. Didelot, X., et al. 2012. Microevolutionary analysis of Clostridium difficile genomes to investigate transmission. Genome Biology, 13(12):R118.

27. Didelot, X., R. Bowden, D. J. Wilson, T. E. Peto, and D. W. Crook. 2012. Transforming clinical microbiology with bacterial genome sequencing. Nature Reviews. Genetics, 13(9):601-12. 28. Dingle, K. E., F. M. Colles, R. Ure, J. A. Wagenaar, B. Duim, F. J. Bolton, A. J. Fox, D. R. Wareing, and M.C. Maiden. 2002. Molecular characterization of Campylobacter jejuni clones: a basis for epidemiologic investigation. Emerging Infectious Diseases, 8(9):949-955. 29. Dorrell, N., J. A. Mangan, K. G. Laing, J. Hinds, D. Linton, H. Al-Ghusein, B. G. Barrell, J. Parkhill, N. G. Stoker, A. V. Karlyshev, P. D. Butcher, and B. W. Wren. 2001. Whole Genome Comparison of Campylobacter jejuni Human Isolates Using a Low-Cost Microarray Reveals Extensive Genetic Diversity. Genome Research, 11(10):1706-1715. 30. Doumith, M., C. Buchrieser, P. Glaser, C. Jacquet, and P. Martin. 2004. Differentiation of the Major Listeria monocytogenes Serovars by Multiplex PCR. Journal of Clinical Microbiology, 42(8):3819-3822. 31. Eisen, J. A. 2000. Horizontal gene transfer among microbial genomes: new insights from complete genome analysis. Current Opinion in Genetics & Development, 10(6):606-611. 32. Eppinger, M., M. K. Mammel, J. E. LeClerc, J. Ravel, and T. A. Cebula. 2011. Genome Signatures of Escherichia coli O157: H7 Isolates from the Bovine Host Reservoir. Applied and Environmental Microbiology, 77(9):2916-2925. 33. Eriksson, U., and A. A. Lindberg. 1977. Adsorption of Phage P22 to Salmonella typhimurium. Journal of General Virology, 34(2):207-221. 34. Erlandson, K., and C. A. Batt. 1997. Strain-specific differentiation of lactococci in mixed starter culture populations using randomly amplified polymorphic DNA-derived probes. Applied and Environmental Microbiology, 63(7):2702-2707. 35. Fakhr, M. K., L. K. Nolan, and C. M. Logue. 2005. Multilocus sequence typing lacks the discriminatory ability of pulsedfield gel electrophoresis for typing Salmonella enterica serovar Typhimurium. Journal of Clinical Microbiology, 43(5):22152219. 36. Feil, E. J., J. E. Cooper, H. Grundmann, D. A. Robinson, M. C. Enright, T. Berendt, S. J. Peacock, J. M. Smith, M. Murphy, B. G. Spratt, C. E. Moore, and N. P. Day. 2003. How clonal is Staphylococcus aureus? Journal of Bacteriology, 185(11):3307-3316. 37. Feil, E. J., B. C. Li, D. M. Aanensen, W. P. Hanage, and B. G. Spratt. 2004. eBURST: inferring patterns of evolutionary descent among clusters of related bacterial genotypes from multilocus sequence typing data. Journal of Bacteriology, 186(5):1518-1530. 38. Fisher, M. M., and E. W. Triplett. 1999. Automated approach for ribosomal intergenic spacer analysis of microbial diversity and its application to freshwater bacterial communities. Applied and Environmental Microbiology, 65(10):4630-4636. 39. Fleischmann, R. D., M. D. Adams, O. White, R. A. Clayton, E. F. Kirkness, A. R. Kerlavage, C. J. Bult, J. F. Tomb, B. A. Dougherty, and J. M. Merrick et al. 1995. Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science, 269(5223):496-512. 40. Foley, S. L., A. M. Lynne, and R. Nayak. 2009. Molecular typing methodologies for microbial source tracking and epidemiological investigations of Gram-negative bacterial foodborne pathogens. Infection, Genetics, and Evolution, 9(4):430-440. 41. Gasanov, U., D. Hughes, and P. M. Hansbro. 2005. Methods for the isolation and identification of Listeria spp. and Listeria monocytogenes: a review. FEMS Microbiology Reviews, 29(5):851-875.

| 167

Compendium of Methods for the Microbiological Examination of Foods |

42. Gault, G., et al. 2011. Outbreak of haemolytic uraemic syndrome and bloody diarrhoea due to Escherichia coli O104: H4, southwest France. Eurosurveillance, 16(26). 43. Gautom, R. K. 1997. Rapid pulsed-field gel electrophoresis protocol for typing of Escherichia coli O157:H7 and other Gram-negative organisms in 1 day. Journal of Clinical Microbiology, 35(11):2977-2980. 44. Gerner-Smidt, P., J. Kincaid, K. Kubota, K. Hise, S. B. Hunter, M. A. Fair, D. Norton, A. Woo-Ming, T. Kurzynski, M. J. Sotir, M. Head, K. Holt, and B. Swaminathan. 2005. Molecular surveillance of shiga toxigenic Escherichia coli O157 by PulseNet USA. Journal of Food Protection, 68(9):1926-1931. 45. Gilmour, M. W., M. Graham, G. Van Domselaar, S. Tyler, H. Kent, K. M. Trout-Yakel, O. Larios, V. Allen, B. Lee, and C. Nadon. 2010. High-throughput genome sequencing of two Listeria monocytogenes clinical isolates during a large foodborne outbreak. BMC Genomics, 11:120. 46. Gilson, E., J. M. Clement, D. Brutlag, and M. Hofnung. 1984. A family of dispersed repetitive extragenic palindromic DNA sequences in E. coli. The EMBO journal, 3(6):1417-1421. 47. Goh, S. H., S. K. Byrne, J. L. Zhang, and A. W. Chow. 1992. Molecular typing of Staphylococcus aureus on the basis of coagulase gene polymorphisms. Journal of Clinical Microbiology, 30(7):1642-1645. 48. Gonzalez-Escalona, N., J. Martinez-Urtaza, J. Romero, R. T. Espejo, L. A. Jaykus, and A. DePaola. 2008. Determination of molecular phylogenetics of Vibrio parahaemolyticus strains by multilocus sequence typing. Journal of Bacteriology, 190(8):2831-40. 49. Grad, Y. H., et al. 2012. Genomic epidemiology of the Escherichia coli O104:H4 outbreaks in Europe, 2011. Proceedings of the National Academy of Sciences of the United States of America, 109(8):3065-3070. 50. Gray, M. J., R. N. Zadoks, E. D. Fortes, B. Dogan, S. Cai, Y. H. Chen, V. N. Scott, D. E. Gombas, K. J. Boor, and M. Wiedmann. 2004. Listeria monocytogenes isolates from foods and humans form distinct but overlapping populations. Applied and Environmental Microbiology, 70(10):58335841. 51. Grif, K., H. Karch, C. Schneider, F. Daschner, L. Beutin, T. Cheasty, H. Smith, B. Rowe, M. Dierich, and F. Allerberger. 1998. Comparative study of five different techniques for epidemiological typing of Escherichia coli O157. Diagnostic Microbiology and Infectious Disease, 32(3):165-176. 52. Griffiths, R. I., A. S. Whiteley, A. G. O’Donnell, and M. J. Bailey. 2000. Rapid method for coextraction of DNA and RNA from natural environments for analysis of ribosomal DNA- and rRNA-based microbial community composition. Applied and Environmental Microbiology, 66(12):54885491. 53. Gurtler, V., and H. D. Barrie. 1995. Typing of Staphylococcus aureus strains by pcr-amplification of variable-length 16S-23S rDNA spacer regions - characterization of spacer sequences. Microbiology-UK, 141:1255-1265. 54. Hagens, S., and M. J. Loessner. 2007. Application of bacteriophages for detection and control of foodborne pathogens. Applied Microbiology and Biotechnology, 76(3):513-519. 55. Harbottle, H., D. G. White, P. F. McDermott, R. D. Walker, and S. Zhao. 2006. Comparison of multilocus sequence typing, pulsed-field gel electrophoresis, and antimicrobial susceptibility typing for characterization of Salmonella enterica serotype Newport isolates. Journal of Clinical Microbiology, 44(7):2449-2457. 56. Hasan, N. A., et al. 2012. Genomic diversity of 2010 Haitian cholera outbreak strains. Proceedings of the National Academy of Sciences of the United States of America, 109(29):E2010-7.

168 |

57. Hedberg, C. W., and J. M. Besser. 2006. Commentary: cluster evaluation, PulseNet, and public health practice. Foodborne Pathogens and Disease, 3(1):32-35. 58. Higgins, C. F., G. F. Ames, W. M. Barnes, J. M. Clement, and M. Hofnung. 1982. A novel intercistronic regulatory element of prokaryotic operons. Nature, 298(5876):760-762. 59. Hopkins, K. L., M. Desai, J. A. Frost, J. Stanley, and J. M. J. Logan. 2004. Fluorescent Amplified Fragment Length Polymorphism Genotyping of Campylobacter jejuni and Campylobacter coli Strains and Its Relationship with Host Specificity, Serotyping, and Phage Typing. Journal of Clinical Microbiology, 42(1):229-235. 60. Hulton, C. S., C. F. Higgins, and P. M. Sharp. 1991. ERIC sequences: a novel family of repetitive elements in the genomes of Escherichia coli, Salmonella typhimurium and other enterobacteria. Molecular Microbiology, 5(4):825-834. 61. Hunter, S. B., P. Vauterin, M. A. Lambert-Fair, M. S. Van Duyne, K. Kubota, L. Graves, D. Wrigley, T. Barrett, and E. Ribot. 2005. Establishment of a universal size standard strain for use with the PulseNet standardized pulsed-field gel electrophoresis protocols: converting the national databases to the new size standard. Journal of Clinical Microbiology, 43(3):1045-1050. 62. Hyytia-Trees, E. K., K. Cooper, E. M. Ribot, and P. GernerSmidt. 2007. Recent developments and future prospects in subtyping of foodborne bacterial pathogens. Future Microbiology, 2(2):175-185. 63. Ichiyama, S., M. Ohta, K. Shimokata, N. Kato, and J. Takeuchi. 1991. Genomic DNA fingerprinting by pulsedfield gel electrophoresis as an epidemiological marker for study of nosocomial infections caused by methicillinresistant Staphylococcus aureus. Journal of Clinical Microbiology, 29(12):2690-2695. 64. Jackson, S. A., M. L. Kotewicz, I. R. Patel, D. W. Lacher, J. Gangiredla, and C. A. Elkins. 2012. Rapid genomic-scale analysis of Escherichia coli O 104: H4 by using high-resolution alternative methods to next-generation sequencing. Applied and Environmental Microbiology, 78(5):1601-1605. 65. Jacobson, M. J., G. Lin, T. S. Whittam, and E. A. Johnson. 2008. Phylogenetic analysis of Clostridium botulinum type A by multi-locus sequence typing. Microbiology, 154(Pt 8):2408-15. 66. Jadhav, S., M. Bhave, and E. A. Palombo. 2012. Methods used for the detection and subtyping of Listeria monocytogenes. Journal of Microbiological Methods, 88(3):327-341. 67. Janssen, P., R. Coopman, G. Huys, J. Swings, M. Bleeker, P. Vos, M. Zabeau, and K. Kersters. 1996. Evaluation of the DNA fingerprinting method AFLP as a new tool in bacterial taxonomy. Microbiology-UK, 142:1881-1893. 68. Jenke, C., B. A. Lindstedt, D. Harmsen, H. Karch, L. T. Brandal, and A. Mellmann. 2011. Comparison of multilocus variable-number tandem-repeat analysis and multilocus sequence typing for differentiation of hemolytic-uremic syndrome-associated Escherichia coli (HUSEC) collection strains. Journal of Clinical Microbiology, 49(10):3644-3646. 69. Jensen, M. A., J. A. Webster, and N. Straus. 1993. Rapid identification of bacteria on the basis of polymerase chain reaction-amplified ribosomal DNA spacer polymorphisms. Applied and Environmental Microbiology, 59(4):945-952. 70. Johnson, J. R., and C. Clabots. 2000. Improved repetitiveelement PCR fingerprinting of Salmonella enterica with the use of extremely elevated annealing temperatures. Clinical and Diagnostic Laboratory Immunology, 7(2):258-64. 71. Keen, J., D. Lester, C. Inglehearn, A. Curtis, and S. Bhattacharya. 1991. Rapid detection of single base mismatches as heteroduplexes on hydrolink gels. Trends in Genetics, 7(1):5.

| Molecular Typing and Differentiation

72. Keim, P., L. Price, A. Klevytska, K. Smith, J. Schupp, R. Okinaka, P. Jackson, and M. Hugh-Jones. 2000. Multiplelocus variable-number tandem repeat analysis reveals genetic relationships within Bacillus anthracis. Journal of Bacteriology, 182(10):2928-2936. 73. Koeuth, T., J. Versalovic, and J. R. Lupski. 1995. Differential subsequence conservation of interspersed repetitive Streptococcus pneumoniae BOX elements in diverse bacteria. Genome Research, 5(4):408-418. 74. Kostman, J. R., T. D. Edlind, J. J. LiPuma, and T. L. Stull. 1992. Molecular epidemiology of Pseudomonas cepacia determined by polymerase chain reaction ribotyping. Journal of Clinical Microbiology, 30(8):2084-2087. 75. Kotewicz, M. L., S. A. Jackson, J. E. LeClerc, and T. A. Cebula. 2007. Optical maps distinguish individual strains of Escherichia coli O157:H7. Microbiology, 153(6):1720-1733. 76. Kozlowski, P., and W. J. Krzyzosiak. 2001. Combined SSCP/ duplex analysis by capillary electrophoresis for more efficient mutation detection. Nucleic Acids Research, 29(14):E71. 77. Legendre, M., N. Pochet, T. Pak, and K. J. Verstrepen. 2007. Sequence-based estimation of minisatellite and microsatellite repeat variability. Genome Research, 17(12):1787-1796. 78. Lewis, T., N. J. Loman, L. Bingle, P. Jumaa, G. M. Weinstock, D. Mortiboy, and M. J. Pallen. 2010. High-throughput wholegenome sequencing to dissect the epidemiology of Acinetobacter baumannii isolates from a hospital outbreak. The Journal of Hospital Infection, 75(1):37-41. 79. Liao, J.-C., C.-C. Li, and C.-S. Chiou. 2006. Use of a multilocus variable-number tandem repeat analysis method for molecular subtyping and phylogenetic analysis of Neisseria meningitidis isolates. BMC Microbiology, 6(1):44. 80. Lim, H., K. H. Lee, C. H. Hong, G. J. Bahk, and W. S. Choi. 2005. Comparison of four molecular typing methods for the differentiation of Salmonella spp. International Journal of Food Microbiology, 105(3):411-418. 81. Lindstedt, B. A. 2005. Multiple-locus variable number tandem repeats analysis for genetic fingerprinting of pathogenic bacteria. Electrophoresis, 26(13):2567-2582. 82. Lista, F., et al. 2006. Genotyping of Bacillus anthracis strains based on automated capillary 25-loci Multiple Locus Variable-Number Tandem Repeats Analysis. BMC Microbiology, 6. 83. Linton, D., A. J. Lawson, R. J. Owen, and J. Stanley. 1997. PCR detection, identification to species level, and fingerprinting of Campylobacter jejuni and Campylobacter coli direct from diarrheic samples. Journal of Clinical Microbiology, 35(10):2568-2572. 84. Liu, D. 2006. Identification, subtyping and virulence determination of Listeria monocytogenes, an important foodborne pathogen. Journal of Medical Microbiology, 55(6):645-659. 85. Liu, W. T., T. L. Marsh, H. Cheng, and L. J. Forney. 1997. Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16S rRNA. Applied and Environmental Microbiology, 63(11):4516-4522. 86. Loessner, M. J. 1991. Improved procedure for bacteriophage typing of Listeria strains and evaluation of new phages. Applied and Environmental Microbiology, 57(3):882884. 87. Loman, N. J., C. Constantinidou, J. Z. Chan, M. Halachev, M. Sergeant, C. W. Penn, E. R. Robinson, and M. J. Pallen. 2012. High-throughput bacterial genome sequencing: an embarrassment of choice, a world of opportunity. Nature Reviews. Microbiology, 10(9):599-606. 88. Loman, N. J., R. V. Misra, T. J. Dallman, C. Constantinidou, S. E. Gharbia, J. Wain, and M. J. Pallen. 2012. Performance

89.

90.

91.

92.

93. 94.

95.

96.

97.

98. 99.

100.

101.

102.

103.

104.

comparison of benchtop high-throughput sequencing platforms. Nature Biotechnology, 30(5):434-439. Louie, M., P. Jayaratne, I. Luchsinger, J. Devenish, J. Yao, W. Schlech, and A. Simor. 1996. Comparison of ribotyping, arbitrarily primed PCR, and pulsed-field gel electrophoresis for molecular typing of Listeria monocytogenes. Journal of Clinical Microbiology, 34(1):15-19. Lupski, J. R. and G. M. Weinstock. 1992. Short, interspersed repetitive DNA sequences in prokaryotic genomes. Journal of Bacteriology, 174(14):4525-4529. Luquez, C., B. H. Raphael, L. A. Joseph, S. R. Meno, R. A. Fernandez, and S. E. Maslanka. 2012. Genetic diversity among Clostridium botulinum strains harboring bont/A2 and bont/A3 genes. Applied and Environmental Microbiology, 78(24):8712-8718. Maiden, M. C., J. A. Bygraves, E. Feil, G. Morelli, J. E. Russell, R. Urwin, Q. Zhang, J. Zhou, K. Zurth, D. A. Caugant, I. M. Feavers, M. Achtman, and B. G. Spratt. 1998. Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic microorganisms. Proceedings of the National Academy of Sciences of the United States of America, 95(6):3140-3145. Maiden, M. C. 2004. Multilocus sequence typing of bacteria. Annual Review of Microbiology, 2006. 60:561-588. Marilley, L., and M. G. CaseyFlavours of cheese products: metabolic pathways, analytical tools and identification of producing strains. International Journal of Food Microbiology, 90(2):139-159. Marshall, S. M., P. L. Melito, D. L. Woodward, W. M. Johnson, F. G. Rodgers, and M. R. Mulvey. 1999. Rapid identification of Campylobacter, Arcobacter, and Helicobacter isolates by PCR-restriction fragment length polymorphism analysis of the 16S rRNA gene. Journal of Clinical Microbiology, 37(12):4158-4160. Mellmann, A., et al. 2011. Prospective genomic characterization of the German enterohemorrhagic Escherichia coli O104: H4 outbreak by rapid next generation sequencing technology. PLoS One, 6(7):e22751. Meng, J., S. Zhao, T. Zhao, and M. P. Doyle. 1995. Molecular characterisation of Escherichia coli O157: H7 isolates by pulsed-field gel electrophoresis and plasmid DNA analysis. Journal of Medical Microbiology, 42(4):258-63. Metzker, M. L. 2010. Sequencing technologies - the next generation. Nature Reviews. Genetics, 11(1):31-46. Miragaia, M., J. C. Thomas, I. Couto, M. C. Enright, and H. de Lencastre. 2007. Inferring a population structure for Staphylococcus epidermidis from multilocus sequence typing data. Journal of Bacteriology, 189(6):2540-52. Multistate outbreak of listeriosis associated with Jensen Farms cantaloupe--United States, August-September 2011. 2011. MMWR Morbidity and Mortality Weekly Reports, 60(39):1357-8. Nagamine, C. M., K. Chan, and Y. F. C. Lau. 1989. A PCR artifact - generation of heteroduplexes. American Journal of Human Genetics, 45(2):337-339. Nair, S., T. K. Lin, T. Pang, and M. Altwegg. 2002. Characterization of Salmonella serovars by PCR-single-strand conformation polymorphism analysis. Journal of Clinical Microbiology, 40(7):2346-2351. Nelson, K. E., et al. 2004. Whole genome comparisons of serotype 4b and 1/2a strains of the food-borne pathogen Listeria monocytogenes reveal new insights into the core genome components of this species. Nucleic Acids Research, 32(8):2386-2395. Noller, A. C., M. C. McEllistrem, O. C. Stine, J. G. Morris, D. J. Boxrud, B. Dixon, and L. H. Harrison. 2003. Multilocus sequence typing reveals a lack of diversity among Escherichia

| 169

Compendium of Methods for the Microbiological Examination of Foods |

105.

106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

170 |

coli O157:H7 isolates that are distinct by pulsed-field gel electrophoresis. Journal of Clinical Microbiology, 41(2):675-679. Noller, A. C., M. C. McEllistrem, A. G. F. Pacheco, D. J. Boxrud, and L. H. Harrison. 2003. Multilocus variable-number tandem repeat analysis distinguishes outbreak and sporadic Escherichia coli O157:H7 isolates. Journal of Clinical Microbiology, 41(12):5389-5397. Oh, M. H., S. H. Paek, G. W. Shin, H. Y. Kim, G. Y. Jung, and S. Oh. 2009. Simultaneous identification of seven foodborne pathogens and Escherichia coli (pathogenic and nonpathogenic) using capillary electrophoresis-based single-strand conformation polymorphism coupled with multiplex PCR. Journal of Food Protection, 72(6):1262-1266. Orita, M., H. Iwahana, H. Kanazawa, K. Hayashi, and T. Sekiya. 1989. Detection of polymorphisms of human DNA by gelelectrophoresis as single-strand conformation polymorphisms. Proceedings of the National Academy of Sciences of the United States of America, 86(8):2766-2770. Orita, M., Y. Suzuki, T. Sekiya, and K. Hayashi. 1989. Rapid and sensitive detection of point mutations and DNA polymorphisms using the polymerase chain-reaction. Genomics, 5(4):874-879. Paillard, D., W. Dubois, R. Duran, F. Nathier, C. Guittet, P. Caumette, and C. Quentin. 2003. Rapid identification of Listeria species by using restriction fragment length polymorphism of PCR-amplified 23S rRNA gene fragments. Applied and Environmental Microbiology, 69(11):6386-6392. Pallen, M. J., N. J. Loman, and C. W. Penn. 2010. Highthroughput sequencing and clinical microbiology: progress, opportunities and challenges. Current Opinion in Microbiology, 13(5):625-31. Pareek, C. S., R. Smoczynski, and A. Tretyn. 2011. Sequencing technologies and genome sequencing. Journal of Applied Genetics, 52(4):413-35. Parkhill, J., et al. 2000. The genome sequence of the foodborne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature, 403(6770):665-668. Parsons, M. B., K. L. Cooper, K. A. Kubota, N. Puhr, S. Simington, P. S. Calimlim, D. Schoonmaker-Bopp, C. Bopp, B. Swaminathan, P. Gerner-Smidt, and E. M. Ribot. 2007. PulseNet USA standardized pulsed-field gel electrophoresis protocol for subtyping of Vibrio parahaemolyticus. Foodborne Pathogens and Disease, 4(3):285-292. Patton, C. M., I. K. Wachsmuth, G. M. Evins, J. A. Kiehlbauch, B. D. Plikaytis, N. Troup, L. Tompkins, and H. Lior. 1991. Evaluation of 10 methods to distinguish epidemic-associated Campylobacter strains. Journal of Clinical Microbiology, 29(4):680-688. Persoh, D., S. Theuerl, F. Buscot, and G. Rambold. 2008. Towards a universally adaptable method for quantitative extraction of high-purity nucleic acids from soil. Journal of Microbiological Methods, 75(1):19-24. Petersen, R. F., E. Litrup, J. T. Larsson, M. Torpdahl, G. Sørensen, L. Mu¨ller, and E. M. Nielsen. 2011. Molecular characterization of Salmonella Typhimurium highly successful outbreak strains. Foodborne Pathogens and Disease, 8(6):655-661. Preston, M. A., W. Johnson, R. Khakhria, and A. Borczyk. 2000. Epidemiologic Subtyping of Escherichia coli Serogroup O157 Strains Isolated in Ontario by Phage Typing and Pulsed-Field Gel Electrophoresis. Journal of Clinical Microbiology, 38(6):2366-2368. Ranjard, L., F. Poly, J. C. Lata, C. Mougel, J. Thioulouse, and S. Nazaret. 2001. Characterization of bacterial and fungal soil communities by automated ribosomal intergenic spacer analysis fingerprints: Biological and methodological variability. Applied and Environmental Microbiology, 67(10):4479-4487.

119. Rasschaert, G., K. Houf, H. Imberechts, K. Grijspeerdt, L. De Zutter, and M. Heyndrickx. 2005. Comparison of five repetitive-sequence-based PCR typing methods for molecular discrimination of Salmonella enterica isolates. Journal of Clinical Microbiology, 43(8):3615-3623. 120. Revazishvili, T., M. Kotetishvili, O. C. Stine, A. S. Kreger, J. G. Morris, and A. Sulakvelidze. 2004. Comparative analysis of multilocus sequence typing and pulsed-field gel electrophoresis for characterizing Listeria monocytogenes strains isolated from environmental and clinical sources. Journal of Clinical Microbiology, 42(1):276-285. 121. Ribot, E. M., C. Fitzgerald, K. Kubota, B. Swaminathan, and T. J. Barrett. 2001. Rapid pulsed-field gel electrophoresis protocol for subtyping of Campylobacter jejuni. Journal of Clinical Microbiology, 39(5):1889-94. 122. Ribot, E. M., M. A. Fair, R. Gautom, D. N. Cameron, S. B. Hunter, B. Swaminathan, and T. J. Barrett. 2006. Standardization of pulsed-field gel electrophoresis protocols for the subtyping of Escherichia coli O157: H7, Salmonella, and Shigella for PulseNet. Foodborne Pathogens and Disease, 3(1):59-67. 123. Rohde, H. et al. 2011. Open-source genomic analysis of Shiga-toxin-producing E. coli O104: H4. New England Journal of Medicine, 365(8):718-724. 124. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proceedings of the National Academy of Sciences of the United States of America, 74(12):5463-7. 125. Savelkoul, P. H. M., H. J. M. Aarts, J. de Haas, L. Dijkshoorn, B. Duim, M. Otsen, J. L. W. Rademaker, L. Schouls, and J.A. Lenstra. 1999. Amplified-fragment length polymorphism analysis: the state of an art. Journal of Clinical Microbiology, 37(10):3083-3091. 126. Sawadogo-Lingani, H., V. Lei, B. Diawara, D. S. Nielsen, P. L. Moller, A. S. Traore, and M. Jakobsen. 2007. The biodiversity of predominant lactic acid bacteria in dolo and pito wort for the production of sorghum beer. Journal of Applied Microbiology, 103(4):765-777. 127. Schmitz, F. J., M. Steiert, H. V. Tichy, B. Hofmann, J. Verhoef, H. P. Heinz, K. Kohrer, and M. E. Jones. 1998. Typing of methicillin-resistant Staphylococcus aureus isolates from Dusseldorf by six genotypic methods. Journal of Medical Microbiology, 47(4):341-351. 128. Scholtens, R. T. 1962. A sub-division ofSalmonella typhimurium into phage types based on the method of craigie and yen; Phages adaptable to species of the B and D group ofSalmonella; Phase adsorption as diagnostic aid. Antonie Van Leeuwenhoek, 28(1):373-381. 129. Schwartz, D. C., and C. R. Cantor. 1984. Separation of yeast chromosome-sized DNAs by pulsed field gradient gel electrophoresis. Cell, 37(1):67-75. 130. Schwartz, M. 1983. Phage l receptor (LamB protein) in Escherichia coli, Methods in Enzymology, 97:100-112. 131. Schwarzkopf, A. and H. Karch. 1994. Genetic variation in Staphylococcus aureus coagulase genes: potential and limits for use as epidemiological marker. Journal of Clinical Microbiology, 32(10):2407-2412. 132. Sharples, G. J. and R. G. Lloyd. 1990. A novel repeated DNA sequence located in the intergenic regions of bacterial chromosomes. Nucleic Acids Research, 18(22):6503-6508. 133. Smith, J. M., C. G. Dowson, and B. G. Spratt. 1991. Localized sex in bacteria. Nature, 349(6304):29-31. 134. Stach, J. E. M., S. Bathe, J. P. Clapp, and R. G. Burns. 2001. PCR-SSCP comparison of 16S rDNA sequence diversity in soil DNA obtained using different isolation and purification methods. FEMS Microbiology Ecology, 36(2-3):139-151.

| Molecular Typing and Differentiation

135. Stern, M. J., G. F. Ames, N. H. Smith, E. C. Robinson, and C. F. Higgins. 1984. Repetitive extragenic palindromic sequences: a major component of the bacterial genome. Cell, 37(3):1015-1026. 136. Stern, N. J., M. A. Myszewski, H. M. Barnhart, and D. W. Dreesen. 1997. Flagellin A gene restriction fragment length polymorphism patterns of Campylobacter spp. isolates from broiler production sources. Avian Diseases, 41(4):899-905. 137. Sun, T. P., and R. E. Webster. 1987. Nucleotide sequence of a gene cluster involved in entry of E colicins and single-stranded DNA of infecting filamentous bacteriophages into Escherichia coli. Journal of Bacteriology, 169(6):2667-2674. 138. Swaminathan, B., and G. M. Matar. 1993. Molecular typing methods. In Diagnostic Molecular Microbiology: Principles and Applications, D.H. Persing et al. (eds.) American Society for Microbiology: Herndon, VA. 26-50. 139. Swaminathan, B., T. J. Barrett, S. B. Hunter, and R. V. Tauxe. 2001. PulseNet: the molecular subtyping network for foodborne bacterial disease surveillance, United States. Emerging Infectious Diseases, 7(3):382-9. 140. Tabet, S. R., G. M. Goldbaum, T. M. Hooton, K. D. Eisenach, M. D. Cave, and C. M. Nolan. 1994. Restriction-fragmentlength-polymorphism analysis detecting a community-based tuberculosis outbreak among persons infected with HumanImmunodeficiency-Virus. Journal of Infectious Diseases, 169(1):189-192. 141. Tamura, K., D. Peterson, N. Peterson, G. Stecher, M. Nei, and S. Kumar. 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Molecular Biology and Evolution, 28(10):2731-9. 142. Tartof, S. Y., O. D. Solberg, A. R. Manges, and L. W. Riley. 2005. Analysis of a uropathogenic Escherichia coli clonal group by multilocus sequence typing. Journal of Clinical Microbiology, 43(12):5860-5864. 143. Tenover, F. C., R. R. Vaughn, L. K. McDougal, G. E. Fosheim, and J. E. McGowan. 2007. Multiple-Locus Variable-Number Tandem-Repeat Assay Analysis of Methicillin-Resistant Staphylococcus aureus Strains. Journal of Clinical Microbiology, 45(7):2215-2219. 144. Tettelin, H., D. Riley, C. Cattuto, and D. Medini. 2008. Comparative genomics: the bacterial pan-genome. Current Opinion in Microbiology, 12:472-477. 145. Threlfall, E. J., M. D. Hampton, L. R. Ward, and B. Rowe. 1996. Application of pulsed-field gel electrophoresis to an international outbreak of Salmonella agona. Emerging Infectious Diseases, 2(2):130-132. 146. Tien, Y.-Y., H. Ushijima, M. Mizuguchi, S.-Y. Liang, and C.-S. Chiou. 2012. Use of multilocus variable-number tandem repeat analysis in molecular subtyping of Salmonella enterica serovar Typhi isolates. Journal of Medical Microbiology, 61(2):223-232. 147. Torpdahl, M., M. N. Skov, D. Sandvang, and D. L. Baggesen. 2005. Genotypic characterization of Salmonella by multilocus sequence typing, pulsed-field gel electrophoresis and amplified fragment length polymorphism. Journal of Microbiological Methods, 63(2):173-184. 148. Urwin, R., and M. C. J. Maiden. 2003. Multi-locus sequence typing: a tool for global epidemiology. Trends in Microbiology, 11(10):479-487. 149. van Belkum, A., S. Scherer, L. van Alphen, and H. Verbrugh. 1998. Short-Sequence DNA Repeats in Prokaryotic Genomes. Microbiology and Molecular Biology Reviews, 62(2):275293.

150. van Belkum, A., P. T. Tassios, L. Dijkshoorn, S. Haeggman, B. Cookson, N. K. Fry, V. Fussing, J. Green, E. Feil, P. Gernersmidt, S. Brisse, and M. Struelens. 2007. Guidelines for the validation and application of typing methods for use in bacterial epidemiology. Clinical Microbiology and Infection, 13(suppl. 3):1-46. 151. van den Berg, R. J., I. Schaap, K. E. Templeton, C. H. W. Klaassen, and E. J. Kuijper. 2007. Typing and Subtyping of Clostridium difficile Isolates by Using Multiple-Locus Variable-Number Tandem-Repeat Analysis. Journal of Clinical Microbiology, 45(3):1024-1028. 152. Vaneechoutte, M., P. Boerlin, H. V. Tichy, E. Bannerman, B. Jager, and J. Bille. 1998. Comparison of PCR-based DNA fingerprinting techniques for the identification of Listeria species and their use for atypical Listeria isolates. International Journal of Systematic Bacteriology, 48:127139. 153. Versalovic, J., T. Koeuth, and J. R. Lupski. 1991. Distribution of repetitive DNA sequences in eubacteria and application to fingerprinting of bacterial genomes. Nucleic Acids Research, 19(24):6823-6831. 154. Vos, P., R. Hogers, M. Bleeker, M. Reijans, T. Vandelee, M. Hornes, A. Frijters, J. Pot, J. Peleman, M. Kuiper, and M. Zabeau. 1995. AFLP - a new technique for DNAfingerprinting. Nucleic Acids Research, 23(21):4407-4414. 155. Wachsmuth, I. K., J. A. Kiehlbauch, C. A. Bopp, D. N. Cameron, N. A. Strockbine, J. G. Wells, and P. A. Blake. 1991. The use of plasmid profiles and nucleic-acid probes in epidemiologic investigations of foodborne, diarrheal diseases. International Journal of Food Microbiology, 12(1):7790. 156. Wagner, M., and F. Allerberger. 2003. Characterization of Listeria monocytogenes recovered from 41 cases of sporadic listeriosis in Austria by serotyping and pulsed-field gel electrophoresis. FEMS Immunology and Medical Microbiology, 35(3):227-234. 157. Wang, X.-M., X.-F. Lu¨, L. Yin, H.-F. Liu, W.-J. Zhang, W. Si, S.-Y. Yu, M.-L. Shao, and S.-G. Liu. 2002. Occurrence and antimicrobial susceptibility of Listeria monocytogenes isolates from retail raw foods. Food Control, 2013. 32(1):153-158. 158. Whittam, T. S., and A. C. Bumbaugh Inferences from wholegenome sequences of bacterial pathogens. Current Opinion in Genetics & Development, 12(6):719-725. 159. Wiedmann, M., J. L. Bruce, C. Keating, A. E. Johnson, P. L. McDonough, and C. A. Batt. 1997. Ribotypes and virulence gene polymorphisms suggest three distinct Listeria monocytogenes lineages with differences in pathogenic potential. Infection and Immunity, 65(7):2707-2716. 160. Wilson, M. K., A. B. Lane, B. F. Law, W. G. Miller, L. A. Joens, M. E. Konkel, and B. A. White. 2009. Analysis of the pan genome of Campylobacter jejuni isolates recovered from poultry by pulsed-field gel electrophoresis, multilocus sequence typing (MLST), and repetitive sequence polymerase chain reaction (rep-PCR) reveals different discriminatory capabilities. Microbial Ecology, 58(4):843-855. 161. Woods, C. R., J. Versalovic, T. Koeuth, and J. R. Lupski. 1993. Whole-cell repetitive element sequence-based polymerase chain reaction allows rapid assessment of clonal relationships of bacterial isolates. Journal of Clinical Microbiology, 31(8349778):1927-1931. 162. Xia, X., S. Zhao, A. Smith, J. McEvoy, J. Meng, and A. A. Bhagwat. 2009. Characterization of Salmonella isolates from retail foods based on serotyping, pulse field gel electrophoresis, antibiotic resistance and other phenotypic properties. International Journal of Food Microbiology, 129(1):93-98.

| 171

Compendium of Methods for the Microbiological Examination of Foods |

163. Yang, B., L. Qiao, X. Zhang, Y. Cui, X. Xia, S. Cui, X. Wang, X. Meng, W. Ge, X. Shi, D. Wang, and J. Meng. 2013. Serotyping, antimicrobial susceptibility, pulse field gel electrophoresis analysis of Salmonella isolates from retail foods in Henan Province, China. Food Control, 32(1):228-235.

172 |

164. Zhou, S., et al. 2004. Single-molecule approach to bacterial genomic comparisons via optical mapping. Journal of Bacteriology, 186(22):7773-7782. 165. Zhou, S., and D. C. Schwartz. 2004. The optical mapping of microbial genomes. ASM News, 70(7):323-330.

|

SECTION II

|

Physiological Groups of Microorganisms

| 173 |

|

CHAPTER 13

|

Psychrotrophic Microorganisms Purnendu C. Vasavada and Faith J. Critzer

13.1 13.11

INTRODUCTION History and Definition of Terms

In 1887, Forster observed microbial growth at 0uC, but it was not until 1902 that the term ‘‘psychrophile’’ was used.96 Psychrophiles have been defined based on growth at low temperature, optimum growth temperature, and temperature of enumeration. Other criteria unrelated to temperature have also been used in defining psychrophilic organisms (e.g., limited to Gram-negative rods only or only bacteria that do not survive pasteurization).223 Mossel and Zwart145 and Eddy43 proposed the term ‘‘psychrotrophs’’ for microorganisms that grow at low temperatures but have higher a temperature optima. Morita143 suggests that the mesophilic microorganisms that grow at 0uC should be called ‘‘psychrotolerant’’ or ‘‘psychrotrophic’’ to contrast with psychrophilic microorganisms, which have a temperature optimum of 15uC, maximum of 20uC, and minimum of 0uC or below. In a recent review, Brenchley17 referred to microorganisms that grow at 5uC or below, regardless of the maximum growth temperature, as psychrophiles. Therefore, there is no consensus on what to call these microorganisms that grow at low temperatures. Microorganisms that grow in foods at refrigeration temperatures (0uC–7uC) but have a temperature optima above 20uC are called psychrotrophs. Psychrotrophs are defined as microorganisms that produce visible growth at 7uC ¡ 1uC within 7–10 days, regardless of their optimum growth temperatures.207 This definition honors the longstanding practice of classifying microorganisms into three temperature groups: thermophiles, mesophiles, and psychrophiles, with psychrotrophs being a subgroup of mesophiles. From a practical standpoint, the microorganisms that are most commonly associated with refrigerated foods and cause food spoilage are psychrotrophs and not psychrophiles, because psychrophiles usually die at room temperature or above. Psychrotrophs grow and spoil foods that are refrigerated, but they grow better at higher temperatures in the mesophilic range. Isolation and characterization of novel cold-resistant and cold-tolerant strains of microorganisms have been reported.110,118,153,173,209 Ming et al.141 isolated a novel cold-resistant bacterium from a peat bog sample in China. The organism was closely

related to several Paenibacillus species. However, because of low deoxyribonucleic acid (DNA)–DNA relatedness levels between the isolate and its closely related phylogenetic neighbors, the bacterium represented a new genomic species and was named Paenibacillus frigoriresistens. Kishore et al.112 report isolating another novel strain of Paenibacillus, called Paenibacillus glacialis sp. nov., from the Kafni glacier of the Himalayas, India. Denner et al.37 reported a novel psychrotrophic halotolerant bacterium—isolated from the Antarctic krill Euphausia superba Dana—that is capable of excreting a cold-adapted metalloprotease. Isolation of psychrotrophic strains of Exiguobacterium sp., such as Exiguobacterium sibiricum, Exiguobacterium undae, and Exiguobacterium antarticum, from Siberian permafrost has been reported.173 Kasana and Yadav110 reported isolating novel psychrophilic strains of Exiguobacterium sp. from the cold environments of the western Himalayas. One of nine psychrotrophic bacterial strains, Exiguobacterium SKPB5, is capable of growing at low temperatures and a high pH, which suggests that this strain could be a psychrotrophic alkali-tolerant bacterium.110,209 The isolation of a psychrotrophic Acinetobacter sp. that grew and had enzyme production at low temperatures and alkaline conditions from western Himalaya has also been reported.179,180 Psychrotrophic bacteria from various genera (e.g., Bacillus,30 Pseudomonas,4 Vibrio,114 Azospirillum,153 and Shewanella118) that produce protease have been reported.

13.12

Growth of Psychrophiles and Psychrotrophs

If a microorganism is to grow at low temperatures, then substrate uptake, cell permeability, enzymatic systems, and synthetic pathways must all function at low temperatures. Some theories concerning the mechanism of the growth of psychrophiles and psychrotrophs focus on the generation of low activation energy for enzymes, the presence of unsaturated fatty acids in the cell membranes and subsequent fluidity, conformational changes in the ribosomal proteins, regulatory enzymes, the presence of cold shock proteins, alterations in substrate uptake, and cell permeability.27,51,70,71,83–85,143,165,171,177,197 Psychrophiles and psychrotrophs grow at 10uC or below, but true psychrophiles have optimum growth rates at 15uC and cannot grow above 25uC.142 Some psychrotrophs (e.g., Clostridium botulinum

| 175 |

Compendium of Methods for the Microbiological Examination of Foods |

type E) have a temperature optima of approximately 35uC and cannot grow above 40uC. Thermoduric psychrotrophs are organisms that survive exposure to relatively high temperatures (e.g., milk pasteurization 63uC for 30 min) but do not necessarily grow at these temperatures and may grow at refrigeration temperatures (,4uC–7uC).106 The cold-tolerant strains of microorganisms can be distinguished from the mesophilic strains by their ability to grow at temperatures of 7uC and below. At low temperatures, psychrotolerant strains have slower metabolic rates and higher catalytic efficiencies compared to mesophiles.102 In contrast to psychrophilic microorganisms, these strains are characteristic of habitats where the temperature fluctuates diurnally and seasonally as they are ‘‘adaptable’’ and can grow over a wide temperature range.109 Such strains are ecologically segregated and able to coexist with mesophilic strains.109

13.13

Psychrophiles Involved in Food Spoilage

Psychrophilic bacteria are primarily Gram-negative and exist in environments where temperatures are constantly below 15uC–20uC.85 Psychrophiles grow in environments where temperatures are fairly constant, whereas psychrotrophs grow in environments where temperatures fluctuate.177 Most psychrophiles in foods are species of Aeromonas, Alcaligenes, Cytophaga, Flavobacterium, Pseudomonas, Serratia, and Vibrio. Some Gram-positive genera that have been isolated from arctic waters, soils, and foods include species of Arthrobacter, Bacillus, Clostridium, and Micrococcus. Makarios-Laham and Levin128,129 isolated psychrophilic Vibrio species from haddock, although their significance in fish spoilage is unknown. Whether psychrophiles are involved in food spoilage has not been determined. Psychrophilic yeasts, molds, and algae have also been identified. Cryptococcus, Leucosporidium, and Torulopsis are psychrophilic genera of yeasts.

13.14

Psychrotrophs Involved in Food Spoilage

The involvement of psychrotrophic bacteria in food spoilage have been well studied. These bacteria include rods and cocci, sporeformers and nonsporeformers, Gramnegative and Gram-positive bacteria, aerobes, facultative anaerobes, and anaerobes. The major psychrotrophic bacteria in milk and dairy products,23,27,106,136 meats and poultry,13,46,106,113,136,137 and fish and seafood 71,89,106,136,211 include species of Acinetobacter, Aeromonas, Alcaligenes, Arthrobacter, Bacillus, Brochothrix, Carnobacterium, Chromobacterium, Citrobacter, Clostridium, Corynebacterium, Enterobacter, Escherichia, Flavobacterium, Klebsiella, Lactobacillus, Leuconostoc, Listeria, Microbacterium, Micrococcus, Moraxella, Pseudomonas, Psychrobacter, Serratia, Shewanella, Streptococcus, and Weissella. In addition, species of Alteromonas (formerly Pseudomonas putrefaciens), Photobacterium, and Vibrio are important in fish spoilage.88,137,211 Species of Bacillus, Clostridium, Enterobacter, Erwinia, Flavobacterium, Pseudomonas, and Yersinia cause soft-rotting of refrigerated vegetables.16,18,126 Psychrotrophic fungi have been isolated from refrigerated fresh animal and marine products and from fruits and vegetables. Mold genera that contain psychrotrophic species include Alternaria, Aspergillus, Botrytis, Cladosporium, 176 |

Colletotrichum, Fusarium, Geotrichum, Monascus, Mucor, Penicillium, Rhizopus, Sporotrichum, Thamnidium, and Trichothecium.27,92,113,194 Fungi predominate in refrigerated food spoilage when low water activity, high acidity, or packaging conditions select for their growth over bacteria in foods such as fruits, jams, dried fruits, and fermented foods (e.g., cheese, sausages, yogurt). Among the yeast genera involved are Candida, Cryptococcus, Debaryomyces, Hansenula, Kluveromyces, Pichia, Saccharomyces, Rhodotorula, Torulopsis, and Trichosporon.15,27,31,92,101 The use of vacuum or modified atmosphere packaging of raw and processed meat, fish, and other foods favors the growth of facultative anaerobes and true anaerobes in the oxygen-reduced environment. The major bacterial genera found in vacuum or modified atmosphere-packaged foods include psychrotrophic species and strains of Brochothrix, Lactobacillus, Leuconostoc, and members of the Enterobacteriaceae with lower populations of Carnobacterium spp. and Weissella viridescens.13,120,172

13.15

Psychrotrophic Bacteria Associated With Milk

Psychrotrophic bacteria from numerous genera have been isolated from milk. These include the lactic acid bacteria (LAB) Lactococcus, Lactobacillus, Leuconostoc, Streptococcus and Enterococcus; the Gram-negative bacteria Achromobacter, Acinetobacter, Aeromonas, Alcaligenes, Enterobacter, Flavobacterium, Pseudomonas, and Serratia; the Gram-positive bacteria Bacillus, Clostridium, Corynebacterium, Microbacterium, Micrococcus, Streptococcus, and Staphylococcus6,146,195; and various yeasts and molds.169 Pseudomonas and Acinetobacter are among the most frequently reported psychrotrophic bacteria in raw milk.38,76,81,168,167,188,206,222 The growth and activity of psychrotrophic bacteria during refrigeration is responsible for spoilage and quality problems and the reduced shelf life of milk and milk products.6,39,81,196,206 The importance of psychrotrophic organisms in the quality and safety of milk and dairy products have been recognized and reviewed elsewhere.27,28,57,188 Recent research has revealed the diversity of psychrotrophic contamination in milk and dairy environment and the isolation and characterization of several novel strains have been reported.80,94,95,169 In a study in Israel focusing on culturable psychrotrophic bacteria in raw milk and their proteolytic and lipolytic traits, approximately 20% of the isolates were novel unidentified species. This was possibly because of the fact that only a few studies have used molecular tools to study the microbial community, especially psychrotrophic bacteria.81 New species of a psychrotrophic bacterium, Chryseobacterium, have been isolated from a dairy source, a dairy environment, raw milk, and a lactic acid beverage.80,82,94,95,190 Novel species of Chryseobacterium have also been isolated from raw chicken in a chicken processing plant32 and from South Atlantic Ocean fish.33 Psychrotrophic and psychrotolerant endospore-forming bacteria, particularly Bacillus and Paenibacillus spp., pose a significant challenge to the dairy industry. These organisms are capable of forming spores and can survive hightemperature, short-time pasteurization. Thus their presence in raw milk represents a major potential cause of milk spoilage.21,86,93,98,139,184,185,186 Molecular studies of Bacillus and Paenibacillus spp. isolated from dairy farm, a raw milk tank

| Psychrotrophic Microorganisms

and truck, and a dairy plant storage silo indicated that the endospore-forming bacterial subtypes are ubiquitous and present in the dairy farm environment and in the processing plant. The extensive diversity of Bacillus and Paenibacillus spp. represents a considerable challenge to fluid milk quality and shelf life.93 In a study of a culturedependent selection strategy and an rpoB sequence-based subtyping method applied to bacterial isolates obtained from environmental samples collected on a New York State dairy farm, Huck et al.93 reported 54 different rpoB allelic types putatively identified as Bacillus (75% of isolates), Paenibacillus (24%), and Sporosarcina spp. (1%) from among 93 isolates. Paenibacillus spp. are increasingly recognized as psychrotolerant sporeformers and as important spoilage bacteria for pasteurized, refrigerated fluid milk.170 They have been isolated from farm environments, raw milk, processing plant environments, and pasteurized fluid milk.170 The importance of aerobic spore-forming bacteria in raw milk that are capable of producing toxins and/ or spoilage enzymes have been reported and reviewed elsewhere.21,24,34,35,86,98,139,185,186,205 De Jonghe et al.34 studied the harmful effects on the quality and safety of dairy products caused by aerobic sporeforming isolates obtained from raw milk. They report that strains of Bacillus subtilis, the Bacillus cereus group, Paenibacillus polymyxa, and Bacillus amyloliquefaciens are strongly proteolytic and that Bacillus licheniformis, Bacillus pumilus, and Lysinibacillus fusiformis strains are less proteolytic. Strains of B. subtilis, B. pumilus, and B. amyloliquefaciens are lipolytic, and strains of P. polymyxa and B. cereus strains produce lecithinase, which causes a ‘‘bitty cream’’ defect in pasteurized milk.34 De Jonghe et al.34 also report that B. amyloliquefaciens and B. subtilis produce a heat-stable cytotoxic component other than the emetic toxin, whereas strains of B. amyloliquefaciens, B. subtilis, B. pumilus and the B. cereus group produce heat-labile cytotoxic substances. They also demonstrated that some strains are capable of growing at room temperature and remaining stable at refrigeration temperatures.34

13.16

Psychrotrophic Bacteria Associated With Meats and Poultry

Psychrotrophic organisms can occur in meats and poultry products and can grow well at chilled temperatures. Psychrotrophic bacteria isolated from meats include species of Citrobacter, Enterobacter, Erwinia, Klebsiella, Kluyvera, and Serratia.26,138 High counts of psychrotrophic Enterobacteriaceae on meat are indicative of poor handling and storage conditions. Pseudomonas spp. are the most common cause of the meat spoilage at refrigeration temperatures.138 Other pseudomonads commonly involved in meat spoilage include Pseudomonas fragi, Pseudomonas lundensis, Pseudomonas fluorescens, and other fluorescent pseudomonads closely related to P. fluorescens. Several other psychrotrophic bacteria such as Acinetobacter, Psychrobacter, Brochothrix thermosphacta, Serratia liquefaciens, Enterobacter agglomerans (now Pantoea agglomerans), and Hafnia alvei reportedly occur in lower numbers in meats and meat environments.138 Important spoilage psychrotrophic organisms involved in spoilage of poultry include the organisms involved in meat spoilage mentioned previously. However, Acinetobacter,

Pseudomonas, and Psychrobacter strains, and occasionally Shewanella, are most commonly involved in the spoilage of poultry. The spoilage flora of meats packaged under vacuum packaging or modified atmosphere packaging (MAP) is different in that common spoilage organisms such as Pseudomonas are inhibited by the packaging conditions. Lactic acid bacteria (e.g., Cranobacterium spp., Enterobacteriaceae, and B. thermosphacta) instead become the predominant organisms in MAP meats. Besides Cranobacterium spp., other lactic acid bacteria (LAB) such as Lactobacillus spp., Leuconostoc spp., and Pediococcus spp. often exist in MAP or vacuum-packed meats, fish, and poultry. Vacuum-packed meats may also contain high levels of Aeromonas91,97 and nonpathogenic Yersinia enterocolitica.61 The potential use of psychrotrophic LAB in controlling the spoilage of seafood products has been investigated. Matamoros et al.133,134 tested inhibitory psychrotrophic lactic acid bacteria isolated from various seafood products against Gram-positive and Gram-negative strains of typical seafood spoiling and pathogenic bacteria. The isolates inhibited spoilage organisms. Theses strains did not produce histamine or tyramine, showed no particular antibiotic resistance profile, and grew at refrigerated temperatures.134 In another study, Matamoros et al.133 inoculated cooked, peeled, and vacuum-packaged shrimp with LAB and estimated the spoilage by sensory analysis after 7 and 28 days of storage at 8uC. The results indicated that LAB strains greatly or moderately extend the shelf life of shrimp. The Lactococcus strains that showed the best results with shrimp also inhibited spoilage organisms and improved the sensory quality of cold-smoked salmon that was stored for 14 days and 28 days at 8uC.133 The LAB strains also significantly inhibited Listeria monocytogenes and Staphylococcus aureus, and reduced the numbers in the product by 2 log colonyforming units per gram (CFU/g) throughout the study for L. monocytogenes and up to 4 weeks for S. aureus.134

13.17

Psychrotrophic Pathogens

The emergence of psychrotrophic foodborne pathogens raises concerns about the safety of refrigerated foods. Pathogenic psychrotrophs that grow at or below 5uC include Aeromonas hydrophila, some strains of B. cereus, C. botulinum type E and nonproteolytic types B and F, L. monocytogenes, Vibrio cholera, Y. enterocolitica, and some strains of enteropathogenic Escherichia coli.7,123,156,158,214 Further information on these pathogenic psychrotrophs can be obtained in their respective chapters. In addition, several reviews on the role of psychrotrophic pathogens in vacuum or in modified atmospherepackaged foods have been published.25,58,88,111,158,175,213,214 Foodborne pathogens such as strains of B. cereus, Clostridium perfringens, the proteolytic strains of C. botulinum, Salmonella serotypes, and S. aureus57,156 have minimal growth at temperatures between 7uC and 15uC; therefore, temperature abuse of refrigerated foods may allow these mesophilic pathogens to resume growth once temperatures rise above 10uC–15uC. In a survey of minimally processed lettuce, nearly 60% of the samples (n 5 71) were contaminated with Y. enterocolitica.204 However, these isolates lacked many of the genetic markers of virulence such as the heat-stable enterotoxin | 177

Compendium of Methods for the Microbiological Examination of Foods |

gene, the attachment and invasion gene locus, the invasin gene locus, and the virulence plasmid. When evaluating pasteurized milk from the Netherlands, 40% (n 5 133) samples were positive for B. cereus. More than 70% of evaluated isolates were positive for enterotoxin production.60 With whole genome sequencing becoming more costeffective, the future may hold increased insights to variation among psychrotrophic pathogens and their associated virulence genes.

13.18

Significance of the Presence and Growth of Psychrotrophs in Foods

Psychrotrophs metabolize carbohydrates, proteins, and lipids across the range of temperatures at which foods are stored, but reaction rates are slower at temperatures at or below 7uC. Minor biochemical changes in the food may occur early during the growth phase of some psychrotrophs, but several days to weeks of refrigeration may be necessary for changes to become organoleptically apparent.28 Spoilage by psychrotrophs may generally result in gas production, slime formation, discoloration, and off odors that may range from pungent to buttery. Information about the spoilage of specific food commodities can be found in their respective chapters. Some general reviews are available for the spoilage of milk and dairy products,10,27,48,136,192,195 meat and poultry,113,136,137 fish and seafoods,72,89,90,136 and fruits and vegetables.16,92,126,194 There is a trend in the United States, Europe, and Japan to market ‘‘minimally processed’’ refrigerated foods that range from deli-type salads to complete dinners.18,123,187 Minimal processing uses procedures such as low heat instead of sterilization; and cleaning, peeling, and cutting of fresh produce instead of leaving it whole.1 Minimal processing includes various heat treatments, vacuum or modified atmosphere-packaging, conventional or microwave heating after product-package assembly, and strict refrigerated distribution systems.123 Two methods involve minimal heat processing in a vacuum package and refrigerated distribution at 2uC–4uC: (1) sous-vide in which the food is placed in an oxygen-impermeable bag and (2) nouvelle carte in which the food is packaged on a plate and placed in a vacuum pouch.123,187 These processes for refrigerated foods create new microbiological concerns for safety and expected shelf life. Packaging in vacuum and modified gaseous atmospheres selects for facultative anaerobes and anaerobes. The minimal heating kills vegetative cells, but not spores. Since these processing and packaging methods are intended only to extend the shelf life, the surviving psychrotrophic spoilage and pathogenic microorganisms can grow and dominate the microbiota of these products. To avoid this situation, proper thermal processing, vacuum packaging plus storage, and distribution at 7uC or below must be maintained.

13.2 13.21

REVIEW OF METHODS USED TO ENUMERATE PSYCHROTROPHS IN FOODS Cultural and Microscopic Methods

General reviews of the methods used to enumerate psychrotrophs have been published for dairy products,27 fish,90 and meats.113 The traditional methods for enumerating 178 |

psychrotrophs have involved either plate counting methods or the use of microscopy.62,63 Examples of the time and temperatures of incubation for psychrotrophic plate counts are 10 days at 7uC for pour plates, or 7–8 days at 7uC for spread plates, and 16 hr at 17uC, followed by 3 days at 7uC. Incubation conditions using shorter times and higher temperatures have included 25 hr at 21uC for milk and cream,75,154,166,193,201 45 hr at 18uC for milk,149 and 24 hr at 25uC for meat.74 Several variations of the plate count procedure provide equivalent accuracy for the enumeration of psychrotrophs. These methods include spiral plating,100,171,192 dry rehydratable film such as Petrifilm (3M Corp., St. Paul, MN),5,64,135 and hydrophobic grid-membrane filter (HGMF).20,189 (See the chapters ‘‘Mesophilic Aerobic Plate Count’’ and ‘‘Rapid Methods for the Detection and Identification of Foodborne Pathogens.’’) The composition of some selective media used for psychrotrophs is based on the assumption that most psychrotrophs are Gram-negative bacteria. In the first two editions of this compendium,62,63 a selective medium was recommended for the enumeration of psychrotrophs that contained crystal violet and triphenyl tetrazolium chloride (CVT) and was incubated for either 48 hr at 30uC or for 5 days at 22uC. Jay and Bue104 found that using CVT agar with incubation for 48 hr at 30uC was unsuitable for the enumeration of Gram-negative psychrotrophs, since many nonpsychrotrophic mesophiles grew well under these conditions. While crystal violet at 2 ppm inhibited three Gram-positive bacteria, but not three Pseudomonas species,191 crystal violet at the recommended usage of 1 ppm61,63 in CVT agar did not inhibit 41 of 44 nonpsychrotrophs or 3 of 45 psychrotrophs that were evaluated.115 Some investigators found that media containing these inhibitors are suitable for assessing Gramnegative psychrotrophs at temperatures that permit the growth of many nonpsychrotrophic mesophiles; however, this is a reflection of the relatively large number of psychrotrophs in the products examined. Most nonpsychrotrophic mesophiles will proliferate at temperatures between 22uC and 40uC when incubated for 48 hr or more. Psychrotrophs have been enumerated by the direct microscopic count (DMC) in which a specific sample size is placed on a defined area of a microscopic slide that is stained before counting the cells in a set number of fields and by the microscopic colony count (MCC) in which a specific amount of agar containing the food sample is incubated on a slide and the microcolonies are counted after a short incubation period. Zall et al.225 used a preincubation of 5 hr at 30uC before doing a DMC and noted that this method could be used because the psychrotrophic value was approximately 1% of the DMC. Juffs and Babel107 did not find a very good correlation between the MCC and the psychrotrophic count performed at 7uC for 10 days; however, they suggested that a slide incubated at 7uC for 48 hr may be useful for enumerating psychrotrophs.

13.22

Rapid and Automated Methods

In a 1997 review, Sørhaug and Stepaniak195 emphasized the need for a rapid and sensitive method to detect psychrotrophs in milk to overcome the disadvantage of the 7-day to 10-day incubation time. Because many psychrotrophic

| Psychrotrophic Microorganisms

bacteria are aerobic and possess the enzyme catalase, an increase in the concentration of this enzyme has been used to estimate the number of psychrotrophic bacteria in foods. The disk flotation method using the Catalasemeter (BioEngineering Group Ltd., Westport, CT) correlated well with the psychrotrophic plate count for raw poultry.218 Dodds et al.38 concluded that the Catalasemeter was unreliable for determining the quality of vacuum-packaged cooked turkey ham when the counts were less than 104 CFU/g. The feasibility of using a catalase-based method for rapid evaluation of raw and pasteurized milk quality has been studied.54,87,164,215 Phillips and Griffiths164 found no correlation between the catalase activity and the total count of milk; however, after a preincubation at 21uC for 25 hr in plate count agar with penicillin, crystal violet, and nisin, the detection limit was 105–106 CFU/mL. From these results, the catalase test would have little value for foods that have low psychrotrophic numbers or that have undergone conditions that select for a psychrotrophic spoilage microbiota that is catalase–negative (e.g., lactobacilli). More information on the catalase and cytochrome c oxidase tests is in section 13.82. Cytochrome c oxidase has been used to estimate the number of psychrotrophs in milk and thereby predicts its keeping quality.115–117 Kroll115 found that more than 104 microbes/mL of milk were needed to detect cytochrome c oxidase. For pasteurized milk, preincubation for 18 hr at 20uC in the presence of benzalkonium chloride to inhibit Gram-positive bacteria was needed for the population to reach 104 microbes/mL.117 The standard plate count was a better predictor of keeping quality than the cytochrome c oxidase test; however, this method may be useful as an initial screening for the presence of more than 104 psychrotrophs per milliliter or per gram of food.116 The reduction of tetrazolium salts in the presence of the Grampositive inhibitor benzalkonium chloride has been suggested as a rapid test for psychrotrophs; however, they need to grow to approximately 107 cells/mL.192 The impedance method41,50 has been used to estimate the number of bacteria in fresh fish,150,210 raw meat,52 raw milk,53,200 and pasteurized milk and cottage cheese.8,9,108 The rapid estimation of psychrotrophs in cod fillets, using brain heart infusion broth at 20uC and impedance measurements for 5–16 hr, correlated well with the standard psychrotrophic plate count.210 The estimation for raw milk showed good correlation with the plate count for psychrotrophs when the samples were analyzed after 16–21 hr at 20uC.53,200 Shelf life testing of milk and cottage cheese requires preincubation for 18 hr at 18uC–21uC before impedance is measured at 21uC.11,108 The impedance method requires only 1–2 days (compared to 7–9 days needed by the Moseley Keeping Quality Test221), it predicts the length of shelf life better, and it requires less labor. When Bishop and White9,10 compared plate counts to rapid methods for estimating the microbial shelf life of pasteurized milk and cottage cheese, impedance and endotoxin detection were both significantly correlated with the shelf life. However, the impedance method produced a better predictive equation than endotoxin determination. Impedance has also been used to detect the growth of yeasts and molds in laboratory media.183 Studies focusing on the so-called impedance-splitting methods

involve separate measurements of impedance change in the electrode and medium components for the rapid detection of bacteria in foods.121,155,178,224 Gram-negative bacteria produce lipopolysaccharides (LPS) as part of their cell envelopes. A lysate produced from amoebocytes of the horseshoe crab (Limulus polyphemus) reacts with LPS, and the reaction can be measured by methods based on gelation, turbidity, or chromogenesis.103 Test results can be obtained in 1 hr by tube gelation or in approximately 30 min by the other methods. The Limulus amoebocyte lysate (LAL) test can detect Gram-negative bacteria in foods. The LAL test has been used to estimate the number of Gram-negative psychrotrophs in refrigerated foods such as meats, milk, fish, and salads.8,10,38,49,79,103,136,140,199,202 Dodds et al.38 reported that LAL values correlated with the number of Enterobacteriaceae in vacuum-packaged cooked turkey with a sensitivity less than 100 cells/g. A sensitivity of 15 bacteria per test was reported for the analysis of milk.199 The LAL values correlated well with the bacterial count for determining the shelf life of nonacidified vegetable salads stored at refrigeration and abuse temperatures.130 In a study of lean fish, Sullivan et al.202 found that LAL values agreed with aerobic plate counts and total volatile bases. Using a microtiter plate method for LAL, Fallowfield and Patterson48 were able to detect 102–103 Pseudomonas species/g in beef and pork stored at 4uC. If LAL is to be used to estimate psychrotrophs in refrigerated foods, then correlation factors for the accept-reject levels need to be established.9,55,79,199 Further information on the use of LAL is in section 13.81. Two methods that have been developed for the detection of psychrotrophs in foods are enzyme-linked immunosorbent assay (ELISA) and the polymerase chain reaction (PCR). For the detection of P. fluorescens in refrigerated meat47,66–69 and milk,66,68 ELISA tests have been developed. The polyclonal antibody only reacts with Pseudomonas species or strains47,66–69 and has low to no recognition of other psychrotrophs.68 A monoclonal antibody recognizes three Pseudomonas species (P. fluorescens, P. fragi, and Pseudomonas putida) and Enterobacter aerogenes.78 A monoclonal antibody developed to detect P. fluorescens has a sensitivity of 105 CFU/mL in milk78 or 104 CFU/cm2 in meat77; however, polyclonal antibodies detect 105 CFU/mL or 105 CFU/cm2 in milk or meat.47,66–69 A PCR method based on 23S ribosomal DNA (rDNA) sequence from P. aeruginosa was developed to detect species of Acinetobacter, Brochothrix, Enterobacter, Flavobacterium, Moraxella, and Pseudomonas in meat.216 Enumeration of psychrotrophic bacteria by PCR has been conducted for organisms most commonly associated with spoilage of milk and meat products. In these food matrices with many PCR inhibitors, limits of detection are log 5–log 7 CFU per gram or per milliliter without enrichment. Therefore, the design of PCR assays targeting these organisms must be improved if they are to be used for direct detection. Other rapid methods that have been studied may not always distinguish psychrotrophs from nonpsychrotrophic mesophiles. These methods include HGMF,20,189 direct epifluorescent filter technique,42,117,147,161,162,163,174,176 estimation of adenosine triphosphate by bioluminescence,2,14,15,125,148,157,196,198,217,220 and calorimetry.73 There are methods that are based on the detection of amines181 and aminopeptidase activity.131,160 | 179

Compendium of Methods for the Microbiological Examination of Foods |

The routine microbiological testing of milk does not involve the identification and characterization of psychrotrophic bacteria in milk, although molecular and phenotypic methods for the identification and characterization of psychrotrophic bacteria from refrigerated milk have been reported.39,65,22,146,219,222 The ecology and transmission of Bacillus spp. and related sporeformers that are important in the quality and safety of milk have been studied by conventional and molecular methods.24,40,170,203,206 Molecular methods for characterizing bacterial strains that are applied for bacterial strain typing and identifying bacteria at the strain level are important for epidemiological surveillance and in studying bacterial population dynamics of complex flora of foods such as milk35,57,98,119,123,151,152 and meat and meat products.29,44,45,105,122 Li et al.124 reviewed current bacterial strain typing methods and classified them into three main categories: DNA banding pattern-based methods, DNA sequencing-based methods, and DNA hybridizationbased methods. Quigley et al.168 reviewed genomic-based methods involved in describing microorganisms in milk and cheese (Table 13-1). Culture-based methods requiring cultivation of microorganisms and PCR are common approaches used in the food industry to detect microorganisms, although many microorganisms cannot be cultured with existing media and methods. Therefore, there is an increased interest in so-called culture independent methods and studying

communities of microorganisms present in a food or ecological environment.36,59,99 Molecular methods have been applied to studying psychrotrophic and mesophilic flora of meats and meat products.

13.3 13.31

GENERAL RECOMMENDATIONS Method Selection

The choice of a method for psychrotroph enumeration will depend on the intended use of the results, time and equipment available, accuracy needed, type of refrigerated food, and degree of processing. When an accurate number of psychrotrophs is needed, plate count methods must be used. It may be necessary to choose time and temperature conditions that simulate the storage conditions of the food or the possible abuse conditions. Selection of the method for enumerating psychrotrophs must involve the consideration of sublethally injured or stressed cells. Absolute conditions cannot be provided for every food or every situation.

13.32

Media Selection

The selection of media for enumeration will depend on the food, type of psychrotrophs expected, recommendations of equipment manufacturers, length of incubation, reactions expected, and other relevant factors. Media other than those listed below may prove valuable.

Table 13-1. Description of the Main Genomic-based Methods Involved in Describing Microorganisms in Milk and Cheese168 Method

Principle

Culture dependent genotyping methods Random Amplified Polymorphic DNA (RAPD) Restriction Fragment Length Polymorphisms (RFLP)

Uses short arbitrary primers and low-stringency hybridization to randomly amplify DNA fragments that are separated to give a fingerprint pattern.

Culture independent molecular methods Denaturing Gradient Gel Electrophoresis (DGGE) or Temporal Temperature Gradient Gel Electrophoresis (TTGE) Single Stranded Conformation Polymorphisms (SSCP) Real-time Polymerase Chain Reaction (PCR; also called quantitative PCR [qPCR]) Intergenic Transcribed Spacer Analysis (ITS) (i.e., ribotyping) Automated Ribosomal Intergenic Spacer Analysis (ARISA) Fluorescence In Situ Hybridization (FISH) Denaturing High Performance Liquid Chromatography (DHPLC) Length Heterogeneity PCR (LH PCR)

180 |

A profiling tool based on digestion of amplified DNA using one or more restriction enzymes. When ribosomal DNA is used as a template, this method is called amplified ribosomal DNA restriction analysis (ARDRA). The separation of small PCR amplicons, which are distinguished by differences in their DNA sequences. Amplicons are separated from a low to high gradient in the direction of the electrophoresis. A chemical gradient (urea or formamide) is used in DGGE. A temperature gradient and a constant concentration of denaturants are used in TTGE. Allows separation of different DNA fragments of similar length on the basis of conformational differences in folded single stranded products. It is visualized on gels or as peaks using an automated sequencer. Uses a fluorescent probe to monitor amplification of the target DNA in real time and enables quantification of a target species. Uses species-specific primers to target a gene or organism. Analyzes the bacterial ITS region located between the 16S and 23S ribosomal genes, thereby allowing differentiation between strains of the same species or closely related species. A method similar to ITS, but it uses a fluorescent primer in the amplification of microbial ribosomal intergenic spacers. It generates peaks that correspond to discrete DNA fragments detected by a fluorescence detection system. Bacterial cells hybridize to a fluorescently labeled DNA probe and can be detected and counted by fluorescence microscopy techniques. Separates PCR amplicons by an ion-pair revered-phase high-performance liquid chromatography automated detection system. Employs a fluorescently labeled oligonucleotide as the forward primer, coupled with an unlabeled reverse pair to amplify hyper-variable regions. Labeled fragments are separated and detected by fluorescence with an automated sequencer.

| Psychrotrophic Microorganisms

13.4 13.41

SAMPLE PREPARATION Sample Collection

Samples must be collected aseptically and analyzed promptly (see the chapter ‘‘Sampling Plans, Sample Collection, Shipment, and Preparation for Analysis’’). Refrigerated storage could allow psychrotrophs to increase in number because some psychrotrophs have generation times as low as 6 hr. Refrigerated foods should ideally be analyzed within 6 hr of their collection. Refrigerated samples should not be frozen because some psychrotrophs are sensitive to freezing and can be injured or killed. If samples must be frozen for shipment, then the possibility of some death must be considered when evaluating the results.

13.42

Sample Homogenization

Samples should be homogenized with diluent in a blender for 2 min or in a stomacher. Because psychrotrophs are sensitive to heat, blending for more than 2 min is discouraged to prevent the generation of heat that can result in cell injury or death. In addition, excessive blending of molds can cause fragmentation of the mycelia, depending on blade sharpness, volume, speed, and time. Using a stomacher in preference to the blender lessens the likelihood of these problems.

13.5

EQUIPMENT, MEDIA, MATERIALS, AND REAGENTS

Refer to the specific section in the recommended methods for equipment that is needed for each method. Incubators that can be maintained at 7uC ¡ 1uC for the traditional psychrotrophic count and 17uC ¡ 1uC to 21uC ¡ 1uC for rapid methods that require incubation are necessary.

13.51

Media

1. Nonselective plate media: standard methods (i.e., plate count) agar or trypticase soy agar for bacteria; dichloran rose bengal chloramphenicol agar or plate count agar plus chloramphenicol or dichloran 18% glycerol for yeasts and molds (see the chapter ‘‘Yeasts and Molds’’). 2. Media and reagents for the rapid methods can be obtained from the manufacturers (see the chapter ‘‘Microbiological Media, Reagents, and Stains’’).

13.6 13.61

psychrotrophs per milliliter, per gram, or per square centimeter, depending on the method of sampling.

PROCEDURES FOR ENUMERATION OF PSYCHROTROPHS Plate Count Method

A plate count method using plate count agar or trypticase soy agar or using a dry rehydratable film such as Petrifilm is recommended for the general enumeration of bacteria (see the chapter ‘‘Mesophilic Aerobic Plate Count’’). Enumeration procedures for yeasts and molds can be found in the chapter ‘‘Yeasts and Molds.’’ Incubate plates for 10 days at 7uC ¡ 1uC because this is the reference definition for psychrotrophs. As an alternative, incubation for 16 hr at 17uC, followed by incubation for 3 days at 7uC, can be used when results are needed in less than 10 days.207 Count the colonies as described in the chapter ‘‘Mesophilic Aerobic Plate Count.’’ Record all counts as the number of

13.7 13.71

PRECAUTIONS Incubation Temperatures

Different types of refrigerated foods are normally processed and held at temperatures of refrigeration that are specific for the food commodity. The incubation temperature used for the enumeration may not lead to adequate assessment of the psychrotrophic population that will grow in the food because microorganisms may grow in laboratory media but not in the food or vice versa. Therefore, caution must be used when interpreting the results of the enumeration of psychrotrophic populations. Sublethally injured psychrotrophs may not be detected analytically but may cause food spoilage or foodborne illness. Therefore, steps to recover sublethally injured cells should be included in enumeration and detection methods.144 See the chapter ‘‘Cultural Methods for the Enrichment and Isolation of Microorganisms’’ for suggestions on methods.

13.72

Pour Plate Versus Spread Plate Techniques

Psychrotrophs are especially susceptible to injury or death when agar that is maintained above 45uC is used for pour plating.212 Hence, a spread/surface plate or spiral plate technique should be used whenever possible. Plates can be prepoured and stored at 5uC for several days or weeks before use. Dry rehydratable films such as Petrifilm may also be used in place of the traditional plating technique.

13.8

RAPID DETECTION

Obtaining results sooner than 10 days is pragmatically desirable, and efforts must continue to find methods that are more rapid for the enumeration of psychrotrophs in foods. Rapid detection of psychrotrophs in raw ingredients, on-line quality control samples, and shelf life samples are important. Three useful rapid methods are impedance, LAL assays, and enzymatic assays.

13.81

Limulus Amoebocyte Lysate Assay

The tube gelation method has been used more extensively than either the turbidity or the chromogenic substrate methods. Its use is described later. However, the chromogenic substrate is the newest of the three basic methods. It provides results in approximately 30 min, it is more sensitive (LPS is detected between 1 pg/mL and 5 pg/mL), and it is becoming more widely accepted. The basic operation and automation of the chromogenic substrate has been described by Tsuji et al.208 and reviewed by Jay.103 The most important considerations in the use of the tube gelation method are (1) the source and sensitivity of the LAL reagent, (2) determining whether to use single-test or multitest vials or reagent, and (3) the choice of endotoxin or LPS standard. Suppliers of freeze-dried LAL reagents such as Associates of Cape Cod (East Falmouth, MA) and Sigma Chemical Co. (St. Louis, MO) provide complete instructions for the proper use of their reagents. These instructions should be followed carefully. The LAL reagents can be obtained with different levels of sensitivity, usually expressed in endotoxin units. | 181

Compendium of Methods for the Microbiological Examination of Foods |

Because LPS from different Gram-negative bacteria varies in its reactivity to the LAL reagent, it is essential that a standard reference endotoxin preparation be used. These are available with complete instructions for use from LAL reagent suppliers. The two reference endotoxins of choice are prepared from E. coli O113:Hl0 (EC-2) or E. coli O55:B5. The tube gelation method described later is from the review by Jay.102 The LAL assay methods require that all utensils and glassware be pyrogen free. Glassware can be depyrogenated by heating in a dry-air oven at temperatures above 180uC for approximately 3 hr. Sterile pipettes and disposable tubes are generally free of pyrogens before use, and pyrogen-free water should be purchased from a vendor that supplies parenteral products. Specific instructions for conducting a tube gelation test are usually provided by the LAL reagent manufacturer. The LAL reagent is supplied in ready-to-use vials, single-test vials, or in multiuse vials. Follow the preparation and storage directions that come with the reagents. It is generally good practice to cover the tubes with aluminum foil until used. The tubes should be used the same day, although some manufacturers indicate that the tubes may be frozen and thawed once. Quality control procedures and negative and positive controls are supplied by the manufacturers. Food samples should be serially diluted, using pyrogenfree water or a suitable buffer, to provide dilutions that will produce negative results. Because two-fold dilutions were not significantly different from ten-fold dilutions, using two-fold dilutions will save reagents and labor.103 Beginning with the highest dilution (i.e., lowest endotoxin concentration), the same pipette tip can be used to transfer 0.1 mL or 0.2 mL of diluted sample to separate LAL reagent tubes. The tubes are vortexed gently, incubated in a water bath at 37uC for 1 hr, and read by inverting 180u and noting gelation. An endotoxin standard should simultaneously be included using an appropriate reference endotoxin. The two-fold diluted endotoxin standard should be treated in the same way as the diluted test preparation; the sensitivity of the LAL preparation to LPS is determined by using this standard to define the lowest quantity that produces a gel. The quantity of endotoxin or LPS in test samples is determined by multiplying the reciprocal of the highest sample dilution by the LAL-determined sensitivity value. For example, if the highest dilution of endotoxin standard that produces a firm gel in the LAL reagent is 0.1 ng, the sensitivity of the LAL reagent is 0.1 ng. If using the aforementioned LAL reagent and the highest dilution of food that produces a firm gel is 103, then the total endotoxin or LPS in food is 0.1 ng 6 1000 5 100 ng/mL.

13.82

Enzymatic Methods: Catalase and Cytochrome Oxidase

Enzymatic methods are not sensitive enough to detect microbial populations below 104 cells/mL or 104 cells/g; therefore, their use is restricted to foods with high microbial loads. A preincubation period of 4–6 hr that may or may not involve selective enrichment media can improve the selectivity and the sensitivity of these methods. The use 182 |

of these methods for solid foods needs further refinement for the extraction of the enzymes from foods. The catalase test can be performed by using instruments that measure oxygen release (e.g., BioTech International, Needville, TX). Catalase that is naturally present in food may need to be inactivated by heating the sample at 50uC for 10 min before doing the test for microbial catalase, which has a higher heat stability.12 The amount of oxygen produced in the catalase test is proportional to the number of microorganisms in the food. The oxidase test is performed by treating 4 mL of sample with 1 mL of a freshly prepared 1% N,N,N9,N9tetramethyl-p-phenylene-diamine dihydrochloride, followed by incubation at 25uC for 5 min. If a sample contains particulate matter, then it can be centrifuged at 7,000 6 g for 10 min before obtaining the reading. The blue color can be evaluated visually against reference color standards or spectrophotometrically. The intensity of the blue color is proportional to the concentration of microorganisms in the food.

13.83

Molecular Methods for Detection and Characterization

As previously discussed, PCR methods have been investigated for the detection of psychrotrophs from pure culture and foods. Table 13-2 shows some reported gene targets, primers, and expected product size for psychrotrophs reported in the literature. As with any PCR method, the concentration of bacterial cells, nucleic acid template extraction, and removal of inhibitors are paramount for improving the detection limit amongst various food matrices. A thorough review of PCR methodology and considerations, which must be considered when designing an assay of this nature, is described in the chapter ‘‘Rapid Methods for the Detection and Identification of Foodborne Pathogens.’’ Many primer sets in Table 13-2 were used primarily for identifying the organism of interest from an isolate. Therefore, they have used traditional PCR as the detection method and have relatively large PCR product sizes (i.e., . 500 bp).

13.9

INTERPRETATION

The enumeration of psychrotrophs in refrigerated foods indicates the potential spoilage, keeping quality, or safety of the food. However, caution should be exercised when trying to make absolute predictions about a food based on these results. Some temperatures of refrigeration may be close to the minimal growth temperature, and the enumeration temperature may be closer to the optimum temperature, particularly in rapid methods that require an incubation period. The temperatures used for food storage and for detection should be closely comparable to achieve meaningful results. In addition, the nature of the food is important. If the food has been refrigerated for some time, the numbers can represent a normal increase in psychrotrophs rather than a poor quality product. Processing can kill or injure psychrotrophs, and analyzing foods immediately after processing may not allow time for injured cells to recover. If processed foods are stored in the refrigerator for extended periods, even a few cells can grow to numbers that are large enough to cause eventual spoilage in a few days or weeks.

| Psychrotrophic Microorganisms

Table 13-2. Gene Targets, Primers, and Product Sizes for Polymerase Chain Reaction Detection of Psychrotrophs Target Organism

Target Gene

Acinetobacter sp.

ser

Aeromonas hydrophila

serA

Bacillius cereus, Bacillius mycoides, and Bacillius thuringiensis

cspA

Brochothrix thermosphacta

16S rDNA

Clostridium estertheticum

16S rDNA

Clostridium gasigenes

16S rDNA

Pseudomonas fluorescens

16S rDNA

Pseudomonas fluorescens

aprX

Pseudomonas fluorescens

aprX

Serratia marcescens

apr

Primer Set

Oligonucleotide Sequence

SerA-F SerA-R SerAh-F SerAh-R BcAPF1

59-GCGGGGTTGCCATTGAAGTA-39 59-TGTGTATGCCGCTTCAAATGT-39 59-TTC CTC CTA CTC CAG CGT CG-39 59-TGA TGA TCC AGG CTC ACG GT-39 59-GAG GAA ATA ATT ATGACA GTT-39

BcAPR1 Bcr3f Bcr3r 16SEF 16SER 16SDBF 16SDBR 16SPSEfluF 16SPSER FP aprI RP aprII SM2F SM3R MetS-F MetS-R

59-CTT (C/T)TT GGC CTT CTT CTA A-59 59-CTC CTC TTC TGT CCT CAA G-39 59-GTT GTC CGG AAT TAT TGG G-39 59-TCG GAA TTT CAC TTT GAG-39 59-AAG GAC TTC ACT CAT CTC TG-39. 59-GAG AGG AGT TCT TCG GAA CGA-39 59-AAG CSA CTT CCC CAA TTA C-39 59-TGCATTCAAAACTGACTG-39 59-AATCACACCGTGGTAACCG-39 59-TAYGGBTTCAAYTCCAAYAC-39 59-VGCGATSGAMACRTTRCC-39 59-AAATCGATAGCTTCAGCCAT-39 59-TTGAGGTTGATCTTCTGGTT-39 59-CGG CGA GAT CTT CAA CCG TT-39 59-GGC GAA GGT GGT CAG AAG TC-39

ACKNOWLEDGMENT Fourth edition authors: Maribeth A. Cousin, James M. Jay, and Purnendu C. Vasavada.

REFERENCES 1. Ahvenainen, R. 1996. New approaches in improving shelf life of minimally processed fruit and vegetables. Trends Food Sci. Technol. 7:179-187. 2. Anderson, R., and F. Labell. 1988. Rapid microbial tests safeguard fresh deli foods. Food Proc. 49(12):90, 92. 3. Bach, H. J., A. Hartmann, M. Schloter, and J. C. Munch. 2001. PCR primers and functional probes for amplification and detection of bacterial genes for extracellular peptidases in single strains and in soil. J. Microbiol. Methods. 44:173-182. 4. Baghel, V.S., R. D. Tripathi, R. W. Ramteke, K. Gopal, S. Dwivedi, R. K. Jain, U. N. Rai, and S. N. Singh. 2005. Psychrotrophic proteolytic bacteria from cold environments of Gangotri glacier, Western Himalaya India. Enz. Microbial. Technol. 36:654-659. 5. Bailey, J. S., and N. A. Cox. 1987. Evaluation of the Petrifilm SM and VRB dry media culture plates for determining microbial quality of poultry. J. Food Prot. 50:643-644. 6. Barbano, D. M., Y. Ma, and M. V. Santos. 2005. Influence of raw milk quality on fluid milk shelf life. J. Dairy Sci. 88:77-77. 7. Beuchat, L. R. 1996. Pathogenic microorganisms associated with fresh produce. J. Food Prot. 59:204-216. 8. Bishop, J. R., and A. B. Bodine. 1986. Quality assessment of pasteurized fluid milk as related to lipopolysaccharide content. J. Dairy Sci. 69:3002-3004. 9. Bishop, J. R., and C. H. White. 1985. Estimation of potential shelf life of cottage cheese utilizing bacterial numbers and metabolites. J. Food Prot. 48:1054-1057.

Predicted Size of PCR Product

References

950 bp

127

650 bp

127

160 bp

56

121 bp

159

790 bp

19

935 bp

19

850 bp

182

194 bp

3

900 bp

132

500 bp

127

10. Bishop, J. R., and C. H. White. 1985. Estimation of potential shelf-life of pasteurized fluid milk utilizing bacterial numbers and metabolites. J. Food Prot. 48:663-667. 11. Bishop, J. R., C. H. White, and R. Firstenberg. 1984. Rapid impedimetric method for determining the potential shelf-life of pasteurized whole milk. J. Food Prot. 47:471-475. 12. Boismenu, D., F. Le´pine, M. Gagnon, and H. Dugas. 1990. Heat inactivation of catalase from cod muscle and from some psychrophilic bacteria. J. Food Sci. 55:581-582. 13. Borch, E., M.-L. Kant-Muermans, and Y. Blixt. 1996. Bacterial spoilage of meat and cured meat products. Int. J. Food Microbiol. 33:103-120. 14. Bossuyt, R. 1981. Determination of bacteriological quality of raw milk by an ATP assay technique. Milchwissenschaft 36:257-260. 15. Bossuyt, R. 1982. A 5-minute ATP platform test for judging the bacteriological quality of raw milk. Neth. Milk Dairy J. 36:355-364. 16. Brackett, R. E. 1987. Microbiological consequences of minimally processed fruits and vegetables. J. Food Qual. 10:195206. 17. Brenchley, J. E. 1996. Psychrophilic microorganisms and their cold-active enzymes. J. Indust. Microbiol. 17:432-437. 18. Brocklehurst, T. F., C. M. Zaman-Wong, and B. M. Lund. 1987. A note on the microbiology of retail packs of prepared salad vegetables. J. Appl. Bacteriol. 63:409-415. 19. Broda, D. M., J. A. Boerema, and R. G Bell. 2003. PCR detection of psychrophilic Clostridium spp. causing ‘blown pack’ spoilage of vacuum-packed chilled meats. J. Appl. Microbiol. 94:515-522. 20. Brodsky, M. H., P. Entis, M. P. Entis, A. N. Sharpe, and G. A. Jarvis. 1982. Determination of aerobic plate and yeast and mold counts in foods using an automated hydrophobic gridmembrane filter technique. J. Food Prot. 45:301-304.

| 183

Compendium of Methods for the Microbiological Examination of Foods |

21. Champagne, C. P., R. R. Laing, D. Roy, and A. A. Mafu. 1994. Psychrotrophs in dairy products: their effects and their control. Curr. Rev. Food Sci. Nutr. 34:1-30. 22. Chang, K. S., H. D. Jang, C. G. Lee, Y. G. Lee, C. J. Yuan, and S. H. Lee. 2006. Series quartz crystal sensor for remote bacteria population monitoring in raw milk via the Internet. Biosens. Bioelectron. 2:1581-1590. 23. Collins, E. B. 1981. Heat resistant psychrotrophic microorganisms. J. Dairy Sci. 64:157-160. 24. Coorevits, A., V. De Jonghe, J. Vandroemme, R. Reekmans, J. Heyrman, W. Messens, P. De Vos, and M. Heyndrickx. 2008. Comparative analysis of the diversity of aerobic sporeforming bacteria in raw milk from organic and conventional dairy farms. System. Appl. Microbiol. 31:126-140. 25. Corlett, D. A. 1989. Refrigerated foods and use of hazard analysis and critical control point principles. Food Technol. 43:91-94. 26. Cory, J. E. L. 2007. Spoilage organisms of red meat and poultry. In ‘‘Microbiological Analysis of Red Meat, Poultry and Eggs.’’ (G. C. Mead, ed.), 101-122. Woodhead Publishing Ltd. Cambridge, UK. 27. Cousin, M. A. 1982. Presence and activity of psychrotrophic microorganisms in milk and dairy products: a review. J. Food Prot. 45:172-207. 28. Cousin, M. A., J. M. Jay and P. C. Vasavada. 2001. Psychrotrophic microorganisms. In ‘‘Compendium of Methods for the Microbiological Examination of Foods.’’ 4th ed. (F. P. Downes and K. Ito, eds.), 159-166. American Public Health Association, Washington, DC. 29. Dainty, R. H., and B. M. Mackey. 1992. The relationship between the phenotypic properties of bacteria from chill-stored meat and spoilage processes. J. Appl. Bacteriol. 73:103S-114S. 30. Davail S, G. Feller, E. Narinx, and C. Greday. 1994. Cold adaptation of proteins. Purification, characterization and sequencing of the heatlabile subtilisin from the Antarctic psychrophile Bacillus TA41. J. Biol. Chem. 269:17448-17453. 31. Davenport, R. R. 1980. Cold-tolerant yeasts and yeast-like organisms. In ‘‘Biology and Activities of Yeasts.’’ (F. A. Skinner, S. M. Passmore, and R. R. Davenport, eds.), pp. 215230, Academic Press, New York. 32. de Beer, H., C. J. Hugo, P. J. Jooste, A. Willems, M. Vancanneyt, T. Coenye, and P. A. R. Vandamme. 2005. Chryseobacterium vrystaatense sp nov., isolated from raw chicken in a chicken-processing plant. Int. J. Syst. Evol. Microbiol. 55:2149-2153. 33. de Beer, H. L., C. J. Hugo, P. J. Jooste, M. Vancanneyt, T. Coenye, and P. Vandamme. 2006. Chryseobacterium piscium sp nov., isolated from fish of the South Atlantic Ocean off South Africa. Int. J. Syst. Evol. Microbiol. 56:1317-1322. 34. De Jonghe, V., A. Coorevits, J. De Block, E. Van Coillie, K. Grijspeerdt, L. Herman, P. De Vos, and M. Heyndrickx. 2010. Toxinogenic and spoilage potential of aerobic sporeformers isolated from raw milk. Int. J. Food Microbiol. 136:318-325. 35. De Jonghe, V., A. Coorevits, J. Vandroemme, J. Heyrman, L. Herman, P. De Vos, and M. Heyndrickx. 2008. Intraspecific genotypic diversity of Bacillus species from raw milk. Int. Dairy J. 18:496-505. 36. Delbe`s, C., L. Ali-Mandjee, and M. C. Monte. 2007. Monitoring bacterial communities in raw milk and cheese by culture-dependent and -independent 16S rRNA genebased analyses. Appl. Environ. Microbiol. 73:1882-1891. 37. Denner, E. B. M., B. Mark, H. J. Busse, M. Turkiewicz, and W. Lubitz. 2001. Psychrobacter proteolyticus sp nov., a psychrotrophic, halotolerant bacterium isolated from the Antarctic krill Euphausia superba Dana, excreting a coldadapted metalloprotease. Syst. Appl. Microbiol. 24:44-53.

184 |

38. Dodds, K. L., R. A. Holley, and A. G. Kempton. 1983. Evaluation of the catalase and Limulus amoebocyte lysate tests for rapid determination of the microbial quality of vacuum-packed cooked turkey. Can. Inst. Food Sci. Technol. J. 16:167-172. 39. Dogan, B., K. J. Boor. 2003. Genetic diversity and spoilage potentials among Pseudomonas spp. isolated from fluid milk products and dairy processing plants. Appl. Environ. Microbiol. 69:130-138. 40. Durak, M. Z., H. I. Fromm, J. R. Huck, R. N. Zadoks, and K. J. Boor. 2006. Development of molecular typing methods for Bacillus spp. and Paenibacillus spp. isolated from fluid milk products. J. Food Sci. 71:M50-M56. 41. Easter, M. C., and D. M. Gibson. 1989. Detection of microorganisms by electrical measurements. Prog. Ind. Microbiol. 26:57-100. 42. Easter, M. C., R. G. Kroll, L. Farr, and A. C. Hunter. 1987. Observations on the introduction of the DEFT for the routine assessment of bacteriological quality. J. Soc. Dairy Technol. 40:100-103. 43. Eddy, B. P. 1960. The use and meaning of the term ‘‘psychrophilic.’’ J. Appl. Bacteriol. 23:189-190. 44. Ercolini, D., F. Russo, G. Blaiotta, O. Pepe, G. Mauriello, and F. Villani. 2007. Simultaneous detection of Pseudomonas fragi, P. lundensis, and P. putida from meat by a multiplex PCR assay targeting the carA gene. Appl. Environ. Microbiol. 73:23542359. 45. Ercolini, D., F. Russo, I. Ferrocino, and F. Villani. 2009. Molecular identification of mesophilic and psychrotrophic bacteria from raw cow’s milk. Food Microbiol. 26:228-231. 46. Ercolini, D., F. Russo, A. Nasi, P. Ferranti, and F. Villani. 1990. Mesophilic and psychrotrophic bacteria from meat and their spoilage potential in vitro and in beef. Appl. Environ. Microbiol. 75:1990-2001. 47. Eriksson P. V, G. N. DiPaola, M. F. Pasetti, and M. A. Manghi. 1995. Inhibition enzyme-linked immunosorbent assay for detection of Pseudomonas fluorescens on meat surfaces. Appl. Environ. Microbiol. 61:397-398. 48. Fairbairn, D. J., and B. A. Law. 1986. Proteinases of psychrotrophic bacteria: their production, properties, effects and control. J. Dairy Res. 53:139-177. 49. Fallowfield, H. J., and J. T. Patterson. 1985. Potential value of the Limulus lysate assay for the measurement of meat spoilage. J. Food Technol. 20:467-479. 50. Felice, C. J., R. E, Madrid, J. M. Olivera, V. I. Rotger, and M. E. Valentinuzzi. 1999. Impedance microbiology: quantification of bacterial content in milk by means of capacitance growth curves. J. Microbiol. Methods 35:37-42. 51. Feller G, E. Narinx, J. L. Arpigny, M. Aittaleb, E. Baise, S. Genicot, and C. Gerday. 1996. Enzymes from psychrophilic organisms. FEMS Microbiol. Rev. 18:189-202. 52. Firstenberg-Eden, R. 1983. Rapid estimation of the number of microorganisms in raw meat by impedance measurement. Food Technol. 37:64-67. 53. Firstenberg-Eden, R., and M. K. Ticarico. 1983. lmpedimetric determination of total, mesophilic and psychrotrophic counts in raw milk. J. Food Sci. 48:1750-1754. 54. Fischer, J. E., and P. C. Vasavada. 1987. Rapid detection of abnormal milk by the catalase test. J. Dairy Sci. 70(Suppl 1):75. 55. Forster, M. A. 1985. Factors affecting the use of the Limulus amoebocyte lysate test in the food industry. N. Z. J. Dairy Sci. Technol. 20:163-172. 56. Francis, K. P., R. Mayr, F. von Stetten, G. Stewart, and S. Scherer. 1998. Discrimination of psychrotrophic and mesophilic strains of the Bacillus cereus group by PCR targeting of major cold shock protein genes. Appl. Environ. Microbiol. 64:3525-3529.

| Psychrotrophic Microorganisms

57. Frank, J. 2007. Milk and dairy products. In ‘‘Food Microbiology: Fundamentals and Frontiers.’’ 3rd ed. (M. P. Doyle and L. R. Beuchat, eds.), pp. 141-155, ASM Press, Washington, DC. 58. Genigeorgis, C. A. 1985. Microbial and safety implications of the use of modified atmospheres to extend the storage life of fresh meat and fish. Int. J. Food Microbiol. 1:237-251. 59. Giannino, M. L., M. Aliprandi, M. Feligini, L. Vanoni, M. Brasca, and F. Fracchetti. 2009. A DNA array based assay for the characterization of microbial community in raw milk. J. Microbiol. Methods. 78:181-188. 60. Giffel, M. C. T., R. R. Beumer, P. E. Granum, and F. M. Rombouts. 1997. Isolation and characterisation of Bacillus cereus from pasteurized milk in household refrigerators in the Netherlands. Int. J. Food Microbiol. 34:307-318. 61. Gill, C. O., and K. G. Newton 1979. Spoilage of vacuumpacked dark firm dry meat at chilled temperatures. Appl. Environ. Microbiol. 37:362-364. 62. Gilliland, S. E., H. D. Michener, and A. A. Kraft. 1976. Psychrotrophic microorganisms. In ‘‘Compendium of Methods for the Microbiological Examination of Foods.’’ 1st ed. (M. L. Speck ed.), pp. 173-178, American Public Health Association, Washington, DC. 63. Gilliland, S. E., H. D. Michener, and A. A. Kraft. 1984. Psychrotrophic microorganisms. In ‘‘Compendium of Methods for the Microbiological Examination of Foods.’’ 2nd ed. (M. L. Speck, ed.), pp. 135-141, American Public Health Association, Washington, DC. 64. Ginn, R. E., V. S. Packard, and T. L. Fox. 1984. Evaluation of the 3M dry medium culture plate (Petrifilm SM) method for determining numbers of bacteria in raw milk. J. Food Prot. 47:753-755. 65. Glass, M. B., and T. Popovic. 2005. Preliminary evaluation of the API 20NE and RapID NF Plus systems for rapid identification of Burkholderia pseudomallei and B. mallei. J. Clin. Microbiol. 43:479-483. 66. Gonza´lez I, R. Martı´n, T. Ga´rcia, P. Morales, B. Sanz, and P. E. Herna´ndez. 1993. A sandwich enzyme-linked immunosorbent assay (ELISA) for detection of Pseudomonas fluorescens and related psychrotrophic bacteria in refrigerated milk. J. Appl. Bacteriol. 74:394-401. 67. Gonza´lez I, R. Martı´n, T. Ga´rcia, P. Morales, B. Sanz, and P. E. Herna´ndez. 1994. Detection of Pseudomonas fluorescens and related psychrotrophic bacteria in refrigerated meat by a sandwich ELISA. J. Food Prot. 57:710-714. 68. Gonzalez I, R. Martin, T. Garcia, P. Morales, B. Sanz, and P. E. Hernandez. 1994. Polyclonal antibodies against live cells of Pseudomonas fluorescens for the detection of psychrotrophic bacteria in milk using a double antibody sandwich enzymelinked immunosorbent assay. J. Dairy Sci. 77:3552-3557. 69. Gonza´lez I, R. Martı´n, T. Ga´rcia, P. Morales, B. Sanz, and P. E. Herna´ndez. 1996. Polyclonal antibodies against protein F from the cell envelope of Pseudomonas fluorescens for detection of psychrotrophic bacteria in refrigerated meat using an indirect ELISA. Meat Sci. 42:305-313. 70. Gounot, A. M. 1986. Psychrophilic and psychrotrophic microorganisms. Experientia 42:1192-1197. 71. Gounot, A. M. 1991. Bacterial life at low temperature: physiological aspects and biotechnical implications. J. Appl. Bacteriol. 71:386-397. 72. Gram L., and H. H. Huss. 1996. Microbiological spoilage of fish and fish products. Int. J. Food Microbiol. 33:121-137. 73. Gram, L., and H. Sogaard. 1985. Microcalorimetry as a rapid method for estimation of bacterial levels in ground meat. J. Food Prot. 48:341-345. 74. Greer, G. G. 1981. Rapid detection of psychrotrophic bacteria in relation to retail beef quality. J. Food Sci. 46:1669-1672.

75. Griffiths, M. W., J. D. Phillips, and D. D. Muir. 1980. Rapid plate counting techniques for enumeration of psychrotrophic bacteria in pasteurized double cream. J. Soc. Dairy Technol. 33:8-10. 76. Gunasekera, T. S., M. R. Dorsch, M. B. Slade, and D. A. Veal. 2003. Specific detection of Pseudomonas spp. in milk by fluoresence in situ hybridization using rRNA directed probes. J. Appl. Microbiol. 94:926-945. 77. Gutierrez, R., T. Ga´ rcia, I. Gonza´ lez, B. Sanz, P. E. Herna´ndez, and R. Martı´n. 1997. Monoclonal antibody detection of Pseudomonas spp. in refrigerated meat by an indirect ELISA. Lett. Appl. Microbiol. 24:5-8. 78. Gutierrez, R., I. Gonza´lez, T. Ga´rcia, E. B. Carrera, Sanz, P. E. Herna´ndez, and R. Martı´n. 1997. Monoclonal antibodies and an indirect ELISA for detection of psychrotrophic bacteria in refrigerated milk. J. Food Prot. 60:23-27. 79. Hansen, K., T. Mikkelsen, and A. Moller-Madsen. 1982. Use of the Limulus test to determine the hygienic status of milk products as characterized by levels of Gram-negative bacterial lipopolysaccharide present. J. Dairy Res. 49:323-328. 80. Hantsis-Zacharov, E., and M. Halpern. 2007. Chryseobacterium haifense sp nov., a psychrotolerant bacterium isolated from raw milk. Int. J. Syst. Evol. Microbiol. 57:2344-2348. 81. Hantsis-Zacharov, E., and M. Halpern. 2007. Culturable psychrotrophic bacterial communities in raw milk and their proteolytic and lipolytic traits. Appl. Environ. Microbiol. 73:7162-7168. 82. Hantsis-Zacharov, E., Y. Senderovich, and M. Halpern. 2008. Chryseobacterium bovis sp nov., isolated from raw cow’s milk. Int. J. Syst. Evol. Microbiol. 58:1024-1028. 83. Hebraud M, E. Dubois, P. Potier, and J. Labadie. 1994. Effect of growth temperatures on the protein levels in a psychrotrophic bacterium, Pseudomonas fragi. J. Bacteriol. 176:40174024. 84. Herbert, R. A. 1981. A comparative study of the physiology of psychrotrophic and psychrophilic bacteria, In ‘‘Psychrotrophic Microorganisms in Spoilage and Pathogenicity.’’ (T. A. Roberts, G. Hobbs, J. H. B. Christian, and N. Skovgaard, eds.), pp. 3-16, Academic Press, New York. 85. Herbert, R. A. 1986. The ecology and physiology of psychrophilic microorganisms. In ‘‘Microbes in extreme environments.’’ (R. A. Herbert and G. A. Codd, eds.), pp. 1-23, Academic Press, New York. 86. Heyndrickx, M., and P. Scheldeman. 2002. Bacilli associated with spoilage in dairy and other food products. In ‘‘Applications and Systematics of Bacillus and Relatives.’’ (R. Berkely, M. Heyndrickx, N. A. Logan, and P. De Vos, eds.), pp. 64-82, Blackwell Science, Oxford, UK. 87. Hill, S. D., R. L. Richter, and C. W. Dill. 1988. Evaluation of a catalase-based method to predict the shelf-life of pasteurized milk. J. Dairy Sci. 71(Suppl 1):112. 88. Hintlian, C. B., and J. H. Hotchkiss. 1986. The safety of modified atmosphere packaging: a review. Food Technol. 40:70-76. 89. Hobbs, G. 1983. Microbial spoilage of fish, In ‘‘Food Microbiology: Advances and Prospects.’’ (T. A. Roberts and F. A. Skinner, eds.), pp. 217-229, Academic Press, New York. 90. Hobbs, G., and W. Hodgkiss. 1982. The bacteriology of fish handling and processing. In ‘‘Developments in Food Microbiology–1.’’ (R. Davies, ed.), pp. 71-117, Applied Science Publishers, Inc., Englewood, NJ. 91. Holly, R.A., M. D. Pierson, J. Lam, and K. B. Tank. 2004. Microbial profiles of commercial vacuum-packaged pork of normal or short storage life. Int. J. Food Microbiol. 97:5362. 92. Hsu, E. J., and L. R. Beuchat. 1986. Factors affecting microflora in processed fruits. In ‘‘Commercial Fruit Processing.’’

| 185

Compendium of Methods for the Microbiological Examination of Foods |

93.

94.

95.

96. 97. 98.

99.

100.

101.

102.

103. 104.

105. 106.

107.

108.

109. 110.

111.

112.

113.

186 |

2nd ed., (J. G. Woodroof and B. S. Luh, eds.),’’ pp. 129-161, AVI Publishing Company, Inc., Westport, CT. Huck, J. R., M. Sonnen, and K. J. Boor. 2008. Tracking heatresistant, cold-thriving fluid milk spoilage bacteria from farm to packaged product. J. Dairy Sci. 91:1218-1228. Hugo, C. J., P. J. Jooste, P. Segers, M. Vancanneyt, and K. Kersters. 1999. A polyphasic taxonomic study of Chryseobacterium strains isolated from dairy sources. Syst. Appl. Microbiol. 22:586-595. Hugo, C. J., P. Segers, B. Hoste, M. Vancanneyt, and K. Kersters. 2003. Chryseobacterium joostei sp nov., isolated from the dairy environment. Int. J. Syst. Evol. Microbiol. 53:771-777. Ingraham, J. L., and J. L. Stokes. 1959. Psychrophilic bacteria. Bacteriol. Rev. 23:97-108. Isonhood, J. H. and M. Drake. 2002. Aeromonas species in foods. J. Food Prot. 65:575-582. Ivy R. A, M. L. Ranieri, N. H. Martin, H. C. den Bakker, B. M. Xavier, M. Wiedmann, and K. J. Boor. 2012. Identification and characterization of psychrotolerant sporeformers associated with fluid milk production and processing. Appl. Environ. Microbiol. 78:1853-1864. Jany, J. L., and G. Barbier. 2008. Culture-independent methods for identifying microbial communities in cheese. Food Microbiol. 25:839-848. Jarvis, B., V. H. Lach, and J. M. Wood. 1977. Evaluation of the spiral plate maker for the enumeration of micro-organisms in foods. J. Appl. Bacteriol. 43:149-157. Jay, J. M. 1987. ‘‘Meats, poultry, and seafoods.’’ 2nd ed. In ‘‘Food and Beverage Mycology.’’ (L. R. Beuchat, ed.), pp. 155-173, Van Nostrand Reinhold Co., New York. Jay, J. M. 1987. The tentative recognition of psychrotrophic Gram-negative bacteria in 48 h by their surface growth at 10 uC. J. Food Microbiol. 4:25-32. Jay, J. M. 1989. The Limulus amoebocyte lysate (LAL) test. Prog. Ind. Microbiol. 26:101-119. Jay, J. M., and M. E. Bue. 1987. Ineffectiveness of crystal violet tetrazolium agar for determining psychrotrophic Gram-negative bacteria. J. Food Prot. 50:147-149. Jay, J. M., M. J. Loessner, and D.A. Golden 2005. Modern Food Microbiology, 7th ed., Springer, New York. Jay, J. M., J. P. Vilai, and M. E. Hughes. 2003. Profile and activity of the bacterial biota of ground beef held from freshness to spoilage at 5–7uC. Int. J. Food Microbiol. 81:105111. Juffs, H. S., and F. J. Babel. 1975. Rapid enumeration of psychrotrophic bacteria in raw milk by the microscopic colony count. J. Milk Food Technol. 38:333-336. Kahn, P., and R. Firstenberg-Eden. 1987. Prediction of shelflife of pasteurized milk and other fluid dairy products in 48 hours. J. Dairy Sci. 70:1544-1150. Kasana, R. C. 2010. Proteases from psychrotrophs: an overview. Crit. Rev. Microbiol. 36:134-145. Kasana R. C., and S. K. Yadav. 2007. Isolation of a psychrotrophic Exiguobacterium sp SKPB5. (MTCC 7803) and characterization of its alkaline protease. Curr. Microbiol. 54:224-229. King, A. D. Jr., and H. R. Bolin. 1989. Physiological and microbiological storage stability of minimally processed fruits and vegetables. Food Technol. 43:132-135, 139. Kishore, K. H., Z. Begum, A. A. K. Pathan, and S. Shivaji. 2010. Paenibacillus glacialis sp nov., isolated from the Kafni glacier of the Himalayas, India. Int. J. Syst. Evol. Microbiol. 60:1909-1913. Kraft, A. A. 1986. Psychrotrophic organisms. In ‘‘Advances in Meat Research. Vol. 2: Meat and Poultry Microbiology.’’ (A. M. Pearson and T. R. Dutson, eds.), pp. 191208, AVI Publishing Company, Inc., Westport, CT.

114. Kristjansson, M. M., O. T. Magnusson, H. M. Gudmundsson, G. A. Alfredsson, and H. Matsuzawa. 1999. Properties of a subtilisin-like proteinase from a psychrotrophic Vibrio species—Comparison with proteinase K and aqualysin I. Eur. J. Biochem. 260:752-760. 115. Kroll, R. G. 1985. The cytochrome c oxidase test for the rapid detection of psychrotrophic bacteria in milk. J. Appl. Bacteriol. 59:137-141. 116. Kroll, R. G., and U. M. Rodrigues. 1986. Prediction of the keeping quality of pasteurized milk by the detection of cytochrome c oxidase. J. Appl. Bacteriol. 60:21-27. 117. Kroll, R. G., and U. M. Rodrigues. 1986. The direct epifluorescent filter technique, cytochrome c oxidase test and plate count method for predicting the keeping quality of pasteurized cream. Food Microbiol. 3:185-194. 118. Kulakova, L., A. Galkin, T. Kurihara, T. Yoshimura, and N. Esaki. 1999. Cold-active serine alkaline protease from the psychrotrophic bacterium Shewanella strain Ac10: Gene cloning and enzyme purification and characterization. Appl. Environ. Microbiol. 65:611-617. 119. Lafarge, V., J. C. Ogier, V. Girard, V. Maladen, J. Y. Leveau, A. Gruss, and A. Delacroix-Buchet. 2004. Raw cow milk bacterial population shifts attributable to refrigeration. Appl. Environ. Microbiol. 70:5644-5650. 120. Lannelongue, M., G. Finne, M. O. Hanna, R. Nickelson II, and C. Vanderzant. 1982. Microbiological and chemical changes during storage of swordfish (Xiphias gladius) steaks in retail packages containing CO2-enriched atmospheres. J. Food Prot. 45:1197-1203. 121. Laureyn, W., D. Nelis, P. Van Gerwen, K. Baert, L. Hermans, R. Magnee, J. J. Pireaux, and G. Maes. 2000. Nanoscaled interdigitated titanium electrodes for impedimetric biosensing. Sens. Actuat. B 68:360-370. 122. Laursen, B. G., L. Bay, I. Cleenwerck, M. Vancanneyt, J. Swings, P. Dalgaard, and J. J. Leisner. 2005. Carnobacterium divergens and Carnobacterium maltaromaticum as spoilers or protective cultures in meat and seafood: phenotypic and genotypic characterization. Syst. Appl. Microbiol. 28:151-164. 123. Lechowich, R. V. 1988. Microbiological challenges of refrigerated foods. Food Technol. 42:84-85, 89. 124. Li, W., D. Raoult, and P. E. Fournier. 2009. Bacterial strain typing in the genomic era. FEMS Microbiol. Rev. 33:892-916. 125. Littel, K. J., S. Pikelis, and A. Spurgash. 1986. Bioluminescent ATP assay for rapid estimation of microbial numbers in fresh meat. J. Food Prot. 49:18-22. 126. Lund, B. M. 1983. Bacterial spoilage. In ‘‘Post-harvest Pathology of Fruits and Vegetables.’’ (C. Dennis, ed.), pp. 219-257, Academic Press, New York. 127. Machado, S. G., D. M. S. Bazzolli, and M. C. D. Vanetti. 2013. Development of a PCR method for detecting proteolytic psychrotrophic bacteria in raw milk. Int. Dairy J. 29:8-14. 128. Makarios-Laham, I., and R. E. Levin. 1984. Isolation from haddock tissue of psychrophilic bacteria with maximum growth temperature below 20uC. Appl. Environ. Microbiol. 48:439-440. 129. Makarios-Laham, I., and R. E. Levin. 1985. Autolysis of psychrophilic bacteria from marine fish. Appl. Environ. Microbiol. 49:997-998. 130. Manvell, P. M., and M. R. Ackland. 1986. Rapid detection of microbial growth in vegetable salads at chill and abuse temperatures. Food Microbiol. 3:59-65. 131. Manzano, S., Ordonez, J. A., Hoz, L., Fernandez, M., 2005. A rapid method for the estimation of the microbiological quality of refrigerated raw milk based on the aminopeptidase activity of Gram-negative bacteria. Int. Dairy J. 15: 79-84.

| Psychrotrophic Microorganisms

132. Marchand, S., G. Vandriesche, A. Coorevits, K. Coudijzer, V. De Jonghe, K. Dewettinck, P. De Vos, B. Devreese, M. Heyndrickx, and J. De Block. 2009. Heterogeneity of heat-resistant proteases from milk Pseudomonas species. Int. J. Food Microbiol. 133:68-77. 133. Matamoros, S., F. Leroi, M. Cardinal, F. Gigout, F. K. Chadli, J. Cornet, H. Prevost, and M. F. Pilet. 2009. Psychrotrophic lactic acid bacteria used to improve the safety and quality of vacuum-packaged cooked and peeled tropical shrimp and cold-smoked salmon. J. Food Prot. 72:365-374. 134. Matamoros S, M. F. Pilet, F. Gigout, H. Pre´vost, and F. Leroi. 2009. Selection and evaluation of seafood-borne psychrotrophic lactic acid bacteria as inhibitors of pathogenic and spoilage bacteria. Food Microbiol. 26:638-644. 135. McGoldrick, K. F., T. L. Fox, and J. S. McAllister. 1986. Evaluation of a dry medium for detecting contamination on surfaces. Food Technol. 40:77-80. 136. McKellar, R. C. (ed.). 1989. Enzymes of psychrotrophs in raw food. CRC Press, Boca Raton, FL. 137. McMeekin, T. A. 1982. Microbial spoilage of meats. In ‘‘Developments in food microbiology–1.’’ (R. Davies, ed.), pp. 1-40, Applied Science Publishers, Inc., Englewood, NJ. 138. Mead G. C. 2007. Fecal indicator organisms for red meat and poultry. In ‘‘Microbiological Analysis of Red Meat, Poultry and Eggs.’’ (G. C. Mead, ed.), pp. 83-100, Woodhead Publishing Ltd., Cambridge, UK. 139. Meer, R. R., J. Bakker, F. W. Bodyfelt, and M. W. Griffiths. 1991. Psychotrophic Bacillus spp. in fluid milk products: a review. J. Food Prot. 54:969-979. 140. Mikolajcik, E. M., and R. B. Brucker. 1983. Limulus amebocyte lysate assay—a rapid test for the assessment of raw and pasteurized milk quality. Dairy Food Sanit. 3:129-131. 141. Ming, H., G.-X. Nie, H.-C. Jiang, T.-T. Yu, E.-M. Zhou, H.-G. Feng, S.-K. Tang, and W.-J. Li. 2012. Paenibacillus frigoriresistens sp nov., a novel psychrotroph isolated from a peat bog in Heilongjiang, Northern China. Antonie Van Leeuwenhoek. Int. J. Gen. Mol. Microbiol. 102:297-305. 142. Montville, T. J., K. R. Matthews, and K. E. Kniel. 2012. Food Microbiology: an Introduction, 3rd ed., ASM Press, Washington, DC. 143. Morita, R. Y. 1975. Psychrophilic bacteria. Bacteriol. Rev. 39:144-167. 144. Mossel, D. A. A., C. M. L. Marengo, and C. B. Struijk. 1994. History of and prospects for rapid and instrumental methodology for the microbiological examination of foods, In ‘‘Rapid Analysis Techniques in Food Microbiology.’’ (P. D. Patel, ed.), pp. 1-28, Blackie Academic and Professional, New York. 145. Mossel, D. A. A., and H. Zwart. 1960. The rapid tentative recognition of psychrotrophic types among Enterobacteriaceae isolated from foods. J. Appl. Bacteriol. 23:185-188. 146. Munsch-Alatossava, P. and T. Alatossava, 2006. Phenotypic characterization of raw milk-associated psychrotrophic bacteria. Microbiol. Res. 161:334-346. 147. Neaves, P., D. I. Jervis, and G. A. Prentice. 1987. A comparison of DEFT clump counts obtained in eight dairy laboratories receiving replicate samples of preserved raw milk. J. Soc. Dairy Technol. 40:53-56. 148. Niza-Ribeiro, J., A. C Louza, P. Santos, and M. Lima. 2000. Monitoring the microbiological quality of raw milk through the use of an ATP bioluminescence method. Food Control 11:209-216. 149. Oehlrich, H. K., and R. C. McKellar. 1983. Evaluation of an 18uC/45-hour plate count technique for the enumeration of psychrotrophic bacteria in raw and pasteurized milk. J. Food Prot. 46:528-529. 150. Ogden, I. D. 1986. Use of conductance methods to predict bacterial counts in fish. J. Appl. Bacteriol. 61:263-268.

151. Ogier, J. C., V. Lafarge, V. Girard, A. Rault, V. Maladen, A. Gruss, J.-Y. Leveau, and A. Delacroix-Buchet. 2004. Molecular fingerprinting of dairy microbial ecosystems by use of temporal temperature and denaturing gradient gel electrophoresis. Appl. Environ. Microbiol. 70:5628-5643. 152. Ogier, J. C., O. Son, A. Gruss, P. Tailliez, and A. DelacroixBuchet. 2002. Identification of the bacterial microflora in dairy products by temporal temperature gradient gel electrophoresis. Appl. Environ. Microbiol. 68:3691-3701. 153. Oh, K. H., C. S. Seong, S. W. Lee, O. S. Kwon, and Y. S. Park. 1999. Isolation of a psychrotrophic Azospirillum sp. and characterization of its extracellular protease. FEMS Microbiol. Lett. 174:173-178. 154. Oliveria, J. S., and C. E. Parmelee. 1976. Rapid enumeration of psychrotrophic bacteria in raw and pasteurized milk. J. Milk Food Technol. 39:269-272. 155. Ong, K. G., J. Wang, R. S. Singh, L. G. Bachas, C. A. Grimes. 2001. Monitoring of bacteria growth using a wireless, remote query resonant-circuit sensor: application to environmental sensing. Biosens. Bioelectron. 16:305-312. 156. Palumbo, S. 1986. A. Is refrigeration enough to restrain foodborne pathogens? J. Food Prot. 49:1003-1009. 157. Patel, P. D., and A. P. Williams. 1985. A note on estimation of food spoilage yeasts by measurement of adenosine triphosphate (ATP) after growth at various temperatures. J. Appl. Bacteriol. 59:133-136. 158. Peck, M. W. 1997. Clostridium botulinum and the safety of refrigerated processed foods of extended durability. Trends Food Sci. Technol. 8:186-192. 159. Pennacchia, C., D. Ercolini, and F. Villani. 2009. Development of a real-time PCR assay for the specific detection of Brochothrix thermosphacta in fresh and spoiled raw meat. Int. J. Food Microbiol. 134:230-236. 160. Perez De Castro, B., M. A. Asensio, B. Sanz, and J. A. Ordon˜ez. 1988. A method to assess the bacterial content of refrigerated meat. Appl. Environ. Microbiol. 54:1462-1465. 161. Pettipher, G. L. 1981. Rapid methods for assessing bacterial numbers in milk. Dairy Ind. Int. 46:15-23. 162. Pettipher, G. L. 1989. The direct epifluorescent filter technique. Prog. Ind. Microbiol. 26:19-56. 163. Pettipher, G. L., R. Mansell, C. H. McKinnon, and C. M. Cousins. 1980. Rapid membrane filtration-epifluorescent microscopy technique for direct enumeration of bacteria in raw milk. Appl. Environ. Microbiol. 39:423-429. 164. Phillips, J. D., and M. W. Griffiths. 1987. A note on the use of the Catalasemetre in assessing the quality of milk. J. Appl. Bacteriol. 62:223-226. 165. Phillips, J. D., and M. W. Griffiths. 1987. The relation between temperature and growth of bacteria in dairy products. Food Microbiol. 4:173-185. 166. Philips, J. D., and M. W. Griffiths, and D. D. Muir. 1983. Accelerated detection of post-heat-treatment contamination in pasteurized double cream. J. Soc. Dairy Technol. 36:41-43. 167. Quigley, L., R. McCarthy, O. O’Sullivan, T. P. Beresford, G. F. Fitzgerald, R. P. Ross, C. Stanton, and P. D. Cotter. 2013. The microbial content of raw and pasteurized cow milk as determined by molecular approaches. J. Dairy Sci. 96:4928-4937. 168. Quigley, L., O. O’Sullivan, T. P. Beresford, R. P. Ross, G. F. Fitzgerald and P. D. Cotter. 2011. Molecular approaches to analysing the microbial composition of raw milk and raw milk cheese. Int. J. Food Microbiol. 150:81-94. 169. Quigley, L., O. O’Sullivan, C. Stanton, T. P. Beresford, R. P. Ross, G. F. Fitzgerald, and P. D. Cotter. 2013. The complex microbiota of raw milk. FEMS Microbiol. Rev. 37:664-698. 170. Ranieri, M. L., R. A. Ivy, W. R. Mitchell, E. Call, S. N. Masiello, M. Wiedmann and K. J. Boor. 2012. Real-time PCR detection of Paenibacillus spp. in raw milk to predict shelf life

| 187

Compendium of Methods for the Microbiological Examination of Foods |

171.

172.

173.

174.

175.

176.

177. 178.

179.

180.

181.

182.

183.

184.

185.

186.

187. 188. 189.

188 |

performance of pasteurized fluid milk products. Appl. Environ. Microbiol. 78:5855-5863. Reichardt, W., and R. Y. Morita. 1982. Temperature characteristics of psychrotrophic and psychrophilic bacteria. J. Gen. Microbiol. 128:565-568. Reuter, G. 1981. Psychrotrophic lactobacilli in meat products. In ‘‘Psychrotrophic Microorganisms in Spoilage and Pathogenicity.’’ (T. A. Roberts, G. Hobbs, J. H. B. Christian, and N. Skovgaard, eds.), pp. 253-258, Academic Press, New York. Rodrigues D. F, J. Goris, T. Vishnivetskay, D. Gilichinsky, M. F. Thomashow, and J. M. Tiedje. 2006. Characterization of Exiguobacterium isolates from the Siberian permafrost: description of Exiguobacterium sibiricum sp. Nov. Extremophiles 10:285-294. Rodrigues, U. M., and R. G. Kroll. 1985. The direct epifluorescent filter technique (DEFT): increased selectivity, sensitivity and rapidity. J. Appl. Bacteriol. 59:493-499. Ronk, R. J., K. L. Carson, and P. Thompson. 1989. Processing, packaging, and regulation of minimally processed fruits and vegetables. Food Technol. 43:136-139. Rosmini, M. R., M. L. Signorini, R. Schneider, and J. C. Bonazza. 2004. Evaluation of two alternative techniques for counting mesophilic aerophilic aerobic bacteria in raw milk. Food Control. 15:39-44. Russell, N. J. 1990. Cold adaptation of microorganisms. Philos. Trans. R. Soc. Lond., B, Biol. Sci. 326:595-611. Russell, N. J. 1998. Molecular adaptations in psychrophilic bacteria, potential for biotechnological applications. Adv. Biochem. Eng. Biotechnol. 61:1-21. Salwan, R., A. Gulati, and R. C. Kasana. 2010. Phylogenetic diversity of alkaline protease-producing psychrotrophic bacteria from glacier and cold environments of Lahaul and Spiti, India. J. Basic Microbiol. 50:150-159. Salwan, R., and R. C. Kasana. 2013. Purification and characterization of an extracellular low temperature-active and alkaline stable peptidase from psychrotrophic Acinetobacter sp MN 12 MTCC (10786). Indian J. Microbiol. 53:63-69. Sayem El Daher, N., and R. E. Simard. 1985. Putrefactive amine changes in relation to microbial counts of ground beef during storage. J. Food Prot. 48:54-58. Scarpellini, M., L. Franzetti, and A. Galli. 2004. Development of PCR assay to identify Pseudomonas fluorescens and its biotype. FFEMS Microbiol. Let. 236:257-260. Schaertel, B. J., and N. Tsang, and R. Firstenberg-Eden. 1987. Impedimetric detection of yeast and mold. Food Microbiol. 4:155-163. Scheldeman, P., K. Goossens, M. Rodriguez-Diaz, A. Pil, J. Goris, L. Herman, P. De Vos, N. A. Logan, and M. Heyndrickx. 2004. Paenibacillus lactis sp nov., isolated from raw and heat-treated milk. Int. J. Syst. Evol. Microbiol. 54:885-891. Scheldeman, P., L. Herman, S. Foster, and M. Heyndrickx. 2006. Bacillus sporothermodurans and other highly heatresistant spore formers in milk. J. Appl. Microbiol. 101:542555. Scheldeman, P., A. Pil, L. Herman, P. De Vos, and M. Heyndrickx. 2005. Incidence and diversity of potentially highly heat-resistant spores isolated at dairy farms. Appl. Environ. Microbiol. 71:1480-1494. Schellekens, M. 1996. New research issues in sous-vide cooking. Trends Food Sci. Technol. 7:256-262. Shah, N. P. 1994. Pyschrotorphs in milk: a review. Milchwissenschaft. 49:432-437. Sharpe, A. N. 1989. The hydrophobic grid-membrane filter. Prog. Ind. Microbiol. 26:169-189.

190. Shimomura, K., S. Kaji, and A, Hiraishi. 2005. Chryseobacterium shigense sp. nov., a yellow pigmented, aerobic bacterium isolated from a lactic acid beverage. Int. J. Syst. Evol. Microbiol. 55:1903-1906. 191. Smith, T. L., and L. D. Witter. 1979. Evaluation of inhibitors for rapid enumeration of psychrotrophic bacteria. J. Food Prot. 42:158-160. 192. Smithwell N., and K. Kailasapathy. 1995. Psychrotrophic bacteria in pasteurized milk: problems with shelf life. Aust. J. Dairy Technol. 50:28-31. 193. Søgaard, H., and R. Lund. 1981. A comparison of three methods for the enumeration of psychrotrophic bacteria in raw milk, In ‘‘Psychrotrophic Microorganisms in Spoilage and Pathogenicity.’’ (T. A. Roberts, G. Hobbs, J. H. B. Christian, and N. Skovgaard, eds.), pp. 109-116, Academic Press, New York. 194. Sommer, N. F. 1985. Strategies for control of postharvest diseases of selected commodities, In ‘‘Postharvest Technology of Horticultural Crops. Special Publication 3311.’’ (A. A. Kader, R. F. Kasmire, F. G. Mitchell, M. S. Reid, N. F. Sommer, and J. F. Thompson, eds.), pp. 83-99, Cooperative Extension, University of California, Davis, CA. 195. Sørhaug, T., and L. Stepaniak. 1997. Psychrotrophs and their enzymes in milk and dairy products: Quality aspects. Trends Food Sci. Technol. 8:35-41. 196. Stannard, C. J. 1989. ATP estimation. Prog. Ind. Microbiol. 26:1-18. 197. Stannard, C. J., A. P. Williams, and P. A. Gibbs. 1985. Temperature/growth relationships for psychrotrophic foodspoilage bacteria. Food Microbiol. 2:115-122. 198. Stannard, C. J., and J. M. Wood. 1983. The rapid estimation of microbial contamination of raw meat by measurement of adenosine triphosphate (ATP). J. Appl. Bacteriol. 55:429-438. 199. Su¨di, J., G. Suhren, W. Heeschen, and A. ToIle. 1981. Entwicklung eines miniaturisierten Limulus-Tests im Mikrotiter-System zum quantitativen Nachweis Gram-negativer Bakterien in Milch und Milchprodukten. Milchwissenschaft 36:193-198. 200. Suhren, G., and W. Heeschen. 1987. Impedance assays and the bacteriological testing of milk and milk products. Milchwissenschaft 42:619-627. 201. Suhren, G., W. Heeschen, and A. Tolle. 1982. Quantitative Bestimmung psychrotropher Mikroorganismen in Roh-und pasteurisierter Milchein Methodenvergleich. Milchwissenschaft 37:594-596. 202. Sullivan, J. D. Jr., P. C. Ellis, R. G. Lee, W. S. Combs Jr., and S. W. Watson. 1983. Comparison of the Limulus amoebocyte lysate test with plate counts and chemical analyses for assessment of the quality of lean fish. Appl. Environ. Microbiol. 45:720-722. 203. Svensson, B., K. Ekelund, H. Ogura, and A. Christiansson, 2004. Characterization of Bacillus cereus isolated from milk silo tanks at eight different dairy plants. Int. Dairy J. 14:17-27. 204. Szabo, E. A., K. J. Scurrah, and J. M. Burrows. 2000. Survey for psychrotrophic bacterial pathogens in minimally processed lettuce. Lett. Appl. Microbiol. 30:456-460. 205. Taylor, J. M. W., A. D. Sutherland, K. E. Aidoo, and N. A. Logan. 2005. Heat-stable toxin production by strains of Bacillus cereus, Bacillus firmus, Bacillus megaterium, Bacillus simplex and Bacillus licheniformis. FEMS Microbiol. Lett. 242:313-317. 206. Ternstro¨m, A., A. M. Lindberg, G. Molin. 1993. Classification of the spoilage flora of raw and pasteurized bovine milk, with special reference to Pseudomonas and Bacillus. J. Appl. Bacteriol. 75:25-34. 207. Thomas, S. B. 1969. Methods of assessing the psychrotrophic bacterial content of milk. J. Appl. Bacteriol. 32:269-296.

| Psychrotrophic Microorganisms

208. Tsuji, K., P. A. Martin, and D. M. Bussey. 1984. Automation of chromogenic substrate Limulus amoebocyte lysate assay method for endotoxin by robotic system. Appl. Environ. Microbiol. 48:550-555. 209. Ulukanli, Z., and M. Digrak. 2002. Alkaliphilic microorganisms and habitats. Turk. J. Biol. 26:181-191. 210. Van Spreekens, K. J. A., and F. K. Stekelenburg. 1986. Rapid estimation of the bacteriological quality of fresh fish by impedance measurements. Appl. Microbiol. Biotechnol. 24:95-96. 211. Van Spreekens, K. J. A., and L. Toepoel. 1981. Quality of fishery products in connection with the psychrophilic and psychrotrophic bacterial flora. In ‘‘Psychrotrophic Microorganisms in Spoilage and Pathogenicity.’’ (T. A. Roberts, G. Hobbs, J. H. B. Christian, and N. Skovgaard, eds.), pp. 283-294, Academic Press, New York. 212. Vanderzant, C., and A. W. Matthys. 1965. Effect of temperature of the plating medium on the viable count of psychrophilic bacteria. J. Milk Food Technol. 28:383-388. 213. Vasavada, P. C. 1988. Low-acid foods defy liabilities. Prep. Foods. 157:122-123, 125. 214. Vasavada, P. C. 1988. Pathogenic bacteria in milk—a review. J. Dairy Sci. 71:2809-2816. 215. Vasavada, P. C., T. A. Bon, and L. Bauman. 1988. The use of Catalasemeter in assessing abnormality in raw milk. J. Dairy Sci. 71(Suppl 1):113. 216. Venkitanarayanan, K. S., M. I. Khan, C. Faustman, and B. W. Berry. 1996. Detection of meat spoilage bacteria by using the polymerase chain reaction. J. Food Prot. 59: 845-848.

217. Waes, G. M., and R. G. Bossuyt. 1982. Usefulness of the benzalkoncrystal violet-ATP method for predicting the keeping quality of pasteurized milk. J. Food Prot. 45:928-931. 218. Wang, G. I. J., and D. Y. C. Fung. 1986. Feasibility of using catalase activity as an index of microbial loads on chicken surfaces. J. Food Sci. 51:1442-1444. 219. Wang, L., B. M. Jayarao. 2001. Phenotypic and genotypic characterization of Pseudomonas fluorescens isolated from bulk tank milk. J. Dairy Sci. 84:1421-1429. 220. Ward, D. R., K. A. LaRocco, and D. J. Hopson. 1986. Adenosine triphosphate bioluminescent assay to enumerate bacterial numbers on fresh fish. J. Food Prot. 49:647-650. 221. White, C. H., J. R. Bishop, and D. M. Morgan. 1992. Microbiological methods for dairy products. In ‘‘Standard Methods for the Examination of Dairy Products.’’ 16th ed. (R. T. Marshall, ed.), pp. 287-308, American Public Health Association, Washington, DC. 222. Wiedmann, M., D. Weilmeier, S. S. Dineen, R. Ralyea, and K. J. Boor. 2000. Molecular and phenotypic characterization of Pseudomonas spp. isolated from milk. Appl. Environ. Microbiol. 66:2085-2095. 223. Witter, L. D. 1961. Psychrophilic bacteria—a review. J. Dairy Sci. 44:983-1015. 224. Yang, L., Y. Li, C. L. Griffis, and M. J. Johnson. 2004. Interdigitated microelectrode (IME) impedance sensor for the detection of viable Salmonella typhimurium. Biosens. Bioelectron. 19:1139-1147. 225. Zall, R. R., J. H. Chen, and S. C. Murphy. 1982. Estimating the number of psychrotrophs in milk using the direct microscopic method. Cult. Dairy Prod. J. 17:24-26, 28.

| 189

|

CHAPTER 14

|

Thermoduric Microorganisms and Heat-Resistance Measurements Tim C. Jackson

14.1

INTRODUCTION

Some non-spore forming bacteria exhibit higher-thanexpected thermal resistance given the physical or biochemical properties of their genera. These organisms are described as thermoduric, that is, having the property of thermotolerance and hence a capacity to survive some pasteurization processes.37 Thermoduric bacteria are capable of surviving heating in a food substrate at a range of 60 to 80uC. Such characteristics also apply to spore-forming bacteria, such as Bacillus and Clostridium spp. This chapter outlines methods for the quantification and determination of heat resistance of thermoduric non-spore-forming and spore-forming bacteria associated with dairy, egg, and meat products. Genera or groups of bacteria reported to contain thermoduric strains are listed in Table 14-1. In addition to the genera listed, heat-resistant coliforms such as Enterobacter aerogenes have been reported to survive pasteurization.61 Most thermoduric organisms grow in the mesophilic range (15–37uC); however, some strains are psychrotrophic. When present from post-process contamination, Gramnegative psychrotrophs will generally outgrow thermoduric psychrotrophs at refrigeration temperatures. However, if Gram-negative bacteria are absent, spoilage may result from the outgrowth of thermoduric psychrotrophs that survived the pasteurization process.27,38,72,75 Studies of thermoduric microorganisms in milk and egg products have concluded that the capacity for surviving pasteurization is not solely dependent upon the properties of the microorganism but may also be influenced by factors such as initial microbial load, product composition, age of the product, and heat treatment method. Studies on market milk supplies in India43,44,62 indicated the importance of the Enterococcus spp. as thermoduric organisms (Table 14-2). A study of raw and commercially pasteurized milk in Japan indicated that Bacillus spp., Microbacterium spp., and Micrococcus spp. were the dominant thermoduric genera in both raw and pasteurized milk, confirming that the thermoduric organisms present in pasteurized milk originated from the raw milk prior to

pasteurization42 (Table 14-3). Other coryneform bacteria, Streptococcus, Lactobacillus, and Actinomycetes were also isolated.42 Thermoduric Streptococcus, Microbacterium, and other coryneform bacteria in milk supplies may originate from milking and creamery equipment.42,70 Spore formers and Micrococcus are often associated with soil, fodder, or hay. Such organisms enter the raw milk during improper handling and cleaning of processing equipment.64 The level of thermophilic microorganisms in milk is generally an indicator of the hygiene of production practices.25 The level of survival of many of these organisms after heat treatment is usually quite low. Survival estimates have been determined in milk for cheese making following laboratory pasteurization at 63uC for 2 min33 (Table 14-4). Reported differences of survival data published in the literature for some thermodurics after heat treatment may simply be the result of variations in resistance or levels in the initial populations. Some genera may be better equipped to survive thermal treatment. Shafi et al.66 demonstrated that the types of organisms surviving the pasteurization of liquid egg were similar to those surviving in milk products, with the exception of finding Staphylococcus as part of the heat-resistant flora. Furthermore, genera like Bacillus and Micrococcus appeared to survive pasteurization regardless of the type of egg product or heat treatment. Payne et al.58 reported that many of the organisms that survived whole-egg pasteurization were of the coryneform group, including Microbacterium lacticum. Two strains of coryneform bacteria survived heat treatments of 20 and 38 min at 80uC in phosphate buffer (pH 7.1). A further characteristic of these thermoduric organisms was that none of the isolates studied grew at 5uC, but all were capable of growing at 10uC. Freezing coryneforms in liquid egg at –18uC before heating had little effect on heat resistance or viability. Foegeding and Stanley24 examined microorganisms surviving ultra-pasteurization of liquid whole egg with subsequent growth at 4 or 10uC. The most heat-resistant isolates recovered were Enterococcus faecalis and Bacillus circulans.

| 191 |

Compendium of Methods for the Microbiological Examination of Foods |

Table 14-3. Incidence and Significance of Thermoduric Bacteriaa in Farm Milk Supplies and Commercial Pasteurized Milk in Japan42

Table 14-1. Bacterial Genera or Groups Reported to Contain Food-Associated Thermoduric Organisms Genera/Group

Associated Foods

Reference

Actinomycetes Alcaligenes Arthrobacter Bacillus Clostridium Coryneform bacteria Lactobacillus

Milk Milk Milk Milk/eggs Milk Eggs Milk/meats/ juices Milk Milk/eggs Eggs Milk/meat/eggs

42 42 42,75 16,24,38,67,70 50 58 29,39,64

Microbacterium Micrococcus Pseudomonas Streptococcus/ Enterococcus

42 24,66,76 24,66 24,43,44,58,64

In meat products, some lactobacilli and enterococci may be recovered after a pasteurization process. Thermoduric E. fecalis and E. faecium have been associated with the spoilage of canned hams and other cured meat products.26 Although not included among the typical thermoduric bacteria, some strains of pathogenic non-spore-formers, such as Listeria monocytogenes and Salmonella may exhibit relatively high heat resistance in certain conditions.3,4,6,24 When using pasteurization processes developed for foods, it is important to consider the initial levels and relative heat resistance of these organisms in the product matrix.2,4,7,45,65 The composition of a foodstuff has been reported to influence the heat resistance of thermoduric spoilage organisms. Ingredients such as pectins have been linked to the survival of Lactobacillus fermentum in tomato juice after heat treatment (55–60uC).39 Factors such as water activity and pH will also influence microorganism survival. Raw milk contains proteinases from both indigenous and bacterial sources that may be involved in gelation or proteolytic activity in ultra-high-temperature (UHT) pasteurized milk. Many of these proteinases are heat resistant or may regenerate activity during storage.1 Spoilage resulting from heat-resistant enzymes may be incorrectly interpreted as having been caused by surviving thermoduric organisms.

Genera/Group

Raw Milk

Commercially Pasteurized Milk

Bacillus Microbacterium Micrococcaceae

30.7% 28.0% 17.4%

33.4% 33.5% 23.4%

a

Isolated, but in lower proportions were other coryneform bacteria, Streptococcus, Lactobacillus, and Actinomyetes.

14.2

HEAT-RESISTANCE MEASUREMENTS

The type of heat-resistance measurement used in an investigation will be determined by the desired use of the results. Where information is needed on the microbiological quality of a foodstuff following pasteurization, a thermoduric count may be appropriate. For this measurement, a food substrate is heated at temperatures and conditions similar to those encountered during processing, and the numbers of surviving organisms by direct plating onto a recovery medium are counted (see Section 14.443). Information may be required on the heat resistance of a specific organism in a specific substrate, (e.g., Listeria monocytogenes in 1% fat milk) in order to determine an appropriate thermal process to apply in commercial production. In such circumstances, a uniform parameter is needed that would allow a comparison of heat resistance between organism or heating menstruum and would enable an estimation of the thermal process necessary to inactivate or reduce a target population. The measurement most often used is DT, the decimal reduction time (also known as D value). DT represents the time required at a specific temperature (T) for a 10-fold, or 90%, reduction of the surviving microbial population in a given menstruum (Table 14-5). DT may be influenced by the intrinsic characteristics of a foodstuff or by the characteristics of the organism itself, including strain, growth phase, temperature, and exposure to sublethal stresses prior to heating. DT may be obtained from quantitative or qualitative data. Quantitative data from successive sampling experiments, in which surviving microbial populations are enumerated at regular intervals during heating, may be used to establish a semi-logarithmic survival curve (Figure 14-1). DT is determined from the linear portion of the survival curve and is calculated as the absolute

Table 14-2. Population Characteristics of Thermoduric Isolates From Market Milk43,44 Organism

% Population

Enterococcus faecalis Enterococcus bovis Enterococcus faecium Streptococcus thermophilus Micrococcus luteus Corynebacteria Microbacterium Bacillus

— 53 — — 11 — 7 29

192 |

Table 14-4. Survival of Thermoduric Organisms in Milk Heated at 63uC for 30 min33 Organism

% Survival Following Pasteurization

Microbacterium Micrococcus Alcaligenes Streptococcus/Enterococcus Lactobacillus Coryneforms

100 1–10 1–10 ,1 ,1 ,1

| Thermoduric Microorganisms and Heat-Resistance Measurements

Table 14-5. Comparison of D Values (DT) Between S. aureus and Two Thermoduric Organisms11 Organism

Temperature of Heating (uC)a

D-Value (min)

Staphylococcus aureus Enterococcus faecalis Streptococcus thermophilus

60 58 58

1.0 3.9 4.2

a

In phosphate buffer (pH 7.0)

reciprocal of the slope of the survival curve, where the slope 5 D number of survivors/D time of heating.58,67 Qualitative data may be obtained as positive or negative growth in enrichment broths following heating. For such data, DT is calculated by determination of the 50% endpoint; that is, the time of heating at a constant temperature where 50% of the samples are positive for growth.73 The holding times for such experiments should be established to bracket the end-point of survival, yielding from allpositive to all-negative results. At least three replicate tests would be performed at each holding time, with a replicate number chosen based upon the precision desired for the estimate. The application of the Spearman-Karber estimation to determine DT from fraction-negative data (such as 50% endpoint, tm) has been demonstrated by Lewis47 and Pflug et al.60 Correcting for the bias in the procedure, DT can be computed as follows47 (assuming the detection method recovers a viable microorganism when present):

DT ~tm =ðlog10 N0 z0:251Þ tm equals the 50% time estimated from the SpearmanKarber procedure, that is, the time when 50% of the analytical units are positive; N0 equals the concentration of bacteria measured by plate count at heating time zero. A challenge to the calculation of DT from quantitative (successive) survivor data is that its application assumes that survivor curves will be linear, following a logarithmic order of death. While survivor curves are frequently linear, they may also be concave, convex, or sigmoidal. They may incorporate initial shoulders or declines or demonstrate tailing in addition to a logarithmic component.31,51,59 Several strategies have been utilized to interpret nonlinear survivor data. Many researchers have applied linear regression analyses to the most linear portions of survivor curves and have excluded initial declines or shoulders. The use of an intercept ratio (IR) or intercept index (II) has been suggested to account for initial shoulders or declines on survivor curves.59,60 Multiple regressions have been used to analyze biphasic survivor curves,36,57 and non-linear regression models have been applied to more complex curves.4,35,48,60 Several authors have suggested that the use of thermal death point measurements, or F values (FT), are a more appropriate tool than D values in situations where survivor curves are non-logarithmic.9,14,49,52 F values are a measure of thermal death time (TDT), the time necessary at a specific temperature to inactivate a microbial population in a specific menstruum. As an example, for an initial population of 5.0 log10 cfu/mL, FT 5 5 6 DT. In practice, FT

Figure 14-1. Log10 survivor curve and determination of DT.

| 193

Compendium of Methods for the Microbiological Examination of Foods |

is most commonly used as an indicator of process lethality in commercial sterilization systems. Another parameter, the z value, is an indication of the change in temperature required to change the decimal reduction time by a factor of 10. It is relatively constant for a given organism under various test conditions.73 The z value may be used to determine equivalent thermal processes at a range of temperatures. Where DT is known for an organism at three temperatures, such data may be plotted as a TDT curve (log10DT 6 temperature, Figure 14-2). The z value may be calculated as the absolute reciprocal of the linear portion of the slope of this curve. Where DT is known at two temperatures (T1, T2), z may be calculated as follows60: z~ðT2 {T1 Þ=ðlog10 D1 {log10 D2 Þ When z for an organism is known, an established DT (at T1) may be used to calculate DT at a different temperature (T2) using the following equation69: log10 D2 ~ðT1 {T2 Þ=zzlog10 D1

14.3

METHODS FOR THE DETERMINATION OF HEAT RESISTANCE

The various methods for determining wet heat destruction rates for microorganisms have been categorized by Pflug and Holcomb60 as successive or multiple-replicate-unit sampling systems. In successive sampling systems, a small volume of a microbial suspension is inoculated into a larger volume of substrate that has been preheated to a specified heating temperature. The suspension is continuously agitated during heating. At established heating intervals, an aliquot of inoculated substrate is aseptically removed by

pipette and dispensed into a sterile tube immersed in ice water for cooling. Because the substrate is already at the specified heating temperature, these systems avoid the need to correct for microbial inactivation during come-up time. Successive sampling systems may use a flask method, heating the substrate in a flask,32,34 canning jar,57 or multiple neck container.46,63 A tank method has been described for the evaluation of temperatures above 100uC.78 In multiple-replicate-unit sampling systems, multiple sample units are prepared and heated concurrently. Heated units are removed at successive time intervals60 and immediately exposed to ice water for cooling. Survivors may be evaluated by the presence or absence of growth or by the enumeration of survivors and the establishment of a survival curve. Sealed glass tubes, vials, ampoules, and capillary tubes have been used in these sampling schemes.5,17,21,28,30,40,53,60 The low-temperature long-time (LTLT) and immersed seal tube (IST) methods described below are multiple-replicate unit sampling systems.

14.4 14.41

Preparation and Handling of Cultures

Where the survival or heat resistance of a specific microorganism is being evaluated, the preparation and handling of cultures should take into consideration factors influencing microbial heat resistance. Several reviewers have summarized the influence of environmental and physiological factors on the heat resistance of bacterial cells.31,60,73 Characteristics of the food substrate such as the presence of nutrients, fat content, water activity, and pH, as well as characteristics of the organism such as strain, growth phase, and age of culture have been demonstrated to influence heat resistance. Relative to culture conditions, incubation and storage temperature before heating may influence a microorganism’s

Figure 14-2. TDT curve and determination of z. TDT 5 thermal death time.

194 |

EQUIPMENT, MATERIALS, AND PROCEDURES

| Thermoduric Microorganisms and Heat-Resistance Measurements

heat resistance. For example, bacterial populations demonstrate greater heat resistance in the stationary rather than the logarithmic phase.12,20,36,60,72,74,77 Greater heat resistance has been reported for vegetative bacteria and spores of cultures grown at higher temperatures20,36,60 Cultures held under refrigeration may exhibit greater heat sensitivity than those stored at higher temperatures.12,34,36,41 Because many factors inherent in a food system may influence the ability of an organism to survive heating, it is often desirable to equilibrate an inoculum to the characteristics of a menstruum before heating. When possible, the inoculated menstruum should be equilibrated to a temperature similar to that encountered in commercial production before undergoing thermal processing. In an evaluation of the heat resistance of Listeria monocytogenes during high-temperature short-time (HTST) pasteurization, Farber et al.22 re-suspended a prepared inoculum pellet into 5 mL sterile whole milk that was then used to inoculate a larger volume of sterile whole milk (5 L) to evaluate the process. The inoculated milk was stored at 4uC overnight before pasteurization to equilibrate the culture to the substrate and to simulate commercial holding practices. A rich, non-selective recovery medium should be used in thermal process studies because selective media may not allow the growth of i