<%BANNER%>

Microbial Composition, Biofilm Formation, and Removal from the Surfaces of the Manway Lid Gaskets of Citrus and Dairy Li...


PAGE 1

MICROBIAL COMPOSITION, BIOFIL M FORMATION, AND REMOVAL FROM THE SURFACES OF THE MANWAY LID GASKETS OF CITRUS AND DAIRY LIQUID TRANSPORTATION TANKERS By MARJORIE RUTH RICHARDS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Marjorie Ruth Richards

PAGE 3

This thesis is dedicated to my family, friends, colleagues, teachers, and committee members for their help and their support.

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank my pare nts (my editors) for their co ntinued love, support, and wisdom. Their undying support a nd belief in my abilities have made the last few years possible. I would like to thank my sib lingsCindy, Heather, Daniel, Jessica, and Benjaminfor being my best friends, my gr eatest supporters, and my toughest critics. I would like to thank USA Tank Wash, Bynu m Transport, Oakley Transport, Inc. and Clewiston Tank and Truck Wash for thei r help and support in conducting this research project. I would also like to thank th e members of the committee. I thank original advisor Dr. Parish who allowed me to participate in this project, largely allowed me the freedom to design and create the sec ond part of this experiment, and for giving me guidance and expanding my love of food microbiology. I th ank my advisor Dr. Goodrich for assisting me after Dr. Parish left for the University of Maryland, in preparations of the thesis, my defense, and future career plans. I thank my co advisor Dr. Archer for his guidance in selecting courses, and for his good-humored atti tude. I thank Dr. Wright for her advice on molecular techniques and for allo wing me to participate in the Salmonella /lake water project to increase my knowledge of molecular techniques. And last but not least I thank Dr. Welt and all the members of his lab for designing the model tanker system.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES...........................................................................................................xi ABSTRACT......................................................................................................................x ii CHAPTER 1 INTRODUCTION........................................................................................................1 Justification for Research.............................................................................................1 Previous Research.........................................................................................................3 Specific Aims and Objectives.......................................................................................4 Part I: Identification and Characteri zation of Microorganisms in Samples..........4 Objectives....................................................................4 Hypotheses.....................................................................................................4 Part II: Biofilm Development and Removal..........................................................5 Objectives.......................................................................................................5 Hypothesis...................................................................5 2 REVIEW OF THE LITERATURE..............................................................................6 The Legal History.........................................................................................................9 Marketing and Food Safety Justification for Citrus Juice Transportation Tanker Research.................................................................................................................17 Previous Research on Transportation Tankers...........................................................19 Citrus Juice and Milk and Th eir Microbial Inhabitants..............................................23 The Environment of Citrus Juice.........................................................................23 Microbial Flora of Citrus Juice............................................................................24 Bacteria.........................................................................................................24 Yeasts...........................................................................................................27 Molds............................................................................................................28 The Environment of Liquid Dairy Products........................................................28 Biofilms......................................................................................................................31 Attachment..........................................................................................................32 Material........................................................................................................33

PAGE 6

vi Cellular components.....................................................................................33 Characteristics of the liquid media...............................................................34 Formation............................................................................................................35 Extracellular polymeric substances (EPS)...................................................36 Architecture..................................................................................................37 Other bacteria and particles..........................................................................37 Maturation...........................................................................................................38 Other bacteria...............................................................................................38 Gene transfer and regulation........................................................................39 Quorum sensing............................................................................................39 Pathogenic organisms...................................................................................42 Resistance.....................................................................................................44 Dispersal..............................................................................................................45 Detergents and Sanitizers...........................................................................................45 Detergents............................................................................................................45 Sanitizers.............................................................................................................47 The Environment of Stainless Steel............................................................................48 Environment of Rubber..............................................................................................52 Review of Methodology.............................................................................................54 Coliforms, Fecal Coliforms, and E coli ...............................................................54 Detection Methods for Salmonella ......................................................................56 Detection Methods for Alicyclobacillus ..............................................................57 Detection of Aciduric, Yeast and Mold, Thermoduric, Mesophilic and Psychroduric Microorganisms.........................................................................57 DNA Sequencing.................................................................................................58 Biofilm Growth Characterization........................................................................59 Observation Methods...........................................................................................59 A Need for More Research.........................................................................................61 3 MATERIALS AND METHODS...............................................................................62 Part I: Identification and Characteri zation of Microorganisms in Samples...............62 Sample Collection...............................................................................................62 Sample Preparation..............................................................................................64 Sample Analysis..................................................................................................64 Psychroduric, mesophilic, thermoduric, yeast and mold; and aciduric enumeration and characterization...........................................................64 Coliform, fecal coliform, and E. coli detection............................................66 Streptococcus spp. and Staphylococcus spp. detection................................67 Salmonella spp. detection.............................................................................67 Alicyclobacillus spp. detection.....................................................................67 16S DNA and 28S rRNA PCR Identification.....................................................68 Statistical Analysis..............................................................................................68 Part II: Biofilm Development and Removal...............................................................69 Liquid Sample Preparation..................................................................................69 Standard growth curves................................................................................69 Model of Liquid Transportation Tanker Manway...............................................71

PAGE 7

vii Gasket Treatment.................................................................................................73 Microbial Analysis of Gasket..............................................................................74 Scanning Electron Microscopy............................................................................75 Fluorescence Microscopy....................................................................................75 Statistical Analysis..............................................................................................76 4 RESULTS...................................................................................................................77 Part I: Sample Identification and Characterization.....................................................77 Psychroduric, Mesophilic, Thermoduri c, Yeast and Mold, and Aciduric Microorganism Enumeration and Characterization.........................................77 Coliform, Fecal Coliform, and E. coli Detection................................................87 Streptococcus and Staphylococcus Detection.....................................................87 Salmonella and Alicyclobacillus Detection.........................................................88 16S DNA and 28S rRNA PCR Identification.....................................................88 Part II: Biofilm Development and Removal...............................................................88 Gasket Analysis...................................................................................................88 Scanning Electron Microscopy............................................................................90 Fluorescence Microscopy....................................................................................93 5 DISCUSSION AND CONCLUSIONS......................................................................95 Part I: Sample Identification and Characterization.....................................................95 Psychroduric, Mesophilic, Thermoduri c, Yeast and Mold; and Aciduric Enumeration and Characterization...................................................................95 Coliform, Fecal Coliform, and E. coli Detection................................................98 Streptococcus and Staphylococcus Detection.....................................................98 Salmonella and Alicyclobacillus Detection............99 Part II: Biofilm Development and Removal.............................................................100 Gasket Analysis.................................................................................................100 Scanning Electron Microscopy..........................................................................104 Fluorescence Microscopy..................................................................................104 Overall Conclusions..................................................................................................105 6 FUTURE WORK......................................................................................................106 Extension..................................................................................................................106 Gasket Washing Video a nd/or Training Manual...............................................106 Workshops.........................................................................................................107 Tank Wash Association.....................................................................................107 Research....................................................................................................................107 Juice Concentrate Research...............................................................................107 Biofilm Research...............................................................................................108 Gasket Alternatives...........................................................................................108

PAGE 8

viii APPENDIX A PART I RAW DATA...............................................................................................110 B PART II RAW DATA..............................................................................................114 C GASKET SURFACE AREA SAMPLE CALCULATION.....................................116 D STANDARD GROWTH CURVES.........................................................................117 E STATISTICAL TABLES I.......................................................................................120 F STATISTICAL TABLES II.....................................................................................124 LIST OF REFERENCES.................................................................................................128 BIOGRAPHICAL SKETCH...........................................................................................142

PAGE 9

ix LIST OF TABLES Table page 2-1 Products and cleaning steps for JPA wash types........................................................7 3-1 Types and number of CF U of microorganisms found in target inoculated milk.....70 4-1 Products effect on aciduric, yeast and mold, psychroduric, and mesophile...........78 4-2 Products effect on aciduric, yeast and mold, psychroduric, and mesophile...........78 4-3 Gaskets effect on aciduric, yeas t and mold, psychroduric, and mesophile.............79 4-4 Gaskets effect on aciduric, yeas t and mold, psychroduric, and mesophile.............79 4-5 Wash temperatures effect on ac iduric, yeast and mold, psychroduric....................79 4-6 Wash temperatures effect on ac iduric, yeast and mold, psychroduric...................80 4-7 Product and Gaskets effects on ac iduric, yeast and mold, psychroduric...............81 4-8 Product and Gaskets effects on ac iduric, yeast and mold, psychroduric...............81 4-9 Product and Wash Temperatures effects on aciduric, yeast and mold...................81 4-10 Product and Wash Temperatures effects on aciduric, yeast and mold...................82 4-11 Gasket and Wash Temperatures effects on aciduric, yeast and mold....................82 4-12 Gasket and Wash Temperatures effects on aciduric, yeast and mold....................83 4-13 Product, Gasket, and Wash Temperatur es effects on aciduric, yeast and mold....84 4-14 Product, Gasket, and Wash Temperatur es effects on aciduric, yeast and mold....85 4-15 The top two bacterial characterizatio ns on different gasket and media types.........86 4-16 Number of coliform, fecal coliform, and E. coli positive gaskets..........................87 4-17 Percentage of Staphylococcus spp. and presumptive Staphylococcus aureus ........88 4-18 Results of 16S DNA and 28S rRNA PCR Identification........................................89

PAGE 10

x 4-19 Log10 reductions among coliform and me sophilic counts for the three wash........89 A-1 Sample type, sample number, gasket washer, aciduric and yeast and mold.........110 A-2 Sample type, sample number, psyc hroduric, mesophilic, and thermoduric..........112 B-1 Sample Letter and results for Part II.....................................................................114

PAGE 11

xi LIST OF FIGURES Figure page 3-1 Type A and Type B manway styles and gasket types.............................................63 3-2 Manway lid set up picture set 1...............................................................................71 3-3 Manway lid set up picture set 2...............................................................................72 3-4 Manway lid set up picture set 3...............................................................................74 4-1 Representative pictures from scanni ng electron microscopy..................................91 4-2 Representative pictures from fluorescent microscopy............................................94 5-1 An example of slow draining bucket.....................................................................101 E-1 Surface area (cm2) of test type vs. product............................................................120 E-2 Surface area (cm2) of test type vs. gasket type......................................................120 E-3 Surface area (cm2) of test type vs. wash type........................................................121 E-4 Surface area (cm2) of test type vs. product, and gasket type.................................121 E-5 Surface area (cm2) of test type vs. product, and wash type...................................122 E-6 Surface area (cm2) of test type vs. gask et type, and wash type.............................122 E-7 Surface area (cm2) of test type vs. product, gasket type, and wash type...............123

PAGE 12

xii Abstract Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science MICROBIAL COMPOSITION, BIOFIL M FORMATION, AND REMOVAL FROM THE SURFACES OF THE MANWAY LID GASKETS OF CITRUS AND DAIRY LIQUID TRANSPORTATION TANKERS By Marjorie R. Richards December 2005 Chair: Rene M. Goodrich Cochair: Douglas L. Archer Major Department: Food Science and Human Nutrition Improved guidelines pertaining to the sa nitation of liquid food transportation tankers are needed to ensure safety and maxi mum shelf life of liquid products sold in the United States. Studies on ATP-bioluminescen ce conducted by Bell and others determined that by sampling the manway lid one could dete rmine if a tanker was dirty or clean. The ATP-bioluminescence study done Pez and others determined that surfaces of the manway lid were the hardest to clean. Da ta collected by Winniczuk and Parish showed that manway lids, specifically gaskets, were the hardest to clean region of the tanker. Rubber gaskets were most likely contaminated with coliform, fecal coliform, and Escherichia coli after cleaning. Therefore, resear ch to better understand microbial activity on manway lid gaskets was deemed necessary. The first part of this research was to characterize distribution of aciduric, thermoduric, mesophilic, psychroduric micr oorganisms, yeast and mold; determine

PAGE 13

xiii coliform and fecal coliform counts; as well as identify the Escherichia coli Salmonella spp., and Alicyclobacillus spp. on surfaces of gasket type A and gasket type B of citrus and dairy tankers after warm or ambient wa sh. Four important results were obtained from this portion of the study: 1) Surface shapes of gaskets are important with respect to cleanability, 2) In cases where significan t differences (P<0.05) exist the Log10/cm2 values of the aciduric, and mesophilic, and Log10 value of the aciduric and mesophilic organisms/total gasket were significantly more for gasket type B from cold washed dairy tanker gaskets than any other gasket type, 3) Salmonella spp. and Alicyclobacillus spp. were not found on any gaskets, 4) E. coli was found on the surface of dairy, type B gaskets. The second part of this research utilized ultra-high temperat ure pasteurized milk inoculated with E. coli other bacteria, and a yeast collected from gaskets; and a model tanker manway to form biofilms on lid gask ets. Effectiveness of three commercial cleaning regimens (detergent wash/water rins e; detergent wash/sanitizer/water rinse; detergent wash/sanitizer/water rinse/hot wate r treatment) on lid gaskets were evaluated using coliform, mesophilic, and yeast counts; E. coli most probable number; and scanning electron and fluorescence microsc opy. Results showed that the detergent wash/sanitizer/water rinse/hot water treatment is more effective than the other two at removing both coliform and mesophilic microorganisms from the gaskets surfaces.

PAGE 14

1 CHAPTER 1 INTRODUCTION Justification for Research There is a problem with food transportation in the United States, which arises from the Sanitary Food Transportation Act (SFTA) of 1990. This act regulates the safety of products transported in motor and rail vehicles; however, Department of Transportation (DOT) is responsible for implementati on and enforcement of SFTA (49 USC 57011514). The Office of the Inspector General has determined that DOT has neither the means nor time to enforce SFTA (Office of the Inspector General 1998). Lack of proper enforcement may have played a role in thr ee recent food transporta tion incidents. In 1994, over 224,000 people were affected by salm onellosis when a tanker truck that had carried unpasteurized liquid eggs was not pr operly cleaned, causing cross-contamination of salmonella to pasteurized ice cream mix that was subsequently transported in the same tanker (Office of the Inspector General 1998) Another problem occurred in 1999 when ice allegedly contaminated with Salmonella was illegally added to orange juice being shipped from Mexico to Arizona (FDA/CF SAN 2001). In 1997, several decomposing bodies were found in three ships entering the U.S. (Office of th e Inspector General 1998). The food transportation industry relies upon voluntary compliance to guidelines, such as Bulk Over-the-Road Food Tanker Transpor t Safety and Security Guidelines (Food Industry Transportation Coalition 2003), and the Juice Products Association Model Tanker Wash Guidelines for the Fruit Juice Industry (2004). Voluntary guidelines vary in quality, and compliance is inconsiste nt. Some of the problems with liquid food

PAGE 15

2 transportation safety occur when transportati on tankers are cleaned. Tankers go to wash stations, but often remain unwashed until just pr ior to the next use. Time periods that they are allowed to be unwashed is not regu lated, so tankers may sit uncleaned for two weeks or more before washing, particularly if transport business is slow. While tankers sit uncleaned insects can infest them, and b acteria, mold, and yeast have the opportunity to multiply and form communities called biofilms. When non-dairy tankers are finally ready to be used, they are washed based upon non-standard washing criteria specific to each particular wash station. Such criteria are not always based on sound scientific data or government standards. Cleaning liquids are applied using a spray washing system that is lowered through the tanker manway. Spray mechanisms are often selected based on cost; these systems may not always provide the best cleaning. Spray systems may be inefficient because sprayed solution may not reach the ends of tankers or because holes in sprayers can and do become plugged. Other pr oblems include cleaning trucks on a slope, which creates an area in the tanker that cannot be properly cleaned, improper or nonexistent manual cleaning of accessory parts su ch as gaskets, and not cleaning filth and debris accumulating around seals during transp ortation. Bell and othe rs (1994) noted that poor water quality, concentration of chemi cals used, and operational temperature could reduce cleaning effectiveness. Also, failure to empty tankers of all wash water can cause recontamination and increase likelihood of microbiological problems (Bell and others 1994). Other problems are evident in the security of wash st ations; some wash stations store dirty and empty tankers in lots without security fences, leaving the tankers open to vandalism, bioterrorism, or accidents.

PAGE 16

3 The U.S. and the State of Florida have reason for concern for the tanker sanitation situation. In 2003/2004 the US proce ssed 37,048,000 metric tons, imported 20,005,000 metric tons and exported 19,955,000 metric to ns of orange juice (USDAs Foreign Agricultural Service (FAS): Production, 2004). Most of the ci trus in the United States comes from Florida (Kader, 2002). In Florida, the $9-billion-per-annum citrus industry is second only to the tourism industry. Citr us generates approximately 90,000 jobs. Ninety-five percent of Flor idas oranges are processed into juice (USDAs Foreign Agricultural Service (FAS): Horticu ltural, 2004). Both concentrates and pasteurized single strength citrus juices are transported in transportati on tankers. There is growing concern about the quality and safety of pr oducts transported in tankers because of bacterial, yeast, and mold formation in tanke rs. Tankers may not be adequately cleaned between loads to prevent cross-contaminati on. Both tanker wash station operators and the citrus industry would lik e to take steps to minimize spoilage and control food pathogens in citrus juice and concentrates. A decrease in qua lity of Florida citrus juice encourages consumers to purchase citrus ju ice from other sources, resulting in economic losses that will eventually affect wash fac ilities. A foodborne outbreak in citrus juice from Florida caused by the inappropriate wash ing could financially ruin a tanker truck wash station as well as reduce consumer trus t in Florida-grown citrus products. Previous Research In 2004, 87 liquid transportation tankers we re sampled for the USDA tanker study (Winniczuk and Parish, unpublished data). Fo rty-eight tankers haul ed a citrus juice product and 26 tankers hauled a dairy produc t immediately prior to the study. The tankers were tested for the presence of Salmonella spp., E. coli and Listeria spp. at different locations within the tanker. Data indicated that citr us and dairy tankers are

PAGE 17

4 equally likely to be contaminated with E. coli although E. coli contamination from the product itself is more likely to come from a da iry product than from a citrus product. The part of tankers most often contaminated wa s manway lid gaskets (Winniczuk and Parish, unpublished data). Specific Aims and Objectives Part I: Identification and Characteri zation of Microorganisms in Samples The overall objectives of Part 1 of this research were to Objectives Sample at least 72 tankers such that ther e was nine of each type created from the eight possible combinations that coul d be created from dairy vs. citrus juice/concentrates, gasket t ype A vs. gasket type B, a nd hot vs. cold temperature wash. Monitor gaskets for coliforms and E. coli Alicyclobacillus and Salmonella spp., and psychroduric; mesoph ilic; thermoduric; yeast and mold; and aciduric microorganism counts. Test for Staphylococcus spp., Streptococcus spp., and/or additional tests for coliforms, fecal coliforms, or E. coli if appropriate. Select, streak for isolation, and characterize using gram staining, the catalase test and the oxidase test on representative co lonies from the psychroduric; mesophilic; thermoduric; yeast and mold; and aciduric enumeration. Determine using the Analysis of Varian ce (ANOVA) and Fishers Least Significant difference, if there was a significant di fference between psychroduric; mesophilic; yeast and mold; and aciduric microorganism counts based on the differences between product, gasket, and wash temperature variables. Hypotheses The number of microflora and the freque ncy of detection of coliforms, fecal coliforms, E. coli Salmonella spp., and Alicyclobacillus spp. on gaskets of cleaned tankers will be different between tankers th at have previously hauled dairy products and juice products. The number of microflora and the freque ncy of detection of coliforms, fecal coliforms, E. coli Salmonella spp., and Alicyclobacillus spp. on gaskets of cleaned juice and dairy tankers will be different between tankers that have Type A, and Type B rubber gaskets.

PAGE 18

5 The number of microflora and the freque ncy of detection of coliforms, fecal coliforms, E. coli Salmonella spp., and Alicyclobacillus spp. on gaskets of cleaned juice and dairy tankers will be different between tankers that have received and have not received a hot water spray. Part II: Biofilm Development and Removal Objectives Develop a manway lid model to simulate splashing of liqui d products onto the manway gasket. Create a cocktail of bacteria and yeast obtained from the first part of the study whose identities of these microorganisms were determined by 16S DNA sequencing and 28S rRNA (D2 expansion segment) sequencing. Develop methods three different methods to mimic treatments to mimic treatments in the tank wash industry. Observe control and three treatments under fluorescent and scanning electron microscopy. Swab the control and three treatments and de termined the presence and quantity of coliform and E. coli were present, and the mes ophilic, yeast, and coliform enumeration. Select, streak for isolation, and characterize using gram staining, the catalase test and the oxidase test representative on co lonies from the mesophilic and the yeast enumeration. Determine, using the Analysis of Variance (ANOVA) and Fishers Least Significant difference, if there was a si gnificant difference between mesophilic and coliform counts based on the differences between Log10 reductions wash types. Hypothesis The E. coli coliform, mesophilic, and yeast c ounts on the gasket from the model system will differ significantly from gasket microflora from the uncleaned sample and the three cleaning regimes.

PAGE 19

6 CHAPTER 2 REVIEW OF THE LITERATURE There is evidence of a microbial problem with liquid food-grade transportation tankers. Transportation tankers in the United States haul such products as pasteurized and unpasteurized milk, single strength or concentrated jui ces, pasteurized and unpasteurized liquid eggs, molasses, liquid yeast, liquefied pork lard, canola oil, citrusol (citric acid), oil (sunflower, vegetable, canola, coconut and cotton seed), olestra, burned syrup, sucrose, vinegar, brown sugar slu rry, cola and other bases for carbonated beverages (Different Products), honey, wate r, corn syrup, maple syrup, peanut butter base, artificial and natural colors and fla vors, pasteurized ice cream premix, soymilk, catsup, and alcohol. After deli vering food products, tankers go to wash stations for cleaning. At the tank wash stations tankers may sit unwashed for hours or days before cleaning. The reason for this pr actice is that wash stations are required to use washed tankers within 48 hours of washing. Also, it can be as expensive as $140 to wash a tanker, which does not make it financially advant ageous to wash a tanker more than once. While tankers sit waiting to be washed, inse cts may infest tankers and molds, bacteria and yeasts can grow, potentially causing food safety and spoilage issues, and forming communities called biofilms. When tankers are ready to be used, those with non-dairy liquids are washed based upon non-standard criterion specific to each wash station. A mode l of wash criteria is the Juice Products Association (JPA) (2004) tank wash types, which is illustrated in Table 1. This scheme provides a recommended set of gui delines for washing and cleaning tankers.

PAGE 20

7 Appropriate washing regimes are chosen ba sed upon product previously shipped in the tanker. Tankers that have carried citrus ju ice are cleaned using either wash type 1 or wash type 2. According to the JPA guidelin es type 1 wash occurs between loads of the same product as long as no one contaminates th e truck in any manner, such as by entering it, and the tanker is never left for more than 12 hours without product in it. If the truck is left empty for more than 12 hours it should rece ive the appropriate wash for the product. Table 2-1. Products and cleaning step s for JPA wash types based on Model Tanker Wash Guidelines for the Fruit Juice Indu stry, 2004 (Juice Products Association 2004) 1 2 3 4 Same Product Water Based Products Water/Oil & Oil Based Products Product with Potential Allergenic Risks Potable water pre-rinse Potable water pre-rinse Potable water prerinse/degrease Potable water prerinse/degrease Inspect Inspect Inspect Warm water rinse (75110oF) Warm water rinse (75110oF) Warm water rinse (75110oF) Manually clean valve and vents Manually clean valve and vents Manually clean valve and vents Hot clean 160F, 15 min Hot degreasing 170 212F for > 15 min Hot degreasing 170 212F for > 15 min Potable water rinse (no spec) Warm water rinse Warm water rinse Inspect Hot clean 160F min, 15 min Hot clean 160F min, 15 min, Sanitize chemical or hot water (185F, 10 min) Potable water rinse (no spec) Hot water rinse 185F for > 20 min Cool down (if hot water used) Inspect Inspect Sanitize chemical or hot water (185F, 10 min) Sanitize chemical or hot water (185F, 10 min) Cool down (if hot water used) Cool down (if hot water used) To maintain the quality of the product, JPA recommends that trucks hauling concentrated juice consecutively should have a Type 2 wash every 7 days; single strength juice, every 72 hours; and unpasteurized juice, every 24 hours (Juice Products Association, 2004). Liquids used to clean ta nker trucks are applie d using one of three

PAGE 21

8 types of spray washing systems (fluid-dri ven, motor-driven, or st ationary) that are lowered from the manway. Fluid-driven wash systems can be purchased in reactionary force, constant speed, and turbine models According to Spraying Systems Company (Wheaton, Illinois): Reactionary Force tank washing nozzles us e the force of the fluid to rotate the spray head. Constant Speed tank washing nozzles use the momentum of the liquid flow to drive the spray head while main taining constant rotating speed. Turbine tank washing nozzles utilize the fluid to spin a turbine, which in turn powers a gear set. Motor-Driven tank washers use high-pr essure solid stream nozzles at pressures from 100 to 1000 psi (7 to 70 bar) with a separate motor for driving the nozzle assembly. Two to four nozzles rotate on a gear hub as they revolve around the central axis of the nozzle assembly. Th e Fixed Spray or stationary tank washing nozzles include multi-nozzle spray assemblie s and individual fixed position spray nozzles. These models (fixed spray) can perform multi-tasks from cleaning tanker trucks to cleaning food-processing tanks. The advantages of these nozzles are: simplicity of design, reliability due to no moving parts, and a wide range of spray coverages. (2004) The problem with spray ball mechanisms is th at they are often chosen on cost and may not be the most appropriate system to clean tankers. Hence, the spray system may be ineffective because it cannot clean hard to re ach areas in the tanker or because holes in the sprayer are plugged. Other cleaning probl ems include cleaning trucks at awkward angles, improper or non-existent hand-cleaning of the accessory parts, and filth entering the tanker around the seals duri ng transportation. Bell et al. (1994) noted that poor water quality, chemical concentration, and operationa l temperature could re duce the ability to clean tankers. Also, failure to empty a ll wash waters from tankers could cause recontamination and increase the likelihood of microbiologica l problems (Bell and others 1994). Other problems are evident in the security of the wash stations ; wash stations will store dirty and empty tankers in lots without security fences, leaving the tankers open to vandalism, bioterrorism, or acciden ts involving curious people.

PAGE 22

9 The Legal History The first government action to correct pr oblems with the U.S. food transportation industry occurred in the late 1980s and early 1990s. In the late 1980s the media reported that chemicals and garbage were being shipped with food products (Office of the Inspector General 1998). The General Acc ounting Office reveal ed that trucking companies were not required to keep reco rds of these mixed loads (Office of the Inspector General 1998). Also, there was c oncern about backhauling, a process where food is delivered in the firs t load, chemicals and/or garb age are shipped in the second load and then food is shipped in the third lo ad (Office of the Inspect or General 1998). In response to these accusations Congress passed the Sanitary Food Transportation Act (SFTA) in 1990. In S ection 5701 Part 2 of the Sanitary Food Transportation Act (SFTA), Congress expres sed its concern for the American public: the United States public is threatened by the transpor tation of products potentially harmful to consumers in motor vehicles and ra il vehicles that are used to transport food and other consumer products. In Part 3 of SFTA, they conclude that the risks to consumers by those transportation practices ar e unnecessary and those practices must be ended. It was hoped that the SFTA would so lve these problems (49 United States Code 5701). SFTA of 1990 is found in 49 USC 5701 to 49 USC 5714. Section 5701 states the findings (discussed in the introduction) that make this act so important. Section 5702 defines terminology used in the act. Secti on 5709 states that it is mandatory for the Secretary of the Department of Transportati on to consult the Secretary of Agriculture, Secretary of Health and Human Services, a nd the Administrator of the Environmental Protection Agency on how to implement sections 5703-5708.

PAGE 23

10 Section 5703 is the general re gulation section. It states that no later than July 31, 1991 the Secretary of Transportatio n is required to prescribe regulations on transportation conditions that would make cosmetics, devi ces, drugs, food, and food additives unsafe for humans or animals. The secret ary is required to consider co smetics, devices, or drugs to be non-food products if they are transported in motor or rail vehicles before or at the same time as a food or food additive, and if they would make the food or food additive unsafe to humans or animals. Other special requirements this section makes of the Secretary are to establish r ecord keeping, identification, ma rking, certification, or other means of verification to comply with sect ions 5704-5706, decontamination, removal, disposal, and isolation to comply with carrying out sections 5704 and 5705 and to produce a list of materials for the constructi on of tank trucks, rail tank cars, cargo tanks, and accessory equipment that will comply with 5704 (49 USC 5702). Also it gives the Secretary the responsibility of considering and establishing the following: determining the extent that packaging can protect cosme tics, devices, drugs, food, and food additives to keep them safe for humans or animals ev en though they are being transported in motor and rail vehicles meant for nonfood products; finding the appropr iate compliance and enforcement measures to carry out this chapter; and establishing appropriate minimum insurance or other liability requirements for a person to whom this chapter applies (49 USC 5702). Lastly, if the Secretary deems that a type of packaging meets certain standards, the rules and regulations on th e transportation conditions for cosmetics, devices, drugs, food, and food additives unsafe for humans or animals may not apply. Section 5704 establishes rules for tank tr ucks, rail tank cars, and cargo tanks. It requires that the Secretary of Transportation publish in the Federal Register a list of

PAGE 24

11 nonfood products that would not make cosmetics, devices, drugs, food, and food additives transported before or with the product unsafe for humans or animals. It prohibits one from using, offering for use, or arranging for the use of a tank truck, rail tank car, or cargo tank to transport cosmetics, devices, drugs, food, and food additives if the vehicle has been used for an unapprove d nonfood product; or providing a vehicle for the purpose of transporting an unapprove d nonfood product when it is marked for cosmetics, devices, drugs, food, and food additives or a nonfood product on the approved list. Also this section requi res individuals arranging for a tank truck or a cargo truck to disclose what they will be shipping if it is or will be used in the preparation of a food additive or if it is listed an approved nonfood item. Section 5705 covers motor and rail tran sportation of nonfood products. This section requires the Secretar y to publish in the Federal Re gistrar a list of unsafe nonfood products. This list should not include food packaging such as cardboard, pallets, beverage containers unless th e Secretary of Transportation determines that transporting these packaging materials in a motor vehicle would make the packaging materials unsafe to humans or animals. It forbids using, offering for use, or arranging for the use of a tank truck, rail tank car, or cargo tank to tr ansport cosmetics, devices, drugs, food, and food additives if it has been used to transpor t nonfood materials listed in this section. Section 5706 forbids the use of a tank truc k to transport food and food additives if the vehicle has been dedicated to transport asbestos, refuse, or other dangerous products (a list of which the Secretary should publish in the Federal Register). Waivers of any part of this chapter are allowed under Section 5707 if the Secretary determines it will not make food and food additives unsafe for humans or animals, or it is in the public interest.

PAGE 25

12 Section 5708 allows state employees to inspect (with funding from the federal government) motor vehicles as long as the stat e agrees to comply with the appropriate federal regulations or compatible state regulati ons. This section also enlists the help of the Secretaries of Agriculture and Health and Human Servi ces; the Administrator of the Environmental Protection Agency, and the h eads of other appropriate departments, agencies, and instrumentalities of the United States Government to help carry out this chapter, including assistance in the training of personnel. Training for federal and state employees would be paid for by the federal government (49 USC 5708). Section 5710 establishes that the Secretar y of Transportation ha s the authority and responsibility to carry ou t this Act. In section 5711, the Secretary is directed to make a list of penalties and procedures and to take civil action against those who violate the regulations set up under the chapter or under sections 5123 and 5124. Section 5712 establishes the relationship be tween this chapter and Sec tion 5125 which is the chapter dealing with the transportation of hazardous material. Section 5713 establishes that sections 5711 and 5712 will only apply after July 31, 1991 (49 USC 5713). And lastly under section 5714 the Secretary after consulting with state officials, will establish procedures to promote more effective coor dination between the departments, agencies, instrumentalities of the United States Governme nt and State Authoritie s (49 USC 5714). After SFTA was enacted the Department of Transportation (DOT) turned over the responsibility of the Act to DOTs Resear ch and Special Programs Administration (RSPA). By July 31, 1991 no final regulati ons were issued under SFTA nor had any been issued by December 1997 when the Chairman of the Senate Committee on Commerce, Science and Transportation asked the Office of Inspector General (OIG) to

PAGE 26

13 investigate how well DOT and RSPA were doing fulfilling its obligation to SFTA (Office of the Inspector General, 1998). The OIG found that RSPA had issued a pr oposed rule in May 1993 to address the safe transportation of food products in highw ay and rail transpor tation but had not issued a final regulation. RSPA had prepar ed a training video for DOT inspectors on the hazards of transporting food but admitted that it was not an adequate safety training program as required by the Act (Office of th e Inspector General, 1998). DOT failed to develop lists of non-food products not uns afe; unsafe non-food products; waivers; and coordination procedures that they were supposed to establish under SFTA (Office of the Inspector General, 1998). However, this was a result of the way the law was written; the categories of non-food products not unsafe and unsafe non-food products were too broadly written and could incl ude every product (Office of the Inspector General, 1998). Therefore, when RSPA coul d not identify any speci fic items to place on the list, no waivers were needed. Meetings of the DOT Secretary with Secr etaries of Agricultu re, and Health and Human Services; and the Administrator of the Environmental Protection Agency on how to implement sections 5703-5708 did not occur until 1995. Meanwhile SFTA was failing its mission to protect the American public be cause improper storage and transportation of food occurred. Three major incidences occurred in 1994, 1997, and 1999. In 1994, 224,000 people were affected by salmonellosis wh en a tanker truck that had carried unpasteurized liquid eggs was not cleaned prope rly resulting in cross-contamination of Salmonella to the pasteurized ice cream mix that was transported in the tanker truck afterwards. In 1997 several decomposing bodies of stowaways were found in three ships

PAGE 27

14 entering the U.S. Decomposi ng bodies in at least one of those instances contaminated food products being imported into the U.S. This incident was the result of the SFTA not including ship and airplane transport in its regulations (Office of the Inspector General 1998). A final incident occurred in 1 999 when 207 people developed cases of salmonellosis because ice alle gedly contaminated with Salmonella was illegally added to orange juice being shipped from Mexico to Arizona (FDA/CFSAN 2001). Since the OIG investigation there have been several uns uccessful attempts to transfer SFTA responsibility to the FDA. In the interim, to maintain the safety of food transpor ted in tanker trucks, four voluntary standards have b een developed: 1) the Bulk Over-the-Road Food Tanker Transport Safety and Security Guidelines (Food Industry Transporta tion Coalition 2003), 2) the Grade A Pasteurized Milk Ordinance (PMO) (FDA/CFSAN: National Conference 2003) /3-A Sanitary Standards for Steel Automotive Transportation Tanks for Bulk Delivery and Farm Pick-Up Service (2002), 3) the FSIS Safety and Security Guidelines for the Transportation and Distri bution of Meat, Poultry, and Egg Products (2003), and 4) the Juice Processors Association Model Tanker Wash Guidelines for the Fruit Juice Industry The Bulk Over-the-Road recomme ndations deal with non-dairy dry and liquid foods and the PMO/3-A Standards de al with dairy foods. The documents are similar in content. The Bulk Over-the-Road Guidelines are divided into an introduction, recommended documentation procedures (exa mple forms are found in the documents appendix), guidelines for receipt and inspection of an empty tanker, how to load tankers, what to do when the tanker has been lo aded, minimum requirements for cleaning non-

PAGE 28

15 dairy food/food grade tankers, conversion of trailers from non-approved, non-food service to approved non-food to food and food to food service, tank requirements for non-dairy and dry-bulk food grade car go tanks, and security guidelines (Food Industry Transportation Coalition 2003). The 3-A Standards provide scope and definitions for the document and metals, fabrication, air venting, mechan ical cleaning, extra fittings, air pressure, temperature, insulation, and design/construc tion requirements for dairy tr ansportation ta nkers (3-A Standards for Transportation Tanks 2002). 3-A equipment produced for dairy complies with the design and construction criteria outlined in the PMO. The PMO is not a mandatory document. The document can be adopted by states or local governments into their legislation; however, once it is adopted or a state ag rees to participate in the National Conference on Interstate of Milk Shipments (NCIMS) the PMO must be followed as part of the law. Appendix B of the PMO provides guidelines for milk sampling, hauling and transportation. It gi ves requirements for the driver, training guidelines for the driver, quality and sampling checks that must be performed before picking up the milk, pumping procedures, inspection procedures, milk tank truck standards, and procedures when standards ar e not met. Appendix B refers to Section 7, 12p the Cleaning and Sanitizing of Containers and Equipment for methodologies on cleaning tankers. The most rele vant points to tank wash fac ilities are 1) Milk containers must be cleaned every 72h, 2) All Grade A milk trucks are to be washed and sanitized at a permitted facility, 3) Washed trucks will carry a wash tag stating the date, time, place and signature or initials of the employee or contract opera tor doing the work, unless the milk tank truck delivers to only one receiving facility where responsibility for cleaning

PAGE 29

16 and sanitizing can be definitely established without tagging and 4) The wash tag will be kept at the next station the tanker is washed and sanitized at for a minimum of fifteen (15) days (FDA/CFSAN: Na tional Conference 2003). The FSIS Safety and Security Guidelines fo r the Transportation and Distribution of Meat, Poultry, and Egg Products is important to tanker truc k transportation with respect to hauling liquid eggs. The gui delines cover the types of vehi cles that may be used to transport meat, poultry and egg products, and procedures related to preloading, loading, in-transit, and unloading. The most important guidelines pe rtaining to the shipment of liquid eggs are the temperature control requ irement, and the guideline under loading which recommends that sealed vehicles shipping egg products (pasteurized, repasteurized, or heat treate d) from one point to anothe r should have a certificate accompanying them. The certificate should state if products have not been pasteurized or if they have tested positive for Salmonella The FSIS Guidelines also includes a section on food security, which offers guidance on es tablishing and implementing a food security plan. This section includes a section for helping to ensure food truck security. (FSIS Safety and Security, 2003) Finally, the Juice Processors Association Model Tanker Wash Guidelines for the Fruit Juice Industry outlines wash station require ments, transportation tanker requirements, 4 types of wash protocols, how to clean accessory parts, a list of commonly transported substances whether or not they are able to be trans ported in a food grade tanker and what wash type they receive, and a set of documenta tion procedures (Juice Products Association, 2004).

PAGE 30

17 The most recent development to improve food transportation was in August 2005 when the president signed into law the Safe, Accountable, Flexible, and Efficient Transportation Equity Act of 2005. Under Ch apter 3 sections 7381-7383 of this Act, effective October 1, 2005, the Secretary of the Department of Health and Human Services (HHS) will take over the many of th e responsibilities outlined in SFTA. In addition the Act adds that the Secretary s hould prescribe practices for sanitation and record keeping. The Act amends Chapter 57 and makes the DOT will still be responsible for the establishing and training DOT inspector s to look for contamination or adulteration of products under section 416 of the Federal Food, Drug, and Cosmetic Act; section 402 of the Federal Meat Inspection Act (21 USC 672); and section 19 of the Poultry Products Inspection Act (21 USC 467a). The objective of the legislation is that by placing the Department of HHS (the branch of the governme nt containing the FDA) in charge of the developing regulation for trans portation and sanitati on liquid transportation tankers that it will improve safety of liquid food in the United States. It will be in teresting in the next few years to see if the Department of HHS is more effective than the DOT. Marketing and Food Safety Justification for Citrus Juice Transportation Tanker Research The United States (U.S.) and the State of Florida have reason for concerns regarding juice transportation. Accord ing to the USDA, the U.S. produced 12,311,000 metric tons of oranges in the 2003/2004 ma rketing year, making it the worlds second largest orange producing na tion following Brazil. The U.S. also produced 1,895,000 metric tons of grapefruit making it the number one producer of grapefruit in the world. Approximately 80% of oranges and 50% of gr apefruit grown in the U.S. are processed into juice products. In the 2003/2004 ma rketing year the U.S. processed 37,048,000

PAGE 31

18 metric tons, imported 20,005,000 metric tons and exported 19,955,000 metric tons of orange juice (USDAs Foreign Agricultural Service (FAS): Production 2004). Most of the citrus in the United States comes from Florida. According to the USDA, Florida accounted for 79 percent of total U.S. citrus production, California totaled 18 percent, while Texas and Arizona produced the rema ining 3 percent (USDAs Citrus Fruit Summary 2004). In Florida the citrus industry has an impact of $9 billion per year and is responsible for 90,000 jobs. It is the largest segment of the agricultural indus try, which is second only to the tourist industry in importance to the states economy. Approximately 95% of Floridas oranges are proce ssed into juice in any specific year (USDAs Foreign Agricultural Service (FAS): Horticultural 2004) Unpasteurized a nd pasteurized citrus juices are transported in transportation tankers. Unpasteurized juice can only be transported in tankers as if the customer d eclares that it will be pasteurized before packaging. Dairy products, both raw and pasteurized are often hauled before juice products in the state of Florida. There is growing awareness that the qua lity and safety of products transported in tankers can become contaminated with bacteria, yeast, and/or mold from inadequately cleaned tankers. Introduction of microorgani sms to products not destined for pasteurization, such as citrus ju ice concentrates, raises concern. Both transportation tanker wash stations and the citrus industry would like to take steps to mitigate food spoilage and to ensu re pathogens do not enter citrus juice and concentrates from tankers. A decrease in quality of citrus ju ice would encourage consumers to purchase other juices or beve rage products, and the ensuing economic loss

PAGE 32

19 over time would affect wash facilities. A f oodborne outbreak in citr us juice from Florida caused by inappropriate washing could financially ruin a tank er truck wash station as well as reduce consumers trust in Florida citrus products. Presently, there is little published res earch completed on microbial aspects of transportation tankers. Research on this topic is discussed below. Also there is no research dealing with biofilm formation caused by citrus juices. Th erefore, the goal of this literature review is to present what is known about microbial aspects of tankers; microorganisms associated with a citrus juice and dairy environment; information on biofilms formation; and methods by which to st udy the above mentioned topics. Previous Research on Transportation Tankers There are a limited number of studies on th e transportation of food in tankers. The vast majority deal with milk transportation in tankers. This may be a result of the nutrient and microbiological make-up of milk but also may reflect the fact that milk is one of the products most frequently transporte d in dedicated tankers. The first study that could be found on transportation tankers a ppeared in Deutsche-Milchwirtschaft (Tamoschus 1979). The author investigated f actors that affected cleaning protocols for milk tankers and concluded that a fully pr ogrammed cleaning and disinfection process, using always freshly prepared solutions, is the most reliable and hygienically most suitable method for cleaning milk tankers (Tam oschus 1979). Most studies since then have included tanker trucks in assessments of residual antib iotic residues on the surface of dairy processing machinery. Also, several research papers have been written on mathematical models to solve tanker truck scheduling problemsUbgabe a nd Sankaran (1994) in India, Butler and Williams (1995) in Ireland, Foulds and Wils on (1997) in New Zealand, and Basnet and

PAGE 33

20 others (1997 and 1999) in New Zeala nd (Basnet and others 1999). The 1999 mathematical model by Basnet and others wa s designed not only to solve the problem of scheduling transportation tanke rs to pick up milk from dairy farms but to set up a schedule so there would not be any congestion back at the dairy processing plant when the milk was pumped from the tankers (Basnet and others 1999). Steele and others (1997) st udied 1,720 pickups of raw milk in transportation tankers for the presence of foodborne pathogens: Salmonella spp., Camplybacter spp., Listeria monocytogenes and toxicgenic Escherichia coli Then they calculated the theoretical probability of 3, 5, and 10 raw milk tankers containing the above-mentioned pathogens. Of the tankers sampled, 8 (0.47%) contained Camplybacter spp.; 3 (0.17%), Salmonella spp.; 15 (0.87%), toxicgenic Escherichia coli ; 47 (2.73%), Listeria monocytogenes Only two tankers contained more than one of the above-mentioned pathogens. They also provided theoretical pr obability of having one pathogen in a bulk tank resulting from the pooling of 1, 3, 5, and 10 tankers is 4.13%, 11.89%, 19.01%, and 34.41% respectively. The author s concluded that although the possibility for an individual tanker to be contam inated is relatively low, th e probability of pooled bulk raw milk containing one or more of the pathogens is fairly high (Steele and others 1997). Another area of study relating to tran sportation tankers is the use of ATPbioluminescence to rapidly determine the am ount of microbiological and other residual contamination (Bell and others 1994). In theo ry, ATP should not be present if a tanker is properly cleaned; however, low levels of ATP may be found ev en in a clean tanker (Bell and others 1994).

PAGE 34

21 ATP-bioluminescence was discovered in 1940 s (Stanley, 1982). All living cells contain the high-energy chemical compound ade nosine triphosphate, ATP. In the ATPbioluminescence method, an enzymatic complex catalyzes conversion of chemical energy of ATP into light through oxidation-re duction reactions (Pez and others 2003). The amount of light generated is measured by a luminometer in relati ve light units (RLU) and is directly proportional to the amount of ATP present in the sample (Pez and others 2003). Bell and others (1994) used two diffe rent commercially available ATPbioluminescence products, from Biotrace Ltd an d Sonco Ltd, for their study. Swabs were taken from 465 milk tankers at 10 cm2 sampling points on the internal surfaces of the manway lid, vessel roof, vessel side wall, ve ssel end wall, flexible hose, and the air elimination vessel after they had been cleane d. For each tanker a visual assessment of clean or dirty was made for each area and sw abs from each area as well as rinse water samples were taken for ATP-bioluminescen ce testing. From every other tanker, microbial counts were taken from the swabs and the rinse water. Parameters were established for whether a site was clean or dirty for each ATP-bioluminescence kit. For internal sites (vessel roof, vessel si de wall, vessel end wall), there was 88.290.6% agreement between the two ATP-biolumines cence kits. For external sites (internal surfaces of the manway lid, flexible hose, a nd the air elimination vessel), there was 77.783.6% agreement between the two kits. In the rinse water, the agreement between the two tests was 65.5%. The study results found th at 93-98% of vessel roofs, vessel side walls, vessel end walls were clean using st andard microbial tec hniques but only 63-89% of the these surfaces were clean using the vi sual or the ATP-bioluminescence kits. For

PAGE 35

22 all areas on the outside of the tanker 56-93% of bacterial swabs were considered clean while 44-66% of the trucks were visually cl ean or clean using ATP-bioluminescence. Lastly, only 60% of the rinse waters were clean using stan dard microbial counts, while less than 40% of the rinse waters were vi sually clean or clean according to the ATPbioluminescence kits. From this study, th e authors reported th e following: 1) the differences between the ATP-bioluminescen ce results and the microbial counts are a result of the fact that the microbial count s reflect the number of microorganisms while ATP-bioluminescence is based on the number of microorganisms plus the soil, 2) ATPbioluminescence is more efficient because it ta kes less than 10 mins to get results while conventional microbial counts take 3 days, 3) tankers which had good drainage generally were found to be clean, and 4) both ATP-biol uminescence test kits correlated well with each other. Therefore, the au thors concluded that ATP-bioluminescence could be used to as an efficient, reliable mechanism to mon itor the cleanliness of transportation tankers and that the results indicate that external su rfaces, in particular the manway lid and the air elimination vessel, provide the more accurate assessment of cleanliness of the vehicle (Bell and others 1994). Pez and others (2003) evaluated three tr ansportation tankers by assaying for ATP on the inner surface of the manw ay lid, outlet pipe and vessel roof of recently cleaned tankers. The final equipment rinse water was sampled and analyzed with ATPbioluminescence and microbial plate count. C lean, caution and dirty ratings were given to ATP-bioluminescence samples based on a scale created by the bioluminescence light manufacturer (BioControl Systems, Bellevu e, WA, USA). The inside surface of the manway was deemed hardest to clean. The outlet pipe had the most variable results when

PAGE 36

23 correlated with the cleanliness of the tanker. Therefore, the authors agreed with Bell and others (1994) that the outlet pipe may be used as an indication of a tankers cleanliness. Few samples were taken from the tankers roof because of irritating odors from the chemical detergents so little could be dr awn from these measurements. Rinse water results from all tankers indicated a good co rrelation exists between the microbial count method and the ATP-bioluminescence method. The authors concluded from this study that ATP-bioluminescence was a good met hod for judging the clean liness of milking equipment, bulk tank and milk transport tankers (Pez and others 2003). Citrus Juice and Milk and Th eir Microbial Inhabitants The Environment of Citrus Juice Fellers and others (1990), in a sampling study of the nutrient content of Florida frozen concentrate orange juice, orange juice from concentrat e, pasteurized orange juice, grapefruit juice, and grapefruit juice from concentrate, discovered that there were significant differences within product categorie s due to differences in the manufacturing plant and time of the year the fruit was proces sed. However, they were able generally to describe the nutrient content of citrus juice products and to reaffirm previous knowledge. They reported that Florida orange juice is a significant source of vitamin C with 90 to 100% of the Recommend Daily Value (RDA) in orange juice products and 70% RDA for grapefruit juice products. Vitami n C (ascorbic acid) plays an im portant role in nutrition as an antioxidant and prevents scurvy in the human body (Smolin and Grosvenor 2000). They reported 6-8% thiamin RDA and 8% folic acid RDA in orange juice products while grapefruit juice products had slightly less of these two nut rients. Magnesium, calcium, copper and phosphorus are found in small but cl aimable levels in citrus products while zinc, iron, and sodium are found at insignifican t levels (Fellers a nd others 1990). Also,

PAGE 37

24 significant levels of potassium exist but no pe rcentage is reported because there is no RDA (Fellers and others 1990). Another notable characteristic of citrus juices is that th ey are highly acidic with a pH of about 4.0 or below. Citrus juice processo rs have modified citrus juices to appeal to different market segments by creating low acid juice, juice with different levels of pulp, fortified juice (with calcium, zinc and vitamins A, B6, B12, C, & E, potassium, and folate), mixed fruit juice, home squeezed, and concentrated. Major processors have also produced a reduced sugar and calorie juice dri nk to serve the low carb market as well as orange juice with sterols for heart-health and orange juice targeted to the needs of children (Tropicana 2004; and Florida s Natural 2004, Minute Maid 2004). Microbial Flora of Citrus Juice Bacteria The major types of bacteria in citrus juic es are spore-forming bacilli, lactic acid bacteria, and, rarely, acetic acid bacteria. Also, bacteria of pub lic health significance, such as Salmonella and E. coli have been found in commercial, unpasteurized citrus juices. Spore-forming bacilli are usually from two genera, Bacillus and Alicyclobacillus Bacillus cells are gram-positive, aerobic or facultatively anaerobic, straight, rod-shaped bacteria with dimensions about 0.5-2.5 1.2-10m Cells from these genera are often found in pairs or chains. They are motile by peritrichous flagella. The bacilli can produce endospores that are very resistant to different conditions such as thermal treatments and sanitizers. Within the genus Bacillus species differ widely in their ability to survive wide range of pHs, temperatures and salinities. Bacillus subtilis is often used as typical example of this species. B. subtilis can survive at pHs <6 and >7,

PAGE 38

25 temperatures between 10oC and 50oC, and salinities of at least 7% (their ability to survive at 10% NaCl has not been tested). Bacillus spp. ferment glucose and are catalase positive. Many Bacillus species are oxidase positive, reduce nitrate to nitrite, and require about 3-12% salt for growth. There are a few pathogenic bacilli (Hensyl 1994); however, they do not germinate and outgrow at pH le vels associated with citrus juices. Alicyclobacillus has 9 species and subspecies. The species A. acidoterrestrius is most associated with spoilage of fruit ju ices. These species are commonly found in soil (Deinhard and others 1987), thermal envir onments (Albuquerque and others 2000), and thermally processed juices such as orange (Uboldi-Eiroa and others 1999). They are gram-positive to gram-variable bacteria that are rod shaped, motile and aerobic (Walls and Chuyate 1998 & Wisse and Parish 1998). They contain -alicyclical fatty acids as part of there membrane (Walls and Chuyate 1998 ). This genus grows best at a pH of 3.54.0; however, it can grow at a minimum of 2.5 and a maximum of 5.5 (Walls and Chuyate 2000 Spoilage and Walls and Chuya te 2000 Isolation) while spores can be produced as low as 3.24 (Walls and Chuyate 1998). A. acidocaldarius prefers to grow at temperatures between 60-65oC, while A. acidoterrestrius and A. cycloheptanicus prefer to grow at 45-50oC (Walls and Chuyate 2000 Spoilage and Walls and Chuyate 2000 Isolation). The spores can survive typical 85-90oC juice processing temperatures (Uboldi-Eiroa and others 1999). None of the bacteria in this genus are pathogenic (Silva and others 1999). Lactobacillus species are gram-positive, rarely motile rods although some may appear as cocco-bacilli. They are about 0.5-1.6m in diameter. They are facultatively anaerobic and occasionally microaerophilic. Most Lactobacillus species grow best when

PAGE 39

26 there is at least 5% carbon dioxide in the at mosphere. They need rich and complex media in which to grow. Products of their ferm entation may include lactate, some acetate, ethanol and carbon dioxide. Not all species produce carbon dioxide. Lactobacillus does not reduce nitrogen, liquefy gela tin, and its cells are catalase and cytochrome negative. This genus is rarely pa thogenic (Hensyl 1994). Salmonella spp., and E. coli cannot grow but may survive for extended periods in chilled unpasteurized citrus juice (Parish 1998) Oyarzbal and othe rs (2003) found that E. coli O157:H7, Salmonella spp., and Listeria monocytogenes inoculated at levels greater than or equal to 103 CFU/g was capable of survivi ng for twelve weeks in orange juice concentrates stored at -10oF. Parish and others (2004) discovered that Salmonella spp. could also survive in grapefruit juice concentrate but its lower pH provided better antimicrobial activity. Also that two days storag e of grapefruit concentrate between 7oF to -12oF would cause a 5-log reduction (Par ish and others 2004). Survival of Salmonella in these products has led to dise ase outbreaks (Parish 1997). In a 1995 outbreak of Salmonella resulting from the consumpti on of unpasteurized orange juice potentially contaminated poorly washed fruit or amphibians in the facility (Parish 1998). Klebsiella pneumoniae and Streptococcus spp. have been found to survive in frozen orange juice concentrate (Fuentes and ot hers 1985, Larkin and others 1955, Patrick 1953). Kaplan and Appleman (1952) studi ed 42 cans of commercially packed concentrate and found that th e enterococci found in the ca n were more prevalent and more resistant to the environment of frozen citrus concentrate than E. coli. Larkin and others (1955) discovered in their research that Streptococcus faecalis, Streptococcus liquefaciens, and E. coli were able to survive in orange juice concentrate stored at -10oF

PAGE 40

27 for 147 days. The number of S. faecalis and S. liquefaciens did not change over time while the numbers of E. coli fluctuated over time. K. pneumoniae isolated from frozen concentrat e processed in Florida and shipped to Puerto Rico was found to survive in lo w temperatures (freez ing), low pH, and low water activity. Researchers ruled that th e contamination occurred before shipping because the pathogen was found in unopened ba rrels of product. Contamination most likely occurred by the mixing of unpasteu rized juice with concentrate, or Klebsiella pneumoniae was present on the machinery used to f ill the concentrate (Fuentes and others 1985). Yeasts Common yeast inhabitants of citrus juices include Saccharomyces, Torulaspora, Candida, Zygosccharomyces, Hanseniaspora, Metschnikowia, Pichia, and Rhodotorula. The genus Candida has 151 species. The vegeta tive cells reproduce by budding and sometimes contain pseudohyphae (such as in the case of C. parapsilosis) or septate hyphae. This genus of yeast does not re produce by sexual reproduc tion. Of the two species of interest in orange juice, C. parapsilosis ferments D-glucose (one of two key carbon sources in citrus juice) and needs D-gl ucose for growth and also may or may not need citrate (the second key carbon source in citrus juice) to survive, while C. stellata does not utilize citrate for growth but will ut ilize D-glucose for fermentation and growth. C. stellata will reproduce best at 25oC and C. parapsilosis will grow at 25-37oC. Hanseniaspora is composed of 6 different species. Hanseniaspora have lemon to oval shaped cells with pseudo -hyphae that reproduce ase xually by polar budding and reproduce sexually by utilizing asci that have one to four ascospores. There are 12 species in the genus Metschnikowia. The yeast cells in the Metschnikowia genus

PAGE 41

28 reproduce asexually by budding and sexually util izing club-shaped asci containing 1 to 2 needle-like ascospores. This yeas t rarely flocculate. The genus Pichia has 89 species. Pichia will have pseudo-hyphae and occasional septate hyphae. This yeast reproduces asexually by budding and sexually by asci with 1 to as many as 8 ascospores. The genus Saccharomyces contains 16 species. The yeast will sometimes have pseudo-hyphae. They reproduce sexually by asci that are formed from directly from a diploid cell with 1 to 4 ascospores. S. cerevisiae ferments and uses D-glucose for growth however citrate is not utilized for growth. Optimal grow th temperature for this yeast are 25-30oC (Barnett and others 2000). Molds Some of the common mold t ypes in citrus juices are Cladosporium cladosporioides, Penicillium citrinum, P. digitatum and P. italicum and Geotrichum spp. (Wyatt and others 1995, Wyatt and Parish 1995). Cladosporim and Penicillium produce spores called conidia while Geotrichum produces spores called arthrospores. Cladosporium is a dark green to black with a black back. The spores are dark, one to two-celled that spread by exposing its dry spor e masses to air currents. This mold is most often found in decaying pl ant matter and in the air. Penicillium has brush-like structures that carry the mold spores. It is commonly found in the soil. This mold produces a green to blue green rot on citrus fruit. Geotrichum are composed of colorless, slimy chains of spores. They can produce strong odors and be pathogenic to humans. They are common in dairy products and in the so il (Malloch 1981). The Environment of Liquid Dairy Products Milk in comparison to juice is a mu ch better medium for the growth of microorganisms because of its almost neutra l pH (Frank 2001). Milk also has wide

PAGE 42

29 variety of available nutrients. Milk is approximately 87.3% water, 4.8% lactose, 3.7% fat, 2.6% casein, 0.6% whey protein, salt cations (0.058% sodium, 0.140% potassium, 0.118% calcium, 0.012% magnesium) and anions (0.176% citrate, 0.104% chloride, 0.074% phosphorus) and nonprotein nitrogen (J enness 1988). The main sources of carbon for microorganisms are lactose, fat, and protein. The amounts of citrate and glucose present in milk are not enough to su stain microbial growth for long; therefore, fermentative microorganisms must be able to utilize lactose. Microorganisms rarely use milk fat as a carbon source because unless fat globules are damaged the microorganisms cannot penetrate the fat globul es protective protein and lip id membrane. Of the two proteins in milk, caseins are ea sy susceptible to proteolysis while whey generally is not. The nonprotein nitrogen that is re adily utilized as a nitrogen s ource is not able to sustain microbial life. Milk is a good source of B v itamins and minerals such as iron, cobalt, copper, and molybdenum. However, many of th e minerals may not be present in a form that can be utilized by bacteria Lastly, milk contains growth stimulants such as orotic acid, which is a metabolic precursor to pyrim idines and which fosters microbial growth (Frank 2001). Milk sold for liquid consumption in the United States must be pasteurized at minimum for 15 sec at 72oC although most processors wi ll use higher temperatures and longer holding times. Milk may be transpor ted raw or pasteurized but it is always pasteurized after transport. Raw milk from healthy animals gene rally has less than 103 microorganisms (Richter and Vedamut hu 2001) that typically consist of Micrococcus, Staphylococcus, Streptococcus, and Corynebacterium spp. Staphylococcus aureus, Streptococcus spp., Pseudomonas spp., and coliforms are related to environmental and

PAGE 43

30 contagious mastitis. Some contamination ca n occur from the cow, the milking room environment or poorly cleaned milking systems. The contaminants are occasionally yeast and mold but generally are Bacillus spp., Clostridium spp., lactic acid bacteria, coliform and other gram-negative bacteria (Olson and Mocquot 1980). When raw milk is cooled the increases in bacteria are caused by Psuedomonas spp. as well as species of Alcaligenes and Flavobacterium (Bishop and White 1986, Cousin 1982, Stadhouders 1975, Thomas 1974). Foodborne outbreaks involving raw milk and Salmonella spp., Campylobacter jejuni, and Yersina enterolitica, and Listeria monocytogenes (Bryan 1983 and 1988) The microorganisms that commonly ar e found in freshly pasteurized milk are gram-positive bacteria that can survive pasteurization: Bacillus, Lactobacillus, Micrococcus, Staphylococcus, Streptococcus, Microbacterium, Enterococcus, Arthrobacter, and Corynebacterium spp. (Cousin 1982). Since th ese bacteria generally do not grow quickly at refrigeration temper atures they are generally outgrown by gramnegative psychroduric colifor ms, and members of the Psuedomonas, Alcaligenes, and Flavobacterium spp. (Cousin 1982, Olson and Mocquot 1980)). Postpasteurization contamination has resulted in listeriosi s and salmonellosis (Byran 1983 and 1988). The postpasteurization contamination of Yersina enterolitica, and Listeria monocytogenes is a major concern because of these pathogens ability to grow at refrigerator temperatures (Richter and Vedamuthu 2001). Other dairy products commonly transporte d in tankers are cream, half-and-half, sweetened condensed milk, liquid ice cream mix, whey and chocolate base. Pathogens that are found in cream and cream-fillings are B. cereus and S. aureus (USDAs Bad Bug Book 1992). Condensed milk contains Bacillus, Lactobacillus, Micrococcus,

PAGE 44

31 Streptococcus, Microbacterium, Enterococcus, Arthrobacter, and Corynebacterium spp., coliform and psychroduric bacteria (Foster and others 1957). Because of the added sugar, sweetened condensed milk generall y contains osmophilic sucrose-fermenting yeasts and molds (Frazier 1958). Bacillus spp. and other postpasteurization contaminants grow in liquid ice cream mix; once the i ce cream mix is frozen growth of most microorganisms stops (Richter and Vedamuthu 2001). However, some bacteria including pathogenic Salmonella spp. and L. monocytogenes have been found to survive in ice cream (Bryan 1983, Rosenow and Marth 1987). Biofilms Biofilms are communities of organisms that contain either bacteria or other higher organisms, such as algae, that are held together by sticky ex tracelluar (polymeric) matrix (Watnick and Kolter 1999) and are irre versibly associated with a surface (Donlan 2002). Biofilms may also contain materials othe r than cells such as blood components, or clay (Donlan 2002). The first person to note biofilms was Van Leeuwenhoek in 1683 when he used his simple microscopes to ex amine bacteria on the surface of teeth (Donlan 2002, University of California 2005). Biofilm s were eventually recognized again in 1940s with the work of scien tists studying marine organism s and noting their ability to attach to surfaces (Heukelekain and Heller 1940 and Zobell 1943). Since then other work has been done on biofilm formation in oil manufacturing, oral caviti es, water sources and refinement operations, medical and industria l settings and very recently the study of biofilms and their relationship to food. The ability to study biofilms has been enhanced by the development of more complex microsc opes, in particular the confocal laser scanning microscope, and th e development of technique s used to study the genes involved in cell adhesion and biof ilm formation (Donlan 2002).

PAGE 45

32 Biofilms have been described as the prev ailing microbial lifestyle because biofilms provide microorganisms with safety, nutrients, protection, and a place to transfer genetic material (Watnick and Kolter 1999). Biofilm s are commonly found on air-water or solidliquid contact surfaces that c ontain a readily available suppl y of nutrients (Stickler 1999 and Donlan 2002). Therefore, biofilms have created problems in both the food industry and the medical community because they at tach to production equipment, prosthetic devices and sterilizing equipmen t (Stickler 1999). Biofilms also serve useful functions in industry by breaking down emulsifiers and oi ls (Pasmore and Costerton 2003) and the environment as well such as breaking down organic matter, degrading pollutants, and cycling nitrogen, sulfur, and many metals (D avey and OToole 2000). There are four major stages in the biofilm life cycle: atta chment, formation, maturation, and dispersal. Attachment The first stage in biofilm formation is attachment. In this stage the bacteria gets close to the surface, slows down its rate of movement, forms a transient attachment to the surface, and searches for a pl ace to settle down and make a st able attachment (Watnick and Kolter 2000). Bacteria may search for suitable sites by twitching or swarming movements if the bacterium is motile (Pasmore and Costerto n 2003). The ability of a bacteria to attach to a surface depends on ma ny factors including the bacterias cellular components, the material the bacteria is at taching to, the surroundi ng environment, gene regulation in the bacterial cell, and the intera ction between the bacteria and other bacteria preexisting on the surfa ce (Donlan 2002). Therefore, where bacteria attach and why they attach may be uniquely different, but one th ing can be certain about all bacteria and surfaces they chose to attach; these surfaces provide an ideal environment to grow and develop and being part of a biofilm has a dvantages over being a free, planktonic cell.

PAGE 46

33 Material Three major factors of materials play a ro le in bacterial attachment to a surface: roughness, hydrophobicity and polarity, and its conditioning. Several authors have noted how surface roughness plays a role in biofilm attachment. Characklis and others noted that an increase in roughness of surfaces in creases attachment (1990). Arnold and others (2001) discovered that the root mean squa re (RMS), which is a measurement of the surfaces roughness, and the center line average, which is the depth from the peak of the sample at which there is a 50% of the samp le area below and a 50% of the area above, can help to predict biofilm formation. A ccording to Donlan most investigators found that microorganisms attach more rapidly to hydrophobic, nonpolar surfaces, such as Teflon and other plastics, than to hydrophilic materials such as glass or metals (2002). Scientists are not sure why this occurs b ecause there has not been a conclusive method for measuring surface hydrophobicity, but it seem s that cells must be able to overcome repulsive forces between itself and the surf ace creating a hydrophobic interaction close to the surface, which allows to th e cell to attach. An explana tion for how this is possible has come from examining the cellular component s role in attachment (discussed below). Finally, conditioning film is created from pa rticles of the media bond to the surface to form a film. This film can affect the rate and amount of attachment (Donlan 2002) because the film has a different chemical co mposition and nutrient value that attracts bacteria. Also the film can reduce the repulsi ve effects of the surface allowing bacteria to easily bind (Pasmore and Costerton 2003). Cellular components The outside of the cell is composed of hydrophobic and hydrophilic regions. Although the outer surface is soluble in water it can form hydrophobic connections with

PAGE 47

34 stratum surface materials, and cell membrane bound proteins and polysaccharides. The extracellular polysaccharides, unl ike proteins, allow cells to form attachments at greater distances from surfaces. In this way bacteria reduce energy needed to attach to surface because by using extracellular polysaccharides the whole cell does not have to enter the surfaces double ionic layer (Pasmore and Costerton 2003). Other cellular components that help the cell attach are the flagella, and pili. The flagella and type IV pili (Pasmore and Cost erton 2003) help to overcome the electrostatic repulsive forces that exist between the su rface and cell (Corpe 1980) in a same way the extracellular polysaccharides he lp to overcome the surfaces double ionic barrier by using the attractive moieties. The type IV pilus has an advantage over the flagella in that it can shoot out and attach to the surface. After it attaches it can reel th e cell back in to the surface (Pasmore and Costerton 2003). Characteristics of the liquid media Characteristics of the media such as pH, nut rient levels, ionic st rength, temperature, and the velocity can influence the rate of attachment. For example, Cowan and others (1991) found an increase in bacter ial attachment occurred as a result of increased nutrient concentration. Others have found that in nutrient rich media b acteria will settle anywhere, while in nutrient poor medium b acteria will only attach to nutrient rich surfaces (Watnick and Kolte r 2000). Barnes and others (1999) found that ionic composition of the suspending medium had the most effect on bacterial adhesion. They discovered that iron and calcium salts presen t in the suspending media increased cellular attachment while potassium, manganese, magne sium, and sodium salts inhibited cellular attachment. It was thought the reason for th e lack of attachment was due the dissolved cations [potassium, manganese, magnesium, and sodium] shielding the surface-negative

PAGE 48

35 charge on bacteria and [stainless] steel while the increased attachment of calcium and iron was believed to be a result of the mol ecules ability to bridge between the bacteria and the surface (or the conditioning film on the surface). Flow velocities affect cellular attachment; when velocities are slow, cellula r attachment depends more on size and cell mobility, and when velocities are high, cells are subject greater turbulence and mixing. Therefore, the cells that attach in these e nvironments are the ones that can make a quick, effective association with the surface and re main attached until the velocities become great enough to exert a shear force on cells th at make them detach (Characklis, Microbial 1990). There are no literature was found on how citr us juice may affect biofilm formation. However, Barnes and others (1999) experiment ed with different concentrations of milk exposed to a stainless steel surface before contact with S. aureus, S. marcescens, and L. monocytogenes. It was found S. aureus increased attachment with samples with 0.1% milk and the control compared to 100, 10, or 1% milk, and S. marcescens, and L. monocytogenes had increased attachment with the control compared to 100, 10, 1 or 0.1% milk. A possible reason is that bacteria are attracted to iron on the surface of the stainless steel if nitrogen from the milk protein blocks the attractive forces (or the electron escape depth) then the bacteria will not readily attach to the stainl ess steel surface. Duddridge and Pritchard (1983) noted that bacteria a ttachment is higher on milk-treated rough surfaces that on milk-t reated smooth surfaces. Formation When a bacterium chooses to attach it ge nerally assimilates it self into part of microcolony as part of biofilm formation. The microcolony develops until it forms a three-dimensional EPS-encased structure at which point it is considered a biofilm

PAGE 49

36 (Watnick and Kolter 2000). A number of me thods such as fluorescently labeled rRNAtargeted oligonucleotides, a variety of microsensors, re al-time image analysis, and confocal microscopy have help ed researchers obse rve bacterial development while other advances have been made to help cultivate bacteria such as chem ostats, continuous-flow slide cultures, microstats, and colonization tracks (Davey and OToole 2000). The development of the biofilm lies in the formation of the extracellular polymeric substances, the architecture and the inaction with other bacteria and particles (Donlan 2002). Extracellular polymeric substances (EPS) EPS makes up the majority of material that forms biofilm matrices. A large portion of EPS is composed of polysaccharides (Donlan 2002). In some bacteria, genes to synthesize flagella are down regulated wh ile the genes to synthesize the EPS are up regulated (Watnick and Kolter 2000). The composition of gram-negative bacterias EPS is largely neutral and in the presence of uronic acids or keta l-linked pryruvat es it can take on an anionic state. Gram-positive bacterias EPS generally are cationic. The EPS of some bacteria, such as those that are coagul ase-negative, contain pr otein. EPS in most cases has hydrophilic and hydrophobi c regions. EPS is benefici al to the biofilm because it can prevent desiccation (Donlan 2002) and can stop antibiotics from being transported in to the biofilm (Donlan, Role 2000), as well as protect against pH shifts, UV radiation, and osmotic shock (Davey and OToole 2000). Composition and variable nature of the EPS are the two properties that have the greatest impact on the biofilm Composition of the EPS can affect the rigidity, the deformation characteristics and the solubility in water. Sutherland gives the example that

PAGE 50

37 EPS that has a backbone made of 1,3or 1,4-linked hexose residues are more rigid, less deformable and not very soluble in wa ter compared to other components of EPS (2001). The amount of EPS can be attributed to the type of organisms that make up the biofilm, the age of the biofilm, the growth rate of the bacteria in the biofilm, the nutrients available to that bacterium from th e liquid medium (Flemming 2000). Architecture As microcolonies increase in number th rough division and addition of new cells they begin to form mushroom-like colonies, as seen in Figure 2-1, that contain a number of channels beneath the mushroom caps th at bring nutrients to cells lower in the biofilm. This shape suggests a controlled grow th pattern that is most likely developed by quorum sensing (a method for communication be tween bacteria) (Costerton 1995). Other bacteria and particles When biofilms are composed of one species of bacteria, the bacteria alter themselves genetically to best survive in the biofilm. In a mixed biofilm, bacterial species will set themselves up in locations that best suit the needs of the different types of bacteria (Watnick and Kolter 2000). It is also important to note that in the development of the stru cture of the biofilm that nonmicrobial components (clay, blood partic les, etc.) may be incorporated from the host or the environment. Mineral build-ups in biofilms can be a problem in medical Figure 2-1. The microocolony on the far right shows typical mushroom cap formation (Costerton 1995 ) .

PAGE 51

38 devices and water systems (Durack 1975, T unney and others 1999, Donlan, Biofilm 2000). Maturation After 12 hours to a few weeks of deve lopment (Pasmore and Costerton 2003), microcolonies become EPS-encased and a matu re biofilm is formed (Davey and OToole 2000). Other bacteria, gene transfer, and, quorum sensing affect the maturation of biofilms. It is also interesting to note the ability for pathogens to be involved in mature biofilms, and resistance that mature biofilms develop. Other bacteria In bacterial communities as in animal co mmunities there exist interactions between species. Mixed biofilms often develop synotrophic relationships where two metabolically different species depend on products the other produces for survival (Davey and OToole 2000). In the scientific literature there are other documented cases of the following types of interactions that exist between bacterial species (Table 2-2). Skillman and others (1998) used four different bacteria from the family Enterobacteriaceae in dual-species biofilm studies. They concluded that the stain of E. coli used out-competed Klebsiella pneumoniae, Serratia marcescens, and Enterobacter agglomerans. However, the E. coli and the S. marcescens used in this experiment were able maintain a neuralistic co-existance; the K. pneumoniae and E. agglomerans, a mutualistic relationship (Skillman and others 1998).

PAGE 52

39 Table 2-2. The types of relationships th at exist between bacterial species. Relationship Name Interaction Neutralism When two populations do not affect each other. Competition When two populations work against each other to achieve a mutually sought after goal (such as nutrients or niche space). Commensalism When one population benefits while the other remains unaffected. Mutualism When both populations benefit as a result of their association. This association can occur in many forms. Symbiosis which are obligatory interactions Protocooperation which are facultative interactions Synergism which enhances the production or consumption of bacterial made derivative. Ammensalism When one population, without having direct contact can have a negative impact on another. Predation When one population feeds on another. Parasitism When one microorga nism is invaded by another. Gene transfer and regulation In biofilms there is a greater rate of ge netic exchange by conj ugation than occurs between planktonic cells. It has been thought that there are plasmids necessary to form biofilm and that bacteria w ill transfer plasmids to one another (Ehlers and Bouwer 1999, Roberts and others 1999, Hausner and Wuertz 1999). Without these plasmids bacteria would only form a microcolony and never develop into a biofilm. The genetic transfer of resistance to antimicrobials is encoded on th e plasmid; therefore, biofilms may be the breeding-ground for antimicrobial resistance (Ghigo 2001). Quorum sensing Our knowledge of how biofilms form is st ill somewhat of a puzzle. Quorum sensing is thought to be necessary to esta blish biofilms (Federle and Bassler, 2003). Quorum sensing was first discovered in th e 1970s by Nealson and Hastings (1979) when they found that Vibrio fischeri was responsible for producing li ght in a flashlight fish.

PAGE 53

40 Figure 2-2. Quorum sensing in a.) Gram-negative bacteria and b.) Gram-positive bacteria (Federle and Bassler, 2003) According to Federle and Bassler, quorum sensing is a process in which bacteria monitor their cell-population de nsity by measuring concentrat ions of small secreted signal molecules called autoinducers ( 2003). It is known that the amount of autoinducers present directly correlates to th e number of bacteria present. Quorum sensing occurs at the interspecies (between species) and intraspecies level (within species). At the intraspeci es level there are quorum sens ing methods for gram-negative and gram-positive bacteria. Figure 2-2 provides a visual representation of these two intraspecies quorum sensing pathways. Gram-neg ative bacteria have one or more type of LuxI-like proteins and each type of LuxI-like, which produce one acylhomoserine lactone (AHL) autoinducers. After the AHLs are pr oduced, they can freely diffuse outside the cell membrane. The concentration of AHL increases outside the membrane until the concentration reaches a certain level; then the AHL molecules are allowed to bind to LuxR-type proteins. Only members of bacteria in the same species can respond to that autoinducer, and therefore, it seems that ther e is little cross talk in mixed gram-negative populations. Gram-positive bacteria have never used AHL; they use oligopeptide autoinducer s that are sometimes referred to as autoinducing peptides (AIP). These AIP compounds are 5-17 amino acids long and may contain side chain modi fications. The gram-

PAGE 54

41 positive membrane is not permeable to AIP so it requires active secretion. AIPs can then be detected by cell surface re ceptors. This detection l eads to the phosporylation of a response regulator, which binds to the DNA promoter to regu late transcription of that gene. At the interspecies level there is one molecule that appears to be universal among most bacteria: autoinducer-2 (AI2). AI-2 is thought to be the key molecule that allows for interbacterial communicat ion in biofilms. AI-2 was first discovered by studying Vibrio harveyi and to this day V. harveyi AI-2 molecular structure is the only one that has been determined. Scientists do know that othe r bacteria can sense this AI-2 molecule and that other bacteria produce AI-2 -like molecules, but they are not sure if these molecules have the same molecular structure or a different but similar structure. The reason for this has to do with the formation of AI-2. Figure 2-3 shows how AI-2 is formed in V. harveyi. It appears that all bact eria known to form AI-2 (o r like molecules) have the luxS gene to convert S-ribosylhomocysteine (SRH) to 4,5dihydroxy-2,3-pentanedione (DPD). However, DPD can be formed into a variety of compounds to which boron can later be added. Therefore, if future studies demonstrate th at AI-2 is a universal chemical, then AI-2 cannot provide bacteria with the knowledge of what species that form the biofilm, but it can let bacteria know how many other bacteria ther e are (Federle and Bassler 2003). If AI-2 can be derived from DPD then a bacterium will know what type of bacteria exist and how many of them there are. AI-2 may not be produced by every bacteria but it is possible that all bacteria may be able to sense and respond to AI-2. Figure 2-3. Biosysthesis of AI-2 . (Federle and Bassler 2003)

PAGE 55

42 Some known bacterial responses to AI-2 in clude virulence, toxin production, and cell division (Federle and Bassler 2003). It has been found that quorum sensing, both intraand interspecies, is necessary fo r biofilm formation. For example Pseudomonas aeruginosa needs AHL-autoinducers to create mature biofilm, and AI-2 seems to be very crucial for the formation of mixed species of biofilms. Therefore, this indicates that quorum sensing is a necessary method for bact eria to set themselv es up in biofilms in ways that are most beneficial to their need s and the needs of the community (Federle and Bassler 2003). Bacteria have also been known to have the ability to remove or add AI-1 or AI-2 molecules to the environment thereby tricking ot her species in believing that they are in a low density or a high density of bacteria. Bacteria that are able to trick others by providing them with false information face a competitive advantage over others. This knowledge also gives an advantage to rese archers looking for a method to discourage biofilm formation (Federle and Bassler 2003). Pathogenic organisms Pathogenic organisms may be able to at tach to biofilms; however, they do not always seem to grow extensively in them. It is surmised that the reason for this is pathogenic organisms fastidious growth requi rements and their inabi lity to compete with other organisms in the biofilm. Legionella pneumophila (Murga and others 2001), S. aureus (Raad and others 1992), Listeria monocytogenes (Wirtanen and others 1996), Campylobacter spp. (Buswell and others 1996), E. coli O157:H7 (Camper 1998), Salmonella typhimurium (Hood and Zottola 1997), Vibrio cholerae (Watnick and Kolter 1999), and Helicobacter pylori (Stark 1999) are pathogens that have been able to grow in biofilms. The reason for their success in pa rt seems to be due to associations and

PAGE 56

43 interactions with organisms preexisting in th e biofilm (Donlan 2002). Some of the more important pathogen inter actions are those of Staphylococcus spp., E. coli, and Salmonella spp. Work by Den Aantrekker and her collegues (2003) shows how Staphylococcus aureus can attach, form a biofilm, and detach from the surfaces of silicone tubing. Gorman and others (1994) found Staphylococcus aureus in mixed cultures with members of Staphylococcus spp. in catheter biofilms. Other ba cteria can disrupt the attachment of Staphylococcus aureus. Reid and others (1995) noted that Lactobacillus spp. can inhibit the ability of S. aureus to attach and displace alre ady established biofilms of S. aureus on fibrous materials and epithelial cells. E. coli has been found to form mixed cultu re biofilms in urinary catheters (Ganderton 1992) and to dua l-species biofilm with Klebsiella spp. in biofilms that block biliary and pancreatic stents (Brant 1996) Research conducted by Banning and others (2003) found that E. coli was capable of establishing itself in mixed culture of indigenous groundwater microorganisms in a laboratory-scale reactor. However, if the nutrient levels were increased the E. coli had difficulty out-competing the indigenous microflora. This demonstrates that the conditions of the system regulate the ability of E. coli to be an integral part of the biofilm (Banning 2003). In a work by Joseph and others ( 2001) examined the ability of two Salmonella spp. isolated from poultry to attach to plastic, cement, and steel surfaces. They found that the bacteria formed the thickest biofilm on pl astic surface followed by cement and steel. The biofilms were then exposed to different qua lities of hypochlorite and iodophor sanitizer for varying lengths of time. The results noted that the biofilms were more resistant to the

PAGE 57

44 sanitizers than the free cells. The authors concluded that Salmonella spp. can form a biofilm on food contact surfaces and can be a source of contamination for food products. Also, if food contact not clean ed with the proper c oncentration of cleaner for the correct amount of time the Salmonella spp. biofilms may persist on the food contact surfaces (Joseph and others 2001). Camper and others (1998) found that Salmonella typhimurium was able to exist in a model water system biofilm with a group of unknown heterotophic organisms for more than 50 days. This indicates that S. typhimurium is capable of integrating itself with other bacteria to form a biofilm (Camper and others 1998). A study by Esteves and others (2005) used Salmonella enterica serovar Typhimurium and E. coli isolated from the natural flora of the gastrointestinal tract to study their biofilm formation on the HEp-2 epithelial cells. Th ey concluded that the Salmonella would predominate over the E. coli if they were exposed to the on the HEp-2 epithelial cells at the same time. If the E. coli was an established bi ofilm on the cells the Salmonella will establish itself in areas where the E. coli has not attached and displace and replace the E. coli biofilm (Esteves and others 2005). Resistance One of the key advantages to being in a biof ilm is that the biofilm can help bacteria resist effects of chlorine, an tibiotics, and detergents. Q uorum sensing (Donlan 2002) and the ability of the biofilm to alter aspects of its local environment such as pH and oxygen concentration may help with this resistance (Pasmore and Costerton 2003). Lewis in the Riddle of Biofilm Resistance points out the three main reason s for biofilm resistance: 1) EPS can diffuse and bind any possible antim icrobials, 2) antimicrobials are more effective at killing rapidly gr owing cells so the slow grow ing cells of the biofilm are harder to kill, and 3) the ad aptation of gene specific traits that help in the resistance

PAGE 58

45 (2001) such as changing cell surface proteins wh ich gives the antibiotics fewer places to bind (Pasmore and Costerton 2003). Dispersal Dispersal happens by one of three mechanis ms: 1) the shedding of daughter cells, 2) detachment that occurs as the result of quorum sensing or nutrient levels, or 3) shearing of biofilm aggregates (continuous re moval of small portions of biofilm) because of flow effects). The mechanisms of disper sal by the shedding of daughter cells is not well understood (Donlan 2002); howev er, according to the resear ch of Gilbert and others shedding occurs because the newly formed daughter cells have the least hydrophobicity (1993). Often, when nutrient levels become low a biofilm will dissociate by breaking down the EPS matrix and in some species it wi ll use the EPS as a nutrient source before seeking a more nutrient rich environment. Detachment due to flow occurs by three methods: erosion or shearing (continuous removal of small portions of the biofilm), sloughing (rapid massive removal), and abrasion (detachment due to collision of particles from the bulk fluid with the biofilm) (Brading and others 1995). Erosion seems to occur most often when the biofilm is thick and th ere is an increased fluid shear (Characklis, Biofilm 1990). Sloughing also occurs in thic k biofilms but occurs more randomly than erosion. It is thought that sloughing occurs as a result of nutrient or oxygen reduction (Brading and others 1995). Detergents and Sanitizers Detergents The job of detergents is to remove gro ss soil and residue. An effective detergent cleaning treatment is based on an analysis of the soil type (lipi d, carbohydrate, protein, mineral deposits, microorganisms, dirt). On e should choose a detergent that will work

PAGE 59

46 most effectively on that particular soil(s). Carbohydrates can be removed from surfaces with water but also alkaline cleaners can be used to remove it as well. Care should be taken to make sure that overheating and dryi ng does not occur because the sugars will caramelize and starches will form a glue-like material. Undenatured proteins are generally water-soluble while denatured prot eins are water insoluble. Both can be removed with an alkaline cleaner. Lipids are insoluble in to water but can be melted with heat, saponified by alkalis and high temperat ures, and emulsified by polyphosphates. Mineral deposits are alkaline in nature and ar e insoluble in water but can be dissolved in acids (Katsuyama 1993). Alkaline detergents saponify fats and fo rm water soluble compound with proteins; however, they are ineffective below a pH of 8.3. Some co mmercial alkalis that are available are sodium hydroxide, sodium carbonate, sodium hydroxide, sodium sesquicarbonate, trisodium phosphate, sodium metasilicate, tetrasodium pyrophosphate, and sodium tetraborate (Parker and Litchfie ld, 1962). Hard water and sodium hydroxide should not be used together as it will cause precipitation. Adding chlorine to an alkaline cleaner allows for better removal of proteins. This leads to better cleaning of milk stone (milk solids and mineral deposits from the milk). Chlorine used in detergents is not the sanitization agent in alkaline cleaners because the pH is too high for the chlorine to be effective. Acid cleaners dissolve mineral de posits. They have pH of less than 2.5 (Katsuyama 1993). Inorganic acids used are hydrochloric, sulfuric, nitric, and phosphoric acids. The disadvantag e to these acid cleaners is that they will corrode soft metals; however, organic acids have are less corrosive and irritati ng (Jennings 1965).

PAGE 60

47 Detergents by themselves may not provide effective cleaning. Based on the level of soil and the equipment to be cleaned, hand scrubbing, high-pressure water, flushing recirculation, and temperature may be needed (Katsuyama, 1993). Achieving the correct time, temperature, and concentration are important for effective detergent cleaning. The deterg ents manufacturers should indicate what temperatures, and concentration are appropriate for the produc t. Other items to evaluate about when choosing a cleaner are corrosivene ss, irritability to personnel, regulatory standards, foaming, and versatility of uses within the facility (Katsuyama, 1993). Sanitizers The goal of sanitizers is to destroy the vegetative cells; however, vegetative cells of resistant bacteria and bacterial spores can surv ive. For sanitizers to work effectively and efficiently, soils must be removed from the surf aces. Sanitizers that can be used in food processing plants are: he at, halogens, quaternary a mmonium compounds (QUATS), acids, alkalis, ultraviolet irra diation, and ozone. There are disadvantages and advantages to using the above-mentioned compounds. Howe ver, a sanitizer should be chosen for its quick kill, customer and employee safety, re gulatory compliance, easy to removal from the surface, cost, ability not affect the food, ease in de termining its concentration, stability, noncorrosiveness, and solubility in water characteristics (Katsuyama 1993). In the tank wash industry, QUATS are common ch oices for sanitizers. Some of the advantages of using QUATS ar e the following: heat-stable, effective over a wide pH range, noncorrosive, nonirritati ng, impart no off flavors to f ood products, not as affected by organic matter than chlorine, and they leav e a non-volatile residue that inhibits molds, yeasts, and bacteria (Clinger 1973 and Ohio St ate University 1967). The disadvantage to using QUATS is that they are not compatible with nonionic wetting agents in detergents,

PAGE 61

48 and they are rendered ineffective by wooden, cotton, nylon, cellulose sponges, and some plastics (Mauer 1974). Wet heat is often used in conjunction with a chemical sanitizer in the tank wash industry. The advantages of using a heat is it is inexpensive, it can be measured, there is no residue, it is not corrosive, it provides a non-selective ki ll, and it penetrates hard to reach surfaces (Jennings 1965). The problem with heat is that to provide effective sanitization it must reach at least 82oC (180oF) (Katsuyama 1993). The Environment of Stainless Steel Stainless steels come in three key gr oups that are based on the microscopic structure and the composition: martensitic, ferritic, and austenitic (Bosio Metal Specialties, 2000). The martensitic group contains AISI metal types 403, 410, 416, 420, 440. This group is composed of about 12-18% chromium, very little nickel (if any), and 0.06 to 1.20% carbon. These stainless steels can be heat-treated and they are magnetic. The ferritic group is composed of AISI metal types 405, 409, 430, 442, and 446. These types of metals contain 12-18% chromium, 0% nickel, and 0.06-35% carbon. This metal group is also magnetic. The austenitic group contains AISI types 201, 202, 301, 302, 303, 304, 316, 321, 347 and most of the 300 series alloys. The group contains up to 730% chromium,6-36% carbon, and 6-36% nickel. Austenitic stainless steels are non magnetic. These metals are not hardened by heat treatment but by cold treatment, which may cause them to be slightly magne tic (Bosio Metal Specialties, 2000). Stainless steel in tankers is generally composed of eith er 304 or 316 stainless steel unless the tanker is used to hold food-grade oils in which case a 407 stainless steel is generally used. 304 and 316 stainless steel is available standard and low carbon (304L and 316L) varieties. 304 is the commonly us ed of stainless steel because of its easy

PAGE 62

49 formability and corrosion resistant nature. The low carbon variation is formulated so there is no carbide precipitation from the we lding process. It has the same corrosion resistance as the standard ve rsion but has lower mechanical properties than the standard 304. The 316 can handle higher temperatures, and is more resistant to pitting and corrosion than the 304. The 316L is used to avoid the carbide part icipation due to the welding process (Bosio Metal Specialties 2000). Some of the most popular fi nishes of stainless steel in the food industry are #2D and #4 stainless steel finish or higher. #2D is used when a manufacturer cannot guarantee a pit free rolled finish but most food processors like at least a #4 finish with a #7 finish being preferred by some (Frank and Chmielewski 2000). A 2D finish is dull manufactured by a cold rolling annealing and descaling (Bosio Meta l Specialties 2000). A #4 finish is one where course abrasives are used initially to gri nd the stainless steel followed by a grinding with 120-150 mesh. This finish is generally used in a wide variety of food applications. Finish #6 is a #4 finish where the last brushing is done with abrasive and oil. The #7 finish is produ ced by baffling finely ground surface, but the grit lines are not removed (Bosio Metal Specialties 2000). The #8 finish is highly reflective and free of grit lines due to the extensive polishing by successive abrasions and baffling (Bosio Metal Specialties 2000). Gauge of stainless steel correlates to the th ickness of the stainless steel. According to American Delphi Stainless Steel Guide 14 gauge stainless is 0.0747inches thick; 16 gauge, 0.0598 inches thick; 18 gauge, 0.0478 inches thick (American Delphi 2004). Stainless steel has become a standard choice for the construction for much food processing machinery for many reasons includ ing durability and its ability to resist

PAGE 63

50 corrosion (Maller 1998). However, the st ainless steel processing environment can become a home to biofilms because of its mi croscopic hills and valleys, which vary with different levels of finish. Studies have been conducted to see how see how different levels of finish affects biofilm formation and biofilm cleanability. Arnold and others (2001) studied 5 different polis h types of 11 gauge, 304 stainl ess steel against untreated stainless steel. They found that electropo lished, and steel-burnis hed was significantly different from the control of untreated stainless steel in th eir ability to resist biofilm formation while glass-beaded, acid dipped, a nd sand-blasted stainl ess steel were not significantly different from the control. It was thought that the reason glass-beaded stainless steel and the sand-blas ted stainless steel had more b acteria attachment was that during the polishing process the glass and sa nd created craters and scars which create regions for bacteria to attach. They also discovered that the leading indicators that biofilm is going to form to the surface are root mean square (RMS) which is a measurement of the surfaces roughness, and th e center line average which is the depth from the peak of the sample at which there is a 50% of the sample area below and a 50% of the area above (Arnold and others 2001). Ho wever, surface finish appears to have no effect on cleanability according to Influence of Surface Finish on the Cleanability of Stainless Steel by Frank and Chmielewski. They first established that Bacillus stearothermophilus spore count was better determin ate of cleanability than the Pseudomonas spp. biofilm. They discovered that the mean peak to valley height RZ(DIN) and maximum peak to valley height Rmax have a significant correla tion to cleanability of Bacillus stearothermophilus spores. It is advi sed that if manufacturer s want to choose a stainless steel that will have maximum cleanability for spores or biofilms, it should be

PAGE 64

51 chosen not necessarily by polish type but by the amount of surface defects (Frank and Chmielewski 2000). The authors also found th at soiling and cleaning creates increased soil build up and decreased the number of Pseudomonas stearothermophilus spores on the stainless steel su rface after 11 soiling and cleaning. The authors suggest that this behavior may be due to that fact that rep eated soiling and cleaning cycles may stimulate heat activation and inactivation of spores (Frank and Chmielewski 2000). A study done by Arnold and Suzuki showed the effect of corrosion on different polishes of stainless steel. They conclude d that the sandblasted and glass-bead polished samples they tested experienced the greatest increase discoloration and biofilm formation after exposure to a corrosive treatment. Wh ile electropolished stainl ess steel experienced the least discoloration and biofilm formation af ter exposure to a corrosive treatment. The researchers thought this was due to the fact that electropolished stainless steel was composed of very few reactive elements. This study emphasizes the importance of understanding composition of the stainless steel as well as the amount of corrosion that occurs (Arnold and Suzuki 2003). In a study evaluating milk prot eins and bacterial adhesion, the interaction between stainless steel, the proteins in milk (a lpha-casein, beta-casein, kappa-casein, alphalactalbumin) or glutaraldehyde treated milk pr oteins, and the amount of biofilm formation of E. coli P. fragi, S. aureus, L. monocytogenes, and S. marcescens were observed (Barnes and others 1999). From th is study the researchers found that S. aureus, L. monocytogenes, and S. marcescens cell attachment is reduced by 20% or less when milk protein is present on a stainless steel surface than when milk is on a clean surface while E. coli and P. fragi show no difference in the amount of biofilm formation in clean or

PAGE 65

52 milk protein coated stainless steel. As concentrations of the milk protein became less the bacterial attachment of S. aureus, L. monocytogenes, and S. marcescens became greater (Barnes and others 1999). The researchers th ink that as the protei n layer on stainless steel surface was thicke r than the iron photoe lectron escape depth but when solution used was below 1% milk there was a sharp increase in iron signal and th is increase bacterial attachment (Barnes and others 1999). Th ey showed the glutaraldehyde treatment increased the attachment because cross-linking the proteins reduced the proteins ability to discourage bacterial attachment. They di scovered that the type of milk protein had no effect on biofilm levels. Ionic composition of the suspending medium had the greatest effect on clean-stainless-steel biofilm forma tion because the dissolved cations from the suspending medium shielded the electronegative charge on the surface of the stainless steel thus reducing biofilm formation. The au thors determined that the suspension media had no effect on attachment when milk protei ns were attached to the stainless steel surface except for CaCl2 and FeCl2 that encouraged biofilm development. The authors concluded that the reason FeCl2 increases absorption is because the ferrous ions can serve as a bridge between the bacteria and the milk proteins or it helps to cross-link proteins. CaCl2 increases biofilm formation because calciu m is a component of milk that is not found in the milk proteins but when it is rein troduced back in with the proteins it causes a conformational change which cau ses the absorbed proteins to facilitate attachment (Barnes and others 1999). Environment of Rubber There are 4 classes of r ubber specified in the 3-A Sanitary Standards for Multiple-Use Rubber and Rubber-Like Materials Used as Product Contact Surfaces. The two classes used on tankers ar e class 1 and class 3. Class 1 gaskets are heat exchanger

PAGE 66

53 gaskets, O-ring, CIP gaskets, flange gaskets, rotary seals and hoses These rubbers can tolerate product exposure and temperature sanitization up 149oC (300oF) and a chemical or bactericidal treatment up to 82oC (180oF). Class 3 rubbers can be used for cold applications such as milk and milk produc ts and air tubing, manhole and door gaskets, seals and hoses. These rubbers can to lerate product exposure and temperature sanitization up 49oC (120oF) and a chemical or bact ericidal treatment up to 82oC (180oF) (3-A Standards for Multiple 1999). All rubbers must meet other standards set out in the 3-A Standards including that they can not be toxic, they must meet cer tain absorption, aging, and compatibility with cleaner and sanitizer standa rds, they must be fabric ated under good manufacturing practices to 3-A Standards (3-A Standards for Multiple 1999). 3-A also states that although gaskets may come in different colors the color does not affect the sanitary conditi ons of the gasket. They note that different conditions in rubbers environment and cleaning regiment will produce different life times. The document states that a rubbers sanitary lifetime should be monitored so rubbers used for a similar purpose so they can be scheduled for replacement before cracks and sloughing appears (3-A Standards for Multiple 1999). Storgards and others (1999) wrote tw o papers on the influence old and new ethylene propylene diene mono mer (EPDM), nitrile butyl (N BR), polytetrafluoroethylene (PTFE) and Viton rubbers had on the formation of B. thuringiensis, Pseudomonas fragi, Pantoea agglomerans, and Pediococcus inopinatus biofilms when used in brewery or dairy processing environments. The author s concluded that new rubbers had different susceptibilities to biofilm formation that is dependent on the type of food processing

PAGE 67

54 environment they were used in and the type of bacteria creating the biofilm. For new the PTFE rubber were most resistant to biofilm build up in dairy conditions and NBR rubber surfaces were most resistant to biofilm bu ild up in brewery conditions. However, both rubbers were as cleanable as stainless stee l when cleaned at both hot dairy or cold brewery conditions. Also the ability to clea n new rubber differed beca use of the different surface properties of the rubber. In some aged rubbers the ability to remove biofilm from the surface was reduced (Storgards and others 1999). NBR that was aged to reflect 432 cleanings was found to more readily support biofilms. Both NBR and Viton have increased cracks and a rougher su rface. Viton was determin ed to be the rubber most quickly affected be aging. It was determined that EPDM was the most durable of the rubbers over time and the hygienic properties of PTFE were found to be almost unaltered over time (Storgards and others 1999). Some rubbers have shown to be bacteriostatic to certain groups of bacteria. NBR is bacteriostatic towards L. monocytogenes, Salmonella typhimurium, Staphylococcus epidermidis, Staphylococcus aureus, Y. enterocolitica and, E. coli O157:H7. EPDM is bacteriostatic towards S. epidermidis and S. aureus. Viton was not bacteriostatic (Ronner and Wong 1993). Review of Methodology Coliforms, Fecal Coliforms, and E coli The BAM directions identify the number of coliforms and fecal coliforms with 3tube most probable number (MPN) dilution seri es in lauryl tryptos e broth (LTB). For each positive tube in the MPN series a loopful is inoculated into br illiant green lactose bile (BGLB) broth to be incubated at 35oC for 24-48 hrs and examined for gas production. Also, each positive MPN tube is i noculated into Escherichia Coli (EC) broth,

PAGE 68

55 incubated at 45.5oC for 24-48 hrs, and examined for gas production. Tubes that test positive for fecal coliforms are streaked to Violet Red Bile Agar (VRBA) and incubated at 35oC for 18-24hrs to test for the presence of E. coli. The colonies should be checked for a flat, dark-centered colony with or without metallic sheen. Positive isolates can be streaked to PCA. These colonies can be fu rther tested for gas from lactose, citrate, methyl red-reactive compounds, Voges-Pros kauer (VP)reactive compounds, indole production, and gram stain (USDAs Bact eriological Analytical 2000). BAM also suggests a solid method for enumerating coliforms by creating pour plates with VRBA and incubating them at 35oC for 18 to 24 h. Ten presumptive coliforms should be inoculated into BGLB a nd incubated as described above to guarantee they are coliforms. As an alternative to VRBA agar, PetrifilmTM Coliform Count Plates (3M; St. Paul, MN) was developed for use by dairies and food production facilities. The Coliform Count Plates are made from VRBA, tetraz olium indicator, and a cold-water-soluble gelling agent. This combination enumerates colifo rms in a similar fashion to VRBA. E. coli also have similar ELISA tests and immunomagnetic beads as Salmonella. Coliform-produced enzyme -galactosidase breaks down X-Gal (5-Bromo-4-Chloro-3Indolyl-D-galactopyranoside) into 5-Bromo,4chloro-indoxyl intermediate which through oxidation produces a blue derivative. E. coli with -glucuronidase produced enzyme breaks down MUG into fluorescent de rivative (4-Methylumbelliferone). Using these two reactions a number of rapid tests for coliform and E. coli detection have been created such as ColiTM Complete (BioControl Systems, Inc.; Bothell, WA) and

PAGE 69

56 E*Colite test (Charm Sciences, Inc.; Lawrence, MA). Both ColiTM Complete can detect and E*Colite test can detect 1 coliform bacteria in 100mL. Detection Methods for Salmonella Salmonella can be detected though a combin ation of standard methods or commercially available test ki ts and extraction measures. The USDAs Bacteriological Analytical Manual (BAM) me thod for the isolation of Salmonella is a commonly used isolation method. BAM has different protocol s for different food products. The method suggests 24hr incubation at 35oC in a universal pre-enrich ment broth, followed by 24hr incubation in a selective enrichment me dia (Rappaport-Vassiliadis (RV) medium, tetrathionate (TT) brot h, selenite cystine (SC) broth), followed by the broth cultures being streaked to preformed plates of hektoen en teric (HE) agar, xylose lysine desoxycholate (XLD) agar, and bismuth sulfite (BS) agar and incubating them at 35oC for 24hrs (USDAs Bacteriological Analytical 2000). Along with standard isolation methods there are also commercial test and extraction kits. Most kits operate under sim ilar principles. The samples are cultured in the preenrichment and selective broths then they are exposed to a surface with antibodies that are specific to Salmonella spp. If Salmonella is present in the sample then the antigens on its surface will bind to the antibodies. Other material is rinsed away then enzyme labeled antibodies are bound to the surface; this generally produces a color reaction that is not present in samples without Salmonella. This is the principle under which the TECRA Salmonella Visual Immuno assay (TECRA International Pty Ltd) and other Enzyme-linked Immuno Assays ( ELISA) operate. Immunomagnetic beads operate on a similar principle as the ELISA test. Antibodies are on the surface of the beads. The Salmonella cells bind to the surface. The magnetic nature of the beads allows

PAGE 70

57 them to be retained during washing without losing them in the washes. The beads and the bacteria can then be plated onto so lid selective media. The TECRA Salmonella Visual Immunoassay can detect 1-5 CFU/ 25g sample and Dynal Anti-Salmonella immunomagnetic beads (Dynal Biotech, Oslo, Norway) can detect 1 CFU/25g of sample. The 1-2 Test (BioControl Systems, Inc. ; Bothell, WA) passes through a selective enrichment through a nonselective motility medium and Salmonella is immobilized by flagellar (Polyvalent H) antibodies. Detection Methods for Alicyclobacillus Some researchers use K agar to isolate Alicyclobacillus while others use AlibrothTM and AliagarTM. For the K agar procedure 3, 1mL porti ons of heat shock sample (10mL at 80oC for 10mins), and 18-20mL of K agar with each heat shocked sample to create three plates to test for Alicyclobacillus. Plates were incubated at 43+/-1oC for 3 days. Colonies will be cream-colored, raised, and opaque (Evancho and Walls 2001). An alternative to K agar is the use of Alibroth and Aliaga r for culturing and isolating Alicyclobacillus. The typical protocol used is to inoculate 100mL of Alibroth with 10mL of heat-shocked culture (75oC for 15 mins) and incubate at 45oC for 72 h. After 72 h Alibroth is streaked to Alia gar plates. If ther e is growth at 45oC after 2 days restreak 2 plates of Plate Count Agar (PCA) and 2 plates of Aliagar. One plate of each media is incubated at 45oC for at least 2 days while one plate of each media is incubated at 25oC for 4 days (Parish and Goodrich 2005). Detection of Aciduric, Yeast and Mold, Thermoduric, Mesophilic and Psychroduric Microorganisms The following protocols are taken from the Compendium of Methods for the Microbial Examination of Foods. Although other methods can be used for isolating and

PAGE 71

58 culturing the above types of microorganisms these are some of the most accepted. Acidophiles can be isolated using on pour pla ting a 1mL sample and Orange Serum Agar (OSA) and then incubating the sample at 30oC for 2 days (Hatcher and others 2001). Yeast and Mold count can be determined by pour plating 1mL of sa mple with acidified Potato Dextrose Agar (aPDA) a nd incubating the plates for 22-25 oC for 5 days (Beuchat and Cousin and others 2001). Mesophilic and Thermoduric organisms are determined by plating 1mL of sample with PCA and incubating the samples at 35 oC and 45-50oC respectively (Olson and Sorrels 2001). Psychr oduric counts are determined by spread plating 0.1mL of sample preformed PCA plates and incubating them at 7oC for 10 days (Vanderzant and Matthys 1965). DNA Sequencing The DNA is separated from the cell thr ough extraction and purification methods (one example of this procedure can be found in Molecular Cloning by Sambrook and others (1989)). Then primers are selected to bind to regions of interest in the DNA sequence. This region is then copied or amplified through cycles of denaturation, annealing and extension over an over agai n with the help of a thermocycler. Denaturation occurs at a high temperature, which separates the DNA into two separate strands. Before annealing begins the temperat ure is lowered and then primers bind to the DNA. The temperature is raised slightly and DNA polymerase bi nds to selected DNA regions and extension begins. During this process PCR polymerase progresses along the strand, replicating creating a copy of the target region. The final result is two double strands. The process is repe ated through several cycles, which leads to an exponential increase in copies of the target region (Entis and others 2001).

PAGE 72

59 These copies are then run through a gel. Th e region of interest cut out of the gel. The fragment is separated from the gel a nd then sequenced. Sequences are compared against a database of know sequences (W orobo 2005). The relationship between the database and the known sequences is determined. Biofilm Growth Characterization There have been many devices designed to grow biofilms such as a mini flow chamber (a variety of flow chambers), the Robbins device (some times referred to as the modified Robbins device), and the Chemostat. According to Ramos and others (2001), they commonly use the mini flow chamber b ecause it is easy to use, easy to monitor microscopically, and the results have been found to be statistically re producible under the same conditions; however, the biofilms in this system are difficult to access and the flow rates that can be used in the system are limited. The Robbins device is composed of several flow chambers that each have holders for holding a piece of material (like a piece of stainless steel) for biofilms to attach to (Ramos and others 2001). The Chemostat is a simple device. The top has openings for th e aseptic insertion of coupons that are hung from wire into a predetermined amount of liquid and bacteria cultu re (Keevil 2001). The interior of the Chemostat is titanium, and does not contain any Fe, Ni, Mn, or Cr that would affect biofilm formation (Keevil 2001). The Chemostat can control such parameters as temperature, dissolved oxygen concentration, and pH. The liquid in the system can be kept moving by a stir bar in th e bottom of the system and if desired liquid and culture can be pumped in and out of the system (Keevil 2001). Observation Methods Several methods have been developed fo r monitoring developi ng biofilms: electron microscopy, epiflourescent microscope, and confocal laser scanning microscopy. There

PAGE 73

60 are advantages and disadvantages to a ll the aforementioned systems. Electron microscopy is a poor choice for studying biofilm due to the dehydration and the eventual destruction of the biofilm st ructure. The foremost disadvantage to epiflourescence microscopy is that as the biofilm gets thicker it becomes harder to see clear images of the biofilm (Christensen and others 1999). Ther efore, epiflourescence microscopy is best used is best used for a single layer of cells in a specific region because all other regions will be out of focus. The epiflourescent microscope can be used with cameras to achieve images of biofilm development (Ramos and others 2001). The scanning confocal laser microscope corrects the problem experi enced by epiflouresce nt microscope by collecting returned fluorescen t light from only the thinnest focal plane afforded by the objective lens. Also, this type of micros copy can generate a three-dimensional image by scanning several planes interspersed at short distances (Christensen and others 1999). These three dimensional images can generally be done by any confocal-based software however the best software is Unix-based systems like IMARIS according to Christensen and others (1999). Other technologies such as Fluorescent in Situ Hybridization (FISH) and wide variety of cameras helped immensely in th e study of biofilm fo rmation. The general principle behind FISH is to fix cells to th e surface to which they are attached, and hybridize specifically targeted genes in the bacterias 16S or 23S rRNA sequence with a fluorescent labeled oligonucleot ides probe. FISH can help researchers identify placement of certain bacteria in the biofilm, as well as help them determine the growth rate of the cells by using the fluorescence intens ity (Ramos and others 2001).

PAGE 74

61 A Need for More Research In light of our limited knowledge on biofilm formation or transportation trucks in general, saying that more studies need to be done on the microbial aspects of transportation tankers is a tremendous understa tement. Future research is needed to understand what species of bact eria are present in citrus juice biofilms, how these biofilms form, what types of quorum sensing occur in these biofilms, how these biofilms form on stainless steel and rubber, how th ese biofilms form in the tanker truck environment, how these biofilms change with the material makeup or the conditions in the transportation tanker, the transportation pr ocess, and the time before cleaning; and how force, pressure, chemicals, age of the bi ofilm affect the cleani ng process. Although not all these questions will be ad dressed in this research, it is the hope of the author that this research will start to address some of th e most basic questions in hope that others will take the opportunity to expand upon this work to bring changes based in scientific fact to the tank wash industry.

PAGE 75

62 CHAPTER 3 MATERIALS AND METHODS Part I: Identification and Characteri zation of Microorganisms in Samples Sample Collection Survey samples were collected from wash station A located in central Florida from January to May of 2005. Tankers sample d carried either citrus juice or dairy products in their most recent load. Tankers ha ve gaskets in a variet y of shapes. Tankers in this study were classified as gasket type A or gasket type B. Figure 3.1 illustrates the differences between gaskets. Some tankers have a lip around the manway (Picture D) while others have a flat surface surrounding th e manway (Picture B). Type A gaskets tested in this study are made out of neoprene and designed to be set onto tankers with a flat manway surface and they have one large lip that extends into the manway to hold the gasket in place (Picture A). Type B gaskets tested in this study are made out a number of different types of rubber incl uding nitrile butyl rubber (NBR) or ethylene propylene diene monomer (EPDM) rubber; however they have th e same design (Picture B). They have a groove between two lips. The groove is de signed to be just big enough to place on manways that have a lip around the edge. Ga skets that meet the two afore-mentioned criteria were washed with a hot or ambien t temperature wash regimens. A hot wash regimen consisted of a hot temperature pressuri zed spray with an alka line detergent, a hot temperature alkaline detergent wash, a hot temperature chorine wash, and an ambient temperature acid sanitizer wash; and an ambien t temperature wash regimen consisted of a ambient temperature pressurized spray with an alkaline detergent, a

PAGE 76

63 A C B D Figure 3-1. Type A and Type B manway styles and gasket types. A.) Gasket type A, B.) Gasket B, C.) Manway lid associated with gasket type A D.) Manway lid associated with gasket type B. ambient temperature alkaline detergent wash, a ambient temperature chorine wash, and a ambient temperature acid sanitizer wash. Surfaces of the gasket that were exposed to the liquid product inside of the tanker we re swabbed with a sterile SpongesicleTM (a sponge on a stick with 10ml of nutrient buffer in a se aled bag) (Biotrace In ternational; Bridgend, Wales) using firm and even pressure. E ach sample was labeled with a consecutive

PAGE 77

64 number. Samples were kept at 4oC until time of analysis (no longer than 24 h after sampling). Sample Preparation Microbiological materials for this re search were manuf actured by Becton, Dickson, and Company (Franklin Lakes, NJ) unl ess otherwise specifie d. 90 mL of sterile buffer peptone water (BPW) were added to each SpongesicleTM bag. Bags were homogenized by hand and then were used for the enumeration of the following types of microorganisms: psychroduric; mesophilic; thermoduric; yeast and mold (YM); and aciduric. Additionally the presence or absence of coliforms, Escherichia coli, Alicyclobacillus spp., and Salmonella spp. was tested. Tests for Streptococcus, and Staphylococcus detection and Most Probable Numb er (MPN) for coliforms, fecal coliforms, E. coli were conducted later if necessary as discussed later in the materials and methods. Sample Analysis Psychroduric, mesophilic, thermoduric, ye ast and mold; and aciduric enumeration and characterization Psychroduric plates were obtained by spr ead plating 0.5 mL of sample on 2 plates of plate count agar (PCA) and incubating them at 6oC for 10 days (Vanderzant and Matthys 1965). Mesophilic plates were obt ained from pour plating 1mL or 0.1mL of sample with PCA and incubating the sample for 5 days at 35oC. This method was altered slightly from Compendium of Methods for the Microbial Examination of Foods (Downes and Ito 2001), which states that PCA plat es should be incubated for 2 days at 35oC. Compendium (Downes and Ito 2001) also notes that 2 days may not be sufficient time to allow for visualization of injured cells (Swans on and others 2001). Since the cells in the

PAGE 78

65 samples may be injured by clean ing agents, during an initial trial of the materials and methods, PCA plates were observed over the firs t eight days of incubation. Observations of PCA plates noted that most colonies had formed by day 5. A paired T-test on the data from part II of this experiment conducted on the mesophilic counts on day 2 and 5 show that there was a significant difference between the counts. The same incubation period was allowed thermoduric plates. Thermoduric plates were obtained from pour plating 1mL of sample with PCA and incubating the sample at 50oC (Olson and Sorrells 2001). Yeast and Mold (YM) plates were obtained from pour plating 2mL of sample with acidified potato dextrose agar (aPDA; pH=3.5) and incubating the sample for 30oC (Redd and others 1986) for 2 days. Redd and ot hers (1986) recommend a three-day incubation period; however, two days is commonly chos en in industry to e xpedite shipping of product if it can be proven that there is no significant di fference between the counts on day 2 and day 3. Two days was chosen over 3 days or 5 days for incubation of YM plates because a T-test of initial trials from pa rt I, and data from part II indicate that there was no significant difference between YM counts at 2, 3 or 5 days. Aciduric plates were obtained from pour plating 3mL or 1mL of sa mple with orange serum agar (OSA) and incubating the sample for 2 days at 30oC (DIFCO Manual 1984). Plate counts were obtaine d and recorded. The quant ity of different colony morphologies present and their descriptions were recorded. Typi cal colonies of the different morphologies were sel ected, and streaked to separate plates of the agar from which they were isolated. Plates were incu bated at the appropriate temperature for 12 days. Colony morphologies and diameter size (mm) were reco rded. Colonies were then gram stained (Tortora and others 1998), a nd catalase/oxidase te sts were performed.

PAGE 79

66 Gram-positive rods were then transferred to s porulation agar (1 L of Nutrient broth, 15 g of Bacto agar, 0.030 g MnCl3, and 0.030 g CaCl2 mixed together and autoclaved at 121oC at 15 psi for 30 m (Huang and others 2001)) for 3 days at the appropr iate temperature. Spore stains were performed (Tortora and others 1998). Coliform, fecal coliform, and E. coli detection E*Colite test (Charm Sciences, Inc.; Lawrence, MA) for the presence or the absence of coliforms or E. coli and a PetrifilmTM Coliform Count Plat e (3M; St. Paul, MN) were performed according to the manufact urers instructions. If coliforms were present in the E*Colite samples or on the PetrifilmTM, the Most Probable Number (MPN) for coliforms would be performed using the orig inal sample as direct ed in chapter 4 part E of the Bacterial Analytical Manual (BAM) (2002). Also if coliforms were present in the E*Colite sample or on the PetrifilmTM; 1ml of the Ecolite sample or colonies from the PetrifilmTM would be inoculated into a tube of E. coli (EC) broth containing 4methylumbelliferyl-D-glucuronide (EC-MUG) and incubated at 44.5oC to determine if fecal coliforms or E. coli were present. If fecal coliforms or E. coli were present an MPN for fecal coliforms or E. coli would be performed as directed in chapter 4 part E of the BAM (September 2002). If presumptive E. coli was present in the E*Colite sample; 1 mL from the E*Colite bag would be inoculated into a tube of EC-MUG and incubated at 44.5oC. If the EC-MUG tube came back positive for E. coli an MPN for E. coli would be performed as directed in chapter 4 part E of the BAM (2002) and a streak on Levines Eosin-Methylene Blue (L-EMB) agar woul d be done to look for a typical colony morphology. All colony morphologies that appe ared typical were c onfirmed presumptive E. coli using a BBLTM EnterotubeTM II. Samples were frozen for subsequent 16S rRNA identification.

PAGE 80

67 Streptococcus spp. and Staphylococcus spp. detection E*Colite bags that were yellow and fluores cent, or blue and fluorescent but were not E. coli were streaked onto PCA agar and incubated at 35oC for 24 h. Representative colonies were selected, and gram stained. Any gram-positive cocci were streaked to Baird-Parker Medium (OXOID LTD.; Basingstoke, Hampshire, United Kingdom) with Egg Yolk-Tellurite Emulsion (OXOID, LTD.; Basingstoke, Hampshire, United Kingdom), or inoculated into Streptococcus Faecalis (SF) Medium as directed by the DIFCO Manual (1984). Samples were frozen down for 16S rRNA identification. Salmonella spp. detection Testing for the presence of Salmonella spp. was done using TECRA Salmonella Visual Immunoassay (TECRA International Pty. Ltd.; Frenchs Forest, Australia) following modified version of Enrichment Protocol 7 and method for Performing the Immunoassay of the manufacturers directions. The enrichment Protocol 7 was modified in the following way: 20mL of the origin al sample in 180mL of BPW for 22 h at 35oC followed by an enrichment of 1mL sample of the BPW in 9mL of Tetrathionate (TT) Broth for 24h at 35oC and 0.1mL of sample to 9.9mL Ra ppaport-Vassiliadis (RV) Broth for 24h at 42oC, followed by a 1mL of each broth culture to be inoculated into 9mL of M broth incubated at 35oC for 8 h. Samples were then heat shocked before the TECRA Salmonella Visual Immunoassay was used. Alicyclobacillus spp. detection Alicyclobacillus spp. were enumerated using the heat shock method described by Parish and Goodrich (2005). The samples were then streaked to duplicate plates of PCA and duplicate plates of Alibro th agar plates (Parish a nd Goodrich 2005). One of each plate type was incubated at 25oC and one of each plate type was incubated at 50oC.

PAGE 81

68 Plates were checked at 24 and 48 h for growth. If growth was only present on the Alibroth agar plates at 50oC, it was assumed to be Alicyclobacillus spp. 16S DNA and 28S rRNA PCR Identification The 16S DNA PCR identification of b acteria and 28S rRNA (D2 expansion segment) rRNA region PCR identification of yeast was done by Accugenix, Inc. (Newark, DE). The company extracts DNA from a pure isolate, the S rRNA gene is amplified, sequenced, the resu ltant extension products are separated and it is then matched in order of increasing genetic dist ance to relevant sequences in a database (Accugenix 2005). The yeast identificati on was done by sequencing of the D2 expansion segment of the large subunit rRNA ge ne and comparing to a database of yeast sequences (Accugenix 2005). The three E. coli strains 36, 87, 113; one presumptive S. aureus, Yeast (OSA) 113B, and gram-positive rod 36C, gram-positive 36D, gram-negative rod 113C from the Mesophilic colonies were sent for identification. Statistical Analysis Using the psychroduric, mesophilic, yeast and mold, and aciduric counts the number of CFU/cm2 and the CFU/total gasket were ca lculated and transformed to Log10 to reduce the effect of outliers on the data (if there were 0 CFU/cm2 or 0 CFU/total gasket, it was changed to a count of 1 CFU/cm2 or 1 CFU/total gasket prior to the log transformation). Analysis of Variance (ANOVA ) was used to determine if samples were significantly different (p>0.05) from each ot her depending on gasket type, product type, wash type, or any combination thereof. Minitab release 14 (Minitab, Inc.; State College, PA) was used for statistical analysis.

PAGE 82

69 Part II: Biofilm Development and Removal Liquid Sample Preparation Whole, homogenenized UHT milk (Par malat Finanziaria S.p.A, Italy) was inoculated with representative bacteria and yeast, obtained from E. coli-positive tanker samples. The milk was also inoculated with a fluorescent-tagged E. coli that was created using TransformAidTM Bacterial Transformation Kit (F ermentas; Burlington, Ontario, Canada) and the E. coli from sample 36. Standard growth curves Standard growth curves were created from 24-hour cultures of one yeast and four bacteria. Dilutions of each sample were done from 100 to 10-9. These dilutions 10-4 to 10-9 were pour plated out with SPC agar onto sterile Petri plates (Fisher Scientific, International; Pittsburg, PA). Plates were incubated at 37oC for 24 h and were then counted. The remainder of the culture wa s used to create a 1/2, 1/4, 1/8, and 1/16 dilution of sample in nutrient broth. Dilute d samples and a pure culture sample where viewed until the spectrophotometer at 600nm. The data from the above plate count was used to determine how many bacteria or yeast was in the original sample. The amount of microorganisms that would be in 1/2, 1/4, 1/8, and 1/16 dilution were calculated. The spectrophotometer measurements were plotted ag ainst the number of bacteria or yeast. The procedure was repeated three times for each sample and a final standard curve was created for each bacteria or yeast (Appendix C). The standard curves were used to help add the approximate quantities of each bacteria or yeast to achieve the following formula outlined in Table 3-1. The formula was crea ted by assuming that the initial day plate count for a sample of milk was 300 colony fo rming units (CFU) per mL. Therefore, a 5 L sample of milk would

PAGE 83

70 Table 3-1. Types and number of CFU of microo rganisms found in target inoculated milk sample. Sample ID Media Type Type (Oxidase/Catalase) CFU/5L 113B OSA Yeast (-/+) 150,000 36D Mesophile Pos Cocci(+/+) 300,000 36C Mesophile Pos Rods (+/+) 1,000,000 113C Mesophile Neg Rods(+/+) 25,000 36 E. coli (E*Colite)Fluorescent 25,000 1,500,000 have 1,500,000 CFU. The representative colonies selected from psychroduric, mesophilic, thermoduric, YM, and aciduric plat es of the three samples containing of E. coli (36, 87, and 113) were compared. The charac terization of colonies in sample 87 was very different from 36 or 113. Sample s 36 and 113 were compared and three characterizations of bacteria and one type of yeast were found to be similar between the two samples. A sample from one of these sets was selected for the culture (see Table 3.1). The quantities of microorganisms inoc ulated into model were chosen by the following criteria: 1) gram-positive spore-form ing bacteria are very likely to be found in pasteurized milk because of thei r ability to survive pasteurization and grow at refrigerator temperatures (Cousin 1982, Washam and othe rs 1977), Frank 2001); therefore, they make up the majority of the microorganisms in the sample, 2) since some gram-positive cocci such as (Micrococcus and Enterococcus) can survive pasteurization and grow at refrigerator temperatures (Richter and Ve damuthu 2001), the gram-positive cocci were placed in the sample in the next largest quality, 3) yeast growth in pasteurized milk is not typical so a low level of yeast was added (Richter and Vedamuthu 2001), 4) since there can be no more that 10 coliforms per mL according to USDAs Grade A Pasteurized Milk Ordinance(FDA/CFSAN National Conference 2003), only 5 CFU of E. coli 36 per

PAGE 84

71 mL (or total 25000 CFU of the E. coli) and 5 CFU of the gram-negative rod 113C per mL (or total 25000 CFU of the gram-negative r od) were added to the sample. Model of Liquid Transportation Tanker Manway Figures 3-2 and 3-3 illustrate the set for the model manway assembly. Sterile gloves were worn throughout this procedure. Pieces of the manway assembly that could not be sterilized were san itized with chlorinated water. A stainless steel manway assembly has been obtained from liquid tr ansportation tanker Manuf acturer B. The manway lid was set on a plastic washbasin (Pictu re A). The rubber gasket (that would be A B C D E F Figure 3-2. Manway lid set up picture set 1. A.) Manway lid on the wash basin, B.) Gasket type A on the manway lid, C.) Olson vent on the manway lid cover, D.) spraying system placed in the center, E.) sterile weights placed on the spraying system, F.) plastic hose conn ecting spraying system to pump. defined as gasket type B according to the defi nition provided in Part I of this experiment) was placed on the lip of the manway (Pictu re B). An Olson vent was placed on the manway lid cover (Picture C). A spraying apparatus was set in the center of the washbasin (Picture D) and weighted down with 4 sterile bottle weights (Picture E). A half-inch diameter plastic hose was connected to the spraying apparatus (Picture F). The

PAGE 85

72 dust cover and manway lid were placed down over the top of the manway (Picture H) and were held in place by five clamps in the manw ay (Picture I). The other end of the halfinch hose was connected to a submersible fountain pump (Peaktop Technologies) that was used to circulate milk inoculum from a container in a 4oC water bath (Picture G) to a spraying apparatus in the cente r of the washbasin. The milk would then drain through a 1-inch plastic tube attached to the drain of the washbasin with PVC piping as seen in Picture K. The spraying device was progr ammed using a timer/controller (Fisher; Pittsburg, PA) (Picture J) to spray the gasket for 5 s every 15 m for three days to mimic sloshing in a moving tanker tr aveling across the United States. The model tanker was G H I J K Figure 3-3. Manway lid set up picture set 2. G.) Incubator were milk inoculum is stored, H.) Dust cover closed over the manw ay lid, I.) manway lid was clamped down, J.) timer/controller, K.) complete model set up. placed in an environmentally controlled chamber that was set up to reproduce the temperature fluctuations on a typical centr al Florida July day in where the high temperature is approximately 90oF (32.2oC) and the low is 70oF (21.1oC) (Southeast Regional Climate Center 2005). Calibrated HOBO H08-002-02 data loggers (Onset

PAGE 86

73 Computer Corporat ion; Bourne, MA) monitored the temperature fluctuations of the room, water bath, and milk inoculum. Gasket Treatment Figure 3-4 illustrates how the gasket was wa shed and prepared. Sterile gloves were worn through out this procedure. After expos ure, the gasket (Picture L) was removed and cut into four pieces using sterile razor blades (Picture M). One piece was left untreated (control), and the remaining pi eces were subjected to three different cleaning regimens: 1) 15s detergent wash (Picture N) followed by a water rinse; 2) 15 s detergent wash, a 10s scrub and 4 m 50 s sanitizer soak, and water rinse; and 3) 15 s detergent wash, a 10 s scrub and 4 m 50 s sanitizer soak, a water rinse, and a hot water treatment (Picture O) of the gasket at 160oF (71.1oC) for 15 m and 185oF (85oC) for 20 m. A food-grade chlorine detergent and a quaternary ammonium based sanitizer commonly used at tank wash stations were used to wash gaskets. The highest concentration of both products that could be used according to label directions was dispensed into bottles of sterile deionized water. Bottles with water, detergent, and sanitizer treatments were held in a 90oF (32.2oC) water bath until they were needed. Detergent and sanitizer treatments were pl aced into premarked tubs with premarked brushes typical of those used at tank wash es. Hot water treatments were created by placing sterile deionized water into two beaker s on separate hot plat es and heating them to the appropriate temperatures. After each gasket piece was washed, two pieces (1/2 inch in length) were removed from the center of the gasket piece (Pictures P&Q). One piece has a thin piece of the top (Picture R), the inside surface, and the outsi de surface removed with a sterile razor blade and placed on a slide to be studied by fluor escence microscopy. The second piece has a

PAGE 87

74 thin piece of the top, and the inside surface rem oved with a sterile razor blade to be used for the scanning electron microscopy. The remaining piece was swabbed using a SpongesicleTM with 10mL of nutrient buffer (Picture S). L M N O P Q R S Figure 3-4. Manway lid set up picture set 3. L.) Model setup after 3 days, M.) Gasket being cut into 4 pieces with a sterile scalpel blade, N.) Gasket washed in detergent, O.) Gasket receiving a h eat treatment, P& Q.) Gasket being prepared for microscopy, R.) Surface section being removed for microscopy, S.) Gasket being swabbed. Microbial Analysis of Gasket The swab was used to determine the total plate count, the total amount of E. coli using the USDAs Bacteriol ogical Analytical Manual Online Chapter 4 Part I Subpart G

PAGE 88

75 for solid media method enumeration of injured coliforms (2002), and a count to determine the number of yeast on aPDA. Pour plates of PCA from 2 mL of sample to a dilution of 10-7 were done to determine the total plate count and pour plates of aPDA from 2 mL of sample to a dilution of 10-1 were done to determine the total yeast count. A random sampling of microorganisms was selected from the total plat e count and acidified potato dextrose agar to determine the a pproximate composition of the sample. An E*Colite bag was prepared for each sample using the method described earlier to check for presence of low levels of E. coli that may not be detected on the Tryptic Soy Agar and Violet Red Bile Agar (VRBA) with M UG (OXOID, LTD.; Basingstoke, Hampshire, United Kingdom) plates. Scanning Electron Microscopy Gasket samples were fixed with 3% glut araldehyde with 1500 ppm Ruthenium Red (RR) in 0.1M cacodylate (CaCo) buffer at pH 7.2. at room temperature. Samples were then washed 3 times with 0.1M CaCo and then were en bloc stained in 1500 ppm RR with 0.1M CaCo buffer at pH 7.2 at 4oC overnight (Luft 1971). Samples were then rinsed twice with 0.1M CaCo buffer for 5 m and dehydrated using ten-step ethanol dilution series for 10 m each dilution. A critical point drier (Ladd Research Industries; Williston, VT) using bone dry CO2 was used to completely dry the samples. Each sample was mounted, coated for 90 s with gold/palladi um (80/20) (Ladd Research Industries; Williston, VT), and viewed and photographed using a Hitachi S-530 scanning electron microscope at 80x, 600x, and 4000x magnifica tion (Chumkhunthod and others 1998). Fluorescence Microscopy Slides containing the inside, top and outside surface of the gasket s inter lip for each of the four samples were viewed an Ol ympus BZ61 Microscope (Olympus Corp.;

PAGE 89

76 Melville, NY) using the 10x objective and a filt er for green fluorescen t protein. Pictures were taken of each piece of the gasket a nd were printed using a Hewlett-Packard Business Inkjet 1200 (Hewlett-P ackard Development Company, L.P.; Palo Alto, CA). Statistical Analysis The experiment was replicated six times to determine if any of the cleaning treatments provided a significant difference th e reduction in the number of the coliforms and mesophilic microorganisms in compar ison to each other using ANOVA (Minitab release 14). Prior to analysis by ANOVA the nu mber of coliforms CFU/total gasket and the number of mesophiles CFU/total gasket were calculated and transformed to Log10 to reduce the effect of outliers on the data (if there were 0 CFU/total gasket it was changed to a count of 1 CFU/total gasket prior to th e log transformation). Log reductions of the three treatments were calculate d. The log reductions were us ed to determine if there was a significant difference in the ANOVA (p>0.05).

PAGE 90

77 CHAPTER 4 RESULTS Part I: Sample Identification and Characterization Psychroduric, Mesophilic, Thermoduric, Yeast and Mold, and Aciduric Microorganism Enumeration and Characterization A total of 126 tankers were sampled in this study. After characterization, nine samples for each of the eight types of gasket combinations were randomly selected. The number of microorganisms per square cm2 and the number of microorganisms on the total gasket were calculated. Appendix A su mmarizes data from the 72 observations, and includes the raw data, the num ber of microorganisms per centimeter squared, and the number of microorganisms per total gask et. ANOVA was run comparing each variable type with the number of microorganisms pe r centimeter squared, and the number of microorganisms on the total gasket for each test. Then Fishers Protected Least Significant Difference Test at a 95% confidence interval was run on groups that had significant differences. The results can be found in Table 4-1 to Table 4-14. A percentage of different characterizations for each media type was obtained for each sample and an overall percentage was calcu lated for each gasket type. The top two components of the microflora found for each me dia type is summarized in Table 4-15.

PAGE 91

78Table 4-1. Products effect on aci duric, yeast and mold, psychrodur ic, and mesophile counts per cm2 of the gasket. Tanker Aciduric/cm2 Yeast and Mold/cm2 Psychroduric/cm2 Mesophile/cm2 Juice 6.7.4Ea 2.5.9Ea 8.4 x101.6Ea 8.5 x101.2Ea Dairy 1.3x101Ea 3.4.1Ea 1.6 x102.3 x101Ea 1.6 x101.2 x101Eb a, b, ceach letter indicates a grouping that is not statistically significantly different Table 4-2. Products effect on aciduri c, yeast and mold, psychroduric, and mesophile counts per total gasket. Tanker Aciduric Total Yeast and Mold TotalPsychroduric Total Mesophile Total Juice 3.4 x103.1 x103Ea 1.0 x103.9 x103Ea 2.8 x102.4 x102Ea 4.3 x103.3 x103Ea Dairy 6.4 x103.9 x103Ea 1.3 x103.6 x103Ea 3.8 x103.6 x104Ea 8.0 x103.0 x103Ea a, b, ceach letter indicates a grouping that is not statistically significantly different

PAGE 92

79Table 4-3. Gaskets effect on aciduric, yeast a nd mold, psychroduric, and mesophile counts per cm2 of gasket. Gasket Shape Aciduric/cm2 Yeast and Mold/cm2 Psychroduric/cm2 Mesophile/cm2 A 5.7.1Ea 4.4.2Ea 2.9.3 x101Ea 7.6.6Ea B 1.4 x101.8 x101Eb 1.4.7Eb 1.1.6Ea 1.7 x102.1 x101Eb a, b, ceach letter indicates a grouping that is not statistically significantly different Table 4-4. Gaskets effect on aciduric, yeast and mol d, psychroduric, and mesophile counts per total gasket. Gasket Shape Aciduric Total Yeast and Mold TotalPsychroduric Total Mesophile Total A 3.2 x103.1 x103Ea 7.6 x102.8 x103Ea 3.5 x103.6 x104Ea 4.2 x103.4 x103Ea B 6.6 x103.3 x103Eb 1.5 x103.7 x103Ea 5.1 x102.2 x103Ea 8.1 x103.9 x103Eb a, b, ceach letter indicates a grouping that is not statistically significantly different

PAGE 93

80Table 4-5. Wash temperatures effect on aciduric, yeast and mold, psyc hroduric, and mesophile counts per cm2 of gasket. Tanker Aciduric/cm2 Yeast and Mold/cm2 Psychroduric/cm2 Mesophile/cm2 Hot 7.1.1Ea 3.3.3Ea 1.1.8Ea 1.2 x101.5 x101Ea Cold 1.3 x101.9 x101Ea 2.4.4Ea 2.9.3 x101Ea 1.2 x101.9 x101Ea a, b, ceach letter indicates a grouping that is not statistically significantly different Table 4-6. Wash temperatures effect on aciduric, yeast and mold, psychroduric, a nd mesophile counts per total gasket. Tanker Aciduric Total Yeast and Mold TotalPsychroduric Total Mesophile Total Hot 3.6 x103.6 x103Ea 1.0 x103.8 x103Ea 3.1 x103.9 x101Ea 6.1 x103.2 x103Ea Cold 6.2 x103.2 x103Ea 1.3 x103.7 x103Ea 9.9 x102.6 x103Ea 6.2 x103.0 x103Ea a, b, ceach letter indicates a grouping that is not statistically significantly different

PAGE 94

81Table 4-7. Product and Gaskets eff ects on aciduric, yeast and mold, psyc hroduric, and mesophile counts per cm2 of the gasket. Tanker Aciduric/cm2 Yeast and Mold/cm2 Psychroduric/cm2 Mesophile/cm2 Juice, A 4.5.2Ea 1.3.3Ea 0.14.47Ea 7.4.4Ea Juice, B 8.9.9Ea 3.7.9Ea 0.84.2Ea 9.6.6Ea Dairy, A 6.8.1 x101Ea 1.4.3Ea 5.6.8 x101Ea 7.7.1 x101Ea Dairy, B 1.9 x101.2 x101Eb 5.1.3Ea 1.4.9Ea 2.5 x101.7 x101Eb a, b, ceach letter indicates a grouping that is not statistically significantly different Table 4-8. Product and Gaskets effects on aciduric, yeast and mold, psychroduric, and mesophile counts per total gasket. Tanker Aciduric Total Yeast and Mold TotalPsychroduric Total Mesophile Total Juice, A 2.5 x103.5 x103Ea 7.4 x102.8 x103Ea 1.7 x102.8 x102Ea 4.1 x103.0 x104Ea Juice, B 4.2 x103.6 x103Ea 1.3 x103.2 x103Ea 3.9 x102.0 x103Ea 4.5 x103.5 x103Ea Dairy, A 3.8 x103.4 x103Ea 7.9 x102.8 x103Ea 6.9 x103.2 x104Ea 4.3 x103.9 x103Ea Dairy, B 9.1 x103.0 x104Eb 1.8 x103.0 x103Ea 6.4 x102.4 x103Ea 1.2 x104.3 x104Eb a, b, ceach letter indicates a grouping that is not statistically significantly different

PAGE 95

82Table 4-9. Product and Wash Temperature s effects on aciduric, yeast and mold, psychroduric, and mesophile counts per cm2 of the gasket. Tanker Aciduric/cm2 Yeast and Mold/cm2 Psychroduric/cm2 Mesophile/cm2 Juice, Hot 7.0.1Ea 2.8.5Ea 0.45.81Ea 1.0 x101.9Ea,b Juice, Cold 6.4.8Ea 2.2.3Ea 0.53.2Ea 6.9.2Eb Dairy, Hot 7.1.2Ea 2.0.2Ea 1.8.8Ea 1.5 x101.9 x101Ea,b Dairy, Cold 1.9 x101.4 x102Eb 4.5.1Ea 5.2.8 x101Ea 1.8 x101.5 x101Ea a, b, ceach letter indicates a grouping that is not statistically significantly different Table 4-10. Product and Wash Temperatures effects on aciduric, yeast and mold, psyc hroduric, and mesophile counts per total ga sket. Tanker Aciduric Total Yeast and Mold TotalPsychroduric Total Mesophile Total Juice, Hot 3.6 x103.8 x103Ea 1.2 x103.1 x103Ea 3.2 x102.4 x102Ea 5.2 x103.6 x103Ea,b Juice, Cold 3.1 x103.6 x103Ea 7.8 x102.8 x103Ea 2.5 x102.0 x103Ea 3.5 x103.7 x103Eb Dairy, Hot 3.6 x103.6 x104Ea 7.9 x102.5 x103Ea 1.7 x103.9 x103Ea 7.1 x103.2 x103Ea,b Dairy, Cold 9.3 x103.1 x104Eb 1.8 x103.4 x103Ea 5.9 x103.2 x104Ea 8.9 x103.2 x104Ea a, b, ceach letter indicates a grouping that is not statistically significantly different

PAGE 96

83Table 4-11. Gasket and Wash Temp eratures effects on aciduric, ye ast and mold, psychroduric, and mesophile counts per cm2 of th e gasket. Tanker Aciduric/cm2 Yeast and Mold/cm2 Psychroduric/cm2 Mesophile/cm2 A, Hot 5.5.1Ea 1.6.4Ea,b 1.2.2Ea 7.9.5Ea A, Cold 5.8Ea 1.2.2Ea 4.5.8 x101Ea 7.3.0 x101Ea B, Hot 8.6.8Ea 3.2.1Ea,b 1.0.4Ea 1.7 x101.8 x101Ea B, Cold 2.0 x101.3 x101Eb 5.6.7Eb 1.2.8Ea 1.8 x101.4 x101Ea a, b, ceach letter indicates a grouping that is not statistically significantly different Table 4-12. Gasket and Wash Temp eratures effects on aciduric, ye ast and mold, psychroduric, and mesophile counts per total gas ket. Tanker Aciduric Total Yeast and Mold TotalPsychroduric Total Mesophile Total A, Hot 3.1 x103.0 x103Ea 9.2 x102.9 x103Ea 1.5 x103.9 x103Ea 4.4 x103.3 x103Ea A, Cold 3.2 x103.1 x103Ea 6.1 x102.8 x103Ea 5.6 x103.2 x104Ea 4.1 x103.6 x103Ea B, Hot 4.1 x103.2 x103Ea 1.9 x103.8 x103Ea 4.7 x102.1 x103Ea 7.9 x103.6 x103Ea B, Cold 9.2 x103.1 x104Eb 1.1 x103.3 x103Ea 5.6 x102.3 x103Ea 8.3 x103.1 x104Ea a, b, ceach letter indicates a grouping that is not statistically significantly different

PAGE 97

84 Table 4-13. Product, Gasket, and Wash Temp eratures effects on aciduric, yeast and mold, psychroduric, and mesophile counts pe r cm2 of the gasket. Tanker Aciduric/cm2 Yeast and Mold/cm2 Psychroduric/cm2 Mesophile/cm2 Juice, A, Hot 6.3.4Ea 2.3.5Ea,b 0.28.66Ea 9.1.8Ea Juice, A, Cold 2.8.0Ea 0.31.38Ea 0.0.0Ea 5.8.1Ea Juice, B, Hot 7.8.0Ea 3.3.8Ea,b 0.61.95Ea 1.1 x101.4Ea Juice, B, Cold 10.1.3 x101Ea 4.0.2Ea,b 1.1.1Ea 7.9.5Ea Dairy, A, Hot 4.7.2Ea 0.94.7Ea 2.2.3Ea 6.7.5Ea Dairy, A, Cold 8.8.5 x101Ea 1.9.4Ea,b 9.1.5Ea 8.8.2 x101Ea Dairy, B, Hot 9.5.7Ea 3.1.7Ea,b 1.4.4Ea 2.2 x101.4 x101Ea,b Dairy, B, Cold 2.9 x101.7 x101Eb 7.1.2 x101Eb 1.3.6Ea 2.7 x101. 1 x101Eb a, b, ceach letter indicates a grouping that is not statistically significantly different

PAGE 98

85Table 4-14. Product, Gasket, and Wash Temper atures effects on aciduric, yeast and mold, psychroduric, and mesophile counts per total gasket. Tanker Aciduric Total Yeast and Mold TotalPsychroduric Total Mesophile Total Juice, A, Hot 3.5 x103.5 x103Ea 1.3 x103.5 x103 Ea 3.4 x102.1 x102Eb,c 5.1 x103.3 x103Ea Juice, A, Cold 1.6 x103.4 x103Ea 1.7 x102.1 x102Ea 0.0.0Eb,c 3.2 x103.6 x104Ea Juice, B, Hot 3.6 x103.6 x103Ea 1.2 x103.7 x103Ea 5.0 x102.5 x102Eb,c 5.3 x103.1 x104Ea Juice, B, Cold 4.7 x103.3 x104Ea 1.4 x103.5 x103Ea 2.9 x102.5 x103 b,c 3.7 x103.5 x104Ea Dairy, A, Hot 3.5 x103.7 x103Ea 5.3 x102.6 x102Ea 2.7 x103.3 x103Ea,c 3.7 x103.5 x103Ea Dairy, A, Cold 8.4 x103.1 x103Ea 1.0 x103.5 x103Ea 1.1 x104.1 x104Ea 4.9 x103.5 x103Ea Dairy, B, Hot 3.6 x103.8 x103Ea 1.1 x103.0 x103Ea 6.5 x102.6 x103Ea,c 1.1 x104.9 x103Ea,b Dairy, B, Cold 1.3 x104.1 x103 Eb 2.5 x103.1 x103 Ea 6.2 x102.2 x103Ea,c 1.3 x104.1 x103Eb a, b, ceach letter indicates a grouping that is not statistically significantly different

PAGE 99

86Table 4-15. The top two bacteria l characterizations on different gasket and media types. Gasket Type Yeast and Mold Aciduric Mesophile Thermoduric Psychroduric Juice, A, Hot Yeast (32.6%) GPC oxcat+ (56%) GNR ox+ cat+ (50%) Yeast (100%) GPC oxcat+ (14%) Yeast (25%) GPR oxcat(100%) GNR oxcat(45%) Juice, A, Cold Yeast (98.5%) Yeast (42%) GPC oxcat+ (53%) GPR oxcat+ (25%) Mold (1.5%) GPC oxcat+ (34%) GPR oxcatspores/ no spores (8.4/8.8%) Mold (13%) Yeast (100%) Juice, B, Hot Yeast (85%) Yeast (51.2%) GPC oxcat+ (36%) GPR oxcat+ (33%) GNR ox+ cat+ (49%) Mold (15%) GPC oxcat+ (14%) Yeast (19.3%) GNC ox+ cat(11%) Yeast (22%) Juice, B, Cold Yeast (89.3%) GPC oxcat+ (36%) GPC oxcat+ (39%) GPR oxcat-, spore (100%) Yeast (47%) Mold (10.7%) Yeast (31.5%) Yeast (26%) GNR ox+ cat+ (19%) Milk, A, Hot Yeast (89.9%) Yeast (36%) GPC oxcat+ (39%) Mold (10.1%) GPC oxcat+ (31.5%) Yeast (21%) GPR ox+ cat+ (100%) GNR ox+ cat+ (83%) Milk, A, Cold Yeast (70.5%) Yeast (32.4%) GP C oxcat+ (41%) Mold (33%) GNR ox+ cat+ (70%) Mold (18.5%) GNR o+ c(14%) GPR ox+cat-, spores (13%) GPC oxcat+ (12%) Yeast (21%) GPR oxcat+ (19%) Yeast (22%) Milk, B, Hot Yeast (87.9%) Yeast (39.2%) GPC oxcat+ (46%) GPR oxcat+ (13%) GNR ox+ cat+ (79%) Mold (5.5%) GPC oxcat+ (20%) Y east (11%) Mold (13%) Yeast (9.1%) Milk, B, Cold Yeast (86.5%) GPC oxcat+ (44%) GPC oxcat+ (48%) GPR oxcat+ (60%) GNR ox+ cat+ (50%) Mold (13.5%) Yeast (13%) GPC ox+ cat+ (14%) Yeast (21%) Mold (7.1%) Yeast (22%)

PAGE 100

87 Coliform, Fecal Coliform, and E. coli Detection Chi-squared tests were run on the frequenc y of coliform detection in E*Colite bags of the 72 gaskets were selected from above on all types of gaskets. Coliforms were found in 30 of these samples; however, it was dete rmined that coliforms were found no more frequently in any particular t ype of gasket over any other. Fecal coliforms were detected in E*Colite samples 36, 74, 81, 87, 113, and 121. E. coli was detected in E*Colite samples 36, 87, and 113. Samples 36, 87, and113 were confirmed E. coli identified by 16S DNA PCR identification. Coliforms we re only detected on the Petrifilm of samples 17, 45, 49,103, 107, 113, 115, 116, 117, and 118. No E. coli or fecal coliforms were ever detected on Petrifilm. Table 4-16. Number of colif orm, fecal coliform, and E. coli positive gaskets determined by PetrifilmTM and E*Colite. Gasket Type % Coliform/Fec. Coli./E. coli on PetrifilmTM % Coliform/Fec. Coli./E. coli in E*Colite Juice, A, Hot 0 / 0 / 0 36.3 / 0 / 0 Juice, A, Cold 0 / 0 / 0 35.7 / 7.1 / 0 Juice, B, Hot 14.3 / 0 / 0 57.1 / 0 / 0 Juice, B, Cold 0 / 0 / 0 33.3 / 0 / 0 Dairy, A, Hot 16.7 / 0 / 0 41.7 / 0 / 0 Dairy, A, Cold 10 / 0 / 0 40.0 / 0 / 0 Dairy, B, Hot 16.7 / 0 / 0 54.2 / 8.3 / 4.2 Dairy, B, Cold 6.3 / 0 / 0 56.3 / 18.8 / 12.5 Streptococcus and Staphylococcus Detection No Streptococcus was found in any of the samples. However, all samples collected from yellow fluorescent E*Colite bags had presumptive Staphylococcus. Chi-squared tests were run on the frequency of Staphylococcus detection in E*Colite bags of the 36 selected from above on all types of gaskets. It appeared that Staphylococcus was found no more frequently in any particular type of gasket over another. 19 samples were presumed to have Staphylococcus and 9 of those samples were presumed to be

PAGE 101

88 Staphylococcus aureus. Samples 65B was presumptive S. aureus and was further evaluated by 16S DNA PCR identification. Table 4-17. Percentage of Staphylococcus spp. and presumptive Staphylococcus aureus on each gasket type. Gasket Type % Staphylococcus spp. % Staphylococcus aureus Juice, A, Hot 33.316.7 Juice, A, Cold 38.515.4 Juice, B, Hot 28.67.1 Juice, B, Cold 18.812.5 Dairy, A, Hot 40.020.0 Dairy, A, Cold 0.00.0 Dairy, B, Hot 14.37.1 Dairy, B, Cold 8.38.3 Salmonella and Alicyclobacillus Detection The presence of Salmonella spp. was determined using the TECRA Salmonella Visual Immunoassay (TECRA International Pty. Ltd.; Frenchs Forest, Australia) and the presence of Alicyclobacillus spp. was determ ined by selective broth and agar. Neither Salmonella nor Alicyclobacillus was detected in any of the samples. 16S DNA and 28S rRNA PCR Identification Table 4-18 shows the bacteria and yeas t identified by Accuge nix, Inc. (Newark, DE). The identification and the isolates with in their database the that sample is most closely related is also included in the table. Results show th at preliminary categorizations were confirmed. Part II: Biofilm Development and Removal Gasket Analysis Raw data for this experiment can be f ound in Appendix B. The experiment was repeated six times to determine if any of the standard cleaning treatments provided a significant difference in the log reduction of the number of the coliforms and mesophilic

PAGE 102

89 Table 4-18. Results of 16S DNA a nd 28S rRNA PCR Identification. ID Number Sample Type Organism Catalase/ Oxidase ID Closest Match 113B Aciduric Yeast NA Kluyveromyces marxianus or lactis Kluyveromyces marxianus or lactis 36C Mesophile Gram-positive rod (+/+) Bacillus badius Bacillus badius 36D Mesophile Gram-positive cocci (+/+) Staphylococcus hominis Staphylococcus hominis hominis 113C Mesophile Gram-negative rod (+/-) Serratia marcescens Serratia marcescens 36 E*Colite E. coli (+/-) E. coli/ Shigella E. coli Sigma W3110 87 E*Colite E. coli (+/-) E. coli/ Shigella E. coli 0157:H7 113 E*Colite E. coli (+/-) E. coli/ Shigella E. coli Sigma W3110 65B E*Colite Staphylococcus spp. NA Staphylococcus pasteuri Staphylococcus pasteuri microorganisms per gasket in comparison to each other using ANOVA (Minitab release 14). Fishers Protected Least Significant Difference Test at a 95% confidence interval was run on groups that had signi ficant differences. The resu lts can be found in Table 419. Table 4-19. Log10 reductions among coliform and mes ophilic counts for the three wash protocols. Log10 Reductions Among Control and Wash Coliform Mesophile Wash 1 3.3 1.2a 3.9 0.88a Wash 2 5.0 2.0a 5.9 2.6b Wash 3 9.3 1.6b 9.7 1.0c Although statistical an alysis could not be perfor med on the yeast count, the E. coli MPN, or the E*Colite bag results, the following pieces of observational data were noted: 1) yeasts are generally present in the cont rol and the Wash 1 but are never present in

PAGE 103

90 wash 2 or 3; 2) E. coli was present in 5 out of 6 contro ls and was present in 1 out of 6 Wash 1 E*Colite bags; 3) coliform bacteria were present in all E*Colite controls and Wash type 1 samples; 4) coliform bacteria were present in 5 out of 6 Wash type W samples but not found in any Wash type 3 samples ; and 5) E. coli was present from <3 to 93 MPN/g. Scanning Electron Microscopy For this portion of the research we defi ned a biofilm as a b acterial residue that could not be removed from the surface of the rubber gasket with gently flowing water. Typical observations made from the SEM an alysis: 1) A thick and well-established biofilm is found on the top (Picture A ) and insi de (Picture B) surface of the gasket of the control; 2) On the inside (Picture C) gasket surface of Wash treatment 1 a well established biofilm is generally still presen t while on the top surface the biofilm may be sparsely scattered (Picture D) to absent (Picture E); 3) The inside surface of Wash treatment 2 had a well-established (Picture F) to sparsely (Picture G) attached biofilm while on the top surface the biofilm may be sp arsely (Picture H) scattered to absent (Picture I); 4) There is generally not biofilm present on the inside (Picture K) or top (Picture L) portion of treatment 3. However, if the bacteria are not properly cleaned off with the detergent or sanitizer treatment they can be a cooked onto the gasket (Picture J); 5) As the biofilm becomes more sparse there appears to be fewer rods and more cocci; 6) Cross sections of the gasket reveal the ab ility of the biofilm to not only form on the surface but to form on the inside of the gasket as well. Pictures M and N was taken from the near the edge and the center cross sect ion of the lip of th e manway gasket. All pictures were taken at 40 00x magnification except for Pict ure N, which was taken at 6000x.

PAGE 104

91 A B C D Figure 4-1. Representative pictures from scanning electron micros copy. Pictures A-M (4000x) and N (6000x). A.) Control top, B.) Control inside, C.) Treatment 1 inside, D.) Treatment 1 top (sparse biofilm), E.) Treatment 1 top (biofilm absent), F.) Treatment 2 inside (well-established biofilm) G.) Treatment 2 inside (sparse biofilm), H.) Treatment 2 inside top (spars e biofilm), I.) Treatment 2 top (biofilm absent), J.) Treatment 3 (dead bacteria present), K. Treatment 3 top (bacteria absent), L.) Treatment 3 insi de (bacteria absent), M.) Control cross section (near the edge), N.) Control cross section (bact eria are circled) (center of the cross section).

PAGE 105

92 E F G H I J Figure 4.1 Continued

PAGE 106

93 K L M N Figure 4.1 Continued Fluorescence Microscopy Due to the large amount of autofluor escence it is hard to determine if E. coli is present on any samples. It does appear that E. coli tends to gather around holes in the gasket. E. coli do not appear to flock together but to be evenly dispersed throughout the biofilm. Figure 4-2 shows a 10x magnification of the insi de, top, and outside surface (Pictures A, B, and C) of the control and 1000x inside surface of the c ontrol (Picture D).

PAGE 107

94 A B C D Figure 4-2. Representative pictures from fl uorescent microscopy. Pictures A-C are 10x and D is 1000x its original size. A.) Control Inside, B.) Control Top, C.) Control Outside, D.) Control Inside

PAGE 108

95 CHAPTER 5 DISCUSSION AND CONCLUSIONS Part I: Sample Identification and Characterization Psychroduric, Mesophilic, Thermoduric, Yeast and Mold; and Aciduric Enumeration and Characterization Based on the 72 selected samples, the ANOVAs, and the significant difference found among samples it appears that the gasket is the variable that contributes most to the significant differences in mesophilic, yeast and mold, aciduric/cm2 and aciduric/total gasket (Table 4-3, 4-4, 4-7, 4-8, 4-11, 4-12, 413, and 4-14). There two possible reasons for this. The first reason is that the gaskets different material types may cause the gasket to be conducive to biofilm formation. St orgards and others (Part I 1999) found that different rubbers varied in their susceptibil ity to biofilm formation based upon whether they were exposed to a dairy industry or br ewing industry set of conditions. Therefore, it is possible that in a juice and dairy envir onment that the neoprene found in the gasket type A gaskets is less suscepti ble to biofilm formation than the any of the rubbers that make up the gasket type B gaskets Storgards and others (Par t 1 1999) also discovered that some biofilms become more resistant to sanitizers on certain t ypes of rubber than on other types of rubber (Storgards and others; Part I 1999). Ther efore, it might be possible to conclude that the species of bacteria attaching to the r ubber of gasket type B gaskets are less susceptible to the effects of the sanitizers used by tank wash A. Finally, Storgards and others (Part I 1999) also discovered that as rubb er deteriorates, it lose its cleanability, and has more biofilm formati on. Hence, perhaps ga sket type B rubber

PAGE 109

96 gaskets may deteriorate faster or may not be re placed as often when they deteriorate then the gasket type A gaskets. The second reason might be that the gasket s conformation affects cleanability. Hence the small space between the lips of the gasket type B gasket may be more difficult to effectively clean than the gasket type A conformation. It is also possible that the design conformation of the gasket type B gask et is subject to more wearing than the gasket type A gasket, which w ould make it an easier place for a biofilm to form. Biofilm has been found to form on difficult-to-clean rubber implants such as rubber urinary catheters (Valraeds and othe r 2000, and Millsap and others 1997) and synthetic rubber voice prostheses (Leunisse and others 2001, Everaert and others 1999, Busscher and others 1997). This portion of the study emphasizes the importance of tank wash employees care in cleaning gaskets and the necessity of having the proper brushes (and/or other equipment) available to them to adequately cl ean inside the two lips of the gasket type B gasket. It also indicates that further research should be done looking at di fferent rubber materials used for manway lid gaskets to de termine which ones are least susceptible to biofilm formation attachment and which ones ar e easiest to clean. Also further studies on how age or gasket conformation affects the differe nt rubber types should be considered. None of the variables tested seemed to account for the differences in psychroduric/cm2; however, psychroduric/total gasket for more than one factor showed that there was always a significant difference be tween juice, type A, cold washed gaskets and dairy, type A, cold washed gaskets. The exception to this is demonstrated in Table 46, which compares gasket and wash temperatur e. This indicates the possibility that

PAGE 110

97 product type plays a significant role in the psychroduric count. It could be the conditions under which citrus juice or dair y products are produced or held may have more influence on the number of psychroduric, and therefore, the count than any of the other variables tested (Frank 2001). Also it is possible that psychroduric organisms group faster in dairy than in citrus juice or that the dairy produc ts cause psychroduric products to adhere to gaskets at a larger level than juice products. Table 4-15 demonstrates the top two microor ganisms in the swabs for each gasket type. The yeast and mold results are as expe cted. The aciduric results indicate that the two types of microflora that can survive in this type of e nvironment are yeasts and acid resistant cocci such as Streptococcus spp. (Fozo and Quivey 2004). The mesophile results indicate that the gram-positive cocci are also the dominate organism in the standard plat e count which indicates most lik ely two things: 1) that the gram-positive cocci that survive pasteurizati on or are postpasteurization contaminants attach readily to the gaskets and are very hard to remove and 2.) gram-positive cocci from the employees hands contaminated the cl ean gasket surface (USDAs Bad Bug Book, 1992). The second possibility is di scussed below in the section on Staphylococcus and Streptococcus spp. In many of the SEM pictures for Part II of this project the control appeared to be covered with rods. After washing, many of th e bacteria that were present appeared to be gram-positive cocci. This indicates that perhaps gram-positive cocci are able to make the first and strongest associa tion with the surfaces of rubber gaskets and they would be the most likely to be found on the surfaces of clean rubber gaskets. The predominate gram-positive rods in the thermoduric count most likely belong to the Bacillus spp. which can survive the pasteuriza tion temperatures (Hensyl 1994). The

PAGE 111

98 gram-negative rods that predominate the psyc hroduric bacteria most likely belong to the Pseudomonas, Alcaligenes, and Flavobacterium spp. and indicate postpasteurization contamination of milk and juice (B ishop and White 1986, Cousin 1982, Stadhouders 1975, Thomas 1974). Coliform, Fecal Coliform, and E. coli Detection The differences between using PetrifilmTM and the E*Colite bags to detect E. coli and coliforms in the samples most likely has to do with the fact that the PetrifilmTM used 1 mL of the sample while the E*Colite bag used 20 mL. Of the three E. coli samples, 36 and 87 were found in the dairy, col d, gasket type B gaskets and seemed to be a result of the cleaning process, the product, and the type of gasket. This emphasizes the importance of three things: 1.) having an appr opriate brush to clean the inside of the two lips of the gasket type B gasket, 2.) havi ng a hot water in the wash sinks and spray cleaning systems (hot water will help liquefy fats making it easier to remove build up from the gaskets surface), and 3.) having em ployees carefully examining the gaskets to make sure no visible residue is left on the gaskets. Contamination of sample 113, which was a dairy, gasket type B, hot washed sample, appeared to be more likely the result of the gasket washers physical condition, which influenced how he washed the gasket. This emphasizes the importance of sendi ng home an employee who is physically incapable of doing his jobs due to illn ess, impairment, or injury. Streptococcus and Staphylococcus Detection Although Staphylococcus was not found in any greater frequency in any particular sampling variable than any other, this is not surprising. Humans and animals are the primary reservoir for staphylococ ci. It is present in the na sal passages and throats and on the hair and skin of 50 percent or more of healthy individuals (USDAs Bad Bug Book,

PAGE 112

99 1992). Food handlers are most often responsible for S. aureus outbreaks (USDAs Bad Bug Book, 1992). At the tank wash facility it was noted that hand washing rarely occurred even after sneezing, coughing, or blow ing ones nose. To re duce possibility of Staphylococcus contamination of gaskets, greater at tention needs to be given to training employees on how, when, and why they should be washing their hands. Also, employers need to provide employees with a clean place to wash their hands that has hot water and is well stocked with soap and paper towels. Reducing or eliminating the amount of S. aureus introduced into tankers is particularly important for tankers carrying cr eam. Cream does not have to be pasteurized when reaching baking facilities. Cream in ba keries is often used to make cream filled desserts. It is questionable if the internal temperatures of these cream filled desserts would become hot enough to kill the S. aureus. Although no such outbreaks of S. aureus have occurred outbreaks in cream filled pastri es as a result of transportation, it has been implicated in foodborne illness. An ex ample of when cream was a cause of a S. aureus infection was in 1983 on a cruise ship where approximately 32% of passengers became ill after eating cream filled pastries available at two separate meals (CDC Morbidity and Mortality Weekly Report 1983). Salmonella and Alicyclobacillus Detection Since Salmonella was not detected in a ny of the samples it can be concluded that either Salmonella was not present on any of the gasket s prior to washing or the washing protocol was adequate to kill any of the Salmonella present on the gaskets surface. Survey data collect by Winniczuk and Parish (u npublished) at this particular facility noted that Salmonella was not present on the surface of juice and dairy gaskets before cleaning and was not found on the surface of juice and dairy gaskets after cleaning.

PAGE 113

100 Therefore, it may be possible to conclude that Salmonella is rarely a contaminant in these food systems and is seldom present on any of the gaskets prior to washing.. There has been no previous research to suggest that Alicyclobacillus can form a biofilm. Hence, it may not be found on the surf ace even if it was present in the orange juice being transported because it lacks the ability to adhere to the surface. It also may be possible that none of the samples transported contai ned or contained Alicyclobacillus at high enough numbers to cause growth and bi ofilm formation on the gaskets surface, or the washing protocol was adequate to kill any of the vegetative Alicyclobacillus present on the gaskets surface. If Alicyclobacillus had been present in the spore form it would have survived the cleaning treatments and r ecovered in the Alibroth. Hence, there was most likely were no Alicyclobacillus spores on the gaskets. Further research should be done to determine if Alicyclobacillus can form a biofilm and the impact this might have on a load of citrus juice if it was not completely removed from the inside surface of tanker. Part II: Biofilm Development and Removal Gasket Analysis Treatment 1 is the least e ffective and should not utili zed. This treatment should not be used in industry. Resu lts from this study demonstrat e that gaskets can still be contaminated with E. coli after Treatment 1. If the E. coli is pathogenic and present in sufficiently high numbers, and the product is no t going to be pasteuri zed again, this could a problematic situation. E. coli and yeasts were not present in any of the second wash type. The sanitizer used in this experiment tested to create a 5-log reducti on in the amount of E. coli and S. aureus. This experiment shows that it could also create a five-log reduction in the

PAGE 114

101 number of total bacteria or coliforms. However, there were still approximately 104 coliforms and 104 mesophilic organisms present on the gasket surface after Treatment 2. The hot water treatment in Treatment 3 wa s extremely effective at sterilizing the gasket. Treatment 3 was also the only treat ment that created a statistically significant reduction in both coliforms and mesophilic orga nisms. Treatment 3 can be conducted in several ways. The gasket can be placed back on the manway lid and allowed to go through the clean-in-place cycle with the tanker; it can be placed in a dishwasher or a sink with steam heated water; or a slow drai ning bucket can be set behind the rear port of the tanker so that the hot water from the cl ean in place cycle can fall into the bucket and sterilize the parts. The slow draining bucket can be create d out of half of a 50-gallon plastic or stainless st eel drum or specially made stai nless steel container with holes drilled into the sides. Figure 5-1 is an example of this slow draining bucket method. Figure 5-1. An example of slow draining bucke t made from half of a 50-gallon plastic drum.

PAGE 115

102 Past research on D and z-values demonstrat es how it would be possible to eliminate the bacteria and yeast in the samples thr ough thermal means. A lthough no research has been done on Bacillus badius, research has been done on Bacillus subtilis spores and it has been determined that at a temperature of 100oC spores can be reduced by one log for 11 mins and that it has a z-value of 7oC (Farkas 2001). Through selecting and identifying representative colonies it was discovered that there were few Bacillus badius colonies in the biofilm. Since the gaskets were sterile af ter the hot water treatment, it appears that the hot water treatment and the sanitizer were adequate to kill in activate the spores. However, if a greater number of spores were present that treatment might not be adequate. E. coli in Ringers solution at pH of 7 at 55oC (131oF) has a D-value of 4 mins (Tomlins and Ordal 1976). Very few E. coli were found in the control samples. All the E. coli were eliminated by the 5-log reduction cau sed by the sanitizer. If not the D-value suggests that during the time of the heat tr eatment at approximate 8-log reduction, which would have been more than adequate to kill the E. coli. The sanitizer was not found to be effective enough to kill the Staphylococcus hominis hominis on the gasket. A D-value of 7.8 mins at 55oC and a z-value of 4.5oC have been calculated for S. aureus in pea soup and custard (Tomlins and Ordal 1976). If the same were to hold true for the gasket the heat treatment would have b een effective enough to kill S. hominis hominis. Put and others (1976) discovered that yeasts can be killed with in 10 to 20 mins at 55oC to 60oC. The heat treatment applied would have been enough to effectivel y eliminate kill the yeasts; however, the sanitizer used for this expe riment killed them first. No D-values for heat treatments of Serratia have been determined. Sutt on and others (1991) determined D-values for killing Serratia on the surfaces of contact lenses using different solutions

PAGE 116

103 some of which contain quaternary ammonium compounds. However, the solutions are coded so no one can determine which D-values belong to cleaning solutions and D-values for the different clean solutions very from 4.9 to 402.6 mins. Farkas (2001) states that vegetative cells of bacteria are genera lly killed after a few minutes at 70oC (158oF) to 80oC (176oF). Katsuyama (1993) states that effectiv e sterilization can oc cur if the system reaches at least 82oC (180oF). Therefore, from the col onies selected from plates following treatment two it indicates that the sanitizers cannot effectively kill the S. marcesens but the heat treatment was effective enough to kill the S. marcesens on the gasket. A potential problem with this method compared to the model system used in this research is that in the model the gasket has be en cut, creating a small piece. This allowed a more effective distribution of heat across th e gaskets surface in the model that would not happen in actual practice. Further studies need to be c onducted to determine if heat of 180oF is effectively distributed across that ga skets entire surface and interior to cause sterilization (Katsuyama 1993). The disadvantage of Treatment 3 is if any b acterial residue is left on the gasket that during the heat treatment it can be cooked to the gaskets surface. The cooked product on the gasket surface could become a base for future microbial attachment. The advantages of using treatment 3 is that the treatment is inexpensive, the temperature (hence its effectiveness) can be eas ily measured, there are no resides, it is not corrosive, it provides a non-selective kill, and it penetrates hard to reach surfaces (Jennings 1965).

PAGE 117

104 Finally one word of caution should be noted when using this method. This method is not effective if cross contamination occu rs after cleaning. To reduce the amount of cross contamination after cleaning employees should make sure that their hands are washed well before handling the gasket. Also if dirty brushes or contaminated water is used to clean the manway lid, the gasket can be contaminated with bacteria when it comes in contact with the lid. This empha sizes the importance of soaking brushes in sanitizer between uses and checking the qual ity of the water being used even using a municipal water source. Scanning Electron Microscopy From these images collected the followi ng conclusions can be drawn. 1) A heavy microbial residue can be found on the gasket on the control. 2) Detergent removes a large portion of the microbial re sidue. And, 3) The inside grove of the gasket is harder to clean than the out side surfaces. This em phasizes the need for good brushes (or other equipment), and sufficient time scrubbing with detergents to remove the microbial residue. Pictures of Treatment 3 indicate in most cases the gaskets are clean; however, in some cases they may not have been properl y cleaned and have cooked microbial and food product residue on the surfaces, allowing opportunity for the development of the next microbial residue. Fluorescence Microscopy The images obtained from the electron microscopy were ambiguous because high magnification images could not be obtained due to the thickness and the rounded edges and thickness of the sample. In the microscopy photos take n at 100x there appear to be small, very bright green sp ecks that could possibly be E. coli, which was the only

PAGE 118

105 microorganism in the sample with the green fluorescent protein. One section of gasket was cut thin enough to place under the fluorescen t microscope and there appear to be several fluorescent ovals that could possibly be E. coli. However, the E*Colite test indicated that there were no E. coli in the control. Possibly r easons for this contradiction include that, the bright specks are not E. coli, that the E. coli were only growing in the region used for the microscopy, and that the E. coli was in such low numbers on the control that an E. coli might not have gotten into an E*Colite bag. Overall Conclusions The results from part II suggest that wash treatments with detergent and a hot water are the most effective at removing soils a nd killing bacteria. To assure maximum effectiveness worker training is necessary to assure that gaskets are being scrubbed well, that the water temper ature is between 104-185oF to melt the milk fat (Katsuyama 1993, Goff 2005), and they are aware and adhering to Good Manufacturing Practices (such as hand washing, wearing their uniform, not smoki ng, etc.). Tank wash managers need to provide employees with the proper brushes to clean the gaskets, and good training on how to clean gaskets. The next chapter s uggests changes that shoul d be implemented to create a better tank wash.

PAGE 119

106 CHAPTER 6 FUTURE WORK Extension Gasket Washing Video and/or Training Manual To help unify the method for gasket wa shing and to teach employees how to properly wash gaskets, a training manual or video should be prepared. The manual and/video should discuss: A discussion of some of the elementa ry microbiology principles behind gasket sanitation. The best tools for cleaning the gaskets, how to use them properly to achieve the cleanest gasket, and how to store th em when they are not being used. An explanation of the different detergents available, the concen tration of detergent necessary to clean a gasket depending on soil on the gasket, and how to test the concentration. An explanation of the different sanitizers available, the concentration of sanitizer necessary to kill 99.999% of E. coli and S. aureus present on the gasket, how to test the concentration, and length of time necessary for the gasket to be exposed to the sanitizer. An explanation of the different methods av ailable to heat-treat the gasket after sanitizer use (dishwasher, on the manway lid during the clean in place cycle with the tanker, a slow draining buck at the back of the tanker, a sink with steam heated water). An explanation of how personal hygiene effects gasket cleanliness and compliance with Good Manufacturing Pr actices (GMP) affects gasket cleanliness. A detailed hand washing demonstration. Following the employee reading the ma nual or watching the video a brief questionnaire should be administered to determine if the employee comprehends the information presented.

PAGE 120

107 Workshops To disseminate information to tank washes on research being conducted, GMP, and Hazard Analysis Critical C ontrol Point (HACCP) programs to tank wash facilities. Meetings should be held in different areas of the country. A lecture manual or guide could be created to give extension educators a guide to run these meetings. Tank Wash Association Task wash owners have noted that there is no professional group for them to belong to and that will provide them with informa tion and support. By uniting tank wash owners materials could be disseminated easier, tank washes could agree on common wash guidelines, and pool financial resources to f und continued research a nd extension work to improve the industry. Research Juice Concentrate Research According to the Juice HACCP, Final Rule (FDA/ CFSAN, Hazard Analysis 2001), concentrates do not need to be pasteuri zed after transport and prior to packaging. It has been found that pathogens such as Listeria monocytogenes, Salmonella spp., and E. coli can survive in concentrate (Oyarzbal 2003 ). Research is n eed to determine the number of colony forming units (CFU) of Listeria monocytogenes, pathogenic Salmonella spp., or E. coli which could potentially lead to food borne illness. This research should be used to calculate the po ssibility that a food borne outbreak will result from a gasket where the surface is contam inated with pathogenic bacteria. The same model used in part II of this research could be used to conduct this experiment. A known amount of Listeria monocytogenes or E. coli in a milk base, or Salmonella in an egg base could be coated onto th e surface of the gasket. Five liters of

PAGE 121

108 concentrate could be sprayed up occasionally on to the gasket surface at the end of three days. The concentrate should be removed at the amount of Salmonella, Listeria monocytogenes, or E. coli present in the 5L should be cal culated. Then the probability for food borne outbreak must be calculated (the researchers will have to take into account that concentrate does not move too much in the tank so the only concentrate that may need to be worried about may be the concentrate near th e manway lid). Biofilm Research There are many directions this research c ould go. The first area that should be looked at is for confirmation that bacteria, yeas t and mold are forming a biofilm on the gaskets surface. A second area would be to take the bacterial stains collec ted from this research and determine if these strains communicate with each other through AI-2 related quorum sensing and which strains form biofilm. This may be help to determine which bacterial species are problematic in terms of ca using a biofilm on tanker gaskets. Another direction would be to look at wh at conditions encourage and discourage bacteria from attaching to th e surface. Looking at different nutrients, temperatures, water activities, gasket materials and product type s to see under what conditions biofilms form. Finally, different types of commercially available cleaners for clean out of place part in the tank wash industry should be studied to determ ine under what length of time, temperature, and concentration are best to us ed the products at to adequately remove a biofilm from different types of materials used for gaskets in the tank wash industry. Gasket Alternatives Three ways of solving the problem with the gasket is to create a manway lid that has a gasketless design, look at new gasket mate rials, or to create a disposable manway

PAGE 122

109 lid. The gasket on the manway lid is to prevent the product from splashing out and to keep the metal pieces from wearing away at each other. The challenge in creating a gasketless design would be to keep the metal pi eces from wearing away at each other and will hold the liquid in without creating a system that is impossible to clean. The second alternative is to study new gasket materials to look for materials that would inhibit biofilm formation or microbial gr owth. Tests will also need to be done to make that the biofilm inhibition continues as the gasket ages. The disposable gasket must be able to be manufactured inexpensively at price wash stations are willing to pay for it. The cost s hould be no more that the wash stations would pay for the materials to clean it plus the value of the gasket over time. It would be ideal if the gasket could be sterilized with irradi ation or by autoclaving. The material must be able to withstand vary hot conditions as well as very cold conditions depending on the time of year, the area of the country it is being used in. Finally the ga sket must be able to keep the metal pieces from wearing away at each other and create a seal hold the liquid in. Also if the gasket could be made from an easy recyclable or easily biodegradable material it would be best for the environment.

PAGE 123

110 APPENDIX A PART I RAW DATA Table A-1. Sample type, sample number, gasket washer, aciduric and yeast and mold raw data. Sample Type Sample # Gasket Washer Acidophilic Acidophilic Total Yeast and Mold Yeast and Mold Total Juice, A, hot 2 N/A 55 18339 450 Juice, A, hot 5 N/A 1 33.330 0 Juice, A, hot 7 N/A 10 333.30 0 Juice, A, hot 28 N/A 113 3767137 6850 Juice, A, hot 53 A 49 16330 0 Juice, A, hot 64 B 58 19331 50 Juice, A, hot 90 B 444 1480086 4300 Juice, A, hot 124 C 15 5001 50 Juice, A, hot 126 B 201 67002 100 Juice, A, cold 52 B 7 233.31 50 Juice, A, cold 56 B 41 13670 0 Juice, A, cold 57 B 17 566.71 50 Juice, A, cold 65 A 63 21001 50 Juice, A, cold 72 B 102 34004 200 Juice, A, cold 93 B 72 240012 600 Juice, A, cold 108 D 48 16003 150 Juice, A, cold 120 C 0 00 0 Juice, A, cold 121 N/A 73 24339 450 Juice, B, hot 42 N/A 45 204014 700 Juice, B, hot 44 B 174 788825 1250 Juice, B, hot 55 A 33 14967 350 Juice, B, hot 84 C 41 185915 750 Juice, B, hot 88 B 62 281110 500 Juice, B, hot 104 C 67 30370 0 Juice, B, hot 106 B 9 4089 450 Juice, B, hot 112 B 109 4941109 5450

PAGE 124

111 Juice, B, hot 116 A 184 834118 900 Juice, B, cold 50 B 57 258420 1000 Juice, B, cold 68 C 12 5440 0 Juice, B, cold 70 B 59 26750 0 Juice, B, cold 75 B 230 1042775 3750 Juice, B, cold 96 B 417 18904143 7150 Juice, B, cold 99 B 5 226.70 0 Juice, B, cold 110 B 41 18590 0 Juice, B, cold 111 C 39 17688 400 Juice, B, cold 123 C 78 35363 150 Dairy, A, hot 8 N/A 11 366.77 350 Dairy, A, hot 13 N/A 191 636760 3000 Dairy, A, hot 22 N/A 31 10331 50 Dairy, A, hot 25 N/A 10 333.32 100 Dairy, A, hot 35 N/A 23 766.70 0 Dairy, A, hot 48 A 112 373311 550 Dairy, A, hot 59 A 28 933.313 650 Dairy, A, hot 80 C 7 233.30 0 Dairy, A, hot 118 A 302 100671 50 Dairy, A, cold 12 N/A 43 14334 200 Dairy, A, cold 26 N/A 4 133.31 50 Dairy, A, cold 49 N/A 60 200021 1050 Dairy, A, cold 66 B 2 66.670 0 Dairy, A, cold 73 B 15 5001 50 Dairy, A, cold 86 N/A 510 170002 100 Dairy, A, cold 94 B 666 22200152 7600 Dairy, A, cold 97 B 26 866.77 350 Dairy, A, cold 102 C 4 133.30 0 Dairy, B, hot 45 A 203 9203101 5050 Dairy, B, hot 54 A 25 11334 200 Dairy, B, hot 62 E 71 32190 0 Dairy, B, hot 67 C 71 32190 0 Dairy, B, hot 109 C 22 997.35 250 Dairy, B, hot 114 B 177 802479 3950 Dairy, B, hot 115 A 109 49410 0 Dairy, B, hot 117 N/A 206 93391 50 Dairy, B, hot 119 A 1 45.330 0 Dairy, B, cold 47 C 43 194918 900 Dairy, B, cold 77 B 426 1931227 1350 Dairy, B, cold 81 B 276 125122 100 Dairy, B, cold 85 B 36 16325 250 Dairy, B, cold 87 B 67 30370 0 Dairy, B, cold 91 D 774 35088148 7400 Dairy, B, cold 95 B 531 24072229 11450 Dairy, B, cold 101 B 570 2584014 700 Dairy, B, cold 105 B 1 45.330 0

PAGE 125

112 Table A-2. Sample type, sample number, ps ychroduric, mesophilic, and thermoduric raw data. Sample Type Sample # Psychroduric Psychroduric Total Mesophiles Mesophiles Total Thermoduric Thermoduric Total Coliform (Ecolite)* Presumptive Staphylococci* Presumptive S.aureus* Juice, A, hot 2 0 0 353500 00 0 Juice, A, hot 5 0 0 00 00 0 Juice, A, hot 7 0 0 181800 00 0 Juice, A, hot 28 0 0 18018000 00 0 Juice, A, hot 53 0 0 515100 00 1 0 0 Juice, A, hot 64 3 660 232300 00 0 1 0 Juice, A, hot 90 11 2420 626200 1100 1 0 0 Juice, A, hot 124 0 0 8800 00 0 1 1 Juice, A, hot 126 0 0 787800 00 1 0 0 Juice, A, cold 52 0 0 7700 00 1 0 0 Juice, A, cold 56 0 0 7700 00 0 0 0 Juice, A, cold 57 0 0 1100 00 1 0 0 Juice, A, cold 65 0 0 131300 1100 0 1 1 Juice, A, cold 72 0 0 11111100 00 0 0 0 Juice, A, cold 93 0 0 202000 2200 1 0 0 Juice, A, cold 108 0 0 191900 00 0 1 0 Juice, A, cold 120 0 0 1100 00 0 0 0 Juice, A, cold 121 0 0 11011000 2200 1 0 0 Juice, B, hot 42 0 0 162176 00 0 1 0 Juice, B, hot 44 9 1224 415576 1136 1 0 0 Juice, B, hot 55 0 0 243264 00 1 0 0 Juice, B, hot 84 0 0 192584 00 0 0 0 Juice, B, hot 88 1 136 6816 00 1 0 0 Juice, B, hot 104 0 0 547344 00 0 1 0 Juice, B, hot 106 3 408 273672 1136 0 1 0 Juice, B, hot 112 0 0 638568 00 0 0 0 Juice, B, hot 116 6 816 9813328 00 1 0 0 Juice, B, cold 50 0 0 324352 00 1 0 0 Juice, B, cold 68 0 0 5680 00 0 0 0 Juice, B, cold 70 0 0 587888 00 0 0 0

PAGE 126

113 Juice, B, cold 75 1 136 527072 00 1 0 0 Juice, B, cold 96 32 4352 152040 00 1 0 0 Juice, B, cold 99 0 0 3408 00 0 1 1 Juice, B, cold 110 0 0 202720 00 0 0 0 Juice, B, cold 111 0 0 7952 00 0 0 0 Juice, B, cold 123 0 0 557480 1100 0 1 1 Dairy, A, hot 8 0 0 4400 00 0 Dairy, A, hot 13 65 14300 14214200 00 1 Dairy, A, hot 22 0 0 323200 00 1 Dairy, A, hot 25 0 0 141400 00 0 Dairy, A, hot 35 2 440 6600 00 1 Dairy, A, hot 48 0 0 8800 00 1 0 0 Dairy, A, hot 59 1 220 8800 00 0 0 0 Dairy, A, hot 80 0 0 5500 00 0 0 0 Dairy, A, hot 118 42 9240 11611600 00 0 1 1 Dairy, A, cold 12 0 0 242400 00 0 Dairy, A, cold 26 0 0 5500 00 0 Dairy, A, cold 49 4 880 161600 00 1 0 0 Dairy, A, cold 66 0 0 00 00 0 0 0 Dairy, A, cold 73 0 0 545400 191900 0 0 0 Dairy, A, cold 86 428 94160 18718700 00 1 0 0 Dairy, A, cold 94 22 4840 13413400 1100 1 0 0 Dairy, A, cold 97 2 440 161600 00 0 0 0 Dairy, A, cold 102 0 0 5500 1100 0 0 0 Dairy, B, hot 45 35 4760 26035360 00 1 0 0 Dairy, B, hot 54 0 0 7710472 1136 0 1 0 Dairy, B, hot 62 8 1088 243264 00 1 0 0 Dairy, B, hot 67 0 0 202720 00 0 0 0 Dairy, B, hot 109 0 0 192584 1136 0 0 0 Dairy, B, hot 114 0 0 9112376 00 0 1 1 Dairy, B, hot 115 0 0 587888 00 1 0 0 Dairy, B, hot 117 0 0 14619856 00 1 0 0 Dairy, B, hot 119 0 0 00 00 0 0 0 Dairy, B, cold 47 0 0 7952 00 0 0 0 Dairy, B, cold 77 1 136 20227472 00 0 0 0 Dairy, B, cold 81 0 0 30040800 00 1 0 0 Dairy, B, cold 85 0 0 131768 00 0 0 0 Dairy, B, cold 87 0 0 172312 00 1 0 0 Dairy, B, cold 91 6 816 13217952 7952 1 0 0 Dairy, B, cold 95 7 952 15020400 00 1 0 0 Dairy, B, cold 101 27 3672 304080 1136 1 0 0 Dairy, B, cold 105 0 0 00 00 0 1 1

PAGE 127

114 APPENDIX B PART II RAW DATA Table B-1. Sample Letter and results for Part II. Sample* Ecolite Results CFU of coliform/buffer mL Log of the # Coliforms/gasket Amount of Change for Coliforms E. coli MPN/g coliform MPN/g CFU Mesophiles/mL Log of the # of Mesophiles/gasket Amount of Change for Mesophiles CFU Yeast/mL Log of the #of Yeast/gasket Amount of Change for Yeast E Control E. coli 18144 6.88 2 1101 26350000 10 15120 6.7973 E1 coliform 61.5 4.41 2.47 0 1100 240 5 5.04 3.5 3.1619 3.635 E2 coliform 3 3.09 3.78 0 3 38 4.2 5.84 0 0 6.797 E3 negative 0 0 6.88 0 0 0 0 10 0 0 6.797 F Control E. coli 3E+07 10.1 2 1101 82000000 10.5 5 3.3168 F1 coliform 170 4.85 5.3 0 240 2200 5.96 4.57 0 0 3.317 F2 coliform 66.5 4.44 5.7 0 140 4.76 5.77 0 0 3.317 F3 negative 0 0 10.1 0 0 0 0 10.5 0 0 3.317 G Control coliform ##### 10.8 0 1101 166,000,000 10.8 3 3.0949 G1 coliform 9400 6.59 4.23 0 23 9500 6.6 4.24 0.5 2.3168 0.778 G2 negative 0.5 2.32 8.5 0 0 0 0 10.8 0 0 3.095 G3 negative 0 0 10.8 0 0 0 0 10.8 0 0 3.095 H Control E. coli 1E+08 10.6 93 23 205000000 10.9 620 5.4102 H1 E. coli 1E+05 7.76 2.88 2 1101 190400 7.9 3.03 10 3.6178 1.792 H2 coliform 903 5.57 5.06 0 1134 5.67 5.26 0 0 5.41 H3 negative 0 0 10.6 0 0 1 2.62 8.31 0 0 5.41 I Control E. coli 1E+07 9.72 2 1101 11700000 9.69 10.5 3.639 I1 coliform 16900 6.85 2.88 0 210 12000 6.7 2.99 1 2.6178 1.021 I2 coliform 5950 6.39 3.33 0 1100 2270 5.97 3.71 0 0 3.639 I3 negative 0 0 9.72 0 0 0 0 9.69 0 0 3.639 J Control 1E+05 7.68 0 1101 940000 8.59 11 3.6592

PAGE 128

115 J1 680 5.45 2.23 0 240 530 5.34 3.25 1 2.6178 1.041 J2 30.5 4.1 3.58 0 38 63 4.42 4.17 0 0 3.659 J3 0 0 7.68 0 0 0 0 8.59 0 0 3.659 *Sample Letters A-D were trial runs to perfect methodology a nd are not shown in the graph.

PAGE 129

116 APPENDIX C GASKET SURFACE AREA SAMPLE CALCULATION Below is a sample calculation of how the su rface area for the gasket used in part I and II were determined. Figure C-1. Cross section of gasket type A from part I of the research project. Surface Area Middle and Bottom r=0.5(d) A1= r2 A1= (8.5in)2 A1=226.4in2 A2= r2 A2= (7.0in)2 A2=153.5in2 A3= r2 A3= (6in)2 A3=113.0in2 A1-2=A1-A2 A1-2=226.4in2-153.4in2 A1-2=72.9in2 A2-3q=A1-A2 A2-3=153.5in2-113.0in2 A2-3=40.5in2 Surface Area of Top C= d C= (17in) C=53.4in A4=(53.4in)2 A4=106.3in2 Total Surface Area Atotal= A1-2+ A2-3+A4 Atotal=72.9in2+40.5 in2+106.3in2 Atotal=219.7 in2 or 1,417.4cm2 d1=17 in d2=14in d3=12in h1=2in h2=0.5in h3=1.5in

PAGE 130

117 APPENDIX D STANDARD GROWTH CURVES y = 5E+09x + 3E+07 R2 = 0.77220 500000000 1000000000 1500000000 2000000000 2500000000 3000000000 00.10.20.30.40.5Absorbance (%)CFU/ml 5 Figure D-1. E. coli absorbance vs. CFU/mL y = 2E+10x 9E+07 R2 = 0.7832-1000000000 0 1000000000 2000000000 3000000000 4000000000 5000000000 00.050.10.150.2Absorbance (%)CFU/ml Figure D-2. Gram-negative rods absorbance vs. CFU/mL

PAGE 131

118 y = 7E+07x + 6E+06 R2 = 0.51120 10000000 20000000 30000000 40000000 50000000 60000000 00.10.20.30.40.50.6Absorbance (%)CFU/ml Figure D-3. Gram-positive rods absorbance vs. CFU/mL y = 6E+08x 9E+06 R2 = 0.9157 0 50000000 100000000 150000000 200000000 250000000 300000000 00.10.20.30.40.5Absorbance (%)CFU/ml Figure D-4. Gram-positive cocci absorbance vs. CFU/mL

PAGE 132

119 y = 2E+08x + 6546.4 R2 = 0.9765 0 20000000 40000000 60000000 80000000 100000000 120000000 140000000 160000000 180000000 00.20.40.60.81Absorbance (%)CFU/ml Figure D-5. Yeast absorbance vs. CFU/mL

PAGE 133

120 APPENDIX E STATISTICAL TABLES I Figure E-1. Surface area (cm2) of test type vs. product. Figure E-2. Surface area (cm2) of test type vs. gasket type.

PAGE 134

121 Figure E-3. Surface area (cm2) of test type vs. wash type. Figure E-4. Surface area (cm2) of test type vs. product, and gasket type.

PAGE 135

122 Figure E-5. Surface area (cm2) of test type vs. product, and wash type. Figure E-6. Surface area (cm2) of test type vs. gasket type, and wash type.

PAGE 136

123 Figure E-7. Surface area (cm2) of test type vs. product, gasket type, and wash type.

PAGE 137

124 APPENDIX F STATISTICAL TABLES II Figure F-1. Surface area (total gask et) of test type vs. product. Figure F-2. Surface area (total gasket ) of test type vs. gasket type.

PAGE 138

125 Figure F-3. Surface area (total gasket ) of test type vs. wash type. Figure F-4. Surface area (total gasket) of test type vs. product and gasket type.

PAGE 139

126 Figure F-5. Surface area (total gasket) of test type vs. product and wash type. Figure F-6. Surface area (total gasket) of te st type vs. gasket type and wash type.

PAGE 140

127 Figure F-7. Surface area (total gasket) of te st type vs. product, gasket type, and wash type.

PAGE 141

128 LIST OF REFERENCES 3-A Sanitary Standards for Steel Automotive Transportation Tanks for Bulk Delivery and Farm Pick-Up Service. 2002. 3-A Sanita ry Standards, Inc. McLean, VA. 3-A Sanitary Standards for Multiple-Use Rubber and rubber-Like Materials Used as Product Contact Surfaces in Dairy Equipmen t, 18-03. 3-A Sanitary Standards, Inc. McLean, VA. Accugenix, Inc. Available from: http:// www.accugenix.com/. Accessed November 5, 2005. American Delphi. ~2000. American Delphi Stainless Steel Guide. Available from: http://www.americandelphi.com/s ink.htm. Accessed June 31, 2004. Albuquerque, L. Rainey, F.A., Chung, A.P., Sunna, A., Nobre, M.F., Grote, R., Antranikian, G., da Costa, M.S. 2000. Alicyclobacillus hesperidum sp. nov. a related genomic species from solfataric soils of Sao Miguel in the Azores. International Journal of Systematic Evolutionary Microbiology. 50:451-457. Arnold, J.W., Boothe, D.H., Bailey, G.W. 2001. Parameters of Treated Stainless Steel Surfaces Important for Resistance to Bacter ial Contamination. Transactions of the ASAE. 44(2): 347-356. Arnold, J.W., Suzuki, O. 2003. Effects of corro sive treatment on stainless steel surface finishes and bacterial attachment. Amer ican Society of Agricultural Engineers. 46(6):1595-1602. Barnes, L.M., Lo, M.F., Adams, M.R., Chamberlain, A.H.L. 1999. Effect of Milk Proteins on Adhesion of Bacteria to Stainless Steel Surfaces. Applied and Environmental Microb iology. 65(10):4543-4548. Barnett, J.A., Payne, R.W., Yarrow, D. 2000. Yeasts: Characteristics and Identification. Cambridge: Cambridge University Press. Banning, N., Toze, S., Mee, B.J. 2004. Persistence of biofilm-associated Escherichia coli and Pseudomonas aeruginosa in groundwater and treated effluent in a laboratory model system. Microbiology. 149(Pt 1):47-55.

PAGE 142

129 Basnet, C., Foulds, L.F., Wilson, J.M. 1999. An Exact Algorithm for a Milk Scheduling and Sequencing Problem. Annals of Operations Research. 86:559-568. Basnet, C., Foulds, L.F., Wilson, J.M. 1997. A deci sion aid for milk tanker run allocation. The Journal for the Operational Re search Society. 48(8):786-792. Bell, C., Stallard, S.E., Brown, S.E., St andley, J.T.E. 1994. ATP-Bioluminescence Techniques for Assessing the Hygienic C ondition of Milk Transport Tankers. International Dairy Journal. 4:629-640. Beuchat, L.R., Cousin, M.A. Chapter 20 Yeas t and Mold. Downes, F.P. and Ito, K. 2001. Compendium of Methods for the Micr obiological Examination of Foods. 4th ed. American Public Health Asso ciation, Washington, D.C. Bishop, J.R. and White, C.H. 1986. Assessment of dairy product quality and potential shelf-lifeA review. Journal of Food Protection. 49:739-753. Bosio Metal Specialties. 2000. Stainless Steel Information. Available from: http://www.bosiometalspecialties.com/e lectropolish.htm. Accessed June 31, 2004. Brading, M.G., Jass, J., Lappin-Scott, H. M. 1995. Dynamics of bacterial biofilm formation. In: Lappin-Scott, H.M., Coster ton, J.W., editors. Microbial biofilms. Cambridge: Cambridge University Press, p. 46. Brant, C.Q., Silva Jr., M.R., Siqueira, E.S., Castro, R.R., Rohr, M.R., Libera Jr., E.D., Morais, M., Freymuller, E., Ferrari Jr., A.P. 1996. Biliary a nd pancreatic stent blockage by bacterial biofilm: presentati on of two cases. Br azilian Journal of Medical and Biological Re search. 29(11):1455-1459. Bryan, F.L. 1983. Epidemiology of milk-borne diseases. Journal of Food Protection. 46:637-649. Bryan, F.L. 1988. Risks associated with vehi cles of foodborne pa thogens and toxins. Journal of Food Protection. 51:498-508. Busscher, H.J., Free, R.H., Weissenbruch, R. Van., Albers, F.W.J., van der Mei, H.C. 2000. Preliminary observations on influence of dairy products on biofilm removal from silicone rubber voice prostheses in vi tro. Journal of Dairy Science. 83(4):641647. Buswell, C.M., Herlihy, Y.M., Lawrence, L.M., McGuiggan, J.T.M., Marsh, P.D., Keevil, C.W., Leach, S.A. 1998. Extended survival and persistence of Campylobacter spp. in water and aquatic biofilms and their detection by immunofluorescent-antibody and -rRNA staining. Applied and Environmental Microbiology. 64:733.

PAGE 143

130 Butler, M., Williams, H.P. 1995. A problem of m ilk delivery: a variation of the traveling salesman problem. Research Report, Depart ment of Mathematics, University of Southampton, England, to a ppear in Comp. Opt. Camper, A.K., Warnecke, M., Jones, W.L., McFeters, G.A. 1998. Pathogens in model distribution system biofilms. Denver: American Water Works Association Research Foundation. CDC Morbidity and Mortality Weekly Report. 1983. Staphylococcal Food Poisoning on a Cruise Ship. 32(22)294-295. Available from: http://www.cdc.gov/mmwr/preview/mmw rhtml/00000093.htm. Accessed October 20, 2005. Characklis, W.G. 1990. Biofilm pr ocesses. In: Characklis, W.G., Marshall, K.C., editors. Biofilms. New York: John Wiley & Sons, p. 195. Characklis, W.G. 1990. Microbial fouling. In: Characklis, W.G., Marshall, K.C., editors. Biofilms. New York: John Wiley & Sons, p. 523. Characklis, W.G., McFeters, G.A., Marshall, K.C. 1990. Physiological ecology in biofilm systems. In: Characklis, W.G., and Mars hall, K.C., Editors Biofilms. New York: John Wiley & Sons. p.341-394 Christensen, B.B. Sternberg, C., Andersen, J. B., Palmer, Jr., R.J., Nielsen, A.T, Givskov, M., and Molin, S. 1999. Molecular Tool fo r Study of Biofilm Physiology Growth. In: R.J. Doyle, editor Methods in Enzymology. Vol. 310. Biofilm. p.20-42. Chumkhunthod, P., Schraft, H., Griffiths, M. W. 1998. Rapid monitoring method to assess efficacy of sanitizers against Pseudomonas putida biofilms. Journal of Food Protection. 61(8):1043-1046. Clinger, W.C. 1973. Cleaning and sanitizing operation. In: Current Concepts in Food Protection. FDA, USDHHS Cincinnati, OH. p. 8. Minute Maid. Division of Coca-Cola North America. 2005. Available from: http://www.minutemaid.com/index.jsp. Accessed: November 5, 2005. Corpe, W.A. 1980. Microbial surface components involved in adsorption of microorganisms onto surfaces. In: Bitton, G., Marshall, K.C., editors. Adsorption of microorganisms to surfaces. New York: John Wiley & Sons, p. 105. Costerton, J.W. 1995. Overview of microbi al biofilms. Journal of Industrial Microbiology. 15(3):137-40. Cousin, M.A. 1982. Presence and activity of ps ychrotrophic microorganisms in milk and dairy products: A review. Jour nal of Food Protection. 45:172-207.

PAGE 144

131 Cousin, M.A. Jay, J.M., Vasavada, P.C. Ch apter 13 Psychrotrophic Microorganisms. Downes, F.P. and Ito, K. 2001. Compendium of Methods for the Microbiological Examination of Foods. 4th ed. American Public Health Association, Washington, D.C. Cowan, M.M., Warren, T.M., Fletcher, M. 199 1. Mixed species colonization of solid surfaces in laboratory biofilms. Biofouling. 3:23-34. Davey, M.E., OToole, G.A. 2000. Microbial biofilms: from ecology to molecular genetics. Microbiology and Molecu lar Biology Review. 64(4):847-867. Deinhard, G., Blanz, P., Poralla, K. 1984. Bacillus acidoterrestris sp. nov., a new thermotolerant acidophile isolated from different soils. Systematic and Applied Microbiology. 10:47-53. Den Aantrekker, E.D., Vernooij, W.W., Re ij, M.W., Zwietering, M.H., Beumer, R.R., Van Schothorst, M., and Boom, R.M. 2003. A Biofilm Model for Flowing Systems in the Food Industry. Journal of Food Protection. 66(8):1432-1438. Different Products Listed as Being Hauled on Tankers Washed at an Indiana Wash Facility. (~1999). DIFCO Manual: Dehydrated Culture Media Reagents for Microbiology. 10th ed. 1984. DIFCO Laboratories. Detroit, MI. Donlan R.M. 2000. Biofilm control in industr ial water systems: approaching an old problem in new ways. In: Evans LV, edito r. Biofilms: recent advances in their study and control. Amsterdam: Harwood Academic Publishers; p. 333. Donlan, R.M. 2000. Role of biofilms in an timicrobial resistance. ASAIO Journal, 46;S47S52. Donlan, R.M. 2002. Biofilms: Microbial Life on Surfaces. Emerging Infectious Diseases. 8(9):881-890. Downes, F.P. and Ito, K. 2001. Compendium of Methods for the Microbiological Examination of Foods. 4th ed. American Public Health Association, Washington, D.C. Duddridge, J. E., and Pritchard, A.M. 1983. Microbial corrosion, p. 28-35. The Metals Society, London, United Kingdom. Durack, D.T. 1975. Experimental bacterial endo carditis. IV Structure and evolution of very early lesions. Journal of Pathology. 115:81. Ehlers, L.J., Bouwer, E.J. 1999. RP4 plasmid transfer among species of Pseudomonas in a biofilm reactor. Water Science Technology. 7:163.

PAGE 145

132 Entis, P., Fung, D.Y.C., Griffiths, M.W., McIntyre, L. Russell, S., Sharpe, A.N., Tortorello. Chapter 10. Rapid Methods for Identification, and Enumeration. Downes, F.P. and Ito, K. 2001. Compendium of Methods for the Microbiological Examination of Foods. 4th ed. American Public Health Association, Washington, D.C. Esteves, C.L., Jones, B.D., Clegg, S. 2005. Biofilm formation by Salmonella enterica serovar Typhimurium and Escherichia coli on epithelial cells following mixed inoculations. Infection a nd Immunity. 73(8):5198-5203. Evancho, G.M., Walls, I. 2001. Aciduric Flat Sour Sporeformers. p.239-244. In: F. P. Downes and K. Ito, editors Compendium of methods for microbial examination of food. American Public Health Association, Washington, D.C. Everaert, E.P., Mahieu, H.F., van de Belt-Gritt er, B., Peeters, A.J., Verkerke, G.J., van der Mei, H.C., and Busscher, H.J. 19 99. Biofilm formation in vivo on perfluoroalkylsiloxane-modified voice prostheses Archives of OtolaryngologyHead and Neck Surgery. 125(12):1329-1332. Farkas, J. Physical Methods of Food Pres ervation. p.567-591. In: Doyle, M.P., Beuchat, L.R, and Montville, T.J. Food Microb iology: Fundamentals and Frontiers. 2nd ed. ASM Press: Washington D.C. p. 111-114. FDA/CFSAN: Chapter IV: Outbreaks Associated with Fresh Produce: Incidence, Growth, and Survival of Pat hogens in Fresh and Fresh-Cut Produce. 2001. Analysis and Evaluation of Preventive Control Measures for the Control and Reduction/Elimination of Microbial Hazar ds on Fresh and Fresh-Cut Produce. Retrieved April 22, 2005. Available: h ttp://vm.cfsan.fda.gov/~comm/ift3-4o.html. FDA/CFSAN. 2001. Hazard Analysis and Critic al Control Point (HACCP); Procedures for the Safe and Sanitary Processing and Importing of Juice; Final Rule. Available from: http://www.cfsan.fda.gov/~lrd/f r01119a.html. Accessed December 2, 2005. FDA/CFSAN: National Conference on Inters tate of Milk Shipments (NCIMS). 2003. Grade A Pasteurized Milk Ordinance. Available from: http://vm.cfsan.fda.gov/~ear/p-nci.h tml. Accessed: November 5, 2005. Federle, M.J., Bassler, B.L. 2003. Interspeci es Communication in B acteria. The Journal of Clinical Investiga tion. 112(9):1291-1299. Fellers, P.J., Nikdel, S., Lee, H.S. 1990. Nutrient content and nutrition labeling of several processed Florida citr us juice products. Journal of the American Dietetic Association. 90(8):1079-1084. Floridas Natural. 2004. Products. Availabl e from: www.floridanat ural.com. Accessed June 31, 2004.

PAGE 146

133 Flemming, H-C., Wingender, J., Griegbe, Ma yer, C. 2000. Physico-chemical properties of biofilms. In: Evans, L.V., editor. Bi ofilms: recent advances in their study and control. Amsterdam: Harwood Academic Publishers, p. 19. Food Industry Transportation Coaliti on. 2003. Bulk Over-the-Road Food Tanker Transport Safety and Security Guidelines. 2003. Available from: http://vm.cfsan.fda.gov/~comm/transafe.html. Accessed April 17, 2004. Foster, E.M. Nelson, F.E. Speck, M.L. Doetsch, R.N., Olson, J.C. 1957. Dairy microbiology. Practice-Hall., Englewood Cliffs, N.J. Foulds, L.R., Wilson, J.M. 1997. A variation of the generalized assignment problem arising in the New Zealand dairy industr y. Annals of Operations Research. 69 (1997):105-114. Fozo, E.M., Quivey, RG Jr. 2004. Shifts in the membrane fatty acid profile of Streptococcus mutans enhance survival in acidic environments. Applied and Environmental Microbiology. 70(2):929-936. Frank, J.F. 2001. Milk and dairy products. In: D oyle, M.P., Beuchat, L.R, and Montville, T.J. Food Microbiology: F undamentals and Frontiers. 2nd ed. ASM Press: Washington D.C. p. 111-114. Frank, J.F., Chmielewski, R. 2001. Influen ce of Surface Finish on the Cleanability of Stainless Steel. Journal of Food Protection. 64(8):1178-1182. Frazier, W.C. 1958. Food microbi ology. McGraw-Hill, New York. FSIS Safety and Security Guidelines for th e Transportation and Di stribution of Meat, Poultry, and Egg Products. 2003. Available from: http://www.ams.usda.gov/tmd/TSB/truck.htm. Accessed April 17, 2004. Fuentes, J.W., Hazen, T.C., Lopez-Torres, A.J., Rechani. P. 1985. Klebsiella pneumoniae in orange juice concentrate. App lied and Environmental Microbiology. 49:15271529. Ganderton, L., Chawla, J., Wint ers, C., Wimpenny, J., Stic kler, D. Scanning electron microscopy of bacterial biofilms on indwelli ng bladder catheters. European Journal of Clinical Microbiology and In fectious Diseases. 11(9):789-796. Ghigo, J.M. Natural conjugative plasmids induce bacterial biofilm development. 2001. Nature. 412:442-445. Gilbert, P., Evans, D.J., Brown, M.R.W. 1993. Formation and dispersal of bacterial biofilms in vivo and in situ. Journal of Applied Bacteriology Symposium Supplement, 74:67SS.

PAGE 147

134 Goff, D. Dairy Science and Technology E ducation, University of Guelph, Canada. Dairy Chemistry and Physics. Available from: http://www.foodsci.uoguelph.ca/dairyedu/chem.html. Accessed November 5, 2005. Gorman, S.P., Adair, C.G., Mawhinney, W.M. 1994. Incidence and nature of peritoneal catheter biofilm determined by electron and confocal laser scanning microscopy. Epidemiological Infection.112(3):551-559. Hatcher, W. S. Jr., Parish, M.E., Weihe, J.L., Splittstoesser, D.F., Woodward, B.B. 2001. Fruit beverages, p. 565-568. In: F. P. Downes and K. Ito, editors, Compendium of methods for microbial examination of f ood. American Public Health Association, Washington, D.C. Hauser, M., Wuertz S. 1999. High rates of conjugation in bacterial biofilms as determined by in situ anaylsis. App lied and Environmental Microbiology. 65:37103713. Hensyl, W.R. 1994. Bergeys Manual of Determinative Bacteriology. 9th ed. Baltimore: Williams and Wilkins. Heukelekian, H., Hellar, A. 1940. Relation be tween food concentration and surface for bacterial growth. Journal of Bacteriology. 40:547-558. Hood, S.K., Zottola, E.A. 1997. Adheren ce to stainless steel by foodborne microorganisms during growth in model f ood systems. International Journal of Food Microbiology. 37:145. HR.2088: Safe, Accountable, Flexible, and E fficient Transportation Equality Act of 2003. Available from: http://thomas.loc.gov/cgibin/bdquery/D?d108:1:./temp/~bd X7Aq::|/bss/d108query.html|. Accessed April 20, 2004. Huang, C-C, Cheng, T-C, Wei-Chung, T ., Jeng, J-Y., Wang, T-K. 2001. Study on the flat-sour spoilage of the canned Hsian-Ts ao gel. Journal of the Agricultural Association of China. 2(5):423-435. Jenness, R. 1988. Composition of milk. In: W ong, N.P. editor. Fundamentals of Dairy Chemistry. Van Nostrand Reinhol d Co., New York, N.Y. p.1-38. Jennings, W.G. 1963. An interpretive review of detergency for the food technologist. Food Technology. 17:863-881. Joseph, B., Otta, S.K., Karunasagar, I., Ka runasagar, I. 2001. Biofilm formation by salmonella spp. on food contact surfaces a nd their sensitivity to sanitizers. International Journal of Food Microbiology. 64(3):367-372.

PAGE 148

135 Juice Products Association. 2004. Model Tanker Wash Guidelines For the Fruit Juice Industry. Available from: http://www.juiceproducts.org/pdf/Model TankerWashGuidelines.pdf. Accessed: October 11, 2005. Kader, A.A. Ed. 2002. Postharvest Technology of Horticultural Crops. 2nd ed. Oakland, CA: The Regents of the Univ ersity of California. Kaplan, M.T. Appleman, M.D. 1952. Microbiol ogy of frozen orange concentrate. III. Studies of enterococci in frozen ora nge concentrate. Food Technology. 6:167-170. Katsuyama, A.M. Ed. 1993. Principles of Food Processing Sanitation. 2nd ed. Food Processor Institute. Washington, D.C. Keevil, C.W. 2001. Continuous Culture Models to Study Pathogens in Biofilms. In: R.J. Doyle Editor. Methods in Enzymology. Vol. 337. Microbial Growth in Biofilm Part B: Special Environments and Physicochemical Aspects. p. 104-122. Larkin, E.P., Litsky, W., Fuller, J.E. 1955. Fecal streptococci in frozen foods III. Effect of freezing on Escherichia coli, Streptococcus faecalis and Streptococcus liquefaciens inoculated into orange concen trate. Applied Microbiology. 3:104-106. Leunisse, C., van Weissenbruch, R., Busscher, H. J., van der Mei, H.C., Dijk, F., Albers, F.W. 2001. Biofilm formation and design f eatures of indwelling silicone rubber tracheoesophageal voice prostheses--an electron microscopical study. Journal of Biomedical Materials Re search. 58(5):556-563. Luft, J.H. 1971. Ruthenium red and violet. I. Chemistry, purification, methods of use for electron microscopy and mechanism of action. Anatomical Record. 171:347-368. Maller, R.R. 1998. Passivation of Stainle ss Steel. Trends in Food Science and Technology. 9: 28-32. Malloch, D. 1981. Moulds: Their Isolation, Cu ltivation and Identification. Toronto: University of Toronto Press. Mauer, I.M. 1974. Sterilizati on and disinfection. In: Hobbs, B.C,. editor. Food Poisoning and Food Hygiene, 3rd ed. Edward Arnold Ltd., London. p. 183-191. Millsap, K.W., Reid, G., van der Mei, H.C., Busscher, H.J. 1997. Adhesion of Lactobacillus species in urine and phospha te buffer to silicone rubber and glass under flow. Biomaterials. 18(1):87-91. Murga, R., Forster, T.S., Brown, E., Pruc kler, J.M., Fields, B.S., Donlan, R.M.. 2001. The role of biofilms in the survival of Legionella pneumophila in a model potable water system. Microbiology, 147:3121.

PAGE 149

136 Nealson, K.H., Hastings J.W. 1979. Bacterial bioluminescence: its control and ecological significance. Microb iology Review. 43:496. Ohio State University. 1968. Cleaning compoun ds: characteristics and functions, WJH68-17-1. Department of Dairy Technology, The Ohio State University, Columbus, OH. Office of Inspector General: Audit Re port. 1998. Department of Transportation. Available from: http://www.oig.dot.gov/s how_txt.php?id=316. Accessed April 17, 2004. Olson, J.C. Mocquot, G. 1980. Milk and milk products. In: Skinner, J.H., Elliot, R.P. Baird-Parker, A.C., Bryan, F.L., Christ ian, J.H., Clark, D.S., Olson, J.C., and Roberts, T.A., editor. Micr obial ecology of foods. Acad emic Press, New York. p. 470-520. Olson, K.E., Sorrells, K.M. Chapter 25 Ther mophilic Flat Sour Sporeformers. Downes, F.P. and Ito, K. 2001. Compendium of Methods for the Microbiological Examination of Foods. 4th ed. American Public Health Association, Washington, D.C. Oyarzbal, O.A., Nogueira, M.C., Gombas, D.E. 2003. Survival of Escherichia coli O157:H7, Listeria monocytogenes, and Salm onella in juice concentrates. Journal of Food Protection. 66(9):1595-1598. Pez, R., Taverna, M., Charlon, V., Cuatrin, A., Etcheverry, F., da Costa, L. H. 2003. Application of ATP-bioluminescence t echnique for assessing cleanliness of milking equipment, bulk tank and milk tran sport tankers. Food Protection Trends. 23(4): 308-314. Parish, M.E. 1997. Public Health and Nonpast eurized Fruit Juices. Critical Reviews in Microbiology. 23(2):109-119. Parish, M.E. 1998. Coliforms, Escherichia coli, and Salmonella Serovars Associated with Citrus-Processing Facility Implicated in a Salmonellosis Outbreak. Journal of Food Protection. 61(3):280-284. Parish, M.E., Goodrich, R.M. 2005. Recovery of Presumptive Alicyclobacillus Strains on Orange Fruit Surfaces. Submitted to the Journal of Food Protection Parish, M.E., Goodrich, R., Miller, W. 2004. Fate of Salmonellae in orange and grapefruit concentrates during cold storage. Journal of Food Protection. 67(12):2671-2674. Parker, M.E., Litchfield, J.H. 1962. Effectiv e detergency and cleaning practices. Food plant sanitation. Reinhold Publishing Corporation.

PAGE 150

137 Pasmore, M., Costerton, J.W. 2003. Biofilms, bacterial signaling, and their ties to marine biology. Journal of Industrial Microbio logy and Biotechnology. 30(7):407-13. Patrick, R. 1953. Coliform bacteria from orange juice concentrate and damaged oranges. Food Technology. 7:157-159. Put, H.M.C., de Jong, J., Sand, F.E.M. van Grinsven, A.M. 1976. Heat resistance studies on yeast spp. causing spoilage in soft drinks. Journal of Applied bacteriology. 40:135-152. Raad, I.I., Sabbagh, M.F., Rand, K.H., Sherer tz, R.J. 1992. Quantitative tip culture methods and the diagnosis of central venous catheter-related infections. Diagnostic Microbiology and Infectious Disease.15:13. Ramos, C. Licht, T.R., Sternberg, C., Kr ogfelt, K.A., and Molin, S. 2001. Monitoring bacterial growth activity in biofilms from Laboratory Flow Chambers, Plant Rhizosphere, and Animal Intestine. In: R.J. Doyle, Editor. Methods in Enzymology. Vol. 337. Microbial Growth in Biofilm Part B: Special Environments and Physicochemical Aspects. p.21-42. Redd, J.B., Hendrix, C.M., Hendrix, D.L. 1986. Quality Control Manual For Citrus Processing Plants. Volume I: Regul ation, Citrus Methodology, Microbiology, Conversion Charts, Tables, Other. In tercit, Inc., Safety Harbor, FL. Reid, G., Tieszer, C., Lam, D. 1995. Influe nce of lactobacilli on the adhesion of Staphylococcus aureus and Candi da albicans to fibers and epithelial cells. Journal of Industrial Microbiology.15(3):248-253. Richter, R.L., Vedamuthu, E.R. Chapter 47 Milk and Milk Products. Downes, F.P. and Ito, K. 2001. Compendium of Methods for the Microbiological Examination of Foods. 4th ed. American Public Health Association, Washington, D.C. Roberts, A.P., Pratten, J., Wilson, M., Mu llany, P. 1999. Transfer of a conjugative transposon, Tn5397 in a model oral bi ofilm. FEMS Microbiology Letters. 177:636. Ronner, A.B., Wong, A.C.L. 1993. Biofilm deve lopment and sanitizer, inactivation of Listeria monocytogenes and Salmonella typhimurium, on stainless steel and BunaN rubber. Journal of Food Protection. 56:750-758. Rosenow, E.M., Marth, E.H. 1987. Listeria, listeriosis a nd dairy foods: A review. Cultured Dairy Products Journal. 22:13-17. Sambrook, J., Fritsch, E.F., Maniatis, T. 1989. Molecular Cloning: A Laboratory Manual. 2nd ed. Cold Spring Harbor Press. New York.

PAGE 151

138 Silva, F.M., Gibbs, P., Vieira, M.C., S ilva, M.C. 1999. Thermal inactivation of Alicyclobacillus acidoterrestris spores under different te mperature, soluble solids and pH conditions for the design of fruit pr ocesses. Internatio nal Journal of Food Microbiology. 51:95-103. Skillman, L.C., Suthland, I.S., Jones, M.V., Goulsbra, A. 1998. Green fluorescent protein as a novel species-specific marker in en teric dual-species biofilms. Microbiology. 144:2095-2101. Smolin, L.A., Grosvenor, M.B. 2000. Nutr ition: Science and Applications. (3rd ed.) New York: Saunders College Publishing. Southeast Regional Climate Center. Winter Haven, FL Period of Record Monthly Climate Summary Period of Record : 7/ 1/1948 to 3/31/2004. Available from: http://cirrus.dnr.state.sc.us/cgi-bin/ser cc/cliMAIN.pl?fl9707. Accessed November 7, 2005. Spray Systems Compan y. 2004. Available from http://www.spray.com/products/tank_w ash.asp. Accessed June 12, 2004. Stadhouders, J. 1975. Microbes in milk and dairy products: An ecological approach. Netherlands Milk Dairy Journal. 29:104-126. Stanley, P.E. 1982. Rapid microbial coun ting using ATP technology. Food. October, 2934. Stark, R.M., Gerwig, G.J., Pitman, R.S., Potts, L.F., Williams, N.A., Greenman, J., Weinzweig, I.P. Hirst, T.R., Milla r, M.R. 1999. Biofilm formation by Helicobacter pylori. Letters in Applie d Microbiology. 28:121. Steele, M.L., McNab, W.B.; Poppe, C., Griff iths, M.W., Shu-Chen, Degrandis, S.A., Fruhner,L.C., Larkin, C.A., Lynch, J.A., Odumeru, J.A. 1997. Survey of Ontario bulk tank raw milk for food-borne pathoge ns. Journal-of-Food-Protection. 60(11): 1341-1346. Stickler, D. 1999. Biofilms. Current Op inion in Microbiology. 2:270-275. Storgards, E., Simola, H., Sjberg, A-M ., Wirtanen, G. 1999. Hygiene of Gasket Materials Used in Food Processing Equi pment Part 1: New Materials. Trans IChemE. 77(C): 137-145. Storgards, E., Simola, H., Sjberg, A-M ., Wirtanen, G. 1999. Hygiene of Gasket Materials Used in Food Processing Equipm ent Part 2: Aged Materials. Trans IChemE. 77(C): 146-155.

PAGE 152

139 Sutton, S.V., Franco, R.J., Porter, D.A., Mowrey-McKee, M.F., Busschaert, S.C., Hamberger, J.F., Proud, D.W. 1991. D-value determinations are an inappropriate measure of disinfecting ac tivity of common contact le ns disinfecting solutions. Applied Environmental Mi crobiology. 57(7):2021-6. Sutherland, I.W. 2001. Biofilm exopolysacchar ides: a strong and sticky framework. Microbiology,147:3. Tamoschus. 1979. The milk tanker a risk to milk quality? Deutsc he-Milchwirtschaft 30(22): 820, 823, 826. Thomas, S.B. 1974. The microflora of bulk collected milkPart I. Dairy Industry International. 39:237-240. Thomas, S.B. 1974. The microflora of bulk collected milkPart II. Dairy Industry International. 39:279-282. Tomlins, R.L. Ordal, Z.J. 1976. Thermal inju ry and inactivation in vegetative bacteria, p.153-190. In: Skinner, F.A., and Hugo, W.B ., editors, Inhibition and Inactivation of Vegetative Microbes. So ciety of Applied Bacteriology Symposia Series, no. 5. Academic Press, Inc., New York. Tropicana. Division of Pepsico. 2004. Product Information. Available from: www.tropicana.com. Accessed June 31, 2004. Tortora, G.J., Funke, B.R., Case, C. L. 1998. Microbiology: An Introduction. 6th ed. Benjamin/Cummings Publishing Company, Inc. Menlo Park, CA. Tunney, M.M., Jones, D.S., Gorman, S.P. 1999. Biofilm and biofilm-related encrustations of urinary tract devices. In: Doyle RJ, editor. Methods in Enzymology, vol. 310. Biofilms. San Diego: Academic Press, p. 558. Ubgabe, R.R., Sankaran, J.K. 1994. Routing tank ers for dairy milk pickup. Interfaces. 24(5):59-66. Uboldi-Eiroa, M.N, Amstalden-Junqueira,V.C., Schmidt, F.L. 1999. Alicyclobacillus in orange juice: occurrence and heat resist ance of spores. Journal of Food Protection. 62(8): 883-886. United States Code (USC). 2000. Available from: http://www.gpoaccess.gov/uscode/. Accessed April 17, 2004. University of California Museum of Paleontology. Antony van Leeuwenhoek (16321723). Available from: http://www.ucmp.be rkeley.edu/history/leeuwenhoek.html. Accessed November 5, 2005. USDAs Bacteriologica l Analytical Manual Online. 2000. Available from: http://vm.cfsan.fda.gov/~ebam/bam-toc.html. Accessed November 5, 2005.

PAGE 153

140 USDAs Bad Bug Book: Foodborne Pathogeni c Microorganisms and Natural Toxins Handbook. 1992. Available from: http://v m.cfsan.fda.gov/~mow/intro.html. Accessed November 5, 2005. USDAs Citrus Fruit Summary. 2004. Available from: http://usda.mannlib.cornell.edu/reports/nassr/fruit/zcf-bb/cfr0904.txt. Accessed November 5, 2005. USDAs Foreign Agricultural Service (FAS): Horticultural and Tropi cal Fruits Division. 2004. Situation and Outlook for Orange Juice. 2004. Available from: http://www.fas.usda.gov/ht p/circular/2004/1-3004%20Web%20Art.%20Updates/1-3004%20Orange%20Juice%20Feature.pdf. Accessed June 6, 2004. USDAs Foreign Agricultural Service (FAS): Production Supply and Distribution. 2004. Horticultural Products pre defined tables fo r fresh oranges and grapefruit and their juices. Available: http://www.fas.usda .gov/psd/intro.asp?circ_id=9. Accessed June 6, 2004. Vanderzant, C., Matthys, A.W. 1965. Effect of temperature of plating medium on the viable count of psychrophilic bacteria Journal of Milk Food Technology. 28:383388. Velraeds, M.M., van de Belt-Gritter, B., Bussc her, H.J., Reid, G., van der Mei, H.C. 2000. Inhibition of uropathogenic biofilm grow th on silicone rubber in human urine by lactobacilli--a teleologic approach. World Journal of Urology. Dec;18(6):422426. Walls, I., and Chuyate, R. 1998. Alicyclobacillushistorical perspective and primary characterization study. Dairy, Food, and Environmental Sanitation. 18:1-5. Walls, I., Chuyate, R. 2000. Spoilage of fruit juices by Alicyclobacillus acidoterrestris. Food Australia. 52:286-288. Walls, I., Chuyate, R. 2000. Isolation of Alicyclobacillus acidoterrestris from fruit juices. Journal of the AOAC International. 83:115-120. Washam, C.J. Olson, H.C., Vedamuthu, E.R. 197 7. Heat-resistant psychrotrophic bacteria isolated from pasteurized milk. Journal of Food Protection. 40:101-108. Watnick, P., Kolter, R. 1999. Steps in the development of Vibrio cholerae El Tor biofilm. Molecular Mi crobiology, 34(3):586-595. Watnick, P., Kolter, R. 2000. Minireview: Biofilm, City of Mi crobes. Journal of Bacteriology. 182(10): 2675-2679 Winniczuk, P.P., Parish, M.E. 2005. Unpublished Baseline Data from the USDA Tanker Project.

PAGE 154

141 Wirtanen, G., Alanko, T., Mattila-Sandholm, T. 1996. Evaluation of epifluorescence image analysis of biofilm growth on stainl ess steel surfaces. Colloids and Surfaces B: Biointerfaces, 5:319326. Wisse, C.A., Parish, M.E. 1998. Isolation and enumeration of sporforming, thermoacidophilic, rod-shaped bacteria fr om citrus processing environments. Dairy, Food, and Environmen tal Sanitation. 8:504-509. Worobo, R. 2005. Identifying Alicyclobacillus cultures by 16S rRNA gene comparison. Personal communication. Wyatt, M.K., Parish, M.E., Widmer,W.W ., Kimbrough, J. 1995. Characterization of mould growth in orange juice. Food Microbiolo gy. 12(4): 347-355. Wyatt, M.K., Parish, M.E. 1995. Spore germinat ion of citrus-juice re lated fungi at low temperatures. Food Microbiology. 12(3): 237-243. Zobell C.E. 1943. The effect of solid su rfaces on bacterial ac tivity. Journal of Bacteriology. 46:39-56.

PAGE 155

142 BIOGRAPHICAL SKETCH Marjorie Ruth Richards wa s born in Elmira, NY. In 1999 she graduated with high honors from Corning-Painted Post East High Sc hool. After she receiving an Associates of Science degree in Liberal Arts and Sciences: Mathematics and Science from Corning Community College in 2001 she enrolled in Cornell University s Food Science program. During her time at Cornell she particip ated in the Food Science Club and the Product Development Team. She had externsh ips with the Nabisco Division of Kraft Foods and with the Lipton Division of Unilever-Best Foods. She worked with the Milk Quality Improvement Program preparing materials and learning the sensory, microbiological, and chemical tests for dair y products. She worked on two research projects during her undergraduate studies. One with the Milk Quality Improvement Program to determine what microorganisms we re responsible for causing chocolate milk to spoil faster than non-flavored milk. The second was with the USDAs Plant Soil Nutrition Laboratory where she learned to cr eate and maintain CACO-2 cell cultures for in vitro iron uptake studies. Her research pr oject was to determine how different concentrations phytic acid would affect intestinal absorption of different concentrations of iron in Nishiki rice using the CACO-2 ce ll cultures. She gra duated in January 2003 from Cornell University with a Food Sc ience degree with an emphasis in food processing. Starting in January 2003 she came to the Un iversity of Florida to work on the USDA tanker sanitation grant. Marjorie ha s been a member and the secretary for the

PAGE 156

143 Gator Citrus Club, and the student representative to the Citrus Products Division of IFT. She has worked on a research project to isolat e, confirm, and determ ine the relatedness of Salmonella from agricultural irrigati on water from central Florida lakes. Marjorie is the recipient of the George Truitt Scholarship from the IFT Citrus Products Division for the benefit of her research to the citrus industry and D. Glynn Davies Scholarship from the Juice Products Association for the successful completion of her inte rnship with Blue Lake Citrus Products.


Permanent Link: http://ufdc.ufl.edu/UFE0013129/00001

Material Information

Title: Microbial Composition, Biofilm Formation, and Removal from the Surfaces of the Manway Lid Gaskets of Citrus and Dairy Liquid Transport Tankers
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013129:00001

Permanent Link: http://ufdc.ufl.edu/UFE0013129/00001

Material Information

Title: Microbial Composition, Biofilm Formation, and Removal from the Surfaces of the Manway Lid Gaskets of Citrus and Dairy Liquid Transport Tankers
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013129:00001


This item has the following downloads:


Full Text












MICROBIAL COMPOSITION, BIOFILM FORMATION, AND REMOVAL FROM
THE SURFACES OF THE MANWAY LID GASKETS OF CITRUS AND DAIRY
LIQUID TRANSPORTATION TANKERS















By

MARJORIE RUTH RICHARDS


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Marjorie Ruth Richards

































This thesis is dedicated to my family, friends, colleagues, teachers, and committee
members for their help and their support.















ACKNOWLEDGMENTS

I would like to thank my parents (my editors) for their continued love, support, and

wisdom. Their undying support and belief in my abilities have made the last few years

possible. I would like to thank my siblings-Cindy, Heather, Daniel, Jessica, and

Benjamin-for being my best friends, my greatest supporters, and my toughest critics.

I would like to thank USA Tank Wash, Bynum Transport, Oakley Transport, Inc.

and Clewiston Tank and Truck Wash for their help and support in conducting this

research project.

I would also like to thank the members of the committee. I thank original advisor

Dr. Parish who allowed me to participate in this project, largely allowed me the freedom

to design and create the second part of this experiment, and for giving me guidance and

expanding my love of food microbiology. I thank my advisor Dr. Goodrich for assisting

me after Dr. Parish left for the University of Maryland, in preparations of the thesis, my

defense, and future career plans. I thank my co advisor Dr. Archer for his guidance in

selecting courses, and for his good-humored attitude. I thank Dr. Wright for her advice

on molecular techniques and for allowing me to participate in the .,\ /l,,nll', /d ike water

project to increase my knowledge of molecular techniques. And last but not least I thank

Dr. Welt and all the members of his lab for designing the model tanker system.

















TABLE OF CONTENTS


A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ........ ........................ ......... .................. ix

LIST OF FIGURES ......... ....... .................... .......... ....... ............ xi

ABSTRACT .............. .............................................. xii

CHAPTER

1 IN TR O D U C T IO N ............................................................. .. ......... ...... .....

Justification for R research ............................................ ...................................... 1
P previous R research ............................................... 3
Specific A im s and O bjectives............................................................................... .. 4
Part I: Identification and Characterization of Microorganisms in Samples ..........4
Objectives ................. ......... .... ......... .. .............. ..4
H y p oth eses .............. .. ............... ...................... ................ .. 4
Part II: Biofilm Development and Removal .....................................................5
O b je ctiv e s ................... ......................................................... 5
H y p oth esis ............................................. ................. 5

2 REVIEW OF THE LITERATURE ........................................ .......................... 6

T he L egal H history ......................... ............................. ... ........ ........ .. 9
Marketing and Food Safety Justification for Citrus Juice Transportation Tanker
R e search ....................................................................... ................ 17
Previous Research on Transportation Tankers .................................. ...............19
Citrus Juice and Milk and Their Microbial Inhabitants ...........................................23
The Environm ent of Citrus Juice...................................... ......................... 23
M icrobial Flora of Citrus Juice......................................... ......................... 24
B a cteria ........................................................................... 2 4
Y e a sts ...................................................................................................... 2 7
M old s ...................................2 8
The Environment of Liquid Dairy Products....................................................28
B io fi lm s ............................................................................. 3 1
A ttachm ent ........................................................................32
M material ............................................. 33


v









Cellular com ponents ........................................ ................................... 33
Characteristics of the liquid m edia............... ..............................................34
F orm action .................................................. ............. ...............35
Extracellular polymeric substances (EPS) ................................................36
Architecture .......................... ....................... ................. 37
O their bacteria and particles...................................... ........................ 37
M a tu ratio n ..................................................................................................... 3 8
O their b bacteria .................. ................... ....................................... .. 38
G ene transfer and regulation ............................................. ............... 39
Q u o ru m sen sin g ........... ...... ................................ .................. .. .... .. .. ... 3 9
P athogenic organism s........................................................ ............... 42
R resistance ............................................................................. 44
D isp ersal ................................................................... 4 5
D etergents and Sanitizers ........................................ ............................................4 5
D etergents ............................................................................................. ....... 45
Sanitizers ....................... ........ ................47
The Environm ent of Stainless Steel................................ ......................... ........ 48
Environm ent of Rubber ...........................................................................52
R eview of M ethodology ........................................................................................ 54
Coliforms, Fecal Coliforms, and E coli.......................................................54
D election M ethods for Salm onella .............................. ................................... 56
Detection Methods for Alicyclobacillus ......................................... .................57
Detection of Aciduric, Yeast and Mold, Thermoduric, Mesophilic and
Psychroduric M icroorganism s ........................................ ...... ............... 57
DNA Sequencing ....... .. ....... .... ......... .. .......... ............. 58
Biofilm Growth Characterization..... .................... ...............59
O observation M ethods................................................. .............................. 59
A N eed for M ore R research ........................................ ..... ................. ............... 61

3 M ATERIALS AND M ETHODS ........................................ ......................... 62

Part I: Identification and Characterization of Microorganisms in Samples ...............62
Sam ple C collection .............................. ........................ .. ...... .... ...... ...... 62
Sam ple P reparation ........ ................................................................ .. .... ..... .. 64
Sam ple A analysis .................... .......... ...... ...... .. .. ......... ...............64
Psychroduric, mesophilic, thermoduric, yeast and mold; and aciduric
enum eration and characterization ................................. ............... 64
Coliform, fecal coliform, and E. coli detection..............................66
Streptococcus spp. and Staphylococcus spp. detection..............................67
Salm onella spp. detection....................................... .......................... 67
Alicyclobacillus spp. detection........................................... 67
16S DNA and 28S rRNA PCR Identification ............................................. 68
Statistical A analysis ................... ............. .... .. .......................... .............. 68
Part II: Biofilm Development and Removal............................................................69
Liquid Sam ple Preparation ........................................................ ............... 69
Standard grow th curves .................................................... ............... ... 69
Model of Liquid Transportation Tanker Manway.............................................71









G asket Treatm ent ............................................ .. .......... ..... .. ........ .... 73
M icrobial Analysis of Gasket .................... .. .... ............. ........ ....... 74
Scanning Electron M icroscopy....................................... ......................... 75
Fluorescence M icroscopy ...................... .............. .................... ............... 75
Statistical A analysis .......................... .......... ............... .... ..... .. 76

4 R E S U L T S .............................................................................7 7

Part I: Sample Identification and Characterization................... ........................77
Psychroduric, Mesophilic, Thermoduric, Yeast and Mold, and Aciduric
Microorganism Enumeration and Characterization.......................................77
Coliform, Fecal Coliform, and E. coli Detection ..........................................87
Streptococcus and Staphylococcus Detection ............................................. 87
Salmonella and Alicyclobacillus Detection........................ ................... 88
16S DNA and 28S rRNA PCR Identification ............................................. 88
Part II: Biofilm Development and Removal...........................................................88
Gasket Analysis .................................... .......................... .... ....... 88
Scanning Electron M icroscopy....................................... ......................... 90
Fluorescence M icroscopy ......................................................... ................ 93

5 DISCUSSION AND CONCLUSIONS ............................................................... 95

Part I: Sample Identification and Characterization................... ..........................95
Psychroduric, Mesophilic, Thermoduric, Yeast and Mold; and Aciduric
Enum eration and Characterization........................................ .....................95
Coliform, Fecal Coliform, and E. coli Detection ..........................................98
Streptococcus and Staphylococcus Detection .............................. ..................98
Salmonella and Alicyclobacillus Detection ..........99
Part II: Biofilm Development and Removal ................................ ....................100
G asket A analysis ..................................................... ... .. .......... .. 100
Scanning Electron M icroscopy...................................... ........................ 104
F luorescence M icroscopy ............................................ ...................................... 104
O overall C onclu sions......... .......................................................... ... ... .... .... .. 105

6 FU TU R E W O R K ................................................................................. ........ 106

E xten sion ............................................ ...... ............... ................. 106
Gasket W ashing Video and/or Training M anual..............................................106
Workshops ............... ......... .......................107
Tank W ash Association ........... ............. ............. .......... ..... 107
Research ......... ........... ................................ 107
Juice Concentrate Research.......................................................... ............... 107
B iofilm R research ............ .. .......................... .................. 108
G asket A alternatives .......................... .... .............. ...........................108









APPENDIX

A PA R T I R A W D A TA ......................................................... ............... 110

B P A R T II R A W D A T A ......... ...... ........... ................. .......................................... 114

C GASKET SURFACE AREA SAMPLE CALCULATION ..................................116

D STANDARD GROW TH CURVES .............. .......... ................. ....................117

E STA TISTICAL TA BLES I.............................................. .............................. 120

F STATISTICAL TABLES II ............. .... .....................124

LIST OF REFEREN CES ............................................. ........................ ............... 128

B IO G R A PH ICA L SK ETCH ......... ................. ..........................................................142





































viii
















LIST OF TABLES


Table page

2-1 Products and cleaning steps for JPA wash types................................... .................7

3-1 Types and number of CFU of microorganisms found in target inoculated milk .....70

4-1 Product's effect on aciduric, yeast and mold, psychroduric, and mesophile ...........78

4-2 Product's effect on aciduric, yeast and mold, psychroduric, and mesophile ...........78

4-3 Gasket's effect on aciduric, yeast and mold, psychroduric, and mesophile.............79

4-4 Gasket's effect on aciduric, yeast and mold, psychroduric, and mesophile.............79

4-5 Wash temperature's effect on aciduric, yeast and mold, psychroduric ..................79

4-6 Wash temperature's effect on aciduric, yeast and mold, psychroduric .................80

4-7 Product and Gasket's effects on aciduric, yeast and mold, psychroduric ..............81

4-8 Product and Gasket's effects on aciduric, yeast and mold, psychroduric ..............81

4-9 Product and Wash Temperature's effects on aciduric, yeast and mold...................81

4-10 Product and Wash Temperature's effects on aciduric, yeast and mold................... 82

4-11 Gasket and Wash Temperature's effects on aciduric, yeast and mold.................... 82

4-12 Gasket and Wash Temperature's effects on aciduric, yeast and mold.................... 83

4-13 Product, Gasket, and Wash Temperature's effects on aciduric, yeast and mold ....84

4-14 Product, Gasket, and Wash Temperature's effects on aciduric, yeast and mold ....85

4-15 The top two bacterial characterizations on different gasket and media types.........86

4-16 Number of coliform, fecal coliform, and E. coli positive gaskets ..........................87

4-17 Percentage of Staphylococcus spp. and presumptive Staphylococcus aureus. .......88

4-18 Results of 16S DNA and 28S rRNA PCR Identification .......................................89









4-19 Loglo reductions among coliform and mesophilic counts for the three wash........89

A-1 Sample type, sample number, gasket washer, aciduric and yeast and mold.........110

A-2 Sample type, sample number, psychroduric, mesophilic, and thermoduric..........112

B-l Sample Letter and results for Part II. ............... ............................................. 114















LIST OF FIGURES


Figure page

3-1 Type A and Type B manway styles and gasket types.............................................63

3-2 M anw ay lid set up picture set 1 ........................................ ......................... 71

3-3 M anw ay lid set up picture set 2....................................... .......................... 72

3-4 M anw ay lid set up picture set 3 ............................................................................ 74

4-1 Representative pictures from scanning electron microscopy .............................91

4-2 Representative pictures from fluorescent microscopy ........................................94

5-1 An example of slow draining bucket.................. .............................................. 101

E-1 Surface area (cm2) of test type vs. product............... ....................... ............... 120

E-2 Surface area (cm2) of test type vs. gasket type............... ..................120

E-3 Surface area (cm2) of test type vs. wash type....................................................... 121

E-4 Surface area (cm2) of test type vs. product, and gasket type............................. 121

E-5 Surface area (cm2) of test type vs. product, and wash type.................................. 122

E-6 Surface area (cm2) of test type vs. gasket type, and wash type.......................... 122

E-7 Surface area (cm2) of test type vs. product, gasket type, and wash type..............123















Abstract Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

MICROBIAL COMPOSITION, BIOFILM FORMATION, AND REMOVAL FROM
THE SURFACES OF THE MANWAY LID GASKETS OF CITRUS AND DAIRY
LIQUID TRANSPORTATION TANKERS

By

Marjorie R. Richards

December 2005

Chair: Renee M. Goodrich
Cochair: Douglas L. Archer
Major Department: Food Science and Human Nutrition

Improved guidelines pertaining to the sanitation of liquid food transportation

tankers are needed to ensure safety and maximum shelf life of liquid products sold in the

United States. Studies on ATP-bioluminescence conducted by Bell and others determined

that by sampling the manway lid one could determine if a tanker was dirty or clean. The

ATP-bioluminescence study done Paez and others determined that surfaces of the

manway lid were the hardest to clean. Data collected by Winniczuk and Parish showed

that manway lids, specifically gaskets, were the hardest to clean region of the tanker.

Rubber gaskets were most likely contaminated with coliform, fecal coliform, and

Escherichia coli after cleaning. Therefore, research to better understand microbial

activity on manway lid gaskets was deemed necessary.

The first part of this research was to characterize distribution of aciduric,

thermoduric, mesophilic, psychroduric microorganisms, yeast and mold; determine









coliform and fecal coliform counts; as well as identify the Escherichia coli, Salmonella

spp., and Alicyclobacillus spp. on surfaces of gasket type A and gasket type B of citrus

and dairy tankers after warm or ambient wash. Four important results were obtained

from this portion of the study: 1) Surface shapes of gaskets are important with respect to

cleanability, 2) In cases where significant differences (P<0.05) exist the Loglo/cm2 values

of the aciduric, and mesophilic, and Loglo value of the aciduric and mesophilic

organisms/total gasket were significantly more for gasket type B from cold washed dairy

tanker gaskets than any other gasket type, 3) Salmonella spp. and Alicyclobacillus spp.

were not found on any gaskets, 4) E. coli was found on the surface of dairy, type B

gaskets.

The second part of this research utilized ultra-high temperature pasteurized milk

inoculated with E. coli, other bacteria, and a yeast collected from gaskets; and a model

tanker manway to form biofilms on lid gaskets. Effectiveness of three commercial

cleaning regimens (detergent wash/water rinse; detergent wash/sanitizer/water rinse;

detergent wash/sanitizer/water rinse/hot water treatment) on lid gaskets were evaluated

using coliform, mesophilic, and yeast counts; E. coli most probable number; and

scanning electron and fluorescence microscopy. Results showed that the detergent

wash/sanitizer/water rinse/hot water treatment is more effective than the other two at

removing both coliform and mesophilic microorganisms from the gaskets surfaces.














CHAPTER 1
INTRODUCTION

Justification for Research

There is a problem with food transportation in the United States, which arises from

the Sanitary Food Transportation Act (SFTA) of 1990. This act regulates the safety of

products transported in motor and rail vehicles; however, Department of Transportation

(DOT) is responsible for implementation and enforcement of SFTA (49 USC 5701-

1514). The Office of the Inspector General has determined that DOT has neither the

means nor time to enforce SFTA (Office of the Inspector General 1998). Lack of proper

enforcement may have played a role in three recent food transportation incidents. In

1994, over 224,000 people were affected by salmonellosis when a tanker truck that had

carried unpasteurized liquid eggs was not properly cleaned, causing cross-contamination

of salmonella to pasteurized ice cream mix that was subsequently transported in the same

tanker (Office of the Inspector General 1998). Another problem occurred in 1999 when

ice allegedly contaminated with Salmonella was illegally added to orange juice being

shipped from Mexico to Arizona (FDA/CFSAN 2001). In 1997, several decomposing

bodies were found in three ships entering the U.S. (Office of the Inspector General 1998).

The food transportation industry relies upon voluntary compliance to guidelines,

such as Bulk Over-the-Road Food Tanker Transport Safety and Security Guidelines

(Food Industry Transportation Coalition 2003), and the Juice Products Association

Model Tanker Wash Guidelinesfor the Fruit Juice Industry (2004). Voluntary guidelines

vary in quality, and compliance is inconsistent. Some of the problems with liquid food









transportation safety occur when transportation tankers are cleaned. Tankers go to wash

stations, but often remain unwashed until just prior to the next use. Time periods that

they are allowed to be unwashed is not regulated, so tankers may sit uncleaned for two

weeks or more before washing, particularly if transport business is slow. While tankers

sit uncleaned insects can infest them, and bacteria, mold, and yeast have the opportunity

to multiply and form communities called biofilms. When non-dairy tankers are finally

ready to be used, they are washed based upon non-standard washing criteria specific to

each particular wash station. Such criteria are not always based on sound scientific data

or government standards. Cleaning liquids are applied using a spray washing system that

is lowered through the tanker manway. Spray mechanisms are often selected based on

cost; these systems may not always provide the best cleaning. Spray systems may be

inefficient because sprayed solution may not reach the ends of tankers or because holes in

sprayers can and do become plugged. Other problems include cleaning trucks on a slope,

which creates an area in the tanker that cannot be properly cleaned, improper or non-

existent manual cleaning of accessory parts such as gaskets, and not cleaning filth and

debris accumulating around seals during transportation. Bell and others (1994) noted that

poor water quality, concentration of chemicals used, and operational temperature could

reduce cleaning effectiveness. Also, failure to empty tankers of all wash water can cause

recontamination and increase likelihood of microbiological problems (Bell and others

1994). Other problems are evident in the security of wash stations; some wash stations

store dirty and empty tankers in lots without security fences, leaving the tankers open to

vandalism, bioterrorism, or accidents.









The U.S. and the State of Florida have reason for concern for the tanker sanitation

situation. In 2003/2004 the US processed 37,048,000 metric tons, imported 20,005,000

metric tons and exported 19,955,000 metric tons of orange juice (USDA's Foreign

Agricultural Service (FAS): Production, 2004). Most of the citrus in the United States

comes from Florida (Kader, 2002). In Florida, the $9-billion-per-annum citrus industry is

second only to the tourism industry. Citrus generates approximately 90,000 jobs.

Ninety-five percent of Florida's oranges are processed into juice (USDA's Foreign

Agricultural Service (FAS): Horticultural, 2004). Both concentrates and pasteurized

single strength citrus juices are transported in transportation tankers. There is growing

concern about the quality and safety of products transported in tankers because of

bacterial, yeast, and mold formation in tankers. Tankers may not be adequately cleaned

between loads to prevent cross-contamination. Both tanker wash station operators and

the citrus industry would like to take steps to minimize spoilage and control food

pathogens in citrus juice and concentrates. A decrease in quality of Florida citrus juice

encourages consumers to purchase citrus juice from other sources, resulting in economic

losses that will eventually affect wash facilities. A foodborne outbreak in citrus juice

from Florida caused by the inappropriate washing could financially ruin a tanker truck

wash station as well as reduce consumer trust in Florida-grown citrus products.

Previous Research

In 2004, 87 liquid transportation tankers were sampled for the USDA tanker study

(Winniczuk and Parish, unpublished data). Forty-eight tankers hauled a citrus juice

product and 26 tankers hauled a dairy product immediately prior to the study. The

tankers were tested for the presence of Salmonella spp., E. coli, and Listeria spp. at

different locations within the tanker. Data indicated that citrus and dairy tankers are









equally likely to be contaminated with E. coli, although E. coli contamination from the

product itself is more likely to come from a dairy product than from a citrus product. The

part of tankers most often contaminated was manway lid gaskets (Winniczuk and Parish,

unpublished data).

Specific Aims and Objectives

Part I: Identification and Characterization of Microorganisms in Samples

The overall objectives of Part 1 of this research were to

Objectives

* Sample at least 72 tankers such that there was nine of each type created from the
eight possible combinations that could be created from dairy vs. citrus
juice/concentrates, gasket type A vs. gasket type B, and hot vs. cold temperature
wash.

* Monitor gaskets for coliforms and E. coli, Alicyclobacillus, and Salmonella spp.,
and psychroduric; mesophilic; thermoduric; yeast and mold; and aciduric
microorganism counts.

* Test for Staphylococcus spp., Streptococcus spp., and/or additional tests for
coliforms, fecal coliforms, or E. coli if appropriate.

* Select, streak for isolation, and characterize using gram staining, the catalase test
and the oxidase test on representative colonies from the psychroduric; mesophilic;
thermoduric; yeast and mold; and aciduric enumeration.

* Determine using the Analysis of Variance (ANOVA) and Fisher's Least Significant
difference, if there was a significant difference between psychroduric; mesophilic;
yeast and mold; and aciduric microorganism counts based on the differences
between product, gasket, and wash temperature variables.

Hypotheses

* The number of microflora and the frequency of detection of coliforms, fecal
coliforms, E. coli, Salmonella spp., and Alicyclobacillus spp. on gaskets of cleaned
tankers will be different between tankers that have previously hauled dairy products
and juice products.
* The number of microflora and the frequency of detection of coliforms, fecal
coliforms, E. coli, Salmonella spp., and Alicyclobacillus spp. on gaskets of cleaned
juice and dairy tankers will be different between tankers that have Type A, and
Type B rubber gaskets.









* The number of microflora and the frequency of detection of coliforms, fecal
coliforms, E. coli, Salmonella spp., and Alicyclobacillus spp. on gaskets of cleaned
juice and dairy tankers will be different between tankers that have received and
have not received a hot water spray.


Part II: Biofilm Development and Removal

Objectives

* Develop a manway lid model to simulate splashing of liquid products onto the
manway gasket.

* Create a cocktail of bacteria and yeast obtained from the first part of the study
whose identities of these microorganisms were determined by 16S DNA
sequencing and 28S rRNA (D2 expansion segment) sequencing.

* Develop methods three different methods to mimic treatments to mimic treatments
in the tank wash industry.

* Observe control and three treatments under fluorescent and scanning electron
microscopy.

* Swab the control and three treatments and determined the presence and quantity of
coliform and E. coli were present, and the mesophilic, yeast, and coliform
enumeration.

* Select, streak for isolation, and characterize using gram staining, the catalase test
and the oxidase test representative on colonies from the mesophilic and the yeast
enumeration.

* Determine, using the Analysis of Variance (ANOVA) and Fisher's Least
Significant difference, if there was a significant difference between mesophilic and
coliform counts based on the differences between Loglo reductions wash types.

Hypothesis

* The E. coli, coliform, mesophilic, and yeast counts on the gasket from the model
system will differ significantly from gasket microflora from the uncleaned sample
and the three cleaning regimes.














CHAPTER 2
REVIEW OF THE LITERATURE

There is evidence of a microbial problem with liquid food-grade transportation

tankers. Transportation tankers in the United States haul such products as pasteurized

and unpasteurized milk, single strength or concentrated juices, pasteurized and

unpasteurized liquid eggs, molasses, liquid yeast, liquefied pork lard, canola oil, citrusol

(citric acid), oil (sunflower, vegetable, canola, coconut and cotton seed), olestra, burned

syrup, sucrose, vinegar, brown sugar slurry, cola and other bases for carbonated

beverages ("Different Products"), honey, water, corn syrup, maple syrup, peanut butter

base, artificial and natural colors and flavors, pasteurized ice cream premix, soymilk,

catsup, and alcohol. After delivering food products, tankers go to wash stations for

cleaning. At the tank wash stations tankers may sit unwashed for hours or days before

cleaning. The reason for this practice is that wash stations are required to use washed

tankers within 48 hours of washing. Also, it can be as expensive as $140 to wash a

tanker, which does not make it financially advantageous to wash a tanker more than once.

While tankers sit waiting to be washed, insects may infest tankers and molds, bacteria

and yeasts can grow, potentially causing food safety and spoilage issues, and forming

communities called biofilms.

When tankers are ready to be used, those with non-dairy liquids are washed based

upon non-standard criterion specific to each wash station. A model of wash criteria is the

Juice Products Association (JPA) (2004) tank wash types, which is illustrated in Table 1.

This scheme provides a recommended set of guidelines for washing and cleaning tankers.










Appropriate washing regimes are chosen based upon product previously shipped in the

tanker. Tankers that have carried citrus juice are cleaned using either wash type 1 or

wash type 2. According to the JPA guidelines type 1 wash occurs between loads of the

same product as long as no one contaminates the truck in any manner, such as by entering

it, and the tanker is never left for more than 12 hours without product in it. If the truck is

left empty for more than 12 hours it should receive the appropriate wash for the product.

Table 2-1. Products and cleaning steps for JPA wash types based on Model Tanker Wash
Guidelines for the Fruit Juice Industry, 2004 (Juice Products Association
2004




Potable water pre- Potable water pre-
Potable water pre-rinse Potable water pre-rinse rinse/degrease rinse/degrease

Inspect Inspect Inspect
Warm water rinse (75- Warm water rinse (75- Warm water rinse (75-
110F) 1100F) 110F)
Manually clean valve and Manually clean valve and Manually clean valve and
vents vents vents
Hot degreasing 170 Hot degreasing 170 -
Hot clean 1600F, 15 min 212F for > 15 min 212F for > 15 min
Potable water rinse (no
spec) Warm water rinse Warm water rinse
Hot clean 160F min, 15 Hot clean 1600F min, 15
Inspect min min,
Sanitize chemical or hot Potable water rinse (no Hot water rinse 1850F for
water (1850F, 10 min) spec) > 20 min
Cool down (if hot water
used) Inspect Inspect
Sanitize chemical or hot Sanitize chemical or hot
water (1850F, 10 min) water (1850F, 10 min)
Cool down (if hot water Cool down (if hot water
used) used)


To maintain the quality of the product, JPA recommends that trucks hauling

concentrated juice consecutively should have a Type 2 wash every 7 days; single strength

juice, every 72 hours; and unpasteurized juice, every 24 hours (Juice Products

Association, 2004). Liquids used to clean tanker trucks are applied using one of three









types of spray washing systems (fluid-driven, motor-driven, or stationary) that are

lowered from the manway. Fluid-driven wash systems can be purchased in reactionary

force, constant speed, and turbine models. According to Spraying Systems Company

(Wheaton, Illinois):

"Reactionary Force tank washing nozzles use the force of the fluid to rotate the
spray head. Constant Speed tank washing nozzles use the momentum of the liquid
flow to drive the spray head while maintaining constant rotating speed. Turbine
tank washing nozzles utilize the fluid to spin a turbine, which in turn powers a gear
set. Motor-Driven tank washers use high-pressure solid stream nozzles at pressures
from 100 to 1000 psi (7 to 70 bar) with a separate motor for driving the nozzle
assembly. Two to four nozzles rotate on a gear hub as they revolve around the
central axis of the nozzle assembly. The Fixed Spray or stationary tank washing
nozzles include multi-nozzle spray assemblies and individual fixed position spray
nozzles. These models (fixed spray) can perform multi-tasks from cleaning tanker
trucks to cleaning food-processing tanks. The advantages of these nozzles are:
simplicity of design, reliability due to no moving parts, and a wide range of spray
coverages." (2004)

The problem with spray ball mechanisms is that they are often chosen on cost and may

not be the most appropriate system to clean tankers. Hence, the spray system may be

ineffective because it cannot clean hard to reach areas in the tanker or because holes in

the sprayer are plugged. Other cleaning problems include cleaning trucks at awkward

angles, improper or non-existent hand-cleaning of the accessory parts, and filth entering

the tanker around the seals during transportation. Bell et al. (1994) noted that poor water

quality, chemical concentration, and operational temperature could reduce the ability to

clean tankers. Also, failure to empty all wash waters from tankers could cause

recontamination and increase the likelihood of microbiological problems (Bell and others

1994). Other problems are evident in the security of the wash stations; wash stations will

store dirty and empty tankers in lots without security fences, leaving the tankers open to

vandalism, bioterrorism, or accidents involving curious people.









The Legal History

The first government action to correct problems with the U.S. food transportation

industry occurred in the late 1980s and early 1990s. In the late 1980s the media reported

that chemicals and garbage were being shipped with food products (Office of the

Inspector General 1998). The General Accounting Office revealed that trucking

companies "were not required to keep records of these mixed loads" (Office of the

Inspector General 1998). Also, there was concern about "backhauling," a process where

food is delivered in the first load, chemicals and/or garbage are shipped in the second

load and then food is shipped in the third load (Office of the Inspector General 1998).

In response to these accusations Congress passed the Sanitary Food

Transportation Act (SFTA) in 1990. In Section 5701 Part 2 of the Sanitary Food

Transportation Act (SFTA), Congress expressed its concern for the American public:

"...the United States public is threatened by the transportation of products potentially

harmful to consumers in motor vehicles and rail vehicles that are used to transport food

and other consumer products". In Part 3 of SFTA, they conclude that "the risks to

consumers by those transportation practices are unnecessary and those practices must be

ended." It was hoped that the SFTA would solve these problems (49 United States Code

5701).

SFTA of 1990 is found in 49 USC 5701 to 49 USC 5714. Section 5701 states the

findings (discussed in the introduction) that make this act so important. Section 5702

defines terminology used in the act. Section 5709 states that it is mandatory for the

Secretary of the Department of Transportation to consult the Secretary of Agriculture,

Secretary of Health and Human Services, and the Administrator of the Environmental

Protection Agency on how to implement sections 5703-5708.









Section 5703 is the general regulation section. It states that no later than July 31,

1991 the Secretary of Transportation is required to prescribe regulations on transportation

conditions that would make cosmetics, devices, drugs, food, and food additives unsafe for

humans or animals. The secretary is required to consider cosmetics, devices, or drugs to

be non-food products if they are transported in motor or rail vehicles before or at the

same time as a food or food additive, and if they would make the food or food additive

unsafe to humans or animals. Other special requirements this section makes of the

Secretary are to establish "record keeping, identification, marking, certification, or other

means of verification to comply with sections 5704-5706," "decontamination, removal,

disposal, and isolation to comply with carrying out sections 5704 and 5705" and to

produce a list of materials for the construction of tank trucks, rail tank cars, cargo tanks,

and accessory equipment that will comply with 5704 (49 USC 5702). Also it gives the

Secretary the responsibility of considering and establishing the following: determining

the extent that packaging can protect cosmetics, devices, drugs, food, and food additives

to keep them safe for humans or animals even though they are being transported in motor

and rail vehicles meant for nonfood products; finding "the appropriate compliance and

enforcement measures to carry out this chapter"; and establishing "appropriate minimum

insurance or other liability requirements for a person to whom this chapter applies" (49

USC 5702). Lastly, if the Secretary deems that a type of packaging meets certain

standards, the rules and regulations on the transportation conditions for cosmetics,

devices, drugs, food, and food additives unsafe for humans or animals may not apply.

Section 5704 establishes rules for tank trucks, rail tank cars, and cargo tanks. It

requires that the Secretary of Transportation publish in the Federal Register a list of









nonfood products that would not make cosmetics, devices, drugs, food, and food

additives transported before or with the product unsafe for humans or animals. It

prohibits one from using, offering for use, or arranging for the use of a tank truck, rail

tank car, or cargo tank to transport cosmetics, devices, drugs, food, and food additives if

the vehicle has been used for an unapproved nonfood product; or providing a vehicle for

the purpose of transporting an unapproved nonfood product when it is marked for

cosmetics, devices, drugs, food, and food additives or a nonfood product on the approved

list. Also this section requires individuals arranging for a tank truck or a cargo truck to

disclose what they will be shipping if it is or will be used in the preparation of a food

additive or if it is listed an approved nonfood item.

Section 5705 covers motor and rail transportation of nonfood products. This

section requires the Secretary to publish in the Federal Registrar a list of unsafe nonfood

products. This list should not include food packaging such as cardboard, pallets,

beverage containers unless the Secretary of Transportation determines that transporting

these packaging materials in a motor vehicle would make the packaging materials unsafe

to humans or animals. It forbids using, offering for use, or arranging for the use of a

tank truck, rail tank car, or cargo tank to transport cosmetics, devices, drugs, food, and

food additives if it has been used to transport nonfood materials listed in this section.

Section 5706 forbids the use of a tank truck to transport food and food additives if

the vehicle has been dedicated to transport asbestos, refuse, or other dangerous products

(a list of which the Secretary should publish in the Federal Register). Waivers of any part

of this chapter are allowed under Section 5707 if the Secretary determines it will not

make food and food additives unsafe for humans or animals, or it is in the public interest.









Section 5708 allows state employees to inspect (with funding from the federal

government) motor vehicles as long as the state agrees to comply with the appropriate

federal regulations or compatible state regulations. This section also enlists the help of

the "Secretaries of Agriculture, and Health and Human Services; the Administrator of the

Environmental Protection Agency, and the heads of other appropriate departments,

agencies, and instrumentalities of the United States Government" to help "carry out this

chapter, including assistance in the training of personnel". Training for federal and state

employees would be paid for by the federal government (49 USC 5708).

Section 5710 establishes that the Secretary of Transportation has the authority and

responsibility to carry out this Act. In section 5711, the Secretary is directed to make a

list of penalties and procedures and to take civil action against those who violate the

regulations set up under the chapter or under sections 5123 and 5124. Section 5712

establishes the relationship between this chapter and Section 5125 which is the chapter

dealing with the transportation of hazardous material. Section 5713 establishes that

sections 5711 and 5712 will only apply after July 31, 1991 (49 USC 5713). And lastly

under section 5714 the Secretary after consulting with state officials, "will establish

procedures to promote more effective coordination between the departments, agencies,

instrumentalities of the United States Government and State Authorities" (49 USC 5714).

After SFTA was enacted the Department of Transportation (DOT) turned over the

responsibility of the Act to DOT's Research and Special Programs Administration

(RSPA). By July 31, 1991 no final regulations were issued under SFTA nor had any

been issued by December 1997 when the Chairman of the Senate Committee on

Commerce, Science and Transportation asked the Office of Inspector General (OIG) to









investigate how well DOT and RSPA were doing fulfilling its obligation to SFTA (Office

of the Inspector General, 1998).

The OIG found that RSPA had issued a proposed rule in May 1993 to address "the

safe transportation of food products in highway and rail transportation" but had not

issued a final regulation. RSPA had prepared a training video for DOT inspectors on the

hazards of transporting food but admitted that it was not an adequate safety training

program as required by the Act (Office of the Inspector General, 1998). DOT failed to

develop lists of "non-food products not unsafe"; "unsafe non-food products"; "waivers";

and "coordination procedures" that they were supposed to establish under SFTA (Office

of the Inspector General, 1998). However, this was a result of the way the law was

written; the categories of "non-food products not unsafe" and "unsafe non-food products"

were too broadly written and could include every product (Office of the Inspector

General, 1998). Therefore, when RSPA could not identify any specific items to place on

the list, no waivers were needed.

Meetings of the DOT Secretary with Secretaries of Agriculture, and Health and

Human Services; and the Administrator of the Environmental Protection Agency on how

to implement sections 5703-5708 did not occur until 1995. Meanwhile SFTA was failing

its mission to protect the American public because improper storage and transportation of

food occurred. Three major incidences occurred in 1994, 1997, and 1999. In 1994,

224,000 people were affected by salmonellosis when a tanker truck that had carried

unpasteurized liquid eggs was not cleaned properly resulting in cross-contamination of

Salmonella to the pasteurized ice cream mix that was transported in the tanker truck

afterwards. In 1997 several decomposing bodies of stowaways were found in three ships









entering the U.S. Decomposing bodies in at least one of those instances contaminated

food products being imported into the U.S. This incident was the result of the SFTA not

including ship and airplane transport in its regulations (Office of the Inspector General

1998). A final incident occurred in 1999 when 207 people developed cases of

salmonellosis because ice allegedly contaminated with Salmonella was illegally added to

orange juice being shipped from Mexico to Arizona (FDA/CFSAN 2001). Since the OIG

investigation there have been several unsuccessful attempts to transfer SFTA

responsibility to the FDA.

In the interim, to maintain the safety of food transported in tanker trucks, four

voluntary standards have been developed: 1) the Bulk Over-the-RoadFood Tanker

Transport Safety and Security Guidelines (Food Industry Transportation Coalition 2003),

2) the Grade 'A'PasteurizedMilk Ordinance (PMO) (FDA/CFSAN: National

Conference 2003)/3-A Sanitary Standards for Steel Automotive Transportation Tanksfor

Bulk Delivery and Farm Pick-Up Service (2002), 3) the FSIS Safety and Security

Guidelines for the Transportation and Distribution of Meat, Poultry, and Egg Products

(2003), and 4) the Juice Processors Association Model Tanker Wash Guidelines for the

Fruit Juice Industry. The Bulk Over-the-Road recommendations deal with non-dairy dry

and liquid foods and the PMO/3-A Standards deal with dairy foods. The documents are

similar in content.

The Bulk Over-the-Road Guidelines are divided into an introduction,

recommended documentation procedures (example forms are found in the document's

appendix), guidelines for receipt and inspection of an empty tanker, how to load tankers,

what to do when the tanker has been loaded, minimum requirements for cleaning non-









dairy food/food grade tankers, conversion of trailers from "non-approved, non-food

service" to "approved non-food to food" and "food to food" service, tank requirements

for non-dairy and dry-bulk food grade cargo tanks, and security guidelines (Food

Industry Transportation Coalition 2003).

The 3-A Standards provide scope and definitions for the document and metals,

fabrication, air venting, mechanical cleaning, extra fittings, air pressure, temperature,

insulation, and design/construction requirements for dairy transportation tankers (3-A

Standards for Transportation Tanks 2002). 3-A equipment produced for dairy complies

with the design and construction criteria outlined in the PMO. The PMO is not a

mandatory document. The document can be adopted by states or local governments into

their legislation; however, once it is adopted or a state agrees to participate in the

National Conference on Interstate of Milk Shipments (NCIMS) the PMO must be

followed as part of the law. Appendix B of the PMO provides guidelines for milk

sampling, hauling and transportation. It gives requirements for the driver, training

guidelines for the driver, quality and sampling checks that must be performed before

picking up the milk, pumping procedures, inspection procedures, milk tank truck

standards, and procedures when standards are not met. Appendix B refers to Section 7,

12p the Cleaning and Sanitizing of Containers and Equipment for methodologies on

cleaning tankers. The most relevant points to tank wash facilities are 1) Milk containers

must be cleaned every 72h, 2) All Grade A milk trucks are to be washed and sanitized at

a permitted facility, 3) Washed trucks will carry a wash tag stating "the date, time, place

and signature or initials of the employee or contract operator doing the work, unless the

milk tank truck delivers to only one receiving facility where responsibility for cleaning









and sanitizing can be definitely established without tagging" and 4) The wash tag will be

kept at the next station the tanker is washed and sanitized at for a minimum of fifteen

(15) days (FDA/CFSAN: National Conference 2003).

The FSIS Safety and Security Guidelinesfor the Transportation and Distribution of

Meat, Poultry, andEgg Products is important to tanker truck transportation with respect

to hauling liquid eggs. The guidelines cover the types of vehicles that may be used to

transport meat, poultry and egg products, and procedures related to preloading, loading,

in-transit, and unloading. The most important guidelines pertaining to the shipment of

liquid eggs are the temperature control requirement, and the guideline under loading

which recommends that sealed vehicles shipping egg products (pasteurized,

repasteurized, or heat treated) from one point to another should have a certificate

accompanying them. The certificate should state if products have not been pasteurized or

if they have tested positive for Salmonella. The FSIS Guidelines also includes a section

on food security, which offers guidance on establishing and implementing a food security

plan. This section includes a section for helping to ensure food truck security. (FSIS

Safety and Security, 2003)

Finally, the Juice Processors Association Model Tanker Wash Guidelinesfor the

Fruit Juice Industry outlines wash station requirements, transportation tanker

requirements, 4 types of wash protocols, how to clean accessory parts, a list of commonly

transported substances whether or not they are able to be transported in a food grade

tanker and what wash type they receive, and a set of documentation procedures (Juice

Products Association, 2004).









The most recent development to improve food transportation was in August 2005

when the president signed into law the Safe, Accountable, Flexible, and Efficient

Transportation Equity Act of 2005. Under Chapter 3 sections 7381-7383 of this Act,

effective October 1, 2005, the Secretary of the Department of Health and Human

Services (HHS) will take over the many of the responsibilities outlined in SFTA. In

addition the Act adds that the Secretary should prescribe practices for sanitation and

record keeping. The Act amends Chapter 57 and makes the DOT will still be responsible

for the establishing and training DOT inspectors to look for contamination or adulteration

of products under section 416 of the Federal Food, Drug, and Cosmetic Act; section 402

of the Federal Meat Inspection Act (21 USC 672); and section 19 of the Poultry Products

Inspection Act (21 USC 467a). The objective of the legislation is that by placing the

Department of HHS (the branch of the government containing the FDA) in charge of the

developing regulation for transportation and sanitation liquid transportation tankers that it

will improve safety of liquid food in the United States. It will be interesting in the next

few years to see if the Department of HHS is more effective than the DOT.

Marketing and Food Safety Justification for Citrus Juice Transportation Tanker
Research

The United States (U.S.) and the State of Florida have reason for concerns

regarding juice transportation. According to the USDA, the U.S. produced 12,311,000

metric tons of oranges in the 2003/2004 marketing year, making it the world's second

largest orange producing nation following Brazil. The U.S. also produced 1,895,000

metric tons of grapefruit making it the number one producer of grapefruit in the world.

Approximately 80% of oranges and 50% of grapefruit grown in the U.S. are processed

into juice products. In the 2003/2004 marketing year the U.S. processed 37,048,000









metric tons, imported 20,005,000 metric tons and exported 19,955,000 metric tons of

orange juice (USDA's Foreign Agricultural Service (FAS): Production 2004). Most of

the citrus in the United States comes from Florida. According to the USDA, "Florida

accounted for 79 percent of total U.S. citrus production, California totaled 18 percent,

while Texas and Arizona produced the remaining 3 percent" (USDA's Citrus Fruit

Summary 2004).

In Florida the citrus industry has an impact of $9 billion per year and is responsible

for 90,000 jobs. It is the largest segment of the agricultural industry, which is second

only to the tourist industry in importance to the state's economy. Approximately 95% of

Florida's oranges are processed into juice in any specific year (USDA's Foreign

Agricultural Service (FAS): Horticultural 2004). Unpasteurized and pasteurized citrus

juices are transported in transportation tankers. Unpasteurized juice can only be

transported in tankers as if the customer declares that it will be pasteurized before

packaging.

Dairy products, both raw and pasteurized are often hauled before juice products in

the state of Florida. There is growing awareness that the quality and safety of products

transported in tankers can become contaminated with bacteria, yeast, and/or mold from

inadequately cleaned tankers. Introduction of microorganisms to products not destined

for pasteurization, such as citrus juice concentrates, raises concern.

Both transportation tanker wash stations and the citrus industry would like to take

steps to mitigate food spoilage and to ensure pathogens do not enter citrus juice and

concentrates from tankers. A decrease in quality of citrus juice would encourage

consumers to purchase other juices or beverage products, and the ensuing economic loss









over time would affect wash facilities. A foodborne outbreak in citrus juice from Florida

caused by inappropriate washing could financially ruin a tanker truck wash station as

well as reduce consumers' trust in Florida citrus products.

Presently, there is little published research completed on microbial aspects of

transportation tankers. Research on this topic is discussed below. Also there is no

research dealing with biofilm formation caused by citrus juices. Therefore, the goal of

this literature review is to present what is known about microbial aspects of tankers;

microorganisms associated with a citrus juice and dairy environment; information on

biofilms formation; and methods by which to study the above mentioned topics.

Previous Research on Transportation Tankers

There are a limited number of studies on the transportation of food in tankers. The

vast majority deal with milk transportation in tankers. This may be a result of the

nutrient and microbiological make-up of milk but also may reflect the fact that milk is

one of the products most frequently transported in dedicated tankers. The first study that

could be found on transportation tankers appeared in Deutsche-Milchwirtschaft

(Tamoschus 1979). The author investigated factors that affected cleaning protocols for

milk tankers and concluded that "a fully programmed cleaning and disinfection process,

using always freshly prepared solutions, is the most reliable and hygienically most

suitable method for cleaning milk tankers" (Tamoschus 1979). Most studies since then

have included tanker trucks in assessments of residual antibiotic residues on the surface

of dairy processing machinery.

Also, several research papers have been written on mathematical models to solve

tanker truck scheduling problems-Ubgabe and Sankaran (1994) in India, Butler and

Williams (1995) in Ireland, Foulds and Wilson (1997) in New Zealand, and Basnet and









others (1997 and 1999) in New Zealand (Basnet and others 1999). The 1999

mathematical model by Basnet and others was designed not only to solve the problem of

scheduling transportation tankers to pick up milk from dairy farms but to set up a

schedule so there would not be any congestion back at the dairy processing plant when

the milk was pumped from the tankers (Basnet and others 1999).

Steele and others (1997) studied 1,720 pickups of raw milk in transportation

tankers for the presence of foodborne pathogens: Salmonella spp., Camplybacter spp.,

Listeria monocytogenes, and toxicgenic Escherichia coli. Then they calculated the

theoretical probability of 3, 5, and 10 raw milk tankers containing the above-mentioned

pathogens. Of the tankers sampled, 8 (0.47%) contained Camplybacter spp.; 3 (0.17%),

Salmonella spp.; 15 (0.87%), toxicgenic Escherichia coli; 47 (2.73%), Listeria

monocytogenes. Only two tankers contained more than one of the above-mentioned

pathogens. They also provided theoretical probability of having one pathogen in a bulk

tank resulting from the pooling of 1, 3, 5, and 10 tankers is 4.13%, 11.89%, 19.01%, and

34.41% respectively. The authors concluded that although the possibility for an

individual tanker to be contaminated is relatively low, the probability of pooled bulk raw

milk containing one or more of the pathogens is fairly high (Steele and others 1997).

Another area of study relating to transportation tankers is the use of ATP-

bioluminescence to rapidly determine the amount of microbiological and other residual

contamination (Bell and others 1994). In theory, ATP should not be present if a tanker is

properly cleaned; however, low levels of ATP may be found even in a clean tanker (Bell

and others 1994).









ATP-bioluminescence was discovered in 1940's (Stanley, 1982). All living cells

contain the high-energy chemical compound adenosine triphosphate, ATP. In the ATP-

bioluminescence method, "an enzymatic complex catalyzes conversion of chemical

energy of ATP into light through oxidation-reduction reactions" (Paez and others 2003).

The amount of light generated is measured by a luminometer in relative light units (RLU)

and "is directly proportional to the amount of ATP present in the sample" (Paez and

others 2003).

Bell and others (1994) used two different commercially available ATP-

bioluminescence products, from Biotrace Ltd and Sonco Ltd, for their study. Swabs were

taken from 465 milk tankers at 10 cm2 sampling points on the internal surfaces of the

manway lid, vessel roof, vessel side wall, vessel end wall, flexible hose, and the air

elimination vessel after they had been cleaned. For each tanker a visual assessment of

clean or dirty was made for each area and swabs from each area as well as rinse water

samples were taken for ATP-bioluminescence testing. From every other tanker,

microbial counts were taken from the swabs and the rinse water. Parameters were

established for whether a site was clean or dirty for each ATP-bioluminescence kit.

For internal sites (vessel roof, vessel side wall, vessel end wall), there was 88.2-

90.6% agreement between the two ATP-bioluminescence kits. For external sites (internal

surfaces of the manway lid, flexible hose, and the air elimination vessel), there was 77.7-

83.6% agreement between the two kits. In the rinse water, the agreement between the

two tests was 65.5%. The study results found that 93-98% of vessel roofs, vessel side

walls, vessel end walls were clean using standard microbial techniques but only 63-89%

of the these surfaces were clean using the visual or the ATP-bioluminescence kits. For









all areas on the outside of the tanker 56-93% of bacterial swabs were considered clean

while 44-66% of the trucks were visually clean or clean using ATP-bioluminescence.

Lastly, only 60% of the rinse waters were clean using standard microbial counts, while

less than 40% of the rinse waters were visually clean or clean according to the ATP-

bioluminescence kits. From this study, the authors reported the following: 1) the

differences between the ATP-bioluminescence results and the microbial counts are a

result of the fact that the microbial counts reflect the number of microorganisms while

ATP-bioluminescence is based on the number of microorganisms plus the soil, 2) ATP-

bioluminescence is more efficient because it takes less than 10 mins to get results while

conventional microbial counts take 3 days, 3) tankers which had good drainage generally

were found to be clean, and 4) both ATP-bioluminescence test kits correlated well with

each other. Therefore, the authors concluded that ATP-bioluminescence could be used to

as an efficient, reliable mechanism to monitor the cleanliness of transportation tankers

and that the results indicate that external surfaces, in particular the manway lid and the air

elimination vessel, provide the more accurate assessment of cleanliness of the vehicle

(Bell and others 1994).

Paez and others (2003) evaluated three transportation tankers by assaying for ATP

on the inner surface of the manway lid, outlet pipe and vessel roof of recently cleaned

tankers. The final equipment rinse water was sampled and analyzed with ATP-

bioluminescence and microbial plate count. "Clean", "caution" and "dirty" ratings were

given to ATP-bioluminescence samples based on a scale created by the bioluminescence

light manufacturer (BioControl Systems, Bellevue, WA, USA). The inside surface of the

manway was deemed hardest to clean. The outlet pipe had the most variable results when









correlated with the cleanliness of the tanker. Therefore, the authors agreed with Bell and

others (1994) that the outlet pipe may be used as an indication of a tanker's cleanliness.

Few samples were taken from the tankers roof because of irritating odors from the

chemical detergents so little could be drawn from these measurements. Rinse water

results from all tankers indicated a good correlation exists between the microbial count

method and the ATP-bioluminescence method. The authors concluded from this study

that ATP-bioluminescence was a good method for judging the cleanliness of milking

equipment, bulk tank and milk transport tankers (Paez and others 2003).

Citrus Juice and Milk and Their Microbial Inhabitants

The Environment of Citrus Juice

Fellers and others (1990), in a sampling study of the nutrient content of Florida

frozen concentrate orange juice, orange juice from concentrate, pasteurized orange juice,

grapefruit juice, and grapefruit juice from concentrate, discovered that there were

significant differences within product categories due to differences in the manufacturing

plant and time of the year the fruit was processed. However, they were able generally to

describe the nutrient content of citrus juice products and to reaffirm previous knowledge.

They reported that Florida orange juice is a significant source of vitamin C with 90 to

100% of the Recommend Daily Value (RDA) in orange juice products and 70% RDA for

grapefruit juice products. Vitamin C (ascorbic acid) plays an important role in nutrition as

an antioxidant and prevents scurvy in the human body (Smolin and Grosvenor 2000).

They reported 6-8% thiamin RDA and 8% folic acid RDA in orange juice products while

grapefruit juice products had slightly less of these two nutrients. Magnesium, calcium,

copper and phosphorus are found in small but claimable levels in citrus products while

zinc, iron, and sodium are found at insignificant levels (Fellers and others 1990). Also,









significant levels of potassium exist but no percentage is reported because there is no

RDA (Fellers and others 1990).

Another notable characteristic of citrus juices is that they are highly acidic with a

pH of about 4.0 or below. Citrus juice processors have modified citrus juices to appeal to

different market segments by creating low acid juice, juice with different levels of pulp,

fortified juice (with calcium, zinc and vitamins A, B6, B12, C, & E, potassium, and

folate), mixed fruit juice, home squeezed, and concentrated. Major processors have also

produced a reduced sugar and calorie juice drink to serve the "low carb" market as well

as orange juice with sterols for "heart-health and orange juice targeted to the needs of

children (Tropicana 2004; and Florida's Natural 2004, Minute Maid 2004).

Microbial Flora of Citrus Juice

Bacteria

The major types of bacteria in citrus juices are spore-forming bacilli, lactic acid

bacteria, and, rarely, acetic acid bacteria. Also, bacteria of public health significance,

such as Salmonella and E. coli have been found in commercial, unpasteurized citrus

juices.

Spore-forming bacilli are usually from two genera, Bacillus and Alicyclobacillus.

Bacillus cells are gram-positive, aerobic or facultatively anaerobic, straight, rod-shaped

bacteria with dimensions about 0.5-2.5 x 1.2-10 jm Cells from these genera are often

found in pairs or chains. They are motile by peritrichous flagella. The bacilli can

produce endospores that are very resistant to different conditions such as thermal

treatments and sanitizers. Within the genus Bacillus, species differ widely in their ability

to survive wide range of pHs, temperatures and salinities. Bacillus subtilis is often used

as typical example of this species. B. subtilis can survive at pH's <6 and >7,









temperatures between 10C and 50C, and salinities of at least 7% (their ability to survive

at 10% NaCl has not been tested). Bacillus spp. ferment glucose and are catalase

positive. Many Bacillus species are oxidase positive, reduce nitrate to nitrite, and require

about 3-12% salt for growth. There are a few pathogenic bacilli (Hensyl 1994); however,

they do not germinate and outgrow at pH levels associated with citrus juices.

Alicyclobacillus has 9 species and subspecies. The species A. acidoterrestrius is

most associated with spoilage of fruit juices. These species are commonly found in soil

(Deinhard and others 1987), thermal environments (Albuquerque and others 2000), and

thermally processed juices such as orange (Uboldi-Eiroa and others 1999). They are

gram-positive to gram-variable bacteria that are rod shaped, motile and aerobic (Walls

and Chuyate 1998 & Wisse and Parish 1998). They contain co -alicyclical fatty acids as

part of there membrane (Walls and Chuyate 1998). This genus grows best at a pH of 3.5-

4.0; however, it can grow at a minimum of 2.5 and a maximum of 5.5 (Walls and

Chuyate 2000 "Spoilage" and Walls and Chuyate 2000 "Isolation") while spores can be

produced as low as 3.24 (Walls and Chuyate 1998). A. acidocaldarius prefers to grow at

temperatures between 60-65C, while A. acidoterrestrius, and A. cycloheptanicus prefer

to grow at 45-50C (Walls and Chuyate 2000 "Spoilage" and Walls and Chuyate 2000

"Isolation"). The spores can survive typical 85-90C juice processing temperatures

(Uboldi-Eiroa and others 1999). None of the bacteria in this genus are pathogenic (Silva

and others 1999).

Lactobacillus species are gram-positive, rarely motile rods although some may

appear as cocco-bacilli. They are about 0.5-1.6 jm in diameter. They are facultatively

anaerobic and occasionally microaerophilic. Most Lactobacillus species grow best when









there is at least 5% carbon dioxide in the atmosphere. They need rich and complex media

in which to grow. Products of their fermentation may include lactate, some acetate,

ethanol and carbon dioxide. Not all species produce carbon dioxide. Lactobacillus does

not reduce nitrogen, liquefy gelatin, and its cells are catalase and cytochrome negative.

This genus is rarely pathogenic (Hensyl 1994).

Salmonella spp., and E. coli cannot grow but may survive for extended periods in

chilled unpasteurized citrus juice (Parish 1998). Oyarzabal and others (2003) found that

E. coli 0157:H7, Salmonella spp., and Listeria monocytogenes inoculated at levels

greater than or equal to 103 CFU/g was capable of surviving for twelve weeks in orange

juice concentrates stored at -100F. Parish and others (2004) discovered that Salmonella

spp. could also survive in grapefruit juice concentrate but its lower pH provided better

antimicrobial activity. Also that two days storage of grapefruit concentrate between

7F to -120F would cause a 5-log reduction (Parish and others 2004). Survival of

Salmonella in these products has led to disease outbreaks (Parish 1997). In a 1995

outbreak of Salmonella resulting from the consumption of unpasteurized orange juice

potentially contaminated poorly washed fruit or amphibians in the facility (Parish 1998).

Klebsiellapneumoniae and Streptococcus spp. have been found to survive in frozen

orange juice concentrate (Fuentes and others 1985, Larkin and others 1955, Patrick

1953). Kaplan and Appleman (1952) studied 42 cans of commercially packed

concentrate and found that the enterococci found in the can were more prevalent and

more resistant to the environment of frozen citrus concentrate than E. coli. Larkin and

others (1955) discovered in their research that Streptococcus faecalis, Streptococcus

liquefaciens, and E. coli were able to survive in orange juice concentrate stored at -10F









for 147 days. The number of S. faecalis and S. liquefaciens did not change over time

while the numbers of E. coli fluctuated over time.

K. pneumoniae isolated from frozen concentrate processed in Florida and shipped

to Puerto Rico was found to survive in low temperatures (freezing), low pH, and low

water activity. Researchers ruled that the contamination occurred before shipping

because the pathogen was found in unopened barrels of product. Contamination most

likely occurred by the mixing of unpasteurized juice with concentrate, or Klebsiella

pneumoniae was present on the machinery used to fill the concentrate (Fuentes and others

1985).

Yeasts

Common yeast inhabitants of citrus juices include Saccharomyces, Torulaspora,

Candida, Zygosccharomyces, Hanseniaspora, Metschnikowia, Pichia, and Rhodotorula.

The genus Candida has 151 species. The vegetative cells reproduce by budding and

sometimes contain pseudohyphae (such as in the case of C. parapsilosis) or septate

hyphae. This genus of yeast does not reproduce by sexual reproduction. Of the two

species of interest in orange juice, C. parapsilosis ferments D-glucose (one of two key

carbon sources in citrus juice) and needs D-glucose for growth and also may or may not

need citrate (the second key carbon source in citrus juice) to survive, while C. stellata

does not utilize citrate for growth but will utilize D-glucose for fermentation and growth.

C. stellata will reproduce best at 250C and C. parapsilosis will grow at 25-370C.

Hanseniaspora is composed of 6 different species. Hanseniaspora have lemon to oval

shaped cells with pseudo-hyphae that reproduce asexually by polar budding and

reproduce sexually by utilizing asci that have one to four ascospores. There are 12

species in the genus Metschnikowia. The yeast cells in the Metschnikowia genus









reproduce asexually by budding and sexually utilizing club-shaped asci containing 1 to 2

needle-like ascospores. This yeast rarely flocculate. The genus Pichia has 89 species.

Pichia will have pseudo-hyphae and occasional septate hyphae. This yeast reproduces

asexually by budding and sexually by asci with 1 to as many as 8 ascospores. The genus

Saccharomyces contains 16 species. The yeast will sometimes have pseudo-hyphae.

They reproduce sexually by asci that are formed from directly from a diploid cell with 1

to 4 ascospores. S. cerevisiae ferments and uses D-glucose for growth however citrate is

not utilized for growth. Optimal growth temperature for this yeast are 25-30C (Barnett

and others 2000).

Molds

Some of the common mold types in citrus juices are Cladosporium

cladosporioides, Penicillium citrinum, P. digitatum and P. italicum and Geotrichum spp.

(Wyatt and others 1995, Wyatt and Parish 1995). Cladosporim and Penicillium produce

spores called conidia while Geotrichum produces spores called arthrospores.

Cladosporium is a dark green to black with a black back. The spores are dark, one to

two-celled that spread by "exposing its dry spore masses to air currents". This mold is

most often found in decaying plant matter and in the air. Penicillium has brush-like

structures that carry the mold spores. It is commonly found in the soil. This mold

produces a green to blue green rot on citrus fruit. Geotrichum are composed of colorless,

slimy chains of spores. They can produce strong odors and be pathogenic to humans.

They are common in dairy products and in the soil (Malloch 1981).

The Environment of Liquid Dairy Products

Milk in comparison to juice is a much better medium for the growth of

microorganisms because of its almost neutral pH (Frank 2001). Milk also has wide









variety of available nutrients. Milk is approximately 87.3% water, 4.8% lactose, 3.7%

fat, 2.6% casein, 0.6% whey protein, salt cations (0.058% sodium, 0.140% potassium,

0.118% calcium, 0.012% magnesium) and anions (0.176% citrate, 0.104% chloride,

0.074% phosphorus) and nonprotein nitrogen (Jenness 1988). The main sources of

carbon for microorganisms are lactose, fat, and protein. The amounts of citrate and

glucose present in milk are not enough to sustain microbial growth for long; therefore,

fermentative microorganisms must be able to utilize lactose. Microorganisms rarely use

milk fat as a carbon source because unless fat globules are damaged the microorganisms

cannot penetrate the fat globules' protective protein and lipid membrane. Of the two

proteins in milk, caseins are easy susceptible to proteolysis while whey generally is not.

The nonprotein nitrogen that is readily utilized as a nitrogen source is not able to sustain

microbial life. Milk is a good source of B vitamins and minerals such as iron, cobalt,

copper, and molybdenum. However, many of the minerals may not be present in a form

that can be utilized by bacteria. Lastly, milk contains growth stimulants such as orotic

acid, which is a metabolic precursor to pyrimidines and which fosters microbial growth

(Frank 2001).

Milk sold for liquid consumption in the United States must be pasteurized at

minimum for 15 sec at 72C although most processors will use higher temperatures and

longer holding times. Milk may be transported raw or pasteurized but it is always

pasteurized after transport. Raw milk from healthy animals generally has less than 103

microorganisms (Richter and Vedamuthu 2001) that typically consist of Micrococcus,

Staphylococcus, Streptococcus, and Corynebacterium spp. Staphylococcus aureus,

Streptococcus spp., Pseudomonas spp., and coliforms are related to environmental and









contagious mastitis. Some contamination can occur from the cow, the milking room

environment or poorly cleaned milking systems. The contaminants are occasionally yeast

and mold but generally are Bacillus spp., Clostridium spp., lactic acid bacteria, coliform

and other gram-negative bacteria (Olson and Mocquot 1980). When raw milk is cooled

the increases in bacteria are caused by Psuedomonas spp. as well as species of

Alcaligenes and Flavobacterium (Bishop and White 1986, Cousin 1982, Stadhouders

1975, Thomas 1974). Foodborne outbreaks involving raw milk and Salmonella spp.,

Campylobacterjejuni, and Yersina enterolitica, and Listeria monocytogenes (Bryan 1983

and 1988) The microorganisms that commonly are found in freshly pasteurized milk are

gram-positive bacteria that can survive pasteurization: Bacillus, Lactobacillus,

Micrococcus, Staphylococcus, Streptococcus, Microbacterium, Enterococcus,

Arthrobacter, and Corynebacterium spp. (Cousin 1982). Since these bacteria generally

do not grow quickly at refrigeration temperatures they are generally outgrown by gram-

negative psychroduric coliforms, and members of the Psuedomonas, Alcaligenes, and

Flavobacterium spp. (Cousin 1982, Olson and Mocquot 1980)). Postpasteurization

contamination has resulted in listeriosis and salmonellosis (Byran 1983 and 1988). The

postpasteurization contamination of Yersina enterolitica, and Listeria monocytogenes is a

major concern because of these pathogens ability to grow at refrigerator temperatures

(Richter and Vedamuthu 2001).

Other dairy products commonly transported in tankers are cream, half-and-half,

sweetened condensed milk, liquid ice cream mix, whey and chocolate base. Pathogens

that are found in cream and cream-fillings are B. cereus and S. aureus (USDA's Bad Bug

Book 1992). Condensed milk contains Bacillus, Lactobacillus, Micrococcus,









Streptococcus, Microbacterium, Enterococcus, Arthrobacter, and Corynebacterium spp.,

coliform and psychroduric bacteria (Foster and others 1957). Because of the added

sugar, sweetened condensed milk generally contains osmophilic sucrose-fermenting

yeasts and molds (Frazier 1958). Bacillus spp. and other postpasteurization contaminants

grow in liquid ice cream mix; once the ice cream mix is frozen growth of most

microorganisms stops (Richter and Vedamuthu 2001). However, some bacteria including

pathogenic Salmonella spp. and L. monocytogenes have been found to survive in ice

cream (Bryan 1983, Rosenow and Marth 1987).

Biofilms

Biofilms are communities of organisms that contain either bacteria or other higher

organisms, "such as algae, that are held together by sticky extracelluar polymericc)

matrix" (Watnick and Kolter 1999) and are irreversibly associated with a surface (Donlan

2002). Biofilms may also contain materials other than cells such as blood components, or

clay (Donlan 2002). The first person to note biofilms was Van Leeuwenhoek in 1683

when he used his simple microscopes to examine bacteria on the surface of teeth (Donlan

2002, University of California 2005). Biofilms were eventually recognized again in

1940's with the work of scientists studying marine organisms and noting their ability to

attach to surfaces (Heukelekain and Heller 1940 and Zobell 1943). Since then other work

has been done on biofilm formation in oil manufacturing, oral cavities, water sources and

refinement operations, medical and industrial settings and very recently the study of

biofilms and their relationship to food. The ability to study biofilms has been enhanced

by the development of more complex microscopes, in particular the confocal laser

scanning microscope, and the development of techniques used to study the genes

involved in cell adhesion and biofilm formation (Donlan 2002).









Biofilms have been described as the prevailing microbial lifestyle because biofilms

provide microorganisms with safety, nutrients, protection, and a place to transfer genetic

material (Watnick and Kolter 1999). Biofilms are commonly found on air-water or solid-

liquid contact surfaces that contain a readily available supply of nutrients (Stickler 1999

and Donlan 2002). Therefore, biofilms have created problems in both the food industry

and the medical community because they attach to production equipment, prosthetic

devices and sterilizing equipment (Stickler 1999). Biofilms also serve useful functions in

industry by breaking down emulsifiers and oils (Pasmore and Costerton 2003) and the

environment as well such as breaking down organic matter, degrading pollutants, and

"cycling nitrogen, sulfur, and many metals" (Davey and O'Toole 2000). There are four

major stages in the biofilm life cycle: attachment, formation, maturation, and dispersal.

Attachment

The first stage in biofilm formation is attachment. In this stage the bacteria gets

close to the surface, slows down its rate of movement, forms a transient attachment to the

surface, and searches for a place to settle down and make a stable attachment (Watnick

and Kolter 2000). Bacteria may search for suitable sites by twitching or swarming

movements if the bacterium is motile (Pasmore and Costerton 2003). The ability of a

bacteria to attach to a surface depends on many factors including the bacteria's cellular

components, the material the bacteria is attaching to, the surrounding environment, gene

regulation in the bacterial cell, and the interaction between the bacteria and other bacteria

preexisting on the surface (Donlan 2002). Therefore, where bacteria attach and why they

attach may be uniquely different, but one thing can be certain about all bacteria and

surfaces they chose to attach; these surfaces provide an ideal environment to grow and

develop and being part of a biofilm has advantages over being a free, planktonic cell.









Material

Three major factors of materials play a role in bacterial attachment to a surface:

roughness, hydrophobicity and polarity, and its conditioning. Several authors have noted

how surface roughness plays a role in biofilm attachment. Characklis and others noted

that an increase in roughness of surfaces increases attachment (1990). Arnold and others

(2001) discovered that the root mean square (RMS), which is a measurement of the

surfaces roughness, and the center line average, which is the depth from the peak of the

sample at which there is a 50% of the sample area below and a 50% of the area above,

can help to predict biofilm formation. According to Donlan "most investigators found

that microorganisms attach more rapidly to hydrophobic, nonpolar surfaces, such as

Teflon and other plastics, than to hydrophilic materials such as glass or metals" (2002).

Scientists are not sure why this occurs because there has not been a conclusive method

for measuring surface hydrophobicity, but it seems that cells must be able to overcome

repulsive forces between itself and the surface creating a hydrophobic interaction close to

the surface, which allows to the cell to attach. An explanation for how this is possible

has come from examining the cellular components role in attachment (discussed below).

Finally, conditioning film is created from particles of the media bond to the surface to

form a film. This film can affect the rate and amount of attachment (Donlan 2002)

because the film has a different chemical composition and nutrient value that attracts

bacteria. Also the film can reduce the repulsive effects of the surface allowing bacteria to

easily bind (Pasmore and Costerton 2003).

Cellular components

The outside of the cell is composed of hydrophobic and hydrophilic regions.

Although the outer surface is soluble in water it can form hydrophobic connections with









stratum surface materials, and cell membrane bound proteins and polysaccharides. The

extracellular polysaccharides, unlike proteins, allow cells to form attachments at greater

distances from surfaces. In this way bacteria reduce energy needed to attach to surface

because by using extracellular polysaccharides the whole cell does not have to enter the

surface's double ionic layer (Pasmore and Costerton 2003).

Other cellular components that help the cell attach are the flagella, and pili. The

flagella and type IV pili (Pasmore and Costerton 2003) help to overcome the electrostatic

repulsive forces that exist between the surface and cell (Corpe 1980) in a same way the

extracellular polysaccharides help to overcome the surface's double ionic barrier by using

the attractive moieties. The type IV pilus has an advantage over the flagella in that it can

shoot out and attach to the surface. After it attaches it can reel the cell back in to the

surface (Pasmore and Costerton 2003).

Characteristics of the liquid media

Characteristics of the media such as pH, nutrient levels, ionic strength, temperature,

and the velocity can influence the rate of attachment. For example, Cowan and others

(1991) found an increase in bacterial attachment occurred as a result of increased nutrient

concentration. Others have found that in nutrient rich media bacteria will settle

anywhere, while in nutrient poor medium bacteria will only attach to nutrient rich

surfaces (Watnick and Kolter 2000). Barnes and others (1999) found that ionic

composition of the suspending medium had the most effect on bacterial adhesion. They

discovered that iron and calcium salts present in the suspending media increased cellular

attachment while potassium, manganese, magnesium, and sodium salts inhibited cellular

attachment. It was thought the reason for the lack of attachment was due the "dissolved

cations [potassium, manganese, magnesium, and sodium] shielding the surface-negative









charge on bacteria and [stainless] steel" while the increased attachment of calcium and

iron was believed to be a result of the molecules ability to bridge between the bacteria

and the surface (or the conditioning film on the surface). Flow velocities affect cellular

attachment; when velocities are slow, cellular attachment depends more on size and cell

mobility, and when velocities are high, cells are subject greater turbulence and mixing.

Therefore, the cells that attach in these environments are the ones that can make a quick,

effective association with the surface and remain attached until the velocities become

great enough to exert a shear force on cells that make them detach (Characklis, Microbial

1990).

There are no literature was found on how citrus juice may affect biofilm formation.

However, Barnes and others (1999) experimented with different concentrations of milk

exposed to a stainless steel surface before contact with S. aureus, S. marcescens, and L.

monocytogenes. It was found S. aureus increased attachment with samples with 0.1%

milk and the control compared to 100, 10, or 1% milk, and S. marcescens, and L.

monocytogenes had increased attachment with the control compared to 100, 10, 1 or 0.1%

milk. A possible reason is that bacteria are attracted to iron on the surface of the stainless

steel if nitrogen from the milk protein blocks the attractive forces (or the electron escape

depth) then the bacteria will not readily attach to the stainless steel surface. Duddridge

and Pritchard (1983) noted that bacteria attachment is higher on milk-treated rough

surfaces that on milk-treated smooth surfaces.

Formation

When a bacterium chooses to attach it generally assimilates itself into part of

microcolony as part of biofilm formation. The microcolony develops until it forms a

three-dimensional EPS-encased structure at which point it is considered a biofilm









(Watnick and Kolter 2000). A number of methods such as fluorescently labeled rRNA-

targeted oligonucleotides, a variety of microsensors, real-time image analysis, and

confocal microscopy have helped researchers observe bacterial development while other

advances have been made to help cultivate bacteria such as chemostats, continuous-flow

slide cultures, microstats, and colonization tracks (Davey and O'Toole 2000). The

development of the biofilm lies in the formation of the extracellular polymeric

substances, the architecture and the inaction with other bacteria and particles (Donlan

2002).

Extracellular polymeric substances (EPS)

EPS makes up the majority of material that forms biofilm matrices. A large portion

of EPS is composed of polysaccharides (Donlan 2002). In some bacteria, genes to

synthesize flagella are down regulated while the genes to synthesize the EPS are up

regulated (Watnick and Kolter 2000). The composition of gram-negative bacteria's EPS

is largely neutral and in the presence of uronic acids or ketal-linked pryruvates it can take

on an anionic state. Gram-positive bacteria's EPS generally are cationic. The EPS of

some bacteria, such as those that are coagulase-negative, contain protein. EPS in most

cases has hydrophilic and hydrophobic regions. EPS is beneficial to the biofilm because

it can prevent desiccation (Donlan 2002) and can stop antibiotics from being transported

in to the biofilm (Donlan, Role 2000), as well as protect against pH shifts, UV radiation,

and osmotic shock (Davey and O'Toole 2000).

Composition and variable nature of the EPS are the two properties that have the

greatest impact on the biofilm. Composition of the EPS can affect the rigidity, the

deformation characteristics and the solubility in water. Sutherland gives the example that









EPS that has a backbone made of 1,3- or 1,4-/7 -linked hexose residues are more rigid,

less deformable and not very soluble in water compared to other components of EPS

(2001). The amount of EPS can be attributed to the type of organisms that make up the

biofilm, the age of the biofilm, the growth rate of the bacteria in the biofilm, the nutrients

available to that bacterium from the liquid medium (Flemming 2000).

Architecture

As microcolonies increase in number through division and addition of new cells

they begin to form mushroom-like colonies, as seen in Figure 2-1, that contain a number

of channels beneath the "mushroom caps" that bring nutrients to cells lower in the

biofilm. This shape suggests a controlled growth pattern that is most likely developed by

quorum sensing (a method for communication between bacteria) (Costerton 1995).

Other bacteria and particles

When biofilms are composed of one

species of bacteria, the bacteria alter

themselves genetically to best survive in ?* 6

the biofilm. In a mixed biofilm, bacterial

species will set themselves up in locations

that best suit the needs of the different Figure 2-1. The microocolony on
the far right shows
types of bacteria (Watnick and Kolter typical mushroom cap
formation (Costerton
2000). 1995).

It is also important to note that in the development of the structure of the biofilm

that nonmicrobial components (clay, blood particles, etc.) may be incorporated from the

host or the environment. Mineral build-ups in biofilms can be a problem in medical









devices and water systems (Durack 1975, Tunney and others 1999, Donlan, Biofilm

2000).

Maturation

After 12 hours to a few weeks of development (Pasmore and Costerton 2003),

microcolonies become EPS-encased and a mature biofilm is formed (Davey and O'Toole

2000). Other bacteria, gene transfer, and, quorum sensing affect the maturation of

biofilms. It is also interesting to note the ability for pathogens to be involved in mature

biofilms, and resistance that mature biofilms develop.

Other bacteria

In bacterial communities as in animal communities there exist interactions between

species. Mixed biofilms often develop synotrophic relationships where two

metabolically different species depend on products the other produces for survival

(Davey and O'Toole 2000). In the scientific literature there are other documented cases

of the following types of interactions that exist between bacterial species (Table 2-2).

Skillman and others (1998) used four different bacteria from the family

Enterobacteriaceae in dual-species biofilm studies. They concluded that the stain ofE.

coli used out-competed Klebsiella pneumoniae, Serratia marcescens, and Enterobacter

dahlinl'li However, the E. coli and the S. marcescens used in this experiment were

able maintain a neuralistic co-existance; the K. pneumoniae and E. a'1v'hincitan\, a

mutualistic relationship (Skillman and others 1998).









Table 2-2. The types of relationships that exist between bacterial species.



Neutralism When two populations do not affect each other.
Competition When two populations work against each other to achieve a
mutually sought after goal (such as nutrients or niche space).
Commensalism When one population benefits while the other remains unaffected.
Mutualism When both populations benefit as a result of their association. This
association can occur in many forms.
Symbiosis which are obligatory interactions
Protocooperation which are facultative interactions
Synergism which enhances the production or consumption of
bacterial made derivative.
Ammensalism When one population, without having direct contact can have a
negative impact on another.
Predation When one population feeds on another.
Parasitism When one microorganism is invaded by another.


Gene transfer and regulation

In biofilms there is a greater rate of genetic exchange by conjugation than occurs

between planktonic cells. It has been thought that there are plasmids necessary to form

biofilm and that bacteria will transfer plasmids to one another (Ehlers and Bouwer 1999,

Roberts and others 1999, Hausner and Wuertz 1999). Without these plasmids bacteria

would only form a microcolony and never develop into a biofilm. The genetic transfer of

resistance to antimicrobials is encoded on the plasmid; therefore, biofilms may be the

breeding-ground for antimicrobial resistance (Ghigo 2001).

Quorum sensing

Our knowledge of how biofilms form is still somewhat of a puzzle. Quorum

sensing is thought to be necessary to establish biofilms (Federle and Bassler, 2003).

Quorum sensing was first discovered in the 1970's by Nealson and Hastings (1979) when

they found that Vibriofischeri was responsible for producing light in a flashlight fish.










According to Federle and Bassler, quorum sensing is "a process in which bacteria

monitor their cell-population density by measuring concentrations of small secreted

signal molecules called autoinducers" (2003). It is known that the amount of

autoinducers present directly correlates to the number of bacteria present. Quorum

sensing occurs at the interspecies (between species) and intraspecies level (within

species). At the intraspecies level there are quorum sensing methods for gram-negative

and gram-positive bacteria. Figure 2-2 provides a visual representation of these two

intraspecies quorum sensing pathways. Gram-negative bacteria have one or more type of

LuxI-like proteins and each type of LuxI-like, which produce one acylhomoserine lactone

(AHL) autoinducers. After the AHLs are produced, they can freely diffuse outside the

cell membrane. The concentration of AHL A A
A A A
increases outside the membrane until the A A A A
A AHL
concentration reaches a certain level; then the AHL t

molecules are allowed to bind to LuxR-type L
Genes xyz
proteins. Only members of bacteria in the same ,s I

species can respond to that autoinducer, and v v
Secretion ATP Sensor kinase
therefore, it seems that there is little cross talk in P
N ADP Response
mixed gram-negative populations. Gram-positive tProcessingsep ruato

bacteria have never used AHL; they use --- Genes xyz

oligopeptide autoinducers that are sometimes
Figure 2-2. Quorum sensing in a.)
referred to as autoinducing peptides (AIP). These Gram-negative bacteria
and b.) Gram-positive
AIP compounds are 5-17 amino acids long and may bacteria (Federle and
Bassler, 2003)
contain side chain modifications. The gram-










positive membrane is not permeable to AIP so it requires active secretion. AIPs can then

be detected by cell surface receptors. This detection leads to the phosporylation of a

response regulator, which binds to the DNA promoter to regulate transcription of that

gene.

At the interspecies level there is one molecule that appears to be universal among

most bacteria: autoinducer-2 (AI-2). AI-2 is thought to be the key molecule that allows

for interbacterial communication in biofilms. AI-2 was first discovered by studying

Vibrio harveyi and to this day V. harveyi AI-2 molecular structure is the only one that has

been determined. Scientists do know that other bacteria can sense this AI-2 molecule and

that other bacteria produce AI-2-like molecules, but they are not sure if these molecules

have the same molecular structure or a different, but similar structure. The reason for this

has to do with the formation of AI-2. Figure 2-3 shows how AI-2 is formed in V

harveyi. It appears that all bacteria known to form AI-2 (or like molecules) have the luxS

gene to convert S-ribosylhomocysteine (SRH) to 4,5-dihydroxy-2,3-pentanedione (DPD).

However, DPD can be formed into a variety of compounds to which boron can later be

added. Therefore, if future studies demonstrate that AI-2 is a universal chemical, then

AI-2 cannot provide bacteria with the knowledge of what species that form the biofilm,

but it can let bacteria know how many "other" bacteria there are (Federle and Bassler

2003). If AI-2 can be derived from DPD then a bacterium will know what type of

bacteria exist and how many of them there are. AI-2 may not be produced by every

bacteria but it is possible that all bacteria may be able to sense and respond to AI-2.

NHz -00 -ooc
NH,
N N SAM -N"V

SO "mnm N6 II, HO O-CH3
l 0o LOHuxS O OH HOH H O HO-
SAM SAH SRH DPD Pro-AI-2 AI-2

Figure 2-3. "Biosysthesis of AI-2". (Federle and Bassler 2003)









Some known bacterial responses to AI-2 include virulence, toxin production, and cell

division (Federle and Bassler 2003). It has been found that quorum sensing, both intra-

and interspecies, is necessary for biofilm formation. For example Pseudomonas

aeruginosa needs AHL-autoinducers to create mature biofilm, and AI-2 seems to be very

crucial for the formation of mixed species of biofilms. Therefore, this indicates that

quorum sensing is a necessary method for bacteria to set themselves up in biofilms in

ways that are most beneficial to their needs and the needs of the community (Federle and

Bassler 2003).

Bacteria have also been known to have the ability to remove or add AI-1 or AI-2

molecules to the environment thereby tricking other species in believing that they are in a

low density or a high density of bacteria. Bacteria that are able to trick others by

providing them with false information face a competitive advantage over others. This

knowledge also gives an advantage to researchers looking for a method to discourage

biofilm formation (Federle and Bassler 2003).

Pathogenic organisms

Pathogenic organisms may be able to attach to biofilms; however, they do not

always seem to grow extensively in them. It is surmised that the reason for this is

pathogenic organisms' fastidious growth requirements and their inability to compete with

other organisms in the biofilm. Legionella pneumophila (Murga and others 2001), S.

aureus (Raad and others 1992), Listeria monocytogenes (Wirtanen and others 1996),

Campylobacter spp. (Buswell and others 1996), E. coli 0157:H7 (Camper 1998),

Salmonella typhimurium (Hood and Zottola 1997), Vibrio cholerae (Watnick and Kolter

1999), and Helicobacterpylori (Stark 1999) are pathogens that have been able to grow in

biofilms. The reason for their success in part seems to be due to associations and









interactions with organisms preexisting in the biofilm (Donlan 2002). Some of the more

important pathogen interactions are those of Staphylococcus spp., E. coli, and Salmonella

spp.

Work by Den Aantrekker and her colleagues (2003) shows how Staphylococcus

aureus can attach, form a biofilm, and detach from the surfaces of silicone tubing.

Gorman and others (1994) found Staphylococcus aureus in mixed cultures with members

of Staphylococcus spp. in catheter biofilms. Other bacteria can disrupt the attachment of

Staphylococcus aureus. Reid and others (1995) noted that Lactobacillus spp. can inhibit

the ability of S. aureus to attach and displace already established biofilms of S. aureus on

fibrous materials and epithelial cells.

E. coli has been found to form mixed culture biofilms in urinary catheters

(Ganderton 1992) and to dual-species biofilm with Klebsiella spp. in biofilms that block

biliary and pancreatic stents (Brant 1996). Research conducted by Banning and others

(2003) found that E. coli was capable of establishing itself in mixed culture of indigenous

groundwater microorganisms in a laboratory-scale reactor. However, if the nutrient

levels were increased the E. coli had difficulty out-competing the indigenous microflora.

This demonstrates that the conditions of the system regulate the ability of E. coli to be an

integral part of the biofilm (Banning 2003).

In a work by Joseph and others (2001) examined the ability of two Salmonella spp.

isolated from poultry to attach to plastic, cement, and steel surfaces. They found that the

bacteria formed the thickest biofilm on plastic surface followed by cement and steel. The

biofilms were then exposed to different qualities of hypochlorite and iodophor sanitizer

for varying lengths of time. The results noted that the biofilms were more resistant to the









sanitizers than the free cells. The authors concluded that Salmonella spp. can form a

biofilm on food contact surfaces and can be a source of contamination for food products.

Also, if food contact not cleaned with the proper concentration of cleaner for the correct

amount of time the Salmonella spp. biofilms may persist on the food contact surfaces

(Joseph and others 2001). Camper and others (1998) found that Salmonella typhimurium

was able to exist in a model water system biofilm with a group of unknown heterotophic

organisms for more than 50 days. This indicates that S. typhimurium is capable of

integrating itself with other bacteria to form a biofilm (Camper and others 1998). A

study by Esteves and others (2005) used Salmonella enterica serovar Typhimurium and

E. coli isolated from the natural flora of the gastrointestinal tract to study their biofilm

formation on the HEp-2 epithelial cells. They concluded that the Salmonella would

predominate over the E. coli if they were exposed to the on the HEp-2 epithelial cells at

the same time. If the E. coli was an established biofilm on the cells the Salmonella will

establish itself in areas where the E. coli has not attached and displace and replace the E.

coli biofilm (Esteves and others 2005).

Resistance

One of the key advantages to being in a biofilm is that the biofilm can help bacteria

resist effects of chlorine, antibiotics, and detergents. Quorum sensing (Donlan 2002) and

the ability of the biofilm to alter aspects of its local environment such as pH and oxygen

concentration may help with this resistance (Pasmore and Costerton 2003). Lewis in the

Riddle ofBiofilm Resistance points out the three main reasons for biofilm resistance: 1)

EPS can diffuse and bind any possible antimicrobials, 2) antimicrobials are more

effective at killing rapidly growing cells so the slow growing cells of the biofilm are

harder to kill, and 3) the adaptation of gene specific traits that help in the resistance









(2001) such as changing cell surface proteins which gives the antibiotics fewer places to

bind (Pasmore and Costerton 2003).

Dispersal

Dispersal happens by one of three mechanisms: 1) the shedding of daughter cells,

2) detachment that occurs as the result of quorum sensing or nutrient levels, or 3)

shearing of biofilm aggregates (continuous removal of small portions of biofilm) because

of flow effects). The mechanisms of dispersal by the shedding of daughter cells is not

well understood (Donlan 2002); however, according to the research of Gilbert and others

shedding occurs because the newly formed daughter cells have the least hydrophobicity

(1993). Often, when nutrient levels become low a biofilm will dissociate by breaking

down the EPS matrix and in some species it will use the EPS as a nutrient source before

seeking a more nutrient rich environment. Detachment due to flow occurs by three

methods: erosion or shearing (continuous removal of small portions of the biofilm),

sloughing (rapid massive removal), and abrasion (detachment due to collision of particles

from the bulk fluid with the biofilm) (Brading and others 1995). Erosion seems to occur

most often when the biofilm is thick and there is an increased fluid shear (Characklis,

Biofilm 1990). Sloughing also occurs in thick biofilms but occurs more randomly than

erosion. It is thought that sloughing occurs as a result of nutrient or oxygen reduction

(Brading and others 1995).

Detergents and Sanitizers

Detergents

The job of detergents is to remove gross soil and residue. An effective detergent

cleaning treatment is based on an analysis of the soil type (lipid, carbohydrate, protein,

mineral deposits, microorganisms, dirt). One should choose a detergent that will work









most effectively on that particular soil(s). Carbohydrates can be removed from surfaces

with water but also alkaline cleaners can be used to remove it as well. Care should be

taken to make sure that overheating and drying does not occur because the sugars will

caramelize and starches will form a glue-like material. Undenatured proteins are

generally water-soluble while denatured proteins are water insoluble. Both can be

removed with an alkaline cleaner. Lipids are insoluble in to water but can be melted with

heat, saponified by alkalis and high temperatures, and emulsified by polyphosphates.

Mineral deposits are alkaline in nature and are insoluble in water but can be dissolved in

acids (Katsuyama 1993).

Alkaline detergents saponify fats and form water soluble compound with proteins;

however, they are ineffective below a pH of 8.3. Some commercial alkalis that are

available are sodium hydroxide, sodium carbonate, sodium hydroxide, sodium

sesquicarbonate, trisodium phosphate, sodium metasilicate, tetrasodium pyrophosphate,

and sodium tetraborate (Parker and Litchfield, 1962). Hard water and sodium hydroxide

should not be used together as it will cause precipitation. Adding chlorine to an alkaline

cleaner allows for better removal of proteins. This leads to better cleaning of milk stone

(milk solids and mineral deposits from the milk). Chlorine used in detergents is not the

sanitization agent in alkaline cleaners because the pH is too high for the chlorine to be

effective. Acid cleaners dissolve mineral deposits. They have pH of less than 2.5

(Katsuyama 1993). Inorganic acids used are hydrochloric, sulfuric, nitric, and

phosphoric acids. The disadvantage to these acid cleaners is that they will corrode soft

metals; however, organic acids have are less corrosive and irritating (Jennings 1965).









Detergents by themselves may not provide effective cleaning. Based on the level

of soil and the equipment to be cleaned, hand scrubbing, high-pressure water, flushing

recirculation, and temperature may be needed (Katsuyama, 1993).

Achieving the correct time, temperature, and concentration are important for

effective detergent cleaning. The detergents' manufacturers should indicate what

temperatures, and concentration are appropriate for the product. Other items to evaluate

about when choosing a cleaner are corrosiveness, irritability to personnel, regulatory

standards, foaming, and versatility of uses within the facility (Katsuyama, 1993).

Sanitizers

The goal of sanitizers is to destroy the vegetative cells; however, vegetative cells of

resistant bacteria and bacterial spores can survive. For sanitizers to work effectively and

efficiently, soils must be removed from the surfaces. Sanitizers that can be used in food

processing plants are: heat, halogens, quaternary ammonium compounds (QUATS),

acids, alkalis, ultraviolet irradiation, and ozone. There are disadvantages and advantages

to using the above-mentioned compounds. However, a sanitizer should be chosen for its

quick kill, customer and employee safety, regulatory compliance, easy to removal from

the surface, cost, ability not affect the food, ease in determining its concentration,

stability, noncorrosiveness, and solubility in water characteristics (Katsuyama 1993). In

the tank wash industry, QUATS are common choices for sanitizers. Some of the

advantages of using QUATS are the following: heat-stable, effective over a wide pH

range, noncorrosive, nonirritating, impart no off flavors to food products, not as affected

by organic matter than chlorine, and they leave a non-volatile residue that inhibits molds,

yeasts, and bacteria (Clinger 1973 and Ohio State University 1967). The disadvantage to

using QUATS is that they are not compatible with nonionic wetting agents in detergents,









and they are rendered ineffective by wooden, cotton, nylon, cellulose sponges, and some

plastics (Mauer 1974).

Wet heat is often used in conjunction with a chemical sanitizer in the tank wash

industry. The advantages of using a heat is it is inexpensive, it can be measured, there is

no residue, it is not corrosive, it provides a non-selective kill, and it penetrates hard to

reach surfaces (Jennings 1965). The problem with heat is that to provide effective

sanitization it must reach at least 820C (180F) (Katsuyama 1993).

The Environment of Stainless Steel

Stainless steels come in three key groups that are based on the microscopic

structure and the composition: martensitic, ferritic, and austenitic (Bosio Metal

Specialties, 2000). The martensitic group contains AISI metal types 403, 410, 416, 420,

440. This group is composed of about 12-18% chromium, very little nickel (if any), and

0.06 to 1.20% carbon. These stainless steels can be heat-treated and they are magnetic.

The ferritic group is composed of AISI metal types 405, 409, 430, 442, and 446. These

types of metals contain 12-18% chromium, 0% nickel, and 0.06-35% carbon. This metal

group is also magnetic. The austenitic group contains AISI types 201, 202, 301, 302,

303, 304, 316, 321, 347 and most of the 300 series alloys. The group contains up to 7-

30% chromium,6-36% carbon, and 6-36% nickel. Austenitic stainless steels are non

magnetic. These metals are not hardened by heat treatment but by cold treatment, which

may cause them to be slightly magnetic (Bosio Metal Specialties, 2000).

Stainless steel in tankers is generally composed of either 304 or 316 stainless steel

unless the tanker is used to hold food-grade oils in which case a 407 stainless steel is

generally used. 304 and 316 stainless steel is available standard and low carbon (304L

and 316L) varieties. 304 is the commonly used of stainless steel because of its easy









formability and corrosion resistant nature. The low carbon variation is formulated so

there is no carbide precipitation from the welding process. It has the same corrosion

resistance as the standard version but has lower mechanical properties than the standard

304. The 316 can handle higher temperatures, and is more resistant to pitting and

corrosion than the 304. The 316L is used to avoid the carbide participation due to the

welding process (Bosio Metal Specialties 2000).

Some of the most popular finishes of stainless steel in the food industry are #2D

and #4 stainless steel finish or higher. #2D is used when a manufacturer cannot

guarantee a pit free rolled finish but most food processors like at least a #4 finish with a

#7 finish being preferred by some (Frank and Chmielewski 2000). A 2D finish is dull

manufactured by a cold rolling annealing and descaling (Bosio Metal Specialties 2000).

A #4 finish is one where course abrasives are used initially to grind the stainless steel

followed by a grinding with 120-150 mesh. This finish is generally used in a wide

variety of food applications. Finish #6 is a #4 finish where the last brushing is done with

abrasive and oil. The #7 finish "is produced by baffling finely ground surface, but the

grit lines are not removed" (Bosio Metal Specialties 2000). The #8 finish is highly

reflective and free of grit lines due to the extensive polishing by successive abrasions and

baffling (Bosio Metal Specialties 2000).

Gauge of stainless steel correlates to the thickness of the stainless steel. According

to American Delphi Stainless Steel Guide 14 gauge stainless is 0.0747inches thick; 16

gauge, 0.0598 inches thick; 18 gauge, 0.0478 inches thick (American Delphi 2004).

Stainless steel has become a standard choice for the construction for much food

processing machinery for many reasons including durability and its ability to resist









corrosion (Maller 1998). However, the stainless steel processing environment can

become a home to biofilms because of its microscopic hills and valleys, which vary with

different levels of finish. Studies have been conducted to see how see how different

levels of finish affects biofilm formation and biofilm cleanability. Arnold and others

(2001) studied 5 different polish types of 11 gauge, 304 stainless steel against untreated

stainless steel. They found that electropolished, and steel-burnished was significantly

different from the control of untreated stainless steel in their ability to resist biofilm

formation while glass-beaded, acid dipped, and sand-blasted stainless steel were not

significantly different from the control. It was thought that the reason glass-beaded

stainless steel and the sand-blasted stainless steel had more bacteria attachment was that

during the polishing process the glass and sand created craters and scars which create

regions for bacteria to attach. They also discovered that the leading indicators that

biofilm is going to form to the surface are root mean square (RMS) which is a

measurement of the surfaces roughness, and the center line average which is the "depth

from the peak of the sample" at which there is a 50% of the sample area below and a 50%

of the area above (Arnold and others 2001). However, surface finish appears to have no

effect on cleanability according to Influence of Surface Finish on the Cleanability of

Stainless Steel by Frank and Chmielewski. They first established that Bacillus

stearothermophilus spore count was better determinate of cleanability than the

Pseudomonas spp. biofilm. They discovered that the mean peak to valley height RZ(DIN)

and maximum peak to valley height Rmax have a significant correlation to cleanability of

Bacillus \ie, i/t eil i e ,qhihi1\ spores. It is advised that if manufacturers want to choose a

stainless steel that will have maximum cleanability for spores or biofilms, it should be









chosen not necessarily by polish type but by the amount of surface defects (Frank and

Chmielewski 2000). The authors also found that soiling and cleaning creates increased

soil build up and decreased the number of Pseudomonas stearothermophilus spores on

the stainless steel surface after 11 soiling and cleaning. The authors suggest that this

behavior may be due to that fact that "repeated soiling and cleaning cycles may stimulate

heat activation and inactivation of spores" (Frank and Chmielewski 2000).

A study done by Arnold and Suzuki showed the effect of corrosion on different

polishes of stainless steel. They concluded that the sandblasted and glass-bead polished

samples they tested experienced the greatest increase discoloration and biofilm formation

after exposure to a corrosive treatment. While electropolished stainless steel experienced

the least discoloration and biofilm formation after exposure to a corrosive treatment. The

researchers thought this was due to the fact that electropolished stainless steel was

composed of very few reactive elements. This study emphasizes the importance of

understanding composition of the stainless steel as well as the amount of corrosion that

occurs (Arnold and Suzuki 2003).

In a study evaluating milk proteins and bacterial adhesion, the interaction between

stainless steel, the proteins in milk (alpha-casein, beta-casein, kappa-casein, alpha-

lactalbumin) or glutaraldehyde treated milk proteins, and the amount of biofilm formation

of E. coli P. fragi, S. aureus, L. monocytogenes, and S. marcescens were observed

(Barnes and others 1999). From this study the researchers found that S. aureus, L.

monocytogenes, and S. marcescens cell attachment is reduced by 20% or less when milk

protein is present on a stainless steel surface than when milk is on a clean surface while

E. coli and P. fragi show no difference in the amount of biofilm formation in clean or









milk protein coated stainless steel. As concentrations of the milk protein became less the

bacterial attachment of S. aureus, L. monocytogenes, and S. marcescens became greater

(Barnes and others 1999). The researchers think that as the protein layer on stainless

steel surface was thicker than the iron photoelectron escape depth but when solution used

was below 1% milk there was a sharp increase in iron signal and this increase bacterial

attachment (Barnes and others 1999). They showed the glutaraldehyde treatment

increased the attachment because cross-linking the proteins reduced the proteins' ability

to discourage bacterial attachment. They discovered that the type of milk protein had no

effect on biofilm levels. Ionic composition of the suspending medium had the greatest

effect on clean-stainless-steel biofilm formation because the dissolved cations from the

suspending medium shielded the electronegative charge on the surface of the stainless

steel thus reducing biofilm formation. The authors determined that the suspension media

had no effect on attachment when milk proteins were attached to the stainless steel

surface except for CaC12 and FeC12 that encouraged biofilm development. The authors

concluded that the reason FeC12 increases absorption is because the ferrous ions can serve

as a bridge between the bacteria and the milk proteins or it helps to cross-link proteins.

CaC12 increases biofilm formation because calcium is a component of milk that is not

found in the milk proteins but when it is reintroduced back in with the proteins it causes a

conformational change which causes the absorbed proteins to facilitate attachment

(Barnes and others 1999).

Environment of Rubber

There are 4 classes of rubber specified in the 3-A Sanitary Standards for

Multiple-Use Rubber and Rubber-Like Materials Used as Product Contact Surfaces. The

two classes used on tankers are class 1 and class 3. Class 1 gaskets are heat exchanger









gaskets, O-ring, CIP gaskets, flange gaskets, rotary seals and hoses. These rubbers can

tolerate product exposure and temperature sanitization up 1490C (300F) and a chemical

or bactericidal treatment up to 820C (180F). Class 3 rubbers can be used for cold

applications such as milk and milk products and air tubing, manhole and door gaskets,

seals and hoses. These rubbers can tolerate product exposure and temperature

sanitization up 490C (120F) and a chemical or bactericidal treatment up to 820C (180F)

(3-A Standards for Multiple 1999).

All rubbers must meet other standards set out in the 3-A Standards including that

they can not be toxic, they must meet certain absorption, aging, and compatibility with

cleaner and sanitizer standards, they must be fabricated under good manufacturing

practices to 3-A Standards (3-A Standards for Multiple 1999).

3-A also states that although gaskets may come in different colors the color does

not affect the sanitary conditions of the gasket. They note that different conditions in

rubbers environment and cleaning regiment will produce different life times. The

document states that a rubber's sanitary lifetime should be monitored so rubbers used for

a similar purpose so they can be scheduled for replacement before cracks and sloughing

appears (3-A Standards for Multiple 1999).

Storgards and others (1999) wrote two papers on the influence old and new

ethylene propylene diene monomer (EPDM), nitrile butyl (NBR), polytetrafluoroethylene

(PTFE) and Viton rubbers had on the formation ofB. thuringiensis, Pseudomonas

fragi, Pantoea at'1',,i'i/\, and Pediococcus inopinatus biofilms when used in brewery

or dairy processing environments. The authors concluded that new rubbers had different

susceptibilities to biofilm formation that is dependent on the type of food processing









environment they were used in and the type of bacteria creating the biofilm. For new the

PTFE rubber were most resistant to biofilm build up in dairy conditions and NBR rubber

surfaces were most resistant to biofilm build up in brewery conditions. However, both

rubbers were as cleanable as stainless steel when cleaned at both hot dairy or cold

brewery conditions. Also the ability to clean new rubber differed because of the different

surface properties of the rubber. In some aged rubbers the ability to remove biofilm from

the surface was reduced (Storgards and others 1999). NBR that was aged to reflect 432

cleaning was found to more readily support biofilms. Both NBR and Viton have

increased cracks and a rougher surface. Viton was determined to be the rubber most

quickly affected be aging. It was determined that EPDM was the most durable of the

rubbers over time and the hygienic properties of PTFE were found to be almost unaltered

over time (Storgards and others 1999).

Some rubbers have shown to be bacteriostatic to certain groups of bacteria. NBR is

bacteriostatic towards L. monocytogenes, Salmonella typhimurium, Staphylococcus

epidermidis, Staphylococcus aureus, Y. enterocolitica and, E. coli 0157:H7. EPDM is

bacteriostatic towards S. epidermidis and S. aureus. Viton was not bacteriostatic (Ronner

and Wong 1993).

Review of Methodology

Coliforms, Fecal Coliforms, and E coli.

The BAM directions identify the number of coliforms and fecal coliforms with 3-

tube most probable number (MPN) dilution series in lauryl tryptose broth (LTB). For

each positive tube in the MPN series a loopful is inoculated into brilliant green lactose

bile (BGLB) broth to be incubated at 350C for 24-48 hrs and examined for gas

production. Also, each positive MPN tube is inoculated into Escherichia Coli (EC) broth,









incubated at 45.50C for 24-48 hrs, and examined for gas production. Tubes that test

positive for fecal coliforms are streaked to Violet Red Bile Agar (VRBA) and incubated

at 35C for 18-24hrs to test for the presence of E. coli. The colonies should be checked

for a flat, dark-centered colony with or without metallic sheen. Positive isolates can be

streaked to PCA. These colonies can be further tested for gas from lactose, citrate,

methyl red-reactive compounds, Voges-Proskauer (VP)- reactive compounds, indole

production, and gram stain (USDA's Bacteriological Analytical 2000).

BAM also suggests a solid method for enumerating coliforms by creating pour

plates with VRBA and incubating them at 350C for 18 to 24 h. Ten presumptive

coliforms should be inoculated into BGLB and incubated as described above to guarantee

they are coliforms.

As an alternative to VRBA agar, PetrifilmTM Coliform Count Plates (3M; St. Paul,

MN) was developed for use by dairies and food production facilities. The Coliform

Count Plates are made from VRBA, tetrazolium indicator, and a cold-water-soluble

gelling agent. This combination enumerates coliforms in a similar fashion to VRBA.

E. coli also have similar ELISA tests and immunomagnetic beads as Salmonella.

Coliform-produced enzyme P-galactosidase breaks down X-Gal (5-Bromo-4-Chloro-3-

Indolyl-P-D-galactopyranoside) into 5-Bromo,4-chloro-indoxyl intermediate which

through oxidation produces a blue derivative. E. coli with p-glucuronidase produced

enzyme breaks down MUG into fluorescent derivative (4-Methylumbelliferone). Using

these two reactions a number of rapid tests for coliform and E. coli detection have been

created such as ColiTM Complete (BioControl Systems, Inc.; Bothell, WA) and









E*Colite test (Charm Sciences, Inc.; Lawrence, MA). Both ColiTM Complete can

detect and E*Colite test can detect 1 coliform bacteria in 100mL.

Detection Methods for Salmonella

Salmonella can be detected though a combination of standard methods or

commercially available test kits and extraction measures. The USDA's Bacteriological

Analytical Manual (BAM) method for the isolation of Salmonella is a commonly used

isolation method. BAM has different protocols for different food products. The method

suggests 24hr incubation at 350C in a universal pre-enrichment broth, followed by 24hr

incubation in a selective enrichment media (Rappaport-Vassiliadis (RV) medium,

tetrathionate (TT) broth, selenite cystine (SC) broth), followed by the broth cultures being

streaked to preformed plates of hektoen enteric (HE) agar, xylose lysine desoxycholate

(XLD) agar, and bismuth sulfite (BS) agar and incubating them at 35C for 24hrs

(USDA's Bacteriological Analytical 2000).

Along with standard isolation methods there are also commercial test and

extraction kits. Most kits operate under similar principles. The samples are cultured in

the preenrichment and selective broths then they are exposed to a surface with antibodies

that are specific to Salmonella spp. If Salmonella is present in the sample then the

antigens on its surface will bind to the antibodies. Other material is rinsed away then

enzyme labeled antibodies are bound to the surface; this generally produces a color

reaction that is not present in samples without Salmonella. This is the principle under

which the TECRA Salmonella Visual Immunoassay (TECRA International Pty Ltd)

and other Enzyme-linked Immuno Assays (ELISA) operate. Immunomagnetic beads

operate on a similar principle as the ELISA test. Antibodies are on the surface of the

beads. The Salmonella cells bind to the surface. The magnetic nature of the beads allows









them to be retained during washing without losing them in the washes. The beads and

the bacteria can then be plated onto solid selective media. The TECRA Salmonella

Visual Immunoassay can detect 1-5 CFU/25g sample and Dynal Anti-Salmonella

immunomagnetic beads (Dynal Biotech, Oslo, Norway) can detect 1 CFU/25g of sample.

The 1-2 Test (BioControl Systems, Inc.; Bothell, WA) passes through a selective

enrichment through a nonselective motility medium and Salmonella is immobilized by

flagellar (Polyvalent H) antibodies.

Detection Methods for Alicyclobacillus

Some researchers use K agar to isolate Alicyclobacillus while others use AlibrothTM

and AliagarTM. For the K agar procedure 3, ImL portions of heat shock sample (10mL at

80C for 10mins), and 18-20mL of K agar with each heat shocked sample to create three

plates to test for Alicyclobacillus. Plates were incubated at 43+/-1C for 3 days. Colonies

will be cream-colored, raised, and opaque (Evancho and Walls 2001).

An alternative to K agar is the use of Alibroth and Aliagar for culturing and

isolating Alicyclobacillus. The typical protocol used is to inoculate 100mL of Alibroth

with 10mL of heat-shocked culture (75C for 15 mins) and incubate at 450C for 72 h.

After 72 h Alibroth is streaked to Aliagar plates. If there is growth at 450C after 2 days

restreak 2 plates of Plate Count Agar (PCA) and 2 plates of Aliagar. One plate of each

media is incubated at 450C for at least 2 days while one plate of each media is incubated

at 25C for 4 days (Parish and Goodrich 2005).

Detection of Aciduric, Yeast and Mold, Thermoduric, Mesophilic and Psychroduric
Microorganisms

The following protocols are taken from the Compendium of Methods for the

Microbial Examination ofFoods. Although other methods can be used for isolating and









culturing the above types of microorganisms these are some of the most accepted.

Acidophiles can be isolated using on pour plating a ImL sample and Orange Serum Agar

(OSA) and then incubating the sample at 300C for 2 days (Hatcher and others 2001).

Yeast and Mold count can be determined by pour plating ImL of sample with acidified

Potato Dextrose Agar (aPDA) and incubating the plates for 22-25 C for 5 days (Beuchat

and Cousin and others 2001). Mesophilic and Thermoduric organisms are determined by

plating ImL of sample with PCA and incubating the samples at 35 C and 45-50C

respectively (Olson and Sorrels 2001). Psychroduric counts are determined by spread

plating 0. ImL of sample preformed PCA plates and incubating them at 7+1C for 10 days

(Vanderzant and Matthys 1965).

DNA Sequencing

The DNA is separated from the cell through extraction and purification methods

(one example of this procedure can be found in Molecular Cloning by Sambrook and

others (1989)). Then primers are selected to bind to regions of interest in the DNA

sequence. This region is then copied or amplified through cycles of denaturation,

annealing and extension over an over again with the help of a thermocycler.

Denaturation occurs at a high temperature, which separates the DNA into two separate

strands. Before annealing begins the temperature is lowered and then primers bind to the

DNA. The temperature is raised slightly and DNA polymerase binds to selected DNA

regions and extension begins. During this process PCR polymerase progresses along the

strand, replicating creating a copy of the target region. The final result is two double

strands. The process is repeated through several cycles, which leads to an exponential

increase in copies of the target region (Entis and others 2001).









These copies are then run through a gel. The region of interest cut out of the gel.

The fragment is separated from the gel and then sequenced. Sequences are compared

against a database of know sequences (Worobo 2005). The relationship between the

database and the known sequences is determined.

Biofilm Growth Characterization

There have been many devices designed to grow biofilms such as a mini flow

chamber (a variety of flow chambers), the Robbins device (some times referred to as the

modified Robbins device), and the Chemostat. According to Ramos and others (2001),

they commonly use the mini flow chamber because it is easy to use, easy to monitor

microscopically, and the results have been found to be statistically reproducible under the

same conditions; however, the biofilms in this system are difficult to access and the flow

rates that can be used in the system are limited. The Robbins device is composed of

several flow chambers that each have holders for holding a piece of material (like a piece

of stainless steel) for biofilms to attach to (Ramos and others 2001). The Chemostat is a

simple device. The top has openings for the aseptic insertion of coupons that are hung

from wire into a predetermined amount of liquid and bacteria culture (Keevil 2001). The

interior of the Chemostat is titanium, and does not contain any Fe, Ni, Mn, or Cr that

would affect biofilm formation (Keevil 2001). The Chemostat can control such

parameters as temperature, dissolved oxygen concentration, and pH. The liquid in the

system can be kept moving by a stir bar in the bottom of the system and if desired liquid

and culture can be pumped in and out of the system (Keevil 2001).

Observation Methods

Several methods have been developed for monitoring developing biofilms: electron

microscopy, epiflourescent microscope, and confocal laser scanning microscopy. There









are advantages and disadvantages to all the aforementioned systems. Electron

microscopy is a poor choice for studying biofilm due to the dehydration and the eventual

destruction of the biofilm structure. The foremost disadvantage to epiflourescence

microscopy is that as the biofilm gets thicker it becomes harder to see clear images of the

biofilm (Christensen and others 1999). Therefore, epiflourescence microscopy is best

used is best used for a single layer of cells in a specific region because all other regions

will be out of focus. The epiflourescent microscope can be used with cameras to achieve

images ofbiofilm development (Ramos and others 2001). The scanning confocal laser

microscope corrects the problem experienced by epiflourescent microscope by

"collecting returned fluorescent light from only the thinnest focal plane afforded by the

objective lens". Also, this type of microscopy can generate a three-dimensional image

"by scanning several planes interspersed at short distances" (Christensen and others

1999). These three dimensional images can generally be done by any confocal-based

software however the best software is Unix-based systems like IMARIS according to

Christensen and others (1999).

Other technologies such as Fluorescent in Situ Hybridization (FISH) and wide

variety of cameras helped immensely in the study of biofilm formation. The general

principle behind FISH is to fix cells to the surface to which they are attached, and

hybridize specifically targeted genes in the bacteria's 16S or 23S rRNA sequence with a

fluorescent labeled oligonucleotides probe. FISH can help researchers identify placement

of certain bacteria in the biofilm, as well as help them determine the growth rate of the

cells by using the fluorescence intensity (Ramos and others 2001).









A Need for More Research

In light of our limited knowledge on biofilm formation or transportation trucks in

general, saying that more studies need to be done on the microbial aspects of

transportation tankers is a tremendous understatement. Future research is needed to

understand what species of bacteria are present in citrus juice biofilms, how these

biofilms form, what types of quorum sensing occur in these biofilms, how these biofilms

form on stainless steel and rubber, how these biofilms form in the tanker truck

environment, how these biofilms change with the material makeup or the conditions in

the transportation tanker, the transportation process, and the time before cleaning; and

how force, pressure, chemicals, age of the biofilm affect the cleaning process. Although

not all these questions will be addressed in this research, it is the hope of the author that

this research will start to address some of the most basic questions in hope that others will

take the opportunity to expand upon this work to bring changes based in scientific fact to

the tank wash industry.














CHAPTER 3
MATERIALS AND METHODS

Part I: Identification and Characterization of Microorganisms in Samples

Sample Collection

Survey samples were collected from wash station "A" located in central Florida

from January to May of 2005. Tankers sampled carried either citrus juice or dairy

products in their most recent load. Tankers have gaskets in a variety of shapes. Tankers

in this study were classified as gasket type A or gasket type B. Figure 3.1 illustrates the

differences between gaskets. Some tankers have a lip around the manway (Picture D)

while others have a flat surface surrounding the manway (Picture B). Type A gaskets

tested in this study are made out of neoprene and designed to be set onto tankers with a

flat manway surface and they have one large lip that extends into the manway to hold the

gasket in place (Picture A). Type B gaskets tested in this study are made out a number of

different types of rubber including nitrile butyl rubber (NBR) or ethylene propylene diene

monomer (EPDM) rubber; however they have the same design (Picture B). They have a

groove between two lips. The groove is designed to be just big enough to place on

manways that have a lip around the edge. Gaskets that meet the two afore-mentioned

criteria were washed with a hot or ambient temperature wash regimens. A hot wash

regimen consisted of a hot temperature pressurized spray with an alkaline detergent, a hot

temperature alkaline detergent wash, a hot temperature chorine wash, and an ambient

temperature acid sanitizer wash; and an ambient temperature wash regimen consisted of a

ambient temperature pressurized spray with an alkaline detergent, a











































Figure 3-1. Type A and Type B manway styles and gasket types. A.) Gasket type A, B.)
Gasket B, C.) Manway lid associated with gasket type A D.) Manway lid
associated with gasket type B.

ambient temperature alkaline detergent wash, a ambient temperature chorine wash, and a

ambient temperature acid sanitizer wash. Surfaces of the gasket that were exposed to the

liquid product inside of the tanker were swabbed with a sterile SpongesicleTM (a sponge

on a stick with 10ml of nutrient buffer in a sealed bag) (Biotrace International; Bridgend,

Wales) using firm and even pressure. Each sample was labeled with a consecutive









number. Samples were kept at 40C until time of analysis (no longer than 24 h after

sampling).

Sample Preparation

Microbiological materials for this research were manufactured by Becton,

Dickson, and Company (Franklin Lakes, NJ) unless otherwise specified. 90 mL of sterile

buffer peptone water (BPW) were added to each SpongesicleTM bag. Bags were

homogenized by hand and then were used for the enumeration of the following types of

microorganisms: psychroduric; mesophilic; thermoduric; yeast and mold (YM); and

aciduric. Additionally the presence or absence of coliforms, Escherichia coli,

Alicyclobacillus spp., and Salmonella spp. was tested. Tests for Streptococcus, and

Staphylococcus detection and Most Probable Number (MPN) for coliforms, fecal

coliforms, E. coli were conducted later if necessary as discussed later in the materials and

methods.

Sample Analysis

Psychroduric, mesophilic, thermoduric, yeast and mold; and aciduric enumeration
and characterization

Psychroduric plates were obtained by spread plating 0.5 mL of sample on 2 plates

of plate count agar (PCA) and incubating them at 60C for 10 days (Vanderzant and

Matthys 1965). Mesophilic plates were obtained from pour plating ImL or 0. ImL of

sample with PCA and incubating the sample for 5 days at 350C. This method was altered

slightly from Compendium of Methods for the Microbial Examination ofFoods (Downes

and Ito 2001), which states that PCA plates should be incubated for 2 days at 35C.

Compendium (Downes and Ito 2001) also notes that 2 days may not be sufficient time to

allow for visualization of injured cells (Swanson and others 2001). Since the cells in the









samples may be injured by cleaning agents, during an initial trial of the materials and

methods, PCA plates were observed over the first eight days of incubation. Observations

of PCA plates noted that most colonies had formed by day 5. A paired T-test on the data

from part II of this experiment conducted on the mesophilic counts on day 2 and 5 show

that there was a significant difference between the counts. The same incubation period

was allowed thermoduric plates. Thermoduric plates were obtained from pour plating

ImL of sample with PCA and incubating the sample at 500C (Olson and Sorrells 2001).

Yeast and Mold (YM) plates were obtained from pour plating 2mL of sample with

acidified potato dextrose agar (aPDA; pH=3.5) and incubating the sample for 30C (Redd

and others 1986) for 2 days. Redd and others (1986) recommend a three-day incubation

period; however, two days is commonly chosen in industry to expedite shipping of

product if it can be proven that there is no significant difference between the counts on

day 2 and day 3. Two days was chosen over 3 days or 5 days for incubation of YM

plates because a T-test of initial trials from part I, and data from part II indicate that there

was no significant difference between YM counts at 2, 3 or 5 days. Aciduric plates were

obtained from pour plating 3mL or ImL of sample with orange serum agar (OSA) and

incubating the sample for 2 days at 300C (DIFCO Manual 1984).

Plate counts were obtained and recorded. The quantity of different colony

morphologies present and their descriptions were recorded. Typical colonies of the

different morphologies were selected, and streaked to separate plates of the agar from

which they were isolated. Plates were incubated at the appropriate temperature for 1- 2

days. Colony morphologies and diameter size (mm) were recorded. Colonies were then

gram stained (Tortora and others 1998), and catalase/oxidase tests were performed.









Gram-positive rods were then transferred to sporulation agar (1 L of Nutrient broth, 15 g

ofBacto agar, 0.030 g MnC13, and 0.030 g CaC12 mixed together and autoclaved at 121C

at 15 psi for 30 m (Huang and others 2001)) for 3 days at the appropriate temperature.

Spore stains were performed (Tortora and others 1998).

Coliform, fecal coliform, and E. coli detection

E*Colite test (Charm Sciences, Inc.; Lawrence, MA) for the presence or the

absence of coliforms or E. coli and a PetrifilmTM Coliform Count Plate (3M; St. Paul,

MN) were performed according to the manufacturer's instructions. If coliforms were

present in the E*Colite samples or on the PetrifilmTM, the Most Probable Number (MPN)

for coliforms would be performed using the original sample as directed in chapter 4 part

E of the Bacterial Analytical Manual (BAM) (2002). Also if coliforms were present in

the E*Colite sample or on the PetrifilmTM; lml of the Ecolite sample or colonies from the

PetrifilmTM would be inoculated into a tube ofE. coli (EC) broth containing 4-

methylumbelliferyl-P-D-glucuronide (EC-MUG) and incubated at 44.50C to determine if

fecal coliforms or E. coli were present. If fecal coliforms or E. coli were present an MPN

for fecal coliforms or E. coli would be performed as directed in chapter 4 part E of the

BAM (September 2002). If presumptive E. coli was present in the E*Colite sample; 1

mL from the E*Colite bag would be inoculated into a tube of EC-MUG and incubated at

44.50C. If the EC-MUG tube came back positive for E. coli an MPN for E. coli would be

performed as directed in chapter 4 part E of the BAM (2002) and a streak on Levine's

Eosin-Methylene Blue (L-EMB) agar would be done to look for a typical colony

morphology. All colony morphologies that appeared typical were confirmed presumptive

E. coli using a BBLTM EnterotubeTM II. Samples were frozen for subsequent 16S rRNA

identification.









Streptococcus spp. and Staphylococcus spp. detection

E*Colite bags that were yellow and fluorescent, or blue and fluorescent but were

not E. coli were streaked onto PCA agar and incubated at 35C for 24 h. Representative

colonies were selected, and gram stained. Any gram-positive cocci were streaked to

Baird-Parker Medium (OXOID LTD.; Basingstoke, Hampshire, United Kingdom) with

Egg Yolk-Tellurite Emulsion (OXOID, LTD.; Basingstoke, Hampshire, United

Kingdom), or inoculated into Streptococcus Faecalis (SF) Medium as directed by the

DIFCO Manual (1984). Samples were frozen down for 16S rRNA identification.

Salmonella spp. detection

Testing for the presence of Salmonella spp. was done using TECRA Salmonella

Visual Immunoassay (TECRA International Pty. Ltd.; Frenchs Forest, Australia)

following modified version of Enrichment Protocol 7 and method for Performing the

Immunoassay of the manufacturer's directions. The enrichment Protocol 7 was modified

in the following way: 20mL of the original sample in 180mL of BPW for 22 h at 35C

followed by an enrichment of ImL sample of the BPW in 9mL of Tetrathionate (TT)

Broth for 24h at 35C and O.lmL of sample to 9.9mL Rappaport-Vassiliadis (RV) Broth

for 24h at 42C, followed by a ImL of each broth culture to be inoculated into 9mL of M

broth incubated at 35C for 8 h. Samples were then heat shocked before the TECRA

Salmonella Visual Immunoassay was used.

Alicyclobacillus spp. detection

Alicyclobacillus spp. were enumerated using the heat shock method described by

Parish and Goodrich (2005). The samples were then streaked to duplicate plates of PCA

and duplicate plates of Alibroth agar plates (Parish and Goodrich 2005). One of each

plate type was incubated at 25C and one of each plate type was incubated at 500C.









Plates were checked at 24 and 48 h for growth. If growth was only present on the

Alibroth agar plates at 500C, it was assumed to be Alicyclobacillus spp.

16S DNA and 28S rRNA PCR Identification

The 16S DNA PCR identification of bacteria and 28S rRNA (D2 expansion

segment) rRNA region PCR identification of yeast was done by Accugenix, Inc.

(Newark, DE). The company extracts DNA from a pure isolate, the "16S rRNA gene is

amplified, sequenced,... the resultant extension products are separated" and it "is then

matched in order of increasing genetic distance to relevant sequences in a database"

(Accugenix 2005). The yeast identification was done by "sequencing of the D2

expansion segment of the large subunit rRNA gene" and comparing to a database of yeast

sequences (Accugenix 2005).

The three E. coli strains 36, 87, 113; one presumptive S. aureus, Yeast (OSA)

113B, and gram-positive rod 36C, gram-positive 36D, gram-negative rod 113C from the

Mesophilic colonies were sent for identification.

Statistical Analysis

Using the psychroduric, mesophilic, yeast and mold, and aciduric counts the

number of CFU/cm2 and the CFU/total gasket were calculated and transformed to Logio

to reduce the effect of outliers on the data (if there were 0 CFU/cm2 or 0 CFU/total

gasket, it was changed to a count of 1 CFU/cm2 or 1 CFU/total gasket prior to the log

transformation). Analysis of Variance (ANOVA) was used to determine if samples were

significantly different (p>0.05) from each other depending on gasket type, product type,

wash type, or any combination thereof. Minitab release 14 (Minitab, Inc.; State

College, PA) was used for statistical analysis.









Part II: Biofilm Development and Removal

Liquid Sample Preparation

Whole, homogenenized UHT milk (Parmalat Finanziaria S.p.A, Italy) was

inoculated with representative bacteria and yeast, obtained from E. coli-positive tanker

samples. The milk was also inoculated with a fluorescent-tagged E. coli that was created

using TransformAidTM Bacterial Transformation Kit (Fermentas; Burlington, Ontario,

Canada) and the E. coli from sample 36.

Standard growth curves

Standard growth curves were created from 24-hour cultures of one yeast and four

bacteria. Dilutions of each sample were done from 100 to 10-9. These dilutions 10-4 to

10-9 were pour plated out with SPC agar onto sterile Petri plates (Fisher Scientific,

International; Pittsburg, PA). Plates were incubated at 370C for 24 h and were then

counted. The remainder of the culture was used to create a 1/2, 1/4, 1/8, and 1/16

dilution of sample in nutrient broth. Diluted samples and a pure culture sample where

viewed until the spectrophotometer at 600nm. The data from the above plate count was

used to determine how many bacteria or yeast was in the original sample. The amount of

microorganisms that would be in 1/2, 1/4, 1/8, and 1/16 dilution were calculated. The

spectrophotometer measurements were plotted against the number of bacteria or yeast.

The procedure was repeated three times for each sample and a final standard curve was

created for each bacteria or yeast (Appendix C). The standard curves were used to help

add the approximate quantities of each bacteria or yeast to achieve the following formula

outlined in Table 3-1. The formula was created by assuming that the initial day plate

count for a sample of milk was 300 colony forming units (CFU) per mL. Therefore, a 5

L sample of milk would









Table 3-1. Types and number of CFU of microorganisms found in target inoculated milk
sample.
Sample ID Media Type Type (Oxidase/Catalase) CFU/5L
113B OSA Yeast (-+) 150,000
36D Mesophile Pos Cocci (+/+) 300,000
36C Mesophile Pos Rods (+/+) 1,000,000
113C Mesophile Neg Rods (+/+) 25,000
36 E. coli (E*Colite)Fluorescent 25,000
1,500,000


have 1,500,000 CFU. The representative colonies selected from psychroduric,

mesophilic, thermoduric, YM, and aciduric plates of the three samples containing of E.

coli (36, 87, and 113) were compared. The characterization of colonies in sample 87 was

very different from 36 or 113. Samples 36 and 113 were compared and three

characterizations of bacteria and one type of yeast were found to be similar between the

two samples. A sample from one of these sets was selected for the culture (see Table

3.1). The quantities of microorganisms inoculated into model were chosen by the

following criteria: 1) gram-positive spore-forming bacteria are very likely to be found in

pasteurized milk because of their ability to survive pasteurization and grow at refrigerator

temperatures (Cousin 1982, Washam and others 1977), Frank 2001); therefore, they make

up the majority of the microorganisms in the sample, 2) since some gram-positive cocci

such as (Micrococcus and Enterococcus) can survive pasteurization and grow at

refrigerator temperatures (Richter and Vedamuthu 2001), the gram-positive cocci were

placed in the sample in the next largest quality, 3) yeast growth in pasteurized milk is not

typical so a low level of yeast was added (Richter and Vedamuthu 2001), 4) since there

can be no more that 10 coliforms per mL according to USDA's Grade A Pasteurized

Milk Ordinance(FDA/CFSAN National Conference 2003), only 5 CFU ofE. coli 36 per









mL (or total 25000 CFU of the E. coli) and 5 CFU of the gram-negative rod 113C per mL

(or total 25000 CFU of the gram-negative rod) were added to the sample.

Model of Liquid Transportation Tanker Manway

Figures 3-2 and 3-3 illustrate the set for the model manway assembly. Sterile

gloves were worn throughout this procedure. Pieces of the manway assembly that could

not be sterilized were sanitized with chlorinated water. A stainless steel manway

assembly has been obtained from liquid transportation tanker Manufacturer "B". The

manway lid was set on a plastic washbasin (Picture A). The rubber gasket (that would be

A B C








D E F








Figure 3-2. Manway lid set up picture set 1. A.) Manway lid on the wash basin, B.)
Gasket type A on the manway lid, C.) Olson vent on the manway lid cover,
D.) spraying system placed in the center, E.) sterile weights placed on the
spraying system, F.) plastic hose connecting spraying system to pump.

defined as gasket type B according to the definition provided in Part I of this experiment)

was placed on the lip of the manway (Picture B). An Olson vent was placed on the

manway lid cover (Picture C). A spraying apparatus was set in the center of the

washbasin (Picture D) and weighted down with 4 sterile bottle weights (Picture E). A

half-inch diameter plastic hose was connected to the spraying apparatus (Picture F). The









dust cover and manway lid were placed down over the top of the manway (Picture H) and

were held in place by five clamps in the manway (Picture I). The other end of the half-

inch hose was connected to a submersible fountain pump (Peaktop Technologies) that

was used to circulate milk inoculum from a container in a 4C water bath (Picture G) to a

spraying apparatus in the center of the washbasin. The milk would then drain through a

1-inch plastic tube attached to the drain of the washbasin with PVC piping as seen in

Picture K. The spraying device was programmed using a timer/controller (Fisher;

Pittsburg, PA) (Picture J) to spray the gasket for 5 s every 15 m for three days to mimic

sloshing in a moving tanker traveling across the United States. The model tanker was

G H I








J K








Figure 3-3. Manway lid set up picture set 2. G.) Incubator were milk inoculum is stored,
H.) Dust cover closed over the manway lid, I.) manway lid was clamped
down, J.) timer/controller, K.) complete model set up.

placed in an environmentally controlled chamber that was set up to reproduce the

temperature fluctuations on a typical central Florida July day in where the high

temperature is approximately 90F (32.2C) and the low is 70F (21.1C) (Southeast

Regional Climate Center 2005). Calibrated HOBO H08-002-02 data loggers (Onset









Computer Corporation; Bourne, MA) monitored the temperature fluctuations of the room,

water bath, and milk inoculum.

Gasket Treatment

Figure 3-4 illustrates how the gasket was washed and prepared. Sterile gloves were

worn through out this procedure. After exposure, the gasket (Picture L) was removed and

cut into four pieces using sterile razor blades (Picture M). One piece was left untreated

(control), and the remaining pieces were subjected to three different cleaning regimens:

1) 15s detergent wash (Picture N) followed by a water rinse; 2) 15 s detergent wash, a 10s

scrub and 4 m 50 s sanitizer soak, and water rinse; and 3) 15 s detergent wash, a 10 s

scrub and 4 m 50 s sanitizer soak, a water rinse, and a hot water treatment (Picture O) of

the gasket at 1600F (71.1C) for 15 m and 185F (85C) for 20 m.

A food-grade chlorine detergent and a quaternary ammonium based sanitizer

commonly used at tank wash stations were used to wash gaskets. The highest

concentration of both products that could be used according to label directions was

dispensed into bottles of sterile deionized water. Bottles with water, detergent, and

sanitizer treatments were held in a 90F (32.2C) water bath until they were needed.

Detergent and sanitizer treatments were placed into premarked tubs with premarked

brushes typical of those used at tank washes. Hot water treatments were created by

placing sterile deionized water into two beakers on separate hot plates and heating them

to the appropriate temperatures.

After each gasket piece was washed, two pieces (1/2 inch in length) were removed

from the center of the gasket piece (Pictures P&Q). One piece has a thin piece of the top

(Picture R), the inside surface, and the outside surface removed with a sterile razor blade

and placed on a slide to be studied by fluorescence microscopy. The second piece has a









thin piece of the top, and the inside surface removed with a sterile razor blade to be used

for the scanning electron microscopy. The remaining piece was swabbed using a

SpongesicleTM with 10mL of nutrient buffer (Picture S).


Figure 3-4. Manway lid set up picture set 3. L.) Model setup after 3 days, M.) Gasket
being cut into 4 pieces with a sterile scalpel blade, N.) Gasket washed in
detergent, 0.) Gasket receiving a heat treatment, P& Q.) Gasket being
prepared for microscopy, R.) Surface section being removed for microscopy,
S.) Gasket being swabbed.

Microbial Analysis of Gasket

The swab was used to determine the total plate count, the total amount ofE. coli

using the USDA's Bacteriological Analytical Manual Online Chapter 4 Part I Subpart G









for solid media method enumeration of injured coliforms (2002), and a count to

determine the number of yeast on aPDA. Pour plates of PCA from 2 mL of sample to a

dilution of 10-7 were done to determine the total plate count and pour plates of aPDA

from 2 mL of sample to a dilution of 10-1 were done to determine the total yeast count. A

random sampling of microorganisms was selected from the total plate count and acidified

potato dextrose agar to determine the approximate composition of the sample. An

E*Colite bag was prepared for each sample using the method described earlier to check

for presence of low levels of E. coli that may not be detected on the Tryptic Soy Agar and

Violet Red Bile Agar (VRBA) with MUG (OXOID, LTD.; Basingstoke, Hampshire,

United Kingdom) plates.

Scanning Electron Microscopy

Gasket samples were fixed with 3% glutaraldehyde with 1500 ppm Ruthenium Red

(RR) in 0.1M cacodylate (CaCo) buffer at pH 7.2. at room temperature. Samples were

then washed 3 times with 0.1M CaCo and then were en bloc stained in 1500 ppm RR

with 0.1M CaCo buffer at pH 7.2 at 40C overnight (Luft 1971). Samples were then rinsed

twice with 0.1M CaCo buffer for 5 m and dehydrated using ten-step ethanol dilution

series for 10 m each dilution. A critical point drier (Ladd Research Industries; Williston,

VT) using bone dry CO2 was used to completely dry the samples. Each sample was

mounted, coated for 90 s with gold/palladium (80/20) (Ladd Research Industries;

Williston, VT), and viewed and photographed using a Hitachi S-530 scanning electron

microscope at 80x, 600x, and 4000x magnification (Chumkhunthod and others 1998).

Fluorescence Microscopy

Slides containing the inside, top and outside surface of the gaskets inter lip for each

of the four samples were viewed an Olympus BZ61 Microscope (Olympus Corp.;









Melville, NY) using the 10x objective and a filter for green fluorescent protein. Pictures

were taken of each piece of the gasket and were printed using a Hewlett-Packard

Business Inkjet 1200 (Hewlett-Packard Development Company, L.P.; Palo Alto, CA).

Statistical Analysis

The experiment was replicated six times to determine if any of the cleaning

treatments provided a significant difference the reduction in the number of the coliforms

and mesophilic microorganisms in comparison to each other using ANOVA (Minitab

release 14). Prior to analysis by ANOVA the number of coliforms CFU/total gasket and

the number of mesophiles CFU/total gasket were calculated and transformed to Loglo to

reduce the effect of outliers on the data (if there were 0 CFU/total gasket it was changed

to a count of 1 CFU/total gasket prior to the log transformation). Log reductions of the

three treatments were calculated. The log reductions were used to determine if there was

a significant difference in the ANOVA (p>0.05).














CHAPTER 4
RESULTS

Part I: Sample Identification and Characterization

Psychroduric, Mesophilic, Thermoduric, Yeast and Mold, and Aciduric
Microorganism Enumeration and Characterization

A total of 126 tankers were sampled in this study. After characterization, nine

samples for each of the eight types of gasket combinations were randomly selected. The

number of microorganisms per square cm2 and the number of microorganisms on the

total gasket were calculated. Appendix A summarizes data from the 72 observations, and

includes the raw data, the number of microorganisms per centimeter squared, and the

number of microorganisms per total gasket. ANOVA was run comparing each variable

type with the number of microorganisms per centimeter squared, and the number of

microorganisms on the total gasket for each test. Then Fisher's Protected Least

Significant Difference Test at a 95% confidence interval was run on groups that had

significant differences. The results can be found in Table 4-1 to Table 4-14. A

percentage of different characterizations for each media type was obtained for each

sample and an overall percentage was calculated for each gasket type. The top two

components of the microflora found for each media type is summarized in Table 4-15.












Table 4-1. Product's effect on aciduric, yeast and mold, psychroduric, and mesophile counts per cm2 of the gasket.
Tanker Aciduric/cm Yeast and Mold/cm Psychroduric/cm Mesophile/cm



Juice 6.78.4Ea 2.54.9Ea 8.4 x101+1.6Ea 8.5 x1018.2Ea

Dairy 1.3x10 118Ea 3.47.1Ea 1.6 x0l21.3 x10OEa 1.6 x10 2.2 x10Eb

a, b, c- each letter indicates a grouping that is not statistically significantly different




Table 4-2. Product's effect on aciduric, yeast and mold, psychroduric, and mesophile counts per total gasket.
Tanker Aciduric Total Yeast and Mold Total Psychroduric Total Mesophile Total



Juice 3.4 x103l4.1 xl03a 1.0 x1031.9 x03Ea 2.8 x102+8.4 xl02Ea 4.3 xlO 39.3 xl03Ea

Dairy 6.4 x10 8.9 xlOEa 1.3 x10 2.6 xlOEa 3.8 x103+1.6 xlO4Ea 8.0 x10 l 8.0 x10 Ea

a, b, c- each letter indicates a grouping that is not statistically significantly different












Table 4-3. Gasket's effect on aciduric, yeast and mold, psychroduric, and mesophile counts per cm2 of gasket.
Gasket Shape Aciduric/cm Yeast and Mold/cm Psychroduric/cm Mesophile/cm




A 5.79.1Ea 4.43.2Ea 2.91.3 xlO1Ea 7.69.6Ea

B 1.4 x101+1.8 x101Eb 1.4+7.7Eb 1.1+2.6Ea 1.7 x102+2.1 x101Eb


a, b, c- each letter indicates a grouping that is not statistically significantly different




Table 4-4. Gasket's effect on aciduric, yeast and mold, psychroduric, and mesophile counts per total gasket.
Gasket Shape Aciduric Total Yeast and Mold Total Psychroduric Total Mesophile Total




A 3.2 x103+5.1 x103Ea 7.6 x1021.8 x103Ea 3.5 x1031.6 x104Ea 4.2 x1035.4 x103Ea

B 6.6 x103+8.3 x103Eb 1.5 x1032.7 x103Ea 5.1 x1021.2 x103Ea 8.1 x1039.9 x103Eb

a, b, c- each letter indicates a grouping that is not statistically significantly different












Table 4-5. Wash temperature's effect on aciduric, yeast and mold, psychroduric, and mesophile counts per cm2 of gasket.
Tanker Aciduric/cm2 Yeast and Mold/cm Psychroduric/cm Mesophile/cm



Hot 7.1+7.1Ea 3.34.3Ea 1.1+2.8Ea 1.2 x101+1.5 xlO1Ea

Cold 1.3 x101+1.9 x101E 2.47.4Ea 2.91.3 xlO1Ea 1.2 x101+1.9 xlO1Ea

a, b, c- each letter indicates a grouping that is not statistically significantly different


Table 4-6. Wash temperature's effect on aciduric, yeast and mold, psychroduric, and mesophile counts per total gasket.
Tanker Aciduric Total Yeast and Mold Total Psychroduric Total Mesophile Total



Hot 3.6 x103+3.6 x103E 1.0 x103+1.8 x103Ea 3.1 x103+2.9 xlO1Ea 6.1 x103+7.2 x103Ea

Cold 6.2 x103+9.2 x103E 1.3 x1032.7 x103E 9.9 x1021.6 x103Ea 6.2 x1039.0 x103Ea

a, b, c- each letter indicates a grouping that is not statistically significantly different












Table 4-7. Product and Gasket's effects on aciduric, yeast and mold, psychroduric, and mesophile counts per cm2 of the gasket.
Tanker Aciduric/cm2 Yeast and Mold/cm2 Psychroduric/cm Mesophile/cm


Juice, A 4.56.2Ea 1.33.3Ea 0.140.47Ea 7.47.4Ea

Juice, B 8.99.9Ea 3.75.9Ea 0.842.2Ea 9.69.6Ea

Dairy, A 6.81.1 xlOE"a 1.43.3Ea 5.61.8 xlO0Ea 7.71.1 x0ElOa

Dairy, B 1.9 x10 12.2 x10Eb 5.19.3Ea 1.42.9Ea 2.5 x10 12.7 x10Eb

a, b, c- each letter indicates a grouping that is not statistically significantly different


Table 4-8. Product and Gasket's effects on aciduric, yeast and mold, psychroduric, and mesophile counts per total gasket.
Tanker Aciduric Total Yeast and Mold Total Psychroduric Total Mesophile Total


Juice, A 2.5 x103l3.5 x103Ea 7.4 x102+1.8 xO13Ea 1.7 x1025.8 xO12Ea 4.1 x103l5.0 x10Ea
Juice, B 4.2 x103+4.6 xl03Ea 1.3 x1033.2 xl03Ea 3.9 x1021l.0 xl03Ea 4.5 x103l3.5 xl03Ea

Dairy, A 3.8 x10 36.4 xl03Ea 7.9 x102+1.8 xO13Ea 6.9 x103l2.2 x104Ea 4.3 x103l5.9 x103Ea

Dairy, B 9.1 x101.0 xO14Eb 1.8 x10 2.0 xlO1Ea 6.4 x10l21.4 xlOEa 1.2 x104+1.3 x104Eb

a, b, c- each letter indicates a grouping that is not statistically significantly different












Table 4-9. Product and Wash Temperature's effects on aciduric, yeast and mold, psychroduric, and mesophile counts per cm2 of the
gasket.
Tanker Aciduric/cm2 Yeast and Mold/cm2 Psychroduric/cm2 Mesophile/cm2


Juice, Hot 7.07.1Ea 2.84.5Ea 0.450.81Ea 1.0 x10 8.9Eab

Juice, Cold 6.49.8Ea 2.25.3Ea 0.53+2.2Ea 6.97.2Eb

Dairy, Hot 7.17.2Ea 2.04.2Ea 1.8+3.8Ea 1.5 x1011.9 xlO1Eab

Dairy, Cold 1.9 x101+2.4 x102Eb 4.59.1Ea 5.21.8 xlOlEa 1.8 x10 12.5 xlO Ea

a, b, c- each letter indicates a grouping that is not statistically significantly different


Table 4-10. Product and Wash Temperature's effects on aciduric, yeast and mold, psychroduric.


Aciduric Total


Yeast and Mold Total


Psychroduric Total


and mesophile counts per total gasket.
Mesophile Total


Juice, Hot 3.6 x103+3.8 x03Ea 1.2 x1032.1 xl03Ea 3.2 x102+6.4 xl02Ea 5.2 x1034.6 xl3Eab

Juice, Cold 3.1 x10 4.6 xlO0Ea 7.8 x10l 1.8 xl03Ea 2.5 xO12l.0 xlO0Ea 3.5 x10 5.7 x10 E

Dairy, Hot 3.6 x1033.6 x104E 7.9 x1021.5 x103E 1.7 x1033.9 x103Ea 7.1 x103+9.2 x103Eab

Dairy, Cold 9.3 x10 1.1 x104Eb 1.8 x103 3.4 xlOEa 5.9 x10 2.2 xlO4Ea 8.9 x103+1.2 x104Ea

a, b, c- each letter indicates a grouping that is not statistically significantly different


Tanker












Table 4-11. Gasket and Wash Temperature's effects on aciduric, yeast and mold, psychroduric, and mesophile counts per cm2 of the
gasket.
Tanker Aciduric/cm2 Yeast and Mold/cm2 Psychroduric/cm2 Mesophile/cm2

A, Hot 5.5+7.1Ea 1.6+3.4Ea'b 1.23.2Ea 7.99.5Ea

A, Cold 5.8+11Ea 1.23.2Ea 4.51.8 xlO0Ea 7.31.0 xlOEa
B, Hot 8.66.8Ea 3.25.1Ea'b 1.02.4Ea 1.7 x10 1.8 xlO1Ea
B, Cold 2.0 x10 2.3 xlO Eb 5.6+9.7Eb 1.22.8Ea 1.8 x10 12.4 xlO Ea

a, b, c- each letter indicates a grouping that is not statistically significantly different


00
Table 4-12. Gasket and Wash Temperature's effects on aciduric, yeast and mold, psychroduric, and mesophile counts per total gasket.


Tanker


Aciduric Total


Yeast and Mold Total


Psychroduric Total


Mesophile Total


A, Hot 3.1 x10 +4.0 xl0Ea 9.2 x102+1.9 xlO E 1.5 x10l 3.9 xlO E 4.4 x10lO5.3 xlOEa

A, Cold 3.2 x10 6.1 xl03 Ea 6.1 x102+1.8 x10 Ea 5.6 x10O 2.2 x10lEa 4.1 x10 5.6 x10 Ea

B, Hot 4.1 x10 33.2 xl03 Ea 1.9 x10 +1.8 xl03Ea 4.7 x10 1.1 xl03Ea 7.9 x10 8.6 xlO0Ea

B, Cold 9.2 x103l1.1 x104Eb 1.1 x1033.3 xl03Ea 5.6 x102 1.3 xl03Ea 8.3 x103l1.1 xl04Ea

a, b, c- each letter indicates a grouping that is not statistically significantly different












Table 4-13. Product, Gasket, and Wash Temperature's effects on aciduric, yeast and mold, psychroduric, and mesophile counts per
cm2 of the gasket.
Tanker Aciduric/cm2 Yeast and Mold/cm2 Psychroduric/cm2 Mesophile/cm2


Juice, A, Hot 6.38.4Ea 2.34.5Ea'b 0.280.66Ea 9.1+9.8Ea


Juice, A, Cold 2.82.0Ea 0.31+0.38Ea 0.00.OEa 5.88.1Ea


Juice, B, Hot 7.86.0Ea 3.34.8Ea'b 0.610.95Ea 1.1 x10 8.4Ea


Juice, B, Cold 10.1+1.3 xlO1Ea 4.07.2Ea'b 1.1+3.1Ea 7.96.5Ea


Dairy, A, Hot 4.76.2Ea 0.941.7Ea 2.24.3Ea 6.79.5Ea


Dairy, A, Cold 8.81.5 xlOE a 1.94.4Ea'b 9.1+2.5Ea 8.81.2 xlO1Ea


Dairy, B, Hot 9.57.7Ea 3.15.7Ea'b 1.43.4Ea 2.2 x10O 2.4 xlO10 E


Dairy, B, Cold 2.9 x10 12.7 x10Eb 7.1+1.2 xlO Eb 1.32.6Ea 2.7 x10 13. 1 x10Eb


a, b, c- each letter indicates a grouping that is not statistically significantly different












Table 4-14. Product, Gasket, and Wash Temperature's effects on aciduric, yeast and mold, psychroduric, and mesophile counts per
total gasket.
Tanker Aciduric Total Yeast and Mold Total Psychroduric Total Mesophile Total


Juice, A, Hot 3.5 x10lO3.5 xl03Ea 1.3 x10lO2.5 xl03Ea 3.4 x102+8.1 xl02 Ebc 5.1 x1035.3 xl03Ea


Juice, A, Cold 1.6 x10 +8.4 x10Ea 1.7 x1022.1 xlO E 0.0O0.0Ebc 3.2 x10 6.6 x104E


Juice, B, Hot 3.6 x10 3.6 x10Ea 1.2 x10lO1.7 xlOEa 5.0 x10 4.5 xlO Ebc 5.3 x103+1.1 x104Ea


Juice, B, Cold 4.7 x10 31.3 xlO4Ea 1.4 x10 32.5 xlOEa 2.9 x102+1.5 x103b,c 3.7 x103+1.5 x104Ea


Dairy, A, Hot 3.5 x10 34.7 xl03Ea 5.3 x1029.6 xlO2Ea 2.7 x10 35.3 xl03Eac 3.7 x10 35.5 x103Ea


Dairy, A, Cold 8.4 x10 31.1 xl03Ea 1.0 x10 32.5 xlOEa 1.1 x104l3.1 xl04Ea 4.9 x103 4.5 xl10Ea


Dairy, B, Hot 3.6 x10 2.8 xl0 Ea 1.1 x10 l 2.0 xl03Ea 6.5 x102+1.6 xl0 Ea' 1.1 x104l3.9 x10 Eab


Dairy, B, Cold 1.3 x104 6.1 xl03Eb 2.5 x103l4.1 x03 Ea 6.2 x102l1.2 x103Ea, 1.3 x1043.1 xl03E


a, b, c- each letter indicates a grouping that is not statistically significantly different












Table 4-15. The top two bacterial characterizations on different gasket and media types.
Gasket Type Yeast and Mold Aciduric Mesophile Thermoduric Psychroduric
Juice, A, Hot Yeast (100%) Yeast (32.6%) GPC ox- cat+ (56%) GPR ox- cat- (100%) GNR ox+ cat+ (50%)
GPC ox- cat+ (14%) Yeast (25%) GNR ox- cat- (45%)
Juice, A, Cold Yeast (98.5%) Yeast (42%) GPC ox- cat+ (53%) GPR ox- cat+ (25%) Yeast (100%)
Mold (1.5%) GPC ox- cat+ (34%) GPR ox- cat- spores/no Mold (13%)
spores (8.4/8.8%)
Juice, B, Hot Yeast (85%) Yeast (51.2%) GPC ox- cat+ (36%) GPR ox- cat+ (33%) GNR ox+ cat+ (49%)
Mold (15%) GPC ox- cat+ (14%) Yeast (19.3%) GNC ox+ cat- (11%) Yeast (22%)
Juice, B, Cold Yeast (89.3%) GPC ox- cat+ (36%) GPC ox- cat+ (39%) GPR ox- cat-, spore Yeast (47%)
(100%)
Mold (10.7%) Yeast (31.5%) Yeast (26%) GNR ox+ cat+ (19%)
Milk, A, Hot Yeast (89.9%) Yeast (36%) GPC ox- cat+ (39%) GPR ox+ cat+ (100%) GNR ox+ cat+ (83%)
Mold (10.1%) GPC ox- cat+ (31.5%) Yeast (21%)
Milk, A, Cold Yeast (70.5%) Yeast (32.4%) GPC ox- cat+ (41%) Mold (33%) GNR ox+ cat+ (70%)
Mold (18.5%) GNR o+ c- (14%) Yeast (21%) GPR ox- cat+ (19%) Yeast (22%)
GPR ox+cat-, spores
(13%)
GPC ox- cat+ (12%)
Milk, B, Hot Yeast (87.9%) Yeast (39.2%) GPC ox- cat+ (46%) GPR ox- cat+ (13%) GNR ox+ cat+ (79%)
Mold (5.5%) GPC ox- cat+ (20%) Yeast (11%) Mold (13%) Yeast (9.1%)
Milk, B, Cold Yeast (86.5%) GPC ox- cat+ (44%) GPC ox- cat+ (48%) GPR ox- cat+ (60%) GNR ox+ cat+ (50%)
Mold (13.5%) Yeast (13%) Yeast (21%) Mold (7.1%) Yeast (22%)
GPC ox+ cat+ (14%)









Coliform, Fecal Coliform, and E. coli Detection

Chi-squared tests were run on the frequency of coliform detection in E*Colite bags

of the 72 gaskets were selected from above on all types of gaskets. Coliforms were found

in 30 of these samples; however, it was determined that coliforms were found no more

frequently in any particular type of gasket over any other. Fecal coliforms were detected

in E*Colite samples 36, 74, 81, 87, 113, and 121. E. coli was detected in E*Colite

samples 36, 87, and 113. Samples 36, 87, andl 13 were confirmed E. coli identified by

16S DNA PCR identification. Coliforms were only detected on the PetrifilmTM of

samples 17, 45, 49,103, 107, 113, 115, 116, 117, and 118. NoE. coli or fecal coliforms

were ever detected on PetrifilTM.

Table 4-16. Number of coliform, fecal coliform, and E. coli positive gaskets determined
by PetrifilmTM and E*Colite.
Gasket Type % Coliform/Fec. Coli./E. coli % Coliform/Fec. Coli./E. coli
on PetrifilmTM in E*Colite
Juice, A, Hot 0/0/0 36.3/0/0
Juice, A, Cold 0/0/0 35.7/7.1/0
Juice, B, Hot 14.3/0/0 57.1/0/0
Juice, B, Cold 0/0/0 33.3/0/0
Dairy, A, Hot 16.7/0/0 41.7/0/0
Dairy, A, Cold 10/0/0 40.0/0/0
Dairy, B, Hot 16.7/0/0 54.2/ 8.3 / 4.2
Dairy, B, Cold 6.3 / 0 / 0 56.3 / 18.8 / 12.5

Streptococcus and Staphylococcus Detection

No Streptococcus was found in any of the samples. However, all samples collected

from yellow fluorescent E*Colite bags had presumptive Staphylococcus. Chi-squared

tests were run on the frequency of Staphylococcus detection in E*Colite bags of the 36

selected from above on all types of gaskets. It appeared that Staphylococcus was found

no more frequently in any particular type of gasket over another. 19 samples were

presumed to have Staphylococcus and 9 of those samples were presumed to be