<%BANNER%>

Survival of Inoculated shigella spp. on Tomato and Orange Surfaces


PAGE 1

SURVIVAL OF INOCULATED Shigella spp. ON TOMATO AND ORANGE SURFACES By DIRK M. SAMPATH 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 Dirk M. Sampath

PAGE 3

To all the members of my family, and friends who have encouraged me over the years to strive for better things. And to my nephe w and niece, Quinn and Patrina respectively, I encourage you the same. I hope that as you both move out into the world, you can look back at the time we were in Tr inidad & Tobago with fond memories.

PAGE 4

iv ACKNOWLEDGMENTS Firstly, I would like to thank my major professor and committee chair, Dr. Keith R. Schneider, for the opportunity he presented to me in pursuing this master’s degree. I also thank the two other members of my graduate committee, Dr. Rene M. Goodrich and Dr. Mark A. Ritenour, for all the invaluable help and advice given to me in various aspects of my project. I thank all my family who supported me during this time, especially my younger sister, Martine, my younger br others, Link, Brett and Verne, and last but not least, my mother, Cynthia, who, thankfully, is my lone living parental witness for this, as my father, Dr. Martin Sampath, flew away on Tuesday, December 5, 1995. I also thank my elder sister, Sylvanna, who exposed me to a lo t of progressive things when we were much younger (sometime in the last century), and w hom I still consider to be much more civilized than I may ever attain. I thank all my labmates past and present, for the advice and help they all freely gave to me whenever I needed it during my tenure here at UF. My work here at the University of Florida was supported by the USDA-CREES IFAFS Grant number 00-52102-9637.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii FIGURE......................................................................................................................... ....ix ABSTRACT....................................................................................................................... ..x CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................6 Shigella .........................................................................................................................9 Fresh Tomato Production and Handling in Florida....................................................12 Fresh Orange Production and Handling in Florida.....................................................13 3 MATERIALS AND METHODS...............................................................................16 Initial Preparation of Rifampicin Stock Solution.......................................................16 Growth Study..............................................................................................................16 Preparations of Rifampicin-resistant Shigella .....................................................16 Adaptation of organisms to rifampic in in tryptic soy broth (TSB)..............16 Transfer of organisms onto Rif80 tryp tic soy agar slants and storage.........17 Procedure for the Growth Studies.......................................................................18 Recovery Study...........................................................................................................19 Preparations Prior to the Recovery Study...........................................................19 Acquisition of produce.................................................................................19 Placement of produce prior to inoculation...................................................19 Preparation of PBS rinsate...........................................................................19 Procedure for the Recovery Study.......................................................................20 Preparation of organi sm source inoculum....................................................20 Determination of CFU/ml in the source inoculum.......................................21 Inoculation onto produce surface.................................................................21 Recovery of initial inoc ulum in BPW rinsate..............................................21

PAGE 6

vi Survival Study............................................................................................................22 Statistical Analyses.....................................................................................................22 4 RESULTS...................................................................................................................23 Growth Study of S. sonnei and S. flexneri ..................................................................23 Recovery and Survival Studies for Tomatoes and Oranges.......................................24 Recovery Study for Tomatoes....................................................................................26 Recovery Study for Oranges.......................................................................................27 Survival Study............................................................................................................27 Tomato Survival Study........................................................................................28 Shigella sonnei .............................................................................................28 Shigella flexneri ............................................................................................31 Orange Survival Study........................................................................................34 Shigella sonnei .............................................................................................34 Shigella flexneri ............................................................................................36 5 DISCUSSION.............................................................................................................39 Tomato Survival Study........................................................................................40 Shigella sonnei .............................................................................................40 Shigella flexneri ............................................................................................42 Orange Survival Study........................................................................................43 Shigella sonnei .............................................................................................44 Shigella flexneri ............................................................................................45 6 CONCLUSION...........................................................................................................49 LIST OF REFERENCES...................................................................................................51 BIOGRAPHICAL SKETCH.............................................................................................59

PAGE 7

vii LIST OF TABLES Table page 4-1. Organism, temperature/relative humidity combinations, and time periods for the tomato and orange survival studies. .........................................................................25 4-2. Tomato recovery of Shigella spp. from 0.1% buffer pe ptone water (BPW) after inoculum dried (1.0 to 2.5 hr). .................................................................................26 4-3. Orange recovery of Shigella spp. from 0.1% buffer pe ptone water (BPW) after inoculum dried (1.0 to 2.5 hr). .................................................................................27 4-4. Comparison of S. sonnei survival population decline on Florida 47 tomatoes at 13oC/60%RH and 13oC/90%RH conditions. ...........................................................29 4-5. Comparison of S. sonnei survival population decline on Florida 47 tomatoes at 30oC/60%RH and 30oC/90%RH conditions ............................................................29 4-6. Cross-comparison of S. sonnei survival population decline on Florida 47 tomatoes at 13oC/60%RH and 30oC/60%RH conditions. ........................................30 4-7. Cross-comparison of S. sonnei population decline on Florida 47 tomatoes at 13oC/90%RH and 30oC/90%RH conditions ............................................................31 4-8. Comparison of S. flexneri survival population de cline on tomatoes at 13oC/60%RH and 13oC/90%RH conditions ............................................................32 4-9. Comparison of S. flexneri survival population declin e on Florida 47 tomatoes at 30oC/60%RH and 30oC/90%RH conditions .........................................................32 4-10. Cross-comparison of S. flexneri survival population decline on Florida 47 tomatoes at 13oC/60%RH and 30oC/60%RH conditions .........................................33 4-11. Cross-comparison of S. flexneri survival population decline on Florida 47 tomatoes at 13oC/90%RH and 30oC/90%RH conditions .........................................34 4-12. Comparison of S. sonnei survival population decline on oranges at 13oC/60%RH and 13oC/90%RH conditions ...................................................................................34 4-13. Comparison of S. sonnei survival population decline on oranges at 30oC/60%RH and 30oC/90%RH conditions ...................................................................................35

PAGE 8

viii 4-14. Cross comparison of S. sonnei survival population decline on oranges at 13oC/60%RH and 30oC/60%RH conditions ............................................................36 4-15. Cross-comparison of S. sonnei survival population decline on oranges at 13oC/90%RH and 30oC/90%RH conditions ............................................................36 4-16. Comparison of S. flexneri survival population decline on oranges at 13oC/60%RH and 13oC/90%RH conditions ............................................................37 4-17. Comparison of S. flexneri survival population decline on oranges at 30oC/60%RH and 30oC/90%RH conditions ............................................................37 4-18. Cross-comparison of S. flexneri survival population decline on oranges at 13oC/60%RH and 30oC/60%RH conditions ............................................................38 4-19. Cross-comparison of S. flexneri survival population decline on oranges at 13oC/90%RH and 30oC/90%RH conditions ............................................................38

PAGE 9

ix FIGURE Figure page 4-1. Growth curves of rifampicin-adapted S. sonnei and S. flexneri grown in TSBRif80 at the stationary phase. ...................................................................................24

PAGE 10

x Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science. SURVIVAL OF INOCULATED Shigella spp. ON TOMATO AND ORANGE SURFACES By Dirk M. Sampath December 2005 Chair: Keith R. Schneider. Major Department: Food Science and Human Nutrition This study examined the survival of Shigella sonnei and Shigella flexneri on the surfaces of Florida-grown tomatoes and ora nges at temperature and relative humidity conditions mimicking those that could be encountered during the growing season, at harvest, and during postharvest storage. Florida 47 tomato es and navel and Valencia oranges were used in this study. Inoculat ed samples were placed in one of four temperature/relative humidity environments: 13oC/60%RH, 13oC/90%RH, 30oC/60%RH or 30oC/90%RH. In some instances, the temperature and re lative humidity combination had a marked effect on the survival of inoculated Shigella spp. on produce surfaces. There were no significant differences in S. sonnei and S. flexneri population reduction on tomato surfaces held at 13oC at 60% or 90%RH. On the tomato surfaces at 30oC, S. sonnei populations declined slower at 60%RH compared to 90%RH. In contrast, there was no significant difference in population decline when S. flexneri was observed under the same

PAGE 11

xi conditions. At 60%RH, there was no signi ficant difference in population decline of S. sonnei or S. flexneri at 13 and 30oC. At 90%RH, the S. sonnei population declined significantly slower at 13oC than at 30oC. No significant difference in population decline was observed with S. flexneri under this condition. On orange surfaces stored at 13oC, significantly more S. sonnei survived at 60%RH (indicated by a 1.47 log10 CFU population decline) than at 90%RH (4.71 log10 CFU decline). A similar observation was made for S. flexneri under this same condition, with 1.45 log10 CFU decline per fruit on Day 7 at 60%RH, and 4.06 log10 CFU decline per fruit, at Day 7. At 30oC, the population decline was si gnificantly lower at 60%RH for S. sonnei (4.24 log10 CFU) and S. flexneri (5.66 log10 CFU) than at 90%RH (5.32 and 6.41 log10 CFU decline per fruit, for S. sonnei and S. flexneri respectively). At 60%RH, significantly more S. sonnei survived at 13oC than at 30oC after 7 days. The same pattern of behavior was observed for S. flexneri At 90%RH, survival of both S. sonnei and S. flexneri was greater at 13oC than at 30oC. Both S. sonnei and S. flexneri approached the least limit of detection on the orange su rfaces at Day 7, under storage conditions of 13oC/90%RH. The results indicate that higher temperature (30oC) and higher relative humidity (90%) favored the population reduction of S. sonnei on the tomato surfaces. Storage conditions of 13oC and 85-90%RH for 48 hr inhibits the survival of S. sonnei and S. flexneri on tomatoes, thereby decreasi ng the risk of shigellosis foodborne disease. Both organisms survived better on orange surfaces at 13oC and 60%RH. However, oranges stored at the present conditi ons and times under which the tomatoes are stored could probably represent a greater risk vehicle of foodborne shigellosis, compared to tomatoes.

PAGE 12

1 CHAPTER 1 INTRODUCTION Recent outbreaks of some foodborne illnesses have been traced back to the consumption of contaminated fresh fruits and vegetables (Beuchat 2002). As healthconscious individuals seek a heal thier lifestyle, one of the choices made in this regard is to consume more plant-oriented foodstuffs. In the U.S. and other countries, successful campaigning to this end has inevitably led to the increased consumption of more of the traditional, and a greater variety of, plan t-based foods, especially fresh fruits and vegetables (Beuchat and Ryu 1997). Along with the increased consumption of these foods, there has been an increase in th e reported incidences of foodborne illnesses (Beuchat 2002). Since 1997, as a result of a report to President Clinton of the U.S. entitled, “Food Safety from Farm-to-Table: A National Food Safety Initiative,” increased cooperation between the U.S. Department of Health and Human Services (USHHS) and the U.S. Department of Agriculture (USDA) has led to combined efforts to protect the health of the American consumers, with more emphasis placed on monitoring domestic and imported produce for safety (Food Safe ty and Inspection Service [FSIS] 2000). Increased monitoring, in combination with th e increasing consumption trend, may lead to a higher occurrence of reported produce-a ssociated, foodborne illness outbreaks among the consumer population. When such outbreaks do occur traceback operations, conducted by the health agencies responsible fo r such activities, have to be implemented in order to determine the origin and destina tions of the produce implicated. Some of the produce items for which traceback operations have been conducted as a result of reported

PAGE 13

2 incidences of foodborne illnesses from 1990 to 2000 have been tomatoes, cantaloupes, scallions, leaf lettuce, raspbe rries, basil and basil-contai ning products, various berries, green onions and parsley (Gu zewich and Salsbury 2001). Since the U.S. depends on seasonal imports for an adequate, continuous supply of perishable produce, not only are monitoring operations of such imports importa nt, but so also are accurate traceback information about sources of product, if foodborne illness outbreaks do occur. For example, inaccurate traceback of a foodborne illness incident in June 1996 by the Texas Department of Health initially identified strawberries from California as the product contaminated with Cyclospora when in fact it was Guatemalan raspberries (Calvin 2003). Bacteria can attach to th e surfaces of many produce item s. They may be found in pores, dents and other surface aberrations wher e they are more protected from adverse environmental conditions (Seo and Frank 1999) They can also be found adhering to cuts, cracks, and perforations in the produ ce surfaces (Liao and Cooke 2001; Takeuchi and Frank 2000; Burnett et al. 2000), wher e they may be further protected. Some produce are washed in dump tanks using recirc ulated water. However, proper sanitation procedures must be followed, since such wa ter treatments may present a potential health risk as microbes are washed off fruit surf aces and accumulate in the water used for cleaning. The contaminated water then can contaminate all subsequent fruit passing through it. In addition to contaminating the surfaces, the contamin ated water may also penetrate into the produce by wh atever accessible openings are present on their surfaces. Bartz (1982) showed that Erwinia carotovora (subspecies carotovora ) invasion into tomatoes could be prevented if their subm ersion times and depths were less than two

PAGE 14

3 minutes and 17 centimeters, respectively. In a more recent study, Duffy et al. (2005) reported that parsley submerged for 15 mi nutes in a peptone-i noculated, 3-strain Salmonella suspension had higher populations of l oosely-attached, strongly-attached, and internalized Salmonella cells than parsley submerged for 3 min in the same suspension. Foodborne bacteria have been shown to have a greater potential to survive on freshcut produce than on those with in tact peels or rinds, because the bacteria had access to the nutrients exuded on the cut produc e surfaces (Francis et al. 1 999). Golden et al. (1993) found that Salmonella grew well on the interior tissues of watermelon, honeydew and cantaloupe melons at 23oC, such as would be found at a roadside stand. Even in cold storage, human pathogens such as Escherichia coli O157:H7 have been shown to survive. After 34 hr storage at 5oC, E. coli O157:H7 levels of 3.1 a nd 3.0 CFU/gm, respectively, on cut watermelons and cantaloupes were unc hanged (Del Rosario and Beuchat 1995). Temperature and relative humidity are the mo st important environmental characteristics which affect the populations of the natural microflora and microbial pathogens found on fresh fruit and vegetable produce (Brackett 1987) These environmental factors may thus enhance the bacterial survival on the produce and lead to increased risk of foodborne illness due to contaminated produce. Bacterial contamination of fresh produce can occur at any time during production, harvesting and postharvest handling. For exam ple, contamination can occur in the field by use of contaminated irrigation water, by harvesters and postharvest handlers, in unclean processing facilities, in food-s ervice establishments, and by consumers themselves (Beuchat 1996). Under laborator y conditions, simulated commercial washing and sanitizing procedures typically results in a one to two log reduction in pathogen

PAGE 15

4 numbers, and this reduction is inadequate fo r microbiological safety (Sapers 2001). In some commercial applications the actual reduction in pa thogen numbers may be lower thereby compromising produce safety (Sapers 2001). Large-scale growers, wanting to keep the produce as “fresh” as possible so as to maximize sales, and minimize losses due to drop in quality and spoilage, try to speed up delivery from the farm to the final outlets. As a result, some washing and rinsing procedur es at these facilities may not be adequate to remove pathogens on the produce surfaces to safe levels, and may even increase the microbial load on them if the wash water is contaminated. A study performed by Lang et al. (2004) found that significantly more pa thogens were recovered from the surface tomatoes dip-inoculated in cell suspensions which represented washing produce in water in dump tanks, compared to those recovered when the tomatoes were spotor sprayinoculated. More pathogens were recovered 1 hr after application and were more viable than those recovered after 24 hr after (Lang et al. 2004) Since some foodborne pathogens have low infective doses, the potential danger to consumers exposed to inadequately washed produce is apparent. In preventing contamination of produce, effective procedures must be implemented at the preharvest and postharvest aspects of production (Cliver 1997). In 1999, of the 1,040 cases of Shigella infections causing illnesses that were reported at FoodNet Surveillance sites, 61% were due to S. sonnei and 29% from S. flexneri (Centers for Disease Control and Preven tion [CDC] 1999). Of the 15% of cases requiring hospitalization from bacterial infectio ns in general, 12% of those were due to Shigella (CDC 1999). In the U.S., S. sonnei accounted for greater than 75% of shigellosis cases (Gupta et al. 2004). In its 2000 FoodNet Surveillance Annual Report,

PAGE 16

5 the CDC stated that from 1996 to 2000, the inci dence of salmonellosis declined while the overall incidence of shigellosis increased, with noticeably large increases in California and Minnesota (CDC 2000). This study examined the survival of each of the bacterial pathogens, S. sonnei (ATCC 9290) and S. flexneri (LJH 607) on the surfaces of tomatoes (Florida 47 variety), and Navel and Valencia oranges. Both toma toes and oranges are grown commercially in Florida, with cash receipts in 2003 being $516 and $983 million respectively (National Agricultural Statistics Serv ice [NASS] 2005a). The temp erature and relative humidity combinations used in the current studies were set to mimic commercial field and postharvest conditions experienced by tomatoes and oranges in Flor ida. The objectives in the study were to estab lish the growth characterist ics of rifampicin–adapted S. sonnei and S. flexneri in tryptic soy broth (TSB), to dete rmine the recovery after drying of both Shigella inoculated around the blossom ends of to matoes and oranges, and to observe the survival of Shigella -inoculated tomatoes and oranges at temperature/relative humidity conditions of 13oC/60%RH, 13oC/90%RH, 30oC/60%RH and 30oC/90%RH.

PAGE 17

6 CHAPTER 2 LITERATURE REVIEW In the U.S. it is estimated that over 76 million illnesses and 5,000 deaths are due to diseases perpetuated by foodborne organisms (Mead et al. 1999). Foodborne diseases and conditions deemed nationally reportable are: botulism, brucellosis, cholera, enterohemorrhagic E. coli (EHEC), post-diarrheal hemolytic uremic syndrome, listeriosis, salmonellosis, shigellosis, typhoi d fever, hepatitis A, cryptosporidiosis, cyclosporiasis, and trichinellosis (Morbidi ty and Mortality Weekly Report (MMWR) 2005). Predictions are that from 2000 to 2020, th e market share for citrus and apples will increase by 27% each, grapes 24%, tomatoes 19%, lettuce 24%, other fruit 26%, and other vegetables 22%, with the least incr ease of 8% shown by fried potatoes/chip consumption (Lin et al. 2003). This projected increase in overall produce consumption, combined with the addition of foreign sour ces to satisfy year-round domestic demand has increased the likelihood of reportable foodborne illnesses. Many food poisoning incidences worldwide ha ve been attributed to contaminated fresh fruits and vegetables. In 1989-90, Salmonella -contaminated cantaloupe from Mexico and Central America were responsib le for causing illness in an estimated 25,000 people in the U.S. (CDC 1991; Lund and Snowdon 2000). In 1991, this organism was also responsible for 33 cases of infection vi a watermelon consumed at a picnic and school fair in Michigan (Blostein 1993). Frozen Br azilian mamey [apple] imported to the U.S. in 1998-99 caused 13 cases of Salmonella Typhi food poisoning (Lund and Snowdon 2000). Cantaloupe containing E. coli O157:H7 caused illness in nine people at an

PAGE 18

7 Oregon restaurant in 1993 (del Rosario and Beuchat 1995). At a Minnesota hotel restaurant serving fruit salad, co leslaw and tossed salad, Noroviru s was responsible for 233 reported cases of foodborne illness in 1982. Norovirus caused 206 cases of food poisoning in the United Kingdom in 1987 (Lund and Snowdon 2000). In 1992, Calcivirus in salad eaten at a catered even t in Ontario, Canada caused illness in 27 people (CFSAN-USDA 2001). In England and Wales, almost 6% of intestinal foodborne disease outbreaks occurring in the 1992-2000 period were linked to fruits, vegetables and salads (O’Brien et al. 2000). Campylobacter jejuni in salad eaten in British Columbia, Canada, sickened 330 patrons in a university cafeteria in 1984 (Alle n 1985). This same organism in 1996 caused sickness in 14 people w ho consumed contaminated lettuce in an Oklahoma restaurant (CDC 1998). Listeria spp. have been frequently isolated from vegetables (de Simn et al. 1992; Heisick et al. 1989; Kaneko et al. 1999; Ryu et al. 1992). In 1979, L. monocytogenes contamination of tomatoes lettuce, and celery likely caused 25 cases of illness in a Boston hospital, including two fatalities (Ho et al. 1986; Schelch et al. 1983). In Per u, cabbage compromised with Vibrio cholerae caused 71 deaths in 1991 (Swerdlow et al 1992). As these incidents show, the affected product, the product point of origin, and th e final venue where the contaminated product was passed on to the consumer have all be en variable over the years. Juices derived from minimally processed fr uits and vegetables, wherever these may be grown, are also causes for concern with respect to foodborne illnesses. Surface pathogens on the intact produce can later be passed into the juices when the products are processed, and may survive if th e juices are not pasteurized or improperly handled. Fresh juices are recognized as causes of foodborne illnesses (Parish 1997). Even the acidic

PAGE 19

8 nature of some juices may not be adequate to kill some pathogens. A study by Zaika (2002) examined the survival of S. flexneri in the presence of 0.4 M, pH 4 citric and malic acids at 4oC, and found that the pathogen survived for more than 70 days. At 4oC, both S. sonnei and S. flexneri were found to survive in tomato juice for 14 days, and for 8 days at 22oC (Bagamboula et al. 2002). Apart fro m contamination of juices or juicederived beverages directly by contaminated produce or processing equipment, equipment normally deemed clean by current sanitati on standards can also be a source of contamination. Keller et al. (2004) stated that in experi ment involving cider production performed in a small commercial facility, ae robic plate counts (APC) on incoming apples used in the cider production should correlate linearly with that of the cider APC, indicating that the counts came from organisms on the apples used. However, they found no such linear relationship between the apples used and cider made indicating possible cross-contamination from the cider-processing equipment (Keller et al. 2004). Workers with unhygienic practices can contaminate freshl y-squeezed juices. Bacteria from such workers may be transferred directly into the squeezed juices or indirectly from the hands onto the fruit or vegetable surfaces. Once on th e surface, bacteria can be transferred into the juices during the squeezing operation. In Patiala City, India, Sandeep et al. (2004) studied the quality of street -vended, freshly-squeezed carro t and Kinnow-Mandarin juices and found the presence of coagulase-positive Staphylococcus aureus in 30% of the carrot juice and 12% of the orange juice samples examined. The total fecal coliform counts (TFCC) and total viable counts (TVC) were about 5 and 6 log10 units respectively, for both juice types (Sandeep et al. 2004).

PAGE 20

9 Zeal for a healthier lifestyle has le d some consumers to abstain from conventionally-grown produce in favor of organically-grown food items. However, organic produce may not necessarily be sa fer to eat. In a study performed on a Minnesota farm, Mukherjee et al. (2004) observe d that 8% of fruits and vegetables had no significant difference in co liform counts between the or ganicallyand conventionallygrown tomatoes, leafy greens, lettuce, green peppers, cabbage, cucumbers, broccoli, summer squash, zucchini, bok choi apples, onions and strawbe rries. The researchers did find a six-fold significant increase in E. coli numbers in organically-grown harvested fruits and vegetables, compared to those conventionally-grown. Even produce garnered from smaller-scale, home kitchen gardens face similar problems of bacteria-contaminated water, as those commercially grown on larger scales. In the Campinas Municipality of So Paulo, Brazil, 19.9% of vegetables examin ed from kitchen gardens using irrigation water had fecal coliform counts greater than 200 CFU/g (Simes et al. 2001). Shigella Shigella species are Gram-negative, facu ltatively anaerobic, nonsporulating, nonmotile rods in the family Enterobacteriaceae (Andrews and Jacobson 1998). They are the causal organisms of shigellosis, also known as bacillary dysentery (Lampel et al. 1999). The four species of Shigella which cause disease, in or der of decreasing severity and serologically typed by their somatic O antigen (Downes and Ito 2001) are S. dysenteriae (Serogroup A; Shiga toxin production), S. flexneri (Serogroup B ), S. boydii (Serogroup C), and S. sonnei (Serogroup D) (Lampel et al. 1999). Nearly all the virulence genes lie in a 37-kilobase pair (kbp) cluster within a 180-220-kpb plasmid (Downes and Ito 2001). By manipul ating host-cell phagocytosis, the Shigella enters the host using its Ipa protein antigens (Downe s and Ito 2001). The Ipa proteins are

PAGE 21

10 transported to and concentrated on the bacterial surface by means of the products expressed from the invasion plasmid gene, ipg and those responsible for Ipa expression on the bacterial surface (Downes and Ito 2001). The infection-type of illness presented by Shigella spp. may be caused by 101-108 colony-forming units (CFU), which may have an incubation period of 4-6 week s, with moderate to severe symptoms in those afflicted lasting days to weeks (Council for Agricultu ral Science and Technology (CAST) 1994). Shigella Contamination in Food Shigella spp. are not associated with any one food; contamination is mainly due to mishandling by food handlers with poor pe rsonal hygiene (Downes and Ito 2001). The foods most often associated with Shigella are potato salads, ch icken, shellfish (Downes and Ito 2001) and raw vegetables (Andrews a nd Jacobson 1998). In an experiment to study the survival of S. flexneri in coleslaw, crab salad, ca rrot salad, cabbage salad and potato salad at 4oC, it was found that the organism survived for 11 days minimum in all five salads, and was not killed by the low pH or inhibited by the normal flora inherently present in the salads (Rafi and Lunsford 2002). Many vegetables have an internal pH of 4.5 or higher, and are able to support bacterial growth, while many fru its (apples, oranges, tomatoes ) have a lower internal pH that inhibits bacterial growth (De Roever 1998). During cider pr oduction, apples have been found to allow the slow growth of E. coli O157:H7 because mold growth increased the pH of the apple flesh (Fisher and Gold en 1998). Carter (1989) noted that during fresh-squeezed orange juice operations, micr oorganisms transferred from the orange peel could propagate in the juice under favorable conditions. Microbial contamination of plant tissue is for the most part associated wi th the surfaces of fruits and vegetables, and sound fruits have sterile inte riors (De Roever 1998). However, it has been demonstrated

PAGE 22

11 that bacteria on the surface of sound fruits ev entually can be internalized (Samish 1963). Organisms can enter into produce such as to matoes through naturally occurring openings and impaired skin surfaces (Bartz and S howalter 1979). Buchanan et al. (1999) found elevated levels of E. coli O157:H7 in the inner core region of whole, intact, warm apples that had been dipped in a cool er temperature suspension of the organism. In a parallel experiment performed during that same st udy, when warm apples were submerged in cold water colored with Red Dye # 40, it was observed that the dye, and by implication the E. coli entered into the apples’ inner core re gions through readily-seen open channels (Buchanan et al. 1999). Insects, birds, and dust can act as vector s for plant and human pathogens, especially after fruits and vegetables have been inju red (Beuchat 1996). In one such example, Shigella was isolated from ordinary houseflie s (Olsten 1998). During harvesting and handling, citrus fruits can be damaged via plugging, bruises, punctu res and splits (Almed et al. 1973). The quality of the fruit a nd vegetable thereby di minishes below that acceptable for safe and palatable consumption, and this quality may decrease faster if temperature and humidity conditions pres ent during harvesting and packing-house storage allow and enhance the survival of pathogens like Shigella Data from experiments performed on pack aged sterile and unsterile vegetables indicate that Shigella survived on vegetables tested after 10 days, reaching a 3-7 log10 unit reduction from the initial inoculum (L ampel et al. 1999). It was noted that S. sonnei grew rapidly on chopped parsley and rema ined viable after 14 days at 4oC (Wu et al. 2000). The survivability of Shigella may be more pronounced at higher temperatures and humidities on fruits and vegetables such as oranges and tomatoes. Studies of Salmonella

PAGE 23

12 survivability on harvested tomatoes have been performed mimicking commercial temperature and relative humidity conditi ons found during the Spring and Fall/Winter harvest season, and th e survivability of Salmonella serotypes observed on various surface matrices which match those in Flor ida packing houses (Allen 2003). Shigella that have been acid-adapted to reduced pH have been shown to grow more effectively in some acidic foods than unadapted Shigella Significantly more acid-adapted S. flexneri than non-adapted ones were recovered from unferme nted or fermented porridges made with corn or corn-cowpea do ugh (Teteh et al. 2004). In 1999, of the estimated 448,000 cases of Shig ella infections reported in the U.S., 89,000 were attributed to foodborne sources and 14 cases resulted in death (Mead et al. 1999). Though causing fewer sicknesses each year (approximately 2,500), illness due to L. monocytogenes results in a much greater rate of mortality (approximately 20%) each year (Mead et al. 1999). FoodNet sites have reported that in 3,784 cases of Shigella infections during 1996-1998, S. sonnei accounted for 72%, S. flexneri 22%, S. boydii 0.9%, and S. dysenteriae 0.6%, and that the ov erall rates of inf ection were highest amongst children 1-9 years old (Shiferaw et al. 2000). In January 2000, a multi-state outbreak of S. sonnei shigellosis gastroenteritis, invol ving 406 persons, was traced back to a commercial brand of bean dip consisti ng of cooked beans, salsa, guacamole, nacho cheese and sour cream (Kimura et al. 2004). Fresh Tomato Production and Handling in Florida In Florida, fresh tomatoes accounted for 40% of U.S. domestic production in 2003 (Economic Research Service [ERS] 2004b). With little change for th e last 40 years, the total tomato production in Florida has been about 40,000 acres, with yields that have increased steadily during those times (ERS 2004c ). This increase can be attributed to

PAGE 24

13 increased utilization of drip irrigation t echniques and the development of new tomato varieties (ERS 2004c). Florida usually comp etes with Mexico for the early Spring and Winter U.S. domestic market. The Mexican s upply peaks in the wint er when Florida is also the prominent producer during that time (ERS 2004a). Tomatoes are usually planted in Florida from mid-July to mid-March, a nd harvesting the crops begin at about midOctober and ends in June, with temperat ure and relative humidities averaging 30oC (86oF) and 60-90% respectively, during that time (National Agricultural Statistics Service [NASS] 2005b). Mature-green to matoes are tomatoes at the point of just changing color but are still a uniform light gr een. Mature-green are hand-har vested and then placed in plastic buckets. The filled buckets are then emptied into pallet bins or the larger gondolas. Trucks then carry filled bins or gondolas to the packinghouse. The tomatoes are kept under shade out of dir ect sunlight in order to allo w the tomatoes to cool. The bins or gondolas are then decanted, allowing th e tomatoes to be dumped or flumed into a dumptank containing heated chlorinated wate r. From the dumptank, the tomatoes are rinsed, dried, sized, graded and waxed befo re being packed in cardboard-type boxes. After commercial packing, tomatoes are routinely palletized and stored at 12oC (Sargent et al. 2002). Mature, green tomatoes can be stored for 2-5 weeks at 13oC and 90-95%RH (Cantwell 2001). The ideal ripening room cond itions used at the commercial level are at 19-21oC and 90-95%RH with 50 l ethylene gas pe r liter of air (Sargent et al. 2002). Fresh Orange Production and Handling in Florida Oranges are usually harvested in Florida from mid-September to June, with Navel oranges harvested until the end of January, and Valencia oranges from the end of January to the end of June (NASS 2005a). Temp erature and relative humidities average 30oC (86oF) and 60-90%RH, respectively, during the or ange harvesting time. For short-term

PAGE 25

14 transport and storage (7 days), oranges are stored at 7-10oC and 85-90%RH without chilling injury and decay-causing fungal growth occurring (Thompson et al. 2002). Florida oranges can also be stored at 1-2oC without chilling injury occurring. In the fields, oranges ready for harvest are mos tly picked by hand. The laborers place the oranges into plastic tubs. When filled, a truck called a “goat,” which has a hydraulic boom, picks up these tubs. The boom then picks up the tubs and dumps them into a container at the rear of the “goat.” The goa t then takes its full load and dumps it into a large tractor trailer. The tractor then takes its loads to packinghouse where, after dumping and trash removal, the oranges are pre-sized, washed, pre-graded, treated with fungicide and wax, heated-air dried, grade d, box-packed and palletized. Unwashed and unwaxed oranges may be removed for severa l reasons at the pre-sizing stage in the packinghouse operation. They may go into st orage for later use. Some packinghouses have de-greening rooms in which unwashed, unwaxed oranges are temporarily stored and subjected to ethylene in order to impart a more desirable ye llow or orange color to the fruit. Those oranges destined for the proces sing plant for orange juice are removed prior to fungicide and wax applications. Florida’s production of orange juice in creased in the 1990’s compared to the previous decade. This was due to a direct response to the harsh freeze weather which decreased production in the prior decade, resulting in the replan ting of the affected freeze areas in the northern areas of th e state, increased planting of trees in the southern areas, and reorganizing existing groves to accommoda te a higher orange tree density (Foreign Agricultural Service [FAS] 1997). Few orange s were processed in Florida prior to the 1931-32 season, and at the end of that decade mo st of the output (80%) was still geared

PAGE 26

15 for the fresh market (NASS 2005c). But from 1940 to 1970, there was a steady increase in the percentage of Florida oranges pr ocessed, with most going into the frozen concentrated orange juice (FCOJ), from around 17% to 90% with a steady state being maintained around 90% or higher, from 1970 onwards (NASS 2005c). Not-From-Concentrate (NFC) orange juic e consumption has also increased in recent years, due to it’s perceived healthier benefits over FCOJ or reconstituted orange (FAS 2002). NFC was introduced in the 1950’ s and its production was much smaller compared to that in the 1990’s (Spreen et al. 2001). In the 1999-2000 orange-growing season, 50% of the Valencia oranges went into NFC production, compared to the 40% that went into FCOJ (Spreen et al. 2001).

PAGE 27

16 CHAPTER 3 MATERIALS AND METHODS Initial Preparation of Rifampicin Stock Solution. A stock solution of 10,000 ppm rifampic in was made by dissolving 1.0 g of powdered rifampicin antibiotic (MP Biomed icals; Irvine, CA) in 100 ml HPLC-grade methanol (Fisher Scientific, Fair Lawn, NJ) at room temperature. This solution was filter-sterilized using a Nalgene vacuum filter unit (Nalge Nunc International, Rochester, NY) fitted with a 0.45 m pore-size filter. This stock solution was tightly capped, covered in aluminum foil, and stored at lab temperature. Growth Study Preparations of Rifampicin-resistant Shigella Growth study experiments were performed separately for S. sonnei ATCC 9290 (American Type Culture Collection, Manassa s, VA) from an isolate obtained by Walter Reed Army Medical Center and S. flexneri LJH 607 (obtained from Dr. L.J. Harris, UC Davis). These organisms were adapted to rifampicin prior to conducting the growth studies. In this manner, the recovered rifampicin-adapted Shigella from all subsequent experiments could survive and proliferate ove r any rifampicin-non-adapted organism in tryptic soy agar (TSA) media fortified with rifampicin. The procedures utilized are described below. Adaptation of organisms to rifamp icin in tryptic soy broth (TSB). A cryogenically-frozen bead harboring the S. sonnei or S. flexneri was placed in 10 ml of tryptic soy broth (TSB) (BD-Difco, Spar ks, MD) in a sterile, Py rex test tube. The

PAGE 28

17 prepared tube was placed overnight in a shaker-table incubator at 37oC and 30 rpm. After 17-24 hr of incubation, 30 l of the inoculum was aseptically transferred into 10 ml of freshly-prepared TSB, using a sterile loop (BD, Sparks, MD) and reincubated. A 10-ml TSB tube containing 2.5 ppm ri fampicin (TSB-Rif2.5) obtaine d from the stock solution was inoculated with 30 l from the sec ond incubation. This TSB-Rif2.5 tube was incubated for 17-24 hr. Successive inoculat ions and incubations were performed, each time increasing the rifampic in concentration as follows: 5, 10, 25, 40, 60, 65, 70, 75 and 80 ppm. Once the maximum concentration of 80 ppm rifampicin (Rif80) was reached, successive cultures were maintained at this level. Transfer of organisms onto Rif80 tr yptic soy agar slants and storage. Tryptic soy agar (TSA) (BD-Difco, Spar ks, MD) with 80 ppm rifampicin (TSARif80) was used to make spread plates us ed for maintenance of rifampicin-resistant cultures. Plates were in oculated with 0.1 ml of Shigella once the culture had reached the maximum rifampicin concentration as describe d above. Spread plates were incubated at 37oC. Suitable single colonies were c hosen and streaked for isolation onto S. sonnei/ boydii plating media (Biosynth, Zurich, Switzer land), and incubated overnight. This procedure was repeated two additional times to select for vigorous, representative cultures before final colonies were streaked onto TSA-Rif80 slants ( one colony per slant) in screw-capped, sterile Pyrex tubes. The slant tubes we re incubated, loosely capped, for approximately 8 hr at 37oC, then once growth was confirmed on the slants, capped tightly and transferred into a 4oC cooler for storage and later use. Shigella were also adapted to 200 ppm rifamp icin by the same step-wise adaptation to increasing concentrations of rif-added TSB. The colony isolation procedure was

PAGE 29

18 repeated. The appropriate 200 ppm rif-adap ted colonies were stored on TSA-Rif200 slants at 4oC. Procedure for the Growth Studies For each organism separately, S. sonnei ATCC 9290 and S. flexneri LJH607, the following procedure was performed. The TS A-Rif80 or Rif200 slant containing the required organism was removed from the 4oC cooler. A 30-l sterile Bacto™ loop (BDDifco, Sparks, MD) was used to remove an al iquot of the required rifampicin-adapted Shigella, and inoculate a 10-ml TSB tube with the corresponding rifampicin concentration (either Rif80 or Rif200). This inoculated tube was allowed to incubate overnight at 37oC and 30 rpm. From this tube, two mo re successive transfers into 10-ml TSB tubes and incubations were performed. From the third incubated TSB tube, loop transfers were carried out into three sterile, 125-ml Pyrex flasks, which contained 100 ml of TSB-Rif80. The flask’s spout was covere d with sterile aluminum foil, and all three inoculated flasks were placed in th e shaker-table incubator set at 37oC and 30 rpm. From each flask, starting from the th ird hour of incubation, and each hour thereafter, a 1.0-ml aliquot of culture was seri al-diluted into sterile tubes each containing 9.0 ml of phosphate buffer solution (PBS) (M P Biomedicals, Irvine, CA). From each dilution tube, 1.0 ml was removed, pour plated onto TSA-Rif8 0 pour plates and enumerated after incubation at 37oC for 24 hr. Colony counts from 25 to 250 colony-forming units (CFU) on the pour plates were recorded. In some instances, more or less CF U were also recorded when no other counts were available. These outlier counts were clea rly noted as not to overly bias the data set. The CFU/ml were then calculated from these counts and dilutions, the log10 CFU/ml for the three flasks at each time increments was determined and a mean population at each

PAGE 30

19 time was calculated. From these values, a plot of log10 mean CFU/ml versus incubation time was performed. The lag, exponential, and stationary growth phases were plotted for each organism. Recovery Study Preparations Prior to the Recovery Study Acquisition of produce The tomatoes and oranges used in th is study were commercially produced and obtained directly from the packer. ‘Florida 47’ tomatoes were picked green in the field and packed directly into boxes (field-packed ) that were delivered to the laboratory located in Gainesville, FL. Commercially mature oranges were picked by University personnel under typical harvesting conditions from research groves located at Lake Alfred and packed into commercial 4/5 bushe l bags. No washed or waxed produce were used in this experiment. About 4 to 6 hr after acquisition of produce from the packers or field, the produce items were stored in dr y, ventilated containers in a dark, 4oC/40%RH cooler, prior to inoculation. Placement of produce prior to inoculation. The required number of tomato or orange samples were removed from the cooler and placed blossom end up on pre-autoclaved, cool ed fiberglass trays at lab temperature. To ensure dry surfaces, the samples were air-dried on the trays overnight at room temperature. Preparation of PBS rinsate Sterile Stomacher bags (Seward, Thetford, UK) were each aseptically filled with 100 ml of buffered peptone water (BPW) at pH 7.0. (BD-Difco, Sparks, MD.), and the

PAGE 31

20 tops sealed with sterile bag c lips (Fisher Scientific, Canada). The sealed bags were stored in a 4oC cooler prior to use. Procedure for the Recovery Study For each organism separately, S. sonnei ATCC 9290 and S. flexneri LJH607, the following procedure was performed: From the TSA-Rif200 slant containing the target organism, three successive loop transfers into 10-ml TSB-Rif200 tubes were completed and the tubes were shakerincubated at 37oC and 30 rpm. A sterile, 125-ml Pyrex flask containing 100 ml of TSBRif200 was inoculated with a loop transfer from the final, incubated 10-ml TSB-Rif200 tube. The flask’s spout was loosely covered with sterile aluminum foil, placed in the shaker-table incubator set at 37oC and 30 rpm, then removed fr om the incubator after the organism had reached stationary phase. Th is time had been determined from the previously described growth studies. Preparation of organism source inoculum The stationary phase inc ubate was aseptically rem oved from the 125-ml flask, placed in a sterile, tightly-capped, centrifuge Falcon tube (Fisher Scientific, Pittsburg, PA). The prepared tube was centrifuge d at 4,000 rpm for 10 min in a CentraMP4R centrifuge (Internationa l Equipment Company, Needham Hei ghts, MA). The supernatant was then decanted and discarded. A 10-ml volume of BPW was added to the centrifuge tube and the pellet resuspe nded using a Vortex Genie2 vortex (Scientific Industries, Inc., Bohemia, NY). This procedure was re peated twice followed by a final resuspension in BPW.

PAGE 32

21 Determination of CFU/ml in the source inoculum Triplicate 1.0-ml aliquots of the prepared source inoculum were separately serialdiluted into sterile tubes each containing 9.0 ml of BPW. From chosen dilution tube, 1.0 ml was removed to make the corresponding TSA-Ri f80 pour plate. Pour plates were then incubated at 37oC for 24 hr. For each sample, triplicate replicates were performed and recorded. The CFU/ml were then calculated from these counts and corresponding dilutions, and the average CFU/ml computed for each sample. Inoculation onto produce surface A repeater pipetter (Brinkmann, Westbury, NY) fitted with the 5-100 l sterile tip (Brinkmann, Westbury, NY) and set to deliver 10 l, was aseptically filled with the source inoculum, according to the manufacturer ’s directions. Ten 10 l drops (100 l total) were then aseptically delivered in a circular patte rn onto the surface surrounding, and 1.0-2.0 cm away from, the blossom end of each fruit. Care was taken so as to ensure that the applied drops did not ‘run’ together. The delivered inoculum was then allowed to air-dry for approximately 1-2 hr. Recovery of initial in oculum in BPW rinsate After the inoculum had dried (about 1.5-2.0 hr), the samples of each fruit were aseptically placed into the pre-prepared, BPW rinsate-filled, Stomacher bag and resealed. Each bag contained one fruit for a total of ten samples. The microorganisms were recovered using a ‘rub-shake-rub’ me thod (Burnett et al. 2001) and the initial inoculum level was calculated. A 1.0-ml aliquot was removed from the stomacher bag and serially diluted into st erile tubes each containing 9.0 ml of BPW. For each dilution tube, 1.0 ml was removed to make the corres ponding TSA-Rif80 pour pl ate. Triplicate pour plates were made for each dilution tube. Plates were incubated and counted as in

PAGE 33

22 previous procedures. The CFU/ml were then calculated from these counts and corresponding dilutions, and the average CFU/100 ml computed for each sample. Survival Study Inoculated produce were placed in either a Caron 6030 (Marietta, OH) or Barnstead International Lab-line E-22560-16D (Dubuque, IA ) environmental humidity chamber set at one of the following temperature/relative humidity combinations: 13oC/60%RH; 13oC/90%RH; 30oC/60%RH; or 30oC/90%RH. From time 0 (a fter drying but prior to placement within the chamber) and other time increments within the chamber, an appropriate number of samples were remove d, each fruit transferred into a 100 ml, PBS filled Stomacher bag, then inoculum was recovered using the ‘rub-shake-rub’ method as previously described for the recovery study. Serial dilutions, pour platings, incubation and enumeration also followed those performed in the recovery study. Statistical Analyses Statistica™ (Statsoft, Tulsa, OK) was used to analyze the results. Tukey’s Honest Significant Difference (HSD) test ( P < 0.05) was utilized to examine data from survival studies for significance. Th e recovery studies for both Shigella organisms were analyzed using the calculated t-statistic to determin e whether the two means of the species’ respective initial so urce and recovered populations we re significantly different ( P < 0.05).

PAGE 34

23 CHAPTER 4 RESULTS Growth Study of S. sonnei and S. flexneri Growth studies were performed for both S. sonnei and S. flexneri in order to determine the time from initial incubation wh en the organisms reached their stationary phases. Typically, stationary phase microorga nism are used in order to consistently obtain a similar inoculum size. The approximate start time of the stationary phase for each organism was noted and the CFU/ml exhibited at that phase was recorded. The Shigella spp were adapted to 80 and 200 ppm rifa mpicin before growth studies were performed. Figure 4-1 shows the stationary pha ses of the rifampicin-adapted Shigella organisms. All the organi sms show counts above 8.0 log10 units at their respective stationary phases, which commenced after approximately eight hours incubation. The log10 CFU/ml for each time period were averaged over the course of the 8 to 12-hr incubation for each of the rifampicin-adapted Shigella organisms, and the means used to determine whether significant differences o ccurred amongst them. The Tukey’s Honest Significant Difference (HSD) test was applied to those means. Shigella sonnei adapted to 200 ppm rifampicin grew to signi ficantly higher titers (8.81 log10 CFU/ml) than S. sonnei adapted to 80 ppm rifampicin (8.54 log10 CFU/ml) ( P < 0.05). However, S. sonnei adapted to 200 ppm rifampicin was significantly lower than S. flexneri adapted to 80 ppm rifampicin (8.98 log10 CFU/ml), and not significantly different from S. flexneri adapted

PAGE 35

24 to 200 ppm rifampicin (8.74 log10 CFU/ml). The lowest recorded mean was noted for S. sonnei adapted to 80 ppm rifampicin, and the highest recorded ti ter was seen with S. flexneri also adapted to 80 ppm rifampicin. Growth Curves of Rifampicin-adapted S. sonnei and S. flexneri grown in TSB-Rif 80.8.00 8.50 9.00 9.50 8.09.010.011.012.0Time (hr) Log10 CFU/ml S. sonnei 200 ppm S. sonnei 80 ppm S. flexneri 200 ppm S. flexneri 80 ppm Figure 4-1. Growth curves of rifampicin-adapted S. sonnei and S. flexneri grown in TSB-Rif80 at the stationary phase. Recovery and Survival Studies for Tomatoes and Oranges These studies were performed to examine the survival of Shigella spp. on the surfaces of unwashed, unwaxed tomatoes and or anges obtained directly from the field. For the recovery study, the area surrounding the blossom end of each produce type were inoculated with the species being tested. Af ter drying for approximately 1-2 hr at room temperature, Shigella were recovered from each fruit, using the ‘rub-shake rub’ method utilizing Stomacher bags filled with 100 ml of BPW. The recovered organisms were then enumerated by serial dilution followed by plating onto TSA-Ri f80 plates incubated at 37oC for 24 hr. The average CFU per tomato and orange was recorded.

PAGE 36

25 The survival study utilized the same inoc ulation procedure used for the recovery study. However, in addition to recovering and enumerating the Shigella from produce samples immediately after the drying peri od, the remaining inoculated samples were placed in environmental humidity chambers. The humidity chambers were pre-set and pre-equilibrated at the temper ature-humidity parameters under which the survival of the Shigella species was to be observed, prior to pl acement of the inoculated produce within. After specific residence times in the humid ity chamber, produce samples were removed and analyzed. Recovery of Shigella into BPW-filled Stomacher bags and enumeration of the Shigella therein were carried out in same manne r as with the recovery studies. The following time periods in Table 4-1 were us ed to perform the survival studies. Table 4-1. Organism, temperature/relative humidity combinations, and time periods for the tomato and orange survival studies. Tomato Survival Study Organism Temperature/Relative Humidity Combinations Time periods for each experiment 13oC/60%RH 13oC/90%RH 30oC/60%RH Shigella sonnei ATCC 9290 30oC/90%RH 13oC/60%RH 13oC/90%RH 30oC/60%RH Shigella flexneri LJH607 30oC/90%RH 0, 1, 2, 4, 6 hr Orange Survival Study Organism Temperature/Relative Humidity Combinations Time periods for each experiment 13oC/60%RH 13oC/90%RH 30oC/60%RH Shigella sonnei ATCC 9290 30oC/90%RH 0, 1, 4, 7 days 13oC/60%RH 13oC/90%RH 30oC/60%RH Shigella flexneri LJH607 30oC/90%RH 0, 1, 2, 7 days

PAGE 37

26 For both recovery and survival studies, the CFU recovered per tomato and orange were log-transformed and plotted against time, and the data interpreted from these. For the purposes of data analysis, plates with no countable colonies were given the value of 1.0, yielding a lower detection lim it of 100 CFU/sample or 2.0 log10 units using the formula log10 [100*(CFU + 1)]. Recovery Study for Tomatoes Both Shigella sonnei and Shigella flexneri organisms were recovered off the tomato surfaces in 0.1% buffered peptone water (BPW ) adjusted to pH 7.0. Table 4-2 shows the log10 CFU/ml of Shigella applied onto the tomato surface, the log10 CFU/ml recovered, and the log10 unit reduction for each species af ter an average drying time of approximately 1.0 to 2.5 hr. Using probability tables, the calculate d t-statistic and the degrees of freedom between the species respective initial source and recovered populations were used to compare with the tabulated t-value at P < 0.05. For each Table 4-2. Tomato recovery of Shigella spp. from 0.1% buffer peptone water (BPW) after inoculum dried (1.0 to 2.5 hr). S. sonnei (Log10 CFU/ml) (st. dev) recovered from Tomatoes. Trial (n=3) log10 Initial log10 Recovered log10 Reduction 8.19a 0.22 5.35b 0.80 2.83 S. flexneri (Log10 CFU/ml) (st. dev) recovered off Tomatoes Trial (n=3) log10 Initial log10 Recovered log10 Reduction 8.69a 0.12 6.91b 0.35 1.78 Values are mean SD of three replications. Differe nt letters (ab) within rows indicate a significant difference in microbial counts ( P < 0.05).

PAGE 38

27 species and their recovered numbers, supers cripts with the same letter above their respective means are not significantly diffe rent. Both species recovered were significantly lower from thei r initial inoculum sizes. S. sonnei had a 2.83 log10 CFU/ml reduction in recovery and S. flexneri 1.78 log10 CFU/ml post drying. Recovery Study for Oranges Using calculated t-statistic and probability tables as performed for the tomato recovery study, it wa s found that the log10 CFU recovered from orange surfaces were significantly lower ( P < 0.05) from their inoculum sour ces for both organisms (Table 34). The drying times for the inoculum on th e oranges at room temperature were also approximately 1.0 to 2.5 hr. Both species rec overed were significantly lower from their initial inoculum sizes post-drying. S. sonnei had a 1.46 log10 CFU/ml reduction in recovery and S. flexneri 1.37 log10 CFU/ml. Table 4-3. Orange recovery of Shigella spp. from 0.1% buffer pe ptone water (BPW) after inoculum dried (1.0 to 2.5 hr). S. sonnei (Log10 CFU/ml) (st. dev) recovered from Oranges. Trial (n=3) log10 Initial log10 Recovered log10 Reduction 8.73a 0.01 7.27b 0.21 1.46 S. flexneri (Log10 CFU/ml) (st. dev) recovered off Oranges Trial (n=3) log10 Initial log10 Recovered log10 Reduction 8.75a 0.13 7.38b 0.15 1.37 Values are mean SD of three replications. Differe nt letters (ab) within rows indicate a significant difference in microbial counts ( P < 0.05). Survival Study For both tomatoes and oranges, survival studies of Shigella spp. on fruit surfaces were conducted, and the studies deemed most appropriate were used to investigate the

PAGE 39

28 temperature-relative humidity effects on the Shigella organisms. The temperaturehumidity conditions used in the survival study were: 13oC/60% relative humidity (RH), 13oC/90%RH, 30oC/60%RH and 30oC/90%RH. Log10 CFU/ml values recovered for each inoculum were transformed as log10 [100*(CFU + 1)] CFU/fruit (t omato or orange). The data was analyzed using ANOVA statistics. The log10 CFU/tomato and orange recovered at each time point were transformed to reflect the log10 CFU decline from the initial, dried inoculum (0 to 6 hr for the tomatoes and 0 to 7 days for the oranges). The 6hr and 7-day end-time limits (for tomatoes a nd oranges, respectively) were selected since these sampling times were the first to have no detectable growth. The ANOVA analyses were constructed to show the relationship of the temperature parameter held constant, while observing the effects of humidity. Using the same data, cross-comparisons, pertaining to the relationship of the humidity parameter held constant while observing the temperature effect, were also computed and described. The ANOVA tables showed significant differences between and among each of the effects ( P < 0.05). Tukey’s HSD method was used to compute probabilities for de termining significances between groups. Tomato Survival Study Shigella sonnei For tomatoes inoculated with S. sonnei and stored at 13oC at 60% and 90%RH (Table 4-4), there no significant difference ( P < 0.05) in total re duction (CFU/tomato) was seen between 60 and 90%RH at any of the time periods examined.

PAGE 40

29 Table 4-4. Comparison of S. sonnei survival population declin e on ‘Florida 47’ tomatoes at 13oC/60%RH and 13oC/90%RH conditions. S. sonnei ; Temperature: 13oC Log10 [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr. Hour 60% RH 90% RH 0 0.00 0.00 1 0.55 ( 0.56)a 0.44 ( 0.24)a 2 0.87 ( 0.53)a 0.76 ( 0.25)a 4 0.79 ( 0.50)a 0.85 ( 0.23)a 6 0.91 ( 0.50)a 0.93 ( 0.23)a Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05). The S. sonnei populations held at 30oC and observed at 60% and 90%RH (Table 45) presented significantly diffe rent population declines at the 2-hr sampling period (0.62 log10 CFU at 60%RH and 4.30 log10 CFU at 90%RH). For the 60%RH, the S. sonnei population decline at 2 hr (0.62 log10 CFU) was not significantly greater than those at the 1-, 4and 6-hr time points. However, for the S. sonnei population at 90%RH, there was a significant change in population decline from the 1-hr (1.17 log10 CFU) to the 2-hr (4.30 log10 CFU) time period. S. sonnei was undetected at 90%RH after 4 and 6 hr. A population decline of 1.59 log10 CFU was indicated after 6 hr at 60%RH. Table 4-5. Comparison of S. sonnei survival population declin e on ‘Florida 47’ tomatoes at 30oC/60%RH and 30oC/90%RH conditions. S. sonnei; Temperature: 30oC Log10 [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr Hour 60% RH 90% RH 0 0.00 0.00 1 1.18 ( 0.50)a 1.17 ( 0.22)a 2 0.62 ( 0.54)a 4.30 ( 0.41)b 4 1.40 ( 0.52)a 4.62 ( 0.21)b 6 1.59 ( 0.54)a 4.62 ( 0.21)b Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05).

PAGE 41

30 For S. sonnei populations observed on tomatoes cross-referenced at 60%RH and compared at13 and 30oC (Table 4-6), both populations indicated no significant population decline, either between or within populations, from the 1 to 6-hr time period. The population decrease at 60%RH and 13oC at 1 hr (0.55 log10 CFU) was not significantly different from the decline at 2 hr (0.87 log10 CFU), 4 hr (0.79 log10 CFU) and 6 hr (0.91 log10 CFU). Also, the populati on decrease at 60%RH and 30oC at 1 hr (1.18 log10 CFU) was not significantly different fr om the decline at 2 hr (0.62 log10 CFU), 4 hr (1.40 log10 CFU) and 6 hr (1.59 log10 CFU). At the end of the 6-hr period, S. sonnei presented a population decline of 0.91 log10 CFU at 13oC and 1.59 log10 CFU at 30oC, which were not significantly different. Table 4-6. Cross-comparison of S. sonnei survival population decline on ‘Florida 47’ tomatoes at 13oC/60%RH and 30oC/60%RH conditions. S. sonnei ; Relative Humidity: 60% Log10 [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr. Hour 13oC 30oC 0 0.00 0.00 1 0.55 ( 0.56)a 1.18 ( 0.50)a 2 0.87 ( 0.53)a 0.62 ( 0.54)a 4 0.79 ( 0.50)a 1.40 ( 0.52)a 6 0.91 ( 0.50)a 1.59 ( 0.54)a Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05). For S. sonnei observed on tomatoes at 90%RH and compared at 13 and 30oC (Table 4-7), both populations d eclined significantly ( P < 0.05) after 1 hr. Th e total decline in population recorded for the 13 and 30oC sample groups were 0.44 and 1.17 log10 CFU, respectively. For the 30oC trials, there was a significant difference in total reduction at 2 hr (4.30 log10 CFU) compared to the 1 hr (1.17 log10 CFU) sampling period. No significant differences were observed for the remaining time periods when compared to

PAGE 42

31 the 2-hr time period. At 13oC, there were no significan t population decline for the remaining time periods compared to the 1-hr time period. When comparing between the 13 an d 30C sample groups at 90%RH, S. sonnei recovered values were significan t different at each time for the entire 6-hr period. At 4 and 6 hr, the 30oC populations were below detectable limits. At 6 hr the population decline recorded at 13oC was 0.93 log10 CFU, which was signifi cantly different from the below-detectable-limit counterpart at 30oC. Table 4-7. Cross-comparison of S. sonnei population decline on ‘Florida 47’ tomatoes at 13oC/90%RH and 30oC/90%RH conditions. S. sonnei ; Relative Humidity: 90% Log10 [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr. Hour 13oC 30oC 0 0.00 0.00 1 0.44 ( 0.24)a 1.17 ( 0.22)b 2 0.76 ( 0.25)a 4.30 ( 0.41)b 4 0.85 ( 0.23)a 4.62 ( 0.21)b* 6 0.93 ( 0.23)a 4.62 ( 0.21)b *4.62 log reduction = below detectab le limits for this data set. Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05). Shigella flexneri The S. flexneri populations were compared at 13oC and observed at 60% and 90%RH (Table 4-8). No significant differe nces in population decline were observed, either between or within both populations when all the time points from 1 to 6 hr were compared ( P < 0.05), due to the greater variability of the obtained results.

PAGE 43

32 Table 4-8. Comparison of S. flexneri survival population de cline on tomatoes at 13oC/60%RH and 13oC/90%RH conditions. S. flexneri; Temperature: 13oC Log10 [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr Hour 60% RH 90% RH 0 0.00 0.00 1 2.25 ( 1.46)a 0.85 ( 1.66)a 2 3.55 ( 1.39)a 1.59 ( 1.00)a 4 3.06 ( 1.51)a 1.30 ( 0.74)a 6 2.49 ( 1.14)a 1.16 ( 0.89)a Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05). Within each population at their respective relative humidities, there was no significant difference in popul ation decline from the 2 to 6-hr period. Between S. sonnei populations, at both relative humidities, th ere were no significant differences in population increases observed at 1 hr (0.63 and 1.00 log10 CFU at 60 and 90%RH, respectively). Additionally, for both relative humidities, no significant differences were observed in population declines at 2 hr (0.12 and 1.05 log10 CFU for 60 and 90%RH, respectively) or at 6 hr (2.28 and 2.78 log10 CFU for 60 and 90%RH, respectively). There was a significant difference noted at 4 hr, with a population increase of 0.50 log10 CFU seen for the 60%RH sample group and a population decline of 1.57 log10 CFU for the 90%RH trial. Table 4-9. Comparison of S. flexneri survival population de cline on ‘Florida 47’ tomatoes at 30oC/60%RH and 30oC/90%RH conditions. S. flexneri; Temperature: 30oC Log10 [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr. Hour 60% RH 90% RH 0 0.00 0.00 1 -0.63 ( 1.53)a -1.00 ( 1.06)a 2 0.12 ( 1.57)a 1.05 ( 1.27)a 4 -0.50 ( 0.87)a 1.57 ( 1.04)b 6 2.28 ( 0.84)a 2.78 ( 1.07)a Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05).

PAGE 44

33 There was a significant difference, between S. flexneri groups, in survival population decline at both temp eratures cross-compared at 60%RH, at the 1-hr period (Table 4-10), with the decline at 13oC being the greater (2.25 log10 CFU) than at 30oC. The population decline at 2 hr at 13oC (3.55 log10 CFU) was significan tly different than that at 30oC (0.12 log10 CFU). At the end of 6 hr, th ere was no significant difference between the populati on declines at 13oC (2.49 log10 CFU) and 30oC (2.28 log10 CFU). Within S. sonnei populations at both temperatures, th ere were no significant differences in population declines during the entire 6-hr period. Table 4-10. Cross-comparison of S. flexneri survival population decline on ‘Florida 47’ tomatoes at 13oC/60%RH and 30oC/60%RH conditions. S. flexneri ; Relative Humidity: 60% Log10 [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr. Hour 13oC 30oC 0 0.00 0.00 1 2.25 ( 1.46)a -0.63 ( 1.53)b 2 3.55 ( 1.39)a 0.12 ( 1.57)b 4 3.06 ( 1.51)a -0.50 ( 0.87)b 6 2.49 ( 1.14)a 2.28 ( 0.84)a Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05). For the S. flexneri populations at 90%RH and compared at 13oC and 30oC (Table 411), there were no significant differences found in population declines between the population groups at each of the 1to 6-hr time periods. Although at 1 hr a population gain of 1.00 log10 CFU was detected for the species at 30oC, this was not significant when compared to the decline at the 2-hr time point (1.05 log10 CFU). No significant differences in population decline at 30oC were observed between the 2-hr, 4-hr and 6-hr time periods.

PAGE 45

34 Table 4-11. Cross-comparison of S. flexneri survival population decline on ‘Florida 47’ tomatoes at 13oC/90%RH and 30oC/90%RH conditions. S. flexneri ; Relative Humidity: 90% Log10 [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr. Hour 13oC 30oC 0 0.00 0.00 1 0.85 ( 1.66)a -1.00 ( 1.06)a 2 1.59 ( 1.00)a 1.05 ( 1.27)a 4 1.30 ( 0.74)a 1.57 ( 1.04)a 6 1.16 ( 0.89)a 2.78 ( 1.07)a Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05). Orange Survival Study Shigella sonnei At 13oC and 60% or 90%RH (Table 4-12), S. sonnei had no significant difference in their population declines on Day 1, though after seven days there was a marked difference, with 1.47 log10 CFU and 4.71 log10 CFU presented at 60%RH and 90%RH respectively. Within the population at 60% RH, there were no significant differences between the decline at Day 2 and Day 7. W ithin the population at 90%RH, there were no significant differences between the decline at Day 4 and Day 7. However, reductions seen at Day 1 were signi ficantly different (2.15 log10 CFU) than Day 4 (4.12 log10 CFU). Table 4-12. Comparison of S. sonnei survival population decline on oranges at 13oC/60%RH and 13oC/90%RH conditions. S. sonnei; Temperature: 13oC Log10 [100*(CFU + 1)] (st. dev) decline / orange from Day 0. Day 60%RH 90%RH 0 0.00 0.00 1 0.95 ( 0.44)a 2.15 ( 0.86)a 4 ND* 4.12 ( 0.93) 7 1.47 ( 0.40)a 4.71 ( 0.57)b *ND = No data recorded for these times. Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05).

PAGE 46

35 At 30oC and 60%RH or 90%RH (Table 4-13), there was a significant difference between population declines at Day 4, with the less decline (4.24 log10 CFU) at 60%RH and populations below detection limits at 90%RH (5.32 log10 CFU). No survivors were detected for either 60 or 90%RH at Day 7. Table 4-13. Comparison of S. sonnei survival population decline on oranges at 30oC/60%RH and 30oC/90%RH conditions. S. sonnei; Temperature: 30oC Log10 [100*(CFU + 1)] (st. dev) decline / orange from Day 0. Day 60%RH 90%RH 0 0.00 0.00 1 ND* 1.56 ( 0.21) 4 4.24 ( 0.12)a 5.32 ( 0.09)b** 7 5.32 ( 0.09)a 5.32 ( 0.09)a *ND = No data recorded for this time. **5.32 log reduction = below detectable limits for this data set. Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05). Shigella sonnei populations on oranges held with 60%RH at 13 or 30oC is shown in Table 4-14. The population decline at Day 1 (0.95 log10 CFU) was not significantly different from the declines at Day 7 (1.47 log10 CFU) for the 13oC population. For the 30oC population, the decline at Day 4 was significantly different than that at Day 7 (5.32 log10 CFU), when no organism growth could be detected. At Day 7, both the 13 and 30C sample group had significan t population declines, 1.47 log10 and 5.32 log10 CFU, respectively. When comp ared to each other after 7 days, the 30C group had a significantly greater decline than the one observed at 13oC.

PAGE 47

36 Table 4-14. Cross comparison of S. sonnei survival population decline on oranges at 13oC/60%RH and 30oC/60%RH conditions. S. sonnei; Relative Humidity: 60% Log10 [100*(CFU + 1)] (st. dev) decline / orange from Day 0. Day 13oC 30oC 0 0.00 0.00 1 0.95 ( 0.44) ND* 4 ND 4.24 ( 0.12) 7 1.47 ( 0.40)a 5.32 ( 0.09)b** *ND = No data recorded for this time. **5.32 log reduction = below detectable limits for this data set. Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05). With 90%RH at 30oC, S. sonnei populations dropped to undetectable levels on oranges by Day 4 and after, indicated by a 5.32 log10 CFU decline. At 13oC at the same %RH, there was a significant population decl ine in survivors at Day 4 (4.12 log10 CFU) compared to that at Day 1 (2.15 log10 CFU), but no significant difference in decline was observed between Day 4 and Day 7 (4.71 log10 CFU). Table 4-15. Cross-comparison of S. sonnei survival population decline on oranges at 13oC/90%RH and 30oC/90%RH conditions. S. sonnei; Relative Humidity: 90% Log10 [100*(CFU + 1)] (st. dev) decline / orange from Day 0. Day 13oC 30oC 0 0.00 0.00 1 2.15 ( 0.86)a 1.56 ( 0.21)a 4 4.12 ( 0.93)a 5.32 ( 0.09)b** 7 4.71 ( 0.57)a 5.32 ( 0.09)a** **5.32 log reduction = below detectable limits for this data set. Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05). Shigella flexneri At 13oC, The S. flexneri population declines were signifi cantly less, at all times, in 60% relative humidity conditions than those at 90%RH (Table 4-16). At 90%RH, there was no significant differences in population declines at Days 1, 2 and 7. At 60%RH, there were no further significant difference s in population decline between Day 1, 2 or 7

PAGE 48

37 readings. At 60%RH, there was no significant difference in population decline when Day 1 was compared to Day 2 (1.06 log10 CFU), but it was significant when compared to Day 7 (1.45 log10 CFU) Table 4-16. Comparison of S. flexneri survival population decline on oranges at 13oC/60%RH and 13oC/90%RH conditions. S. flexneri; Temperature: 13oC Log10 [100*(CFU + 1)] (st. dev) decline / orange from Day 0. Day 60%RH 90%RH 0 0.00 0.00 1 0.37 ( 0.36)a 3.07 ( 1.04)b 2 1.06 ( 1.06)a 3.40 ( 0.87)b 7 1.45 ( 0.38)a 4.06 ( 0.84)b Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05). For the S. flexneri held at 30oC, there were significant differences in population declines at Day 1 (1.80 log10 CFU at 60%RH and 3.82 log10 CFU at 90%RH) (Table 417). Day 2 declines were significantly diffe rent between relative humidities, with 2.60 log10 CFU at 60%RH, and the almost complete extinction represented by a decline of 5.97 log10 CFU at 90%RH. The population d ecline at 60%RH on Day 7 (5.66 log10 CFU) was significantly different to that at Day 2. At 90%RH, no survivors were detected on Day 7. Table 4-17. Comparison of S. flexneri survival population decline on oranges at 30oC/60%RH and 30oC/90%RH conditions. S. flexneri; Temperature: 30oC Log10 [100*(CFU + 1)] (st. dev) decline / orange from Day 0. Day 60%RH 90%RH 0 0.00 0.00 1 1.80 ( 0.38)a 3.82 ( 0.47)b 2 2.60 ( 1.02)a 5.97 ( 0.56)b 7 5.66 ( 0.96)a 6.41 ( 0.08)a* *6.41 log reduction = below detectab le limits for this data set. Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05).

PAGE 49

38 At 60%RH, S. flexneri populations were significantly di fferent at 13C compared to 30oC at each time period sampled (Table 4-18) For Day 1 the population decline was 0.37 log10 CFU at 13oC, while a greater decline of 1.80 log10 CFU was noted at 30oC. For Day 2, the population decline was 1.06 log10 CFU at 13oC, while the greater decline at 30oC was 2.60 log10 CFU. For Day 7, the population decline was 1.45 log10 CFU at 13oC, while the greater decline at 30oC was 5.66 log10 CFU. Table 4-18. Cross-comparison of S. flexneri survival population decline on oranges at 13oC/60%RH and 30oC/60%RH conditions. S. flexneri; Relative Humidity: 60% Log10 [100*(CFU + 1)] (st. dev) decline / orange from Day 0. Day 13oC 30oC 0 0.00 0.00 1 0.37 ( 0.36)a 1.80 ( 0.38)b 2 1.06 ( 1.06)a 2.60 ( 1.02)b 7 1.45 ( 0.38)a 5.66 ( 0.96)b Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05). At 90%RH (Table 4-19), S. flexneri population reductions were not significantly different at Day 1 at either temperature. Significant differe nces were observed at Day 2, with populations more than 2.5 log10 CFU lower at 30oC, compared to at 13oC. No survivors were detected by Day 7 at 30oC (6.41 log10 CFU), compared to a 4.06 log10 CFU reduction seen at 13oC. Table 4-19. Cross-comparison of S. flexneri survival population decline on oranges at 13oC/90%RH and 30oC/90%RH conditions. S. flexneri; Relative Humidity: 90% Log10 [100*(CFU + 1)] (st. dev) decline / orange from Day 0. Day 13oC 30oC 0 0.00 0.00 1 3.07 ( 1.04)a 3.82 ( 0.47)a 2 3.40 ( 0.87)a 5.97 ( 0.56)b 7 4.06 ( 0.84)a 6.41 ( 0.08)b* *6.41 log reduction = below detectab le limits for this data set. Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab) within rows indicate a significant difference in population declines ( P < 0.05).

PAGE 50

39 CHAPTER 5 DISCUSSION Consumers moving towards healthier lifesty les are consuming more fresh fruit and vegetables. Concomitant with this shift in consumption is an increased probability of foodborne illness from such produce. Produce growing in the fields or moving to the final consumer undergoes many levels of preand post-production pr ocesses that present opportunities for produce contamination. In response, producers and packer/shippers have implemented and improved measures to assure consumers of a safe product. The goal is to prevent contamination in the first place; if contaminated, all foodborne bacteria cannot be easily removed without injuring th e product itself. Fruit-handling equipment has to be regularly cleaned and sanitized so that they don’t become a source of contamination. In recent years, Shigella has emerged as an impor tant illness-causing pathogen that has been identified as a produ ce contaminant, especi ally in salad-type preparations. This study obser ved the survival behavior of S. sonnei and S. flexneri on the surfaces of unwashed, unwaxed tomatoes and oranges. The inoculated produce were subjected to environmental conditions of temperature and relative humidity commonly experienced while growing in the field, or dur ing postharvest storag e and transport. Growth curves were estab lished in order to identify where and at what time the stationary phases of both occurred The stationary phase was chosen as a time in the growth curve that a constant inoculum leve l could be harvested. Additionally, organisms in the stationary phase of their growth cy cle are more resilient, thus making for an improved model for the study of bacterial decline. Shigella sonnei (Rif-200) was found

PAGE 51

40 to be not significantly different from S. flexneri (Rif-200) based on Tukey’s Honest Significant Difference (HSD) test ( P < 0.05). Both reached stat ionary phases after 8 hr incubation with populations at around 8.5 log10 CFU/ml (Figure 4-1). Tomato Survival Study Shigella sonnei For S. sonnei -inoculated tomatoes subjected to 13oC and 60% or 90%RH, there were no significant differences ( P < 0.05) between populations, as they declined irrespective of the differing relative humidities and times. The comparison of S. sonnei populations maintained at 30oC and observed at either 60% or 90%RH showed that the organism survived better at the lower humidity range. S. sonnei populations declined quickly and were undetected at 90%RH after 4 hr. For S. sonnei observed on tomatoes and cross-compared at 60%RH at 13 and 30oC, both populations declined equally over the 6-hr period. With the S. sonnei observed on tomatoes at 90%RH and compared at 13 and 30oC, the 13oC population declined slower over the 6-hr period. Thus at 13oC the S. sonnei declined slower (s urvived better) than those stored at 30oC. At a storage temperature of 13oC, relative humidity ceased to be a factor in influencing survival of the S. sonnei on the tomato surfaces. When comparing 90%RH and 60%RH stored at 30C, the survival of S. sonnei only those tomatoes held at evaluated temperature and rela tive humidity significantly de clined. Beattie and Lindow (1999) stated that higher humidity enhan ces bacterial survival by making more free moisture available to the organism. Conversel y, the results of this study indicated that relative humidity and higher storage temper atures appeared to take precedence in influencing metabolic activity in S. sonnei causing the organism to inactivate faster on

PAGE 52

41 the tomato surface. This is similar to the results obtained by Allen (2003) in which it was observed that lower quantities of Salmonella were recovered off tomato surfaces held at 30oC/80%RH, compared to those recovered at 20oC/60%RH or 20oC/90%RH. As indicated in Tables 4-6 and 4-7, the total redu ction seen for S. sonnei at 90%RH for 30oC was significantly greater than at 90%RH for 13C, or with either temperature at 60%RH, on the tomato surfaces. The highe r temperature and relative humidity, combined with the lack of nutrients on the tomato surface, could have caused S. sonnei to use up its energy reserves and so die off qui cker. Thus temperature/relative humidity conditions that would normally be more favorable for growth of S. sonnei in the presence of suitable nutrients, reduces the organism’s ability to survive on the tomato surfaces in the absence of those nutrients. In experiments conducted on parsley, S. sonnei has been observed to behave similarly (Wu et al 2000). Whole parsley incubated at 21oC, inoculated with either an initial 6.19 or 3.23 log10 CFU inoculum, increased less than 1.0 log10 CFU for both inocula after 1 day, which was followed by a subsequent decline in numbers after 2 days. However, the organism proliferated on c hopped parsley incubated at 21oC, from initial inocula of 6.48 and 3.49 log10 CFU/g to 9.20 and 6.32 log10 CFU/g, respectively, in 2 days (Wu et al. 2000). These results support the hypothesis that an increase in moisture and/or nutrients is crucial for bacterial growth on produce. Inability to survive at a higher temperatur e for a longer period was also observed for other organisms inoculated on other produce surfaces. When Listeria monocytogenes was inoculated onto unwashed carrots at titers of 2.4 log10 CFU (incubated at 5oC) and 3.0 log10 CFU (incubated at 15oC), less than 1.0 log10 CFU was recovered after the 18 and 7 day sampling periods, respectively (Beuch at and Brackett 1990). However, the

PAGE 53

42 opposite behavior has also been observed with other organisms. From cantaloupe rind surfaces initially inoculated with 5.2 log10 CFU of a 4-strain cocktail of E. coli O157:H7 and incubated at temperature/rela tive humidity conditions of 5oC/93%RH and 25oC/93%RH, Del Rosario et al. (19 95) recovered less than 1.0 log10 CFU (after 8 days) and 7.1 log10 CFU (after 21 days), respectively. The data for this study suggested that the S. sonnei populations survived equally well at 13C, irrespectiv e of relative humidity. Howeve r in contrast, Wu et al. (2000) found that S. sonnei (inoculated at a high and low c oncentrations) on both whole (6.19 and 3.2 log10 CFU/g) and chopped (6.5 and 3.5 log10 CFU/g) parsley declined during a 14-day storage period at 5oC. For the whole, inoculated parsley, the S. sonnei population significantly declined to a pproximately 4.1 and <1.0 log10 CFU/g, respectively. For the chopped parsley, recovered levels were 3.9 and <1 log10 CFU/g, respectively after 14 days of storage (Wu et al. 2000). Shigella flexneri The S. flexneri populations held at either 13 or 30oC and observed at 60% or 90%RH (Tables 4-8 and 4-9) were compared. Neither temperature changes from 13 to 30oC nor humidity changes from 60 to 90%RH in any of the temperature/relative humidity combinations had any significan t effect on population decline seen on the tomato surfaces at the 6-hr sampling peri od. Some early sampling periods showed significance, though these appear to be an artifact of sample variation. Another study by Islam et al. (1993) observed S. flexneri survival on chopped cucumbers. It was found that at 25oC, the S. flexneri population increased from an in itial inoculum size of about 5.7 log10 CFU to approximately 8.5 log10 CFU/g after 6 hr, then declined to 7.2 log10 CFU/g after 72 hr (Islam et al. 1993). At 5oC, the S. flexneri population remained steady at its

PAGE 54

43 initial inoculum size of about 5.8 log10 CFU/g for 72 hr. It is po ssible that the same trend might occur with S. flexneri at the two temperatures us ed in this study (13 and 30oC), if the experiments were performed with chopped or sliced tomatoes that allow nutrients to be available, thereby making a more favorab le environment for bacterial growth. In this study, both S. sonnei and S. flexneri were studied individually on the tomato surfaces. Joy (2005) observed that a Shigella cocktail population, comprised of both S. sonnei and S. flexneri survived longer on tomato surfaces at fall/winter conditions of 27oC/90%RH, than at 27oC/60%RH. Joy (2005) also conc luded that on wounded tomato surfaces, the Shigella cocktail in combination with Erwinia carotova survived least at temperature/relative humidity conditions of 27oC/60%RH, compared to conditions at 27oC/90%RH. Joy (2005) postulated th at the soft rot caused by the Erwinia which was more notable at 90%RH than 60%RH, may incr eased the availability of nutrients and increased the pH of the tomato, which is a pproximately pH 4.5 (Guo et al. 2001). These factors may have contributed to the survival and proliferation of the Shigella Orange Survival Study The most notable observation from this study was the increased survival of Shigella spp. inoculated on orange surfaces as compared to tomatoes. Survival studies conducted on tomatoes were conducted over a 6 hr time period, whereas studies utilizing oranges were conducted over 7 days. It is important to emphasis this point to clarify comparison between tomatoes and oranges. It is also notable to point out that in all cases, Shigella spp. inoculated onto orange surfaces that we re stored at higher temperatures, were eliminated quicker.

PAGE 55

44 Shigella sonnei At the 13oC/60%RH and 13oC/90%RH, S. sonnei populations had different survival behavior. Shigella sonnei survived better at 60%RH than at 90%RH (Table 4-12). Under the same conditions, S. sonnei behaved differently than in the tomato survival study, as no difference in survival behavior was observed at these parameters. It may be that under the same conditions of temperature and rela tive humidity, the orange surfaces exhibit different characteristics than the to mato surfaces, thereby allowing the S. sonnei to survive better at the lower humidity. Both surface texture and structure of vegetables play important roles in the attachment a nd survival of bacteria on them (Kauze and Joseph 2001). For the survival of S. sonnei on the orange surface, the organism may have been better shielded from the environment on the rougher orange surface as compared to the relatively smooth tomato surface. Table 4-13 showed that at th e storage temperature of 30oC, there was significantly less reduction observed at 60%RH (4.24 log10 CFU) on Day 4, compared to the 5.32 log10 CFU reduction seen at 90%RH. Stine et al. (2005) found that the inactivation of S. sonnei on cantaloupe surfaces was unchanged at 22.7oC when the relative humidity was increased from 47.1% compared to 90.3%. However, it was also found that on bell pepper surfaces, the S. sonnei inactivation at 24.8oC/48.8%RH was significantly lower (1.16 log10 CFU/g) than at 24.8oC/86.1%RH (1.48 log10 CFU/g). Since bell peppers have similar surface topography to th at of tomatoes, it is reasonable to expect survival characteristics to be similar. Shigella sonnei was found in greater numbers on Day 7 at 13oC/60%RH (1.47 log10 CFU reduction) than at 13oC/90%RH (4.71 log10 CFU reduction) (Table 4-12). No survivors were detected at Day 7 at 30oC/60%RH or 30oC/90%RH. The overall survival

PAGE 56

45 at 13oC was greater than what was observed for 30oC, for both relative humidities tested. As with the previously mentioned st udy conducted on pepper surfaces, the lower temperature favored better survival. The analysis of S sonnei at 60%RH noted better survival at 13oC as compared to 30oC (Table 4-14). The higher temp erature could have caused the S. sonnei to deplete its nutrients and stores of energy quicker. At a 90%RH S sonnei survived better at 13oC than at 30oC at 4 days, though differences seen at the Day 7 time period were not significant (Table 4-15). Higher temperature may have caused the S. sonnei to deplete its nutrient s and stores of energy quicker, resulting in below detectable leve ls of the organism after 4 days. The S. sonnei also declined in the 13C group, though not as quickly, reaching a level statistically equivalent to the 30C group after 7 days. Despite the high relative humidity that would intuitively support the survival due to an increased availa bility to water, Shigella was effectively reduced from the orange surface in the 7 day study, though not as quickly as seen in the tomato trials. The results indicated that as the temperature increased from 13 to 30oC, survival of S. sonnei on the orange surfaces decreased at 60 %RH. When the relative humidity was increased from 60%RH to 90%RH, the populatio n still showed better survival at 13oC compared to 30oC. Shigella flexneri Shigella flexneri presented similar behavi or as that observed for S. sonnei on orange surfaces. Table 4-16 indicated that for the 13oC group at Day 7, the 1.45 log10 CFU decline for 60%RH was significantly less than the declin e for 90%RH (4.06 log10 CFU). Table 4-17 in dicated that at 30oC at Day 2 there was a significantly greater

PAGE 57

46 decline for the 90%RH group (5.97 log10 CFU) compared to the group at 60%RH (2.60 log10 CFU). It was noted in Table 4-17 th at, although there we re no significant differences in decline between groups at Day 7, for the 90%RH, no survivors were detected, with the 60%RH group presenting a 5.66 log10 CFU decline in survivors. When relative humidity was increased from 60%RH to 90%RH, as seen with the S. sonnei tests, the S. flexneri exhibited better survival at 13oC, compared to 30oC. Table 418 indicated that at 60%RH on Day 7 for the 13oC group, there was significantly less population decline (1.45 log10 CFU) compared to the 30oC group (5.66 log10 CFU). In Table 4-19, at Day 7, the 13oC group showed significantl y better survival (4.06 log10 CFU decline) than the 30oC group, which was below detectable limits. The ability for S. flexneri to survive better at lower te mperatures has been reported in previous studies. Tetteh and Beuchat (2003) reported that S. flexneri acid-adapted to pH 4.5 were seen to survive, in TS B acidified to pH 3.5, for 2 hr at 48oC, less than 1 day at 30oC, and 6 days at 4oC. The acid-adapted S. flexneri showed a 2.5 log10 CFU/ml decrease when held for 6 days at 4oC, compared to a 6.0 log10 CFU/ml reduction seen with the unadapted, control cells (Tetteh and Beuchat 2003) Zaika (2001) found that S. flexneri also survived better at lower temperat ures as pH increased. In brain-heart infusion broth media at pH 4, the organism was undetected after 5, 15, 23, 85 and 85 days at incubation temperatures of 37, 28, 19, 12 and 4oC, respectively (Zaika 2001). In the same media at pH 3, the survival character istics were 1, 7, 9, 16 and 29 days for those same respective temperatures (Zaila 2001). At pH 2, S. flexneri populations dropped to undetectable levels after 1 to 3 Days at 19oC or lower. However, when held at 37oC or 28oC, S. flexneri populations dropped to undetectable levels after only 2 and 8 hr,

PAGE 58

47 respectively (Zaika 2001). When the media was prepared at pH 5, populations decreased by 0.5-1.0 log10 CFU/ml after 75 days at 4oC, were undetectable after 135 days at 12oC, though populations increased rather than decrea sed at the higher temperatures studied (Zaika 2001). Based on the findings of Zaika (2001) and Tetteh and Beuchat (2003), an alternative hypothesis for the survival of S. flexneri on the orange surface could have been the organism’s ability to adapt to th e acidic orange surface, while the initial inoculum was drying. This adaptation may have allowed it to survive better at 13oC compared to 30oC, once the produce was placed in the humidity chamber. This acid adaptation may have attributed a cross-protection effect to en able the organism to better survive at 60%RH compared to 90%RH on the orange surface. In contrast, as there were no significant differences in survival at both temperatures and relative humidities on the tomato surface, one may infer that no acid adaptation of S. flexneri occurred here. Results seen for S. sonnei on the orange surface mimic those seen for S. flexneri where survival was enhanced at lower temperatures and relative humidities. The results also suggest that acid adap tation and cross-protection di d not occur, though further research would be necessary to rule out this possibility. In contrast with the observations of S. flexneri on the tomato surface, S. sonnei only showed significantly better survival at 30oC/60%RH compared to 30oC/90%RH, and 13oC/90%RH compared to 30oC/90%RH. For both organisms, drier field conditions would allow them to survive better on the orange surfaces, but when the oranges are transferred to st orage conditions of 13oC/90%RH, the organisms are less likely to surv ive. There is less chance that fieldcontaminated S sonnei or S. flexneri could survive long enou gh to reach consumers on

PAGE 59

48 surface-contaminated tomatoes as compared to oranges which can survive for up to 7 days. Typically, short storage time for tomato es at this temperature and relative humidity is for 48-hr duration, while transit storage time may be for 1 or 2 weeks. Enough time would thus elapse for the organisms to be extinguished, and so be of limited threat to end-consumers. In this study where oranges were held at 13oC/90%RH, both organisms approached the least unit of detection after Day 7 (data not shown). Oranges, if no degreening time is needed, may be packed w ithin 24 hours of harvest and then spend at least a few days in the transit/distribution chai n. If degreening time is required for better color, the fruit will be held in Florid a at approximately 85F and 95% RH with approximately 2-5 ppm ethylene for 1 to 3 days (sometime more for grapefruit). This represents only slightly greater Shigella foodborne threat to endconsumers, than is found for tomatoes. Also, one must also consider that tomatoes and oranges are consumed differently. Usually tomatoes are consumed cut into slices or smaller, bite-sized chunks with the skin intact, such as would be found in salad preparations. Oranges are not consumed with the skins, which are discarded prior to eating. In this sense, one may perceive a greater threat of Shigella foodborne illness arising from the consumption of tomatoes compared to oranges.

PAGE 60

49 CHAPTER 6 CONCLUSION The growth curves for each rifampicin-adapted Shigella spp. were established, and their stationary phases noted in order to cons istently obtain an appr opriate inoculum size for each to be used in experiments on the produce surfaces. Recovery of Shigella from the inoculated tomato or orange surfaces, by the ‘rub-shake-rub’ method, was performed to ensure consist inoculum recovery. The tomato survival study indicated that both S. sonnei and S. flexneri behaved similarly when compared at 13oC and observed at 60 and 90%RH. At 13oC, both S. sonnei and S. flexneri showed no significant differen ce in population survivability, irrespective of the relative humidities. However humidity significantly affected survival of S. sonnei when held at 30oC and compared at 60 and 90%RH. Shigella sonnei survived longer on the tomato surfaces at 60 than at 90%RH. Shigella flexneri survived equally well on tomato surfaces at all temperat ure/relative humidity combinations used in the study. The orange survival stu dy indicated that both S. sonnei and S. flexneri survived longer at 60 than at 90%RH for both temperatures of 13 and 30oC. Both organisms also survived longer at 13 than at 30oC when cross-compared at either 60 or 90%RH. This indicated that lower temperature and lower humidity aided in the survival of the Shigella on the orange surfaces. Typical storage conditions used in industr y for tomatoes and oranges in are 10 and 13oC, and 85 and 90%RH, respectively. The results of the survival study indicated that

PAGE 61

50 the storage temperature/relative humidity and times presently used should hinder or eliminate, rather than enha nce, the survivability of Shigella on the tomatoes. Hence, tomato-associated Shigella outbreaks are most likely not due to the present storage conditions used in the industry, but rather due to an outside source of contamination, such as contamination from retail or food se rvice workers with po or personal hygiene, inoculating at the point of consumption after the produce has been removed from storage or the retail consumers who may themselves have unhygienic practices while handling such produce.

PAGE 62

50 LIST OF REFERENCES Allen A.B. 1985. Outbreak of Campylobacterio sis in a Large Educational Institution-British Columbia. Can. Dis. Weekly Rep. 2:28-30. Allen, R.L. 2003. A Recovery Study of Salmonella spp. from the Surfaces of Tomatoes and Packing Line Materials. (Master’s thesis, University of Florida, 2003). http://www.uflib.ufl.edu/etd.html Last accessed on July 18, 2005. Almed, E.M., F.G. Martin and R.C. Fl uck. 1973. Damaging Stresses to Fresh and Irradiated Citrus Fruit. J. Food Sci. 38:230-233. Andrews, W. H. and Jacobson, A. 1998. Shigella Bacteriological Analytical Manual. 8th.ed. http://www.cfsan.fda.gov/~ebam/bam-6.html Last accessed on September 7, 2005. Bartz, J.A. 1982. Infiltration of Tomatoes Immersed at Different Temperatures to Different Depths in Suspensions of Erwinia carotovora subsp. carotovora. Plant Dis. 66: 302-306. Bartz, J.A. and R.K. Showalter. 1979. Postha rvest Water Intake and Decay of Tomatoes. Citrus and Vegetable Magazine 3(44):7 & 28. Bagamboula, C.F., M. Uyttendaele and J. Debevere. 2002. Acid Tolerance of Shigella sonnei and Shigella flexneri J. Appl. Microbiol. 93:479-486. Beuchat, L.R.1996. Pathogenic Microorganisms Associated with Fresh Produce. J. Food Prot. 59(2):204-216. Beuchat, L.R. 2002. Ecological Factors Infl uencing Survival and Growth of Human Pathogens on Raw Fruits and Vegetabl es. Microbes and Infection. 4:413-423. Beuchat, L.R. and J.H. Ryu. 1997. Produce Handling and Processing Practices. Emerg. Infect. Dis. 3:459-463. Beuchat, L.R. and R.E. Brackett. 1990. Survival and Growth of Listeria monocytogenes on Lettuce as Influenced by Shredding, Chlo rine Treatment, Modified Atmosphere Packaging and Temperature. J. Food. Sci. 55(3):755-758, 870. Beattie, G.A. and S.E. Lindow. 1999. Bacteria l Colonization of L eaves: A Spectrum of Strategies. Phytopath ol. 89(5):353-359.

PAGE 63

52 Blostein, J. 1993. An Outbreak of Salmonella javiana Associated with Consumption of Watermelon. J. Environ Health. 56(1):29-31. Brackett, R.E. 1987. “Microbiological Conseque nces of Minimally Processed Fruits and Vegetables.” J. Food Qual. 10:195-206. Buchanan, R.L., S.G. Edelson, R.L. Miller and G.M. Sapers. 1999. Contamination of Intact Apples after Immersion in an Aqueous Environment Containing Escherichia coli O157:H7. J. Food Prot. 62(5):444-450. Burnett, A.B. and L.R. Beuchat. 2001. Comp arison of Sample Preparation Methods for Recovering Salmonella from Raw fruits, Vegetables and Herbs. J. Food Prot. 64(10):1459-1465. Burnett, S.L., J. Chen and L.R. Beuchat. 2000. Attachment of Escherichia coli O157:H7 to the surfaces an Internal Structures of Apples as Detected by Confocal Scanning Laser Microscopy. Appl. and E nviron. Microbiol. 66:4679-4687. Calvin, L. 2003. Produce, Food Safety, and International Trade: Response to U.S. Foodborne Illness Outbreaks Associated with Imported Produce. Chapter 5. In : J. Buzby (Ed.), International Trade and Food Safety: Economic Trade and Case Studies. USDA, Economic Research Service, AER-828. http://www.ers.usda.gov/publications/aer828/ Last accessed on July 8, 2005. Cantwell, M. 2001. Properties and Recommended Conditions for Storage of Fresh Fruits and Vegetables. In : Postharvest Technology. http://postharvest.ucdavis.e du/Produce/Storage/prop_ty.shtml Last accessed on May 11, 2005. Carter, R.D. 1989. Technical Manual: Fresh-Squeezed Florida Orange Juice Production/Packaging/Distribution. Florid a Department of Citrus, Scientific Research Department, University of Florida, Lake Alfred, FL. Centers for Disease Control and Preventi on [CDC]. 1991. Multi-State Outbreak of Salmonella poona Infections United States and Canada, 1991. MMWR. 40:549552. Centers for Disease Control and Pr evention [CDC]. 1998. Outbreak of Campylobacter enteritis Associated with Cross-Contam ination of Food--Oklahoma, 1996. MMWR. 47:129-31. Centers for Disease Control and Preventi on [CDC]. 1999. FoodNet Surveillance Report for 1999 (Final Report). http://www.cdc.gov/foodnet/annual /1999/pdf/FoodNet_1999Annual_Report.pdf Last accessed on July 10, 2005.

PAGE 64

53 Centers for Disease Control and Preventi on [CDC]. 2000. FoodNet Surveillance Report for 2000 (Final Report). http://www.cdc.gov/foodnet/annual/2000/2000final_report.pdf Last accessed on July, 10, 2005. Center for Food Safety and Applied Nutr ition – U.S. Food and Drug Administration [CFSAN-USDA]. 2001. http://www.cfsan.fda.gov/~comm/ift3-4o.html Last accessed on April 27, 2005. Cliver, D.O. 1997. Virus Transmission vi a Food. Food Technologist. 51:71-78. Council for Agricultural Science and T echnology (CAST). 1994. Foodborne Pathogens: Risks and Consequences. Task Force Report no. 122. September 1994. pp. 12. De Roever, C. 1998. Review: “Microbi ological Safety Evaluations and Recommendations on Fresh Produce. ” Food Control. 9:321-347. de Simn, M. C. Tarrag and M.D. Ferrer. 1992. Incidence of Listeria monocytogenes in Fresh Foods in Barcelona (Spain). Int. J. Food Microbiol. 16:153-156. Del Rosario, B.A. and L.R. Beuchat. 1995. Su rvival and Growth of Enterohemorrhagic Escherichia coli O157:H7 in Cantaloupe and Watermelon. J. Food Prot. 58(1):105107. Downes, F.P. and K. Ito (eds.). 2001. Mi crobiological Examination of Foods. Washington: Sheridan Books, Inc. pp. 381-382. Duffy, E.A., L. Cisneros-Zevallos, A. Castillo, S.D. Pillai, S.C. Rick and G.R. Acuff. 2005. Survival of Salmonella Transformed to Express Green Fluorescent Protein on Italian Parsley as Affected by Processing a nd Storage. J. Food Prot. 68(4):687-695. Economic Research Service [ERS]. 2002. Or anges: The Most Consumed Fruit in America. http://www.ers.usda.gov/Briefing/Fru itAndTreeNuts/fruitnutpdf/oranges.pdf Last accessed on July 12, 2005. Economic Research Service [ERS]. 2004a. Co mmodity Highlight: Fresh Tomatoes. U.S. Department of Agriculture. http://www.ers.usda.gov/Briefing/ Vegetables/vegpdf/FrTomatoHigh.pdf Last accessed on July 13, 2005. Economic Research Service [ERS]. 2004b. Im pact of Greenhouse Tomatoes on the Fresh Field Tomato Industry. U.S. De partment of Agriculture. http://www.ers.usda.gov/publ ications/err2/err2g.pdf Last accessed on July 13, 2005.

PAGE 65

54 Economic Research Service [ERS]. 2004c. Tomatoes: Background. U.S. Department of Agriculture. http://www.ers.usda.gov/bri efing/tomatoes/background.htm Last accessed on July 13, 2005. Foreign Agricultural Service [FAS]-USDA .1997. World Horticultural Trade and U.S. Export Opportunities. http://www.fas.usda.gov/ htp2/circular/1997/9708/aug97cov.htm Last accessed on July 14, 2005. Foreign Agricultural Service [FAS]-USDA 2002. Recent Developments in the World Orange Juice Trade and the U.S. Competitive Position. http://www.fas.usda.gov/htp2/circ ular/2000/00-02/ojspecial.htm Last accessed on September 18, 2005. Fisher, T.L. and D.A. Golden. 1998. Fate of Escherichia coli O157:H7 in Ground Apples Used in Cider Production. J. Food Prot. 61(10):1372-1374. Food Safety and Inspection Service [FSIS]. 2000. http://www.fsis.usda.gov/OA/background/2000fsi.htm Last accessed on July 7, 2005. Francis, G.A., C. Thomas and D. O'Beirne 1999. The Microbial Safety of Minimally Processed Vegetables. Int. J. Food Sci. Technol 34:1-22. Golden, D.A., E.J. Rhodehamel and D.A. Kautter. 1993. Growth of Salmonella spp. In Cantaloupe, Watermelon and Honeydew Me lons. J. Food Prot. 56:194-196. Guo, X., J. Chen, R.E. Brackett and L.R. Be uchat. 2001. Survival of Salmonellae on and in Tomato Plants from the Time of Inoc ulation at Flowering and Early Stages of Fruit Development through Fruit Ripeni ng. Appl. Environ. Microbiol. 67(10):47604765. Gupta, A., C.S. Polyak, R.D. Bishop, J. Sobel and E.D. Mintz. 2004. LaboratoryConfirmed Shigellosis in the United Stat es, 1989--2002: Epidemiologic Trends and Patterns. Clin. Infect. Dis. 38:1372-1377. Guzewich, J.J. and P. Salsbury. FDA’s Role in Traceback Investigations for Produce. Food Safety Magazine. 2001. Heisick, J.E., D.E. Wagner, M.L. Nieman and J.T. Peeler. 1989. Listeria spp. on Fresh Market Produce. Appl. Environ. Microbiol. 55:1925-1927. Ho, J.L, K.N. Shands, G. Friedland, P. Eckind and D.W. Fraser. 1986. An Outbreak of Type 4b Listeria monocytogenes Infection Involving Pati ents from Eight Boston Hospitals. Arch. Intern. Med. 146:520-523. Joy, J.A. 2005. Survival of Salmonella and Shigella on Tomatoes in the Presence of the Soft Rot Pathogen, Erwinia Carotovora (Master’s thesis, University of Florida, 2003).

PAGE 66

55 Islam, M.S., M.K. Hasan and S.I. Khan. 1993. Growth and Survival of Shigella flexneri in Common Bangladeshi Foods under Various Conditions of Time and Temperature. Appl. Environ. Microbiol. 59(2):652-654. Kaneko, K.I., H. Hayashidami, Y. Ohtomo, J. Kosuge, M. Kato, K. Takahashi, Y. Shiraki and M. Ogawa. 1999. Bacterial Contamina tion of Ready-to-Eat Foods and Fresh Products in Retail Shops and Food Fact ories. J. Food Prot. 62:644-649. Kauze, A.S. and H. Joseph. 2001. Quantitative De termination of the Role of Lettuce Leaf Structure on Protecting E. coli O157:H7 from Chlorine Disinfection. J. Food Prot. 64:147-151. Keller, S.E., S.J. Chirtel, R.J. Merker, K.T. Taylor, H.L. Tan and A.J. Miller. 2004. Influence of Fruit Variety, Harvest Tec hnique, Quality Sorting, and Storage on the Native Microflora of Unpasteurized Appl e Cider. J. Food Prot., 67(10):2240-2247. Kimura, A.C., K. Johnson, M.S. Palumbo, J. Hopkins, J.C. Boase, R. Reporter, M. Goldoft, K.R. Stefonek, J.A. Farrar, T. J. Van Gilder and D.J. Vuglar. 2004. Multistate Shigellosis Outbreak and Comm ercially Prepared Food, United States. Emerg. Infect. Dis. 10(6):1147-1149. Lampel, K.A. 2001. Shigella In : Downes, F.P. and K. Ito. (Eds.). Compendium of Methods for the Microbiological Examination of Foods (4th ed., pp. 381-384). United States of America: Sheridan Books, Inc. Lampel, K., R.C. Sandin and S. Formal. 1999. Shigella : Introduction and Detection by Classical Cultural Techniques. In : Robinson, R.K., C.A. Batt and P.D. Patel (Eds.). 2000. Encyclopedia of Food Microbiology. 3:2015-2020. Lang, M.M., L.J. Harris and L.R. Beuchat. 2004. Evaluation of I noculation Method and Inoculum Drying Time for Their Effects on Su rvival and Efficiency of Recovery of Escherichia coli O157:H7, Salmonella and Listeria monocytogenes Inoculated on the Surface of Tomatoes. J. Food Prot. 67(4):732-741. Liao, C.K. and P.H. Cooke. 2001. Can. J. Microbiol. 47:25-32. Lin, B., J.N. Variyam, J. Allshouse and J. Cromartie. 2003. Food and Agricultural Commodity Consumption in the United States: Looking Ahead to 2020. USDA/ERS, AER-820. http://www.ers.usda.gov/p ublications/aib792/aib7927/aib792-7.pdf Last accessed on July 13, 2005. Lund, B.M. and A.L. Snowdon. 2000. Fresh and Processed Fruits, Chapter 27. In : B.M. Lund, T.C. Baird-Parker and G.W. Gould (Eds.). The Mi crobiological Safety and Quality of Food, Volume I. Gaithersburg (MD): Aspen. pp. 738-758. Mead, P.S., L. Slutsker, V. Dietz, L.F. McCa ig, J.S. Bresee, C. Shapiro, P.M. Griffin and R.V. Tauxe.1999. Food-related Illness and Death in the United States. Emerg. Infect. Dis. 5:607-617.

PAGE 67

56 Morbidity and Mortality Weekly Report (MMWR). 2005. 52(54):1-85. http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5254a1.htm Last accessed on September 18, 2005. Mukherjee, A; D. Speh; E. Dyck and F. Di ez-Gonzales. 2004. Prehar vest Evaluation of Coliforms, Escherichia coli Salmonella and Escherichia coli O157:H7 in Organic and Conventional Produce Grown by Minneso ta Farmers. J. Food Prot. 67(5):894900. National Agricultural Statis tics Service [NASS]-USDA. 2005a. Florida Agricultural Facts. http://www.nass.usda.gov/fl/rtoc0v.htm National Agricultural Statis tics Service [NASS]-USDA. 2005b. Weekly Weather and Crop Bulletin. 92:37. http://usda.mannlib.cornell.edu/repo rts/waobr/weather/2005/full/wwcb3705.pdf Last accessed on September 20, 2005. National Agricultural Statis tics Service [NASS]-USDA. 2005c. Frozen Concentrated Orange Juice. In : Trends in U.S. Agriculture. http://www.usda.gov/nass/pubs/t rends/concentratedoj.htm Last accessed on July 10, 2005. O’Brien, S., R.T. Mitchell, I.A. Gillespi e and G.K. Adak. 2000. The Microbiological Status of Ready-to-Eat Fruit and Vege tables. Discussion paper ACM/476 of the Advisory Committee on the Mi crobiological Safety of Food. http://www.foodstandards.gov.uk/pdf_files/papers/acm476.pdf O’Brien, S., R.T. Mitchell, I.A. Gillespi e and G.K. Adak. 2000. The Microbiological Status of Ready-to-Eat Fruit and Vege tables. Discussion paper ACM/510 of the Advisory Committee on the Mi crobiological Safety of Food. http://www.food.gov.uk/multimedia/pdfs/acm510A.pdf Olsen, A.R. 1998. Regulatory Action Criteria fo r Filth and Other Extr aneous Materials. III. Review of Flies and Foodborne Enteri c Disease. Reg. Toxicol. Pharmacol. 28:199-211. Parish, M.E.1997. Public Health and Nonpasteu rized Fruit Juices. Crit. Rev. Microbiol. 23:109-119. Rafi, E. and P. Lunsford. 1997. Survival and Detection of Shigella flexneri in Vegetables and Commercially Prepared Salads. J. A ssoc. of Anal. Chem. Intl. 80:1191-1197. Ryu, C.–H., S. Igimi, S. Inoue and S. Kumagai. 1992. The Incidence of Listeria Species in Retail Foods in Japan. Int. J. Food Microbiol. 16:157-160. Samish, Z., R. Etinger-Tulczynska and M. Bick. 1963. The Microflora Within the Tissue of Fruits and Vegetables. J. Food Sci. 28:259-266.

PAGE 68

57 Sandeep, M, D. Aggarwal and A. Ganguli. 2004. Microbiological An alysis of Streetvended Fresh Squeezed Carrot and Kinnow-Ma ndarin Juices in Patiala City, India. Internet Journal of Food Safety, V.3, 1-3. http://www.foodhaccp.com/in ternetjournal/ijfsv31.pdf Last accessed on July 27, 2005. Sapers, G.M. 2001. Efficacy of Washi ng and Sanitizing Methods. Food Technol. Biotechnol. 39(4):305-311. Sargent, Steven A., and C.L. Moretti. 2002. Tomato. In : The Commercial Storage of Fruits, Vegetables, and Florist & Nursery Crops (3rd ed.). http://www.ba.ars.usda.gov/hb66/138tomato.pdf Last accessed on September 18, 2005. Schelch, W.F., P.M. Lavigne, R.A. Bortolussi A.C. Allen, E.V. Haldane, A.J. Wort, A.W. Hightower, S.E. Johnson, S.H. King, E.S. Nicholls and C.V. Broome. 1983. Epidemic Listeriosis--Evidence for Transmission by Food. N. Engl. J. Med. 308:203-206. Seo, K.H. and J.F. Frank. 1999. Attachment of Escherichia coli O157:H7 to Lettuce Leaf Surface and Bacterial Viability in Response to Chlorine Treatment as Demonstrated by Using Confocal Scanni ng Laser Microscopy. J. Food Prot. 62:39. Shiferaw, B., S. Shallow, G. Kazi, S. Segl er, D. Soderlund, T. Van Gilder and the EIP FoodNet Working Group. 2000. Shigella Then and Now: Comparing Passive Surveillance for Shigellosis in Five FoodNet Sites, 1996-1998. 2nd International Conference on Emerging Infectious Diseases. Atlanta, GA. Simes, M., B. Pisani, E.G.L. Marques, M.A.G. Prandi, M.H. Martini, P.F.T. Chiarini, J.L.F. Antunes and A.P. Nogueira. 2001. Hygienic-Sanitary Conditions of Vegetables and Irrigation Water from Kitchen Gardens in the Municipality of Campinas, SP. Braz. J. Microbiol. 32:331-333. Spreen, L.T., W. Fernandes, Jr., C. Mo reira and R.P. Muraro. 2001. An Economic Evaluation of Hamlin versus Valencia Ora nge Production in Florida. Department of Food and Resource Economics, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, Universi ty of Florida, Gainesville, FL. http://edis.ifas.ufl.edu/BODY_FE300 Last accessed on September 18, 2005. Stine, S.W., I. Song, C.Y. Choi and C.P. Gerba. 2005. Effect of Relative Humidity on Preharvest Survival of Bacterial and Vi ral Pathogens on the Surface of Cantaloupe, Lettuce, and Bell Peppers. J. Food Prot. 68(7):1352-1358.

PAGE 69

58 Swerdlow, D.L., K. D. Greene, R. V. Tauxe, J. G. Wells, N. H. Bean, A. A. Ries, P. A. Blake, E. D. Mintz, M. Pollack, M. R odriguez, E. Tejada, L. Seminario, C. Ocampo, B. Vertiz, L. Espejo and W. Sa ldana. 1992. Waterborne Transmission of Epidemic Cholera in Trujillo, Peru; Le ssons for a Continent at Risk. Lancet 340(4):28-32. Takeuchi, K. and J.F. Frank. 2000. Penetration of Escherichia coli O157:H7 into Lettuce Tissues as Affected by Inoculum Size and Temperature and the Effect of Chlorine Treatment on Cell Viability. J. Food Prot. 63:434-440. Tetteh, G.L. and L.R. Beuchat. 2003. Survival Growth, and Inactiva tion of Acid-Stressed Shigella flexneri as Affected by pH and Temperatur e. Int. J. Food Microbiol. 87(12):131-138. Tetteh, G.I., S.K. Sefa-Dedeh, R.D. Phill ips and L.R. Beuchat. 2004. Survival and Growth of Acid-adapted and Unadapted Shigella flexneri in a Traditional Fermented Ghanian Weaning Food as Affect ed by Fortification with Cowpea. Int. J. Food Microbiol. 90:189-195. Thompson, J., A. Kader and K. Sylva. 2002. Compatibility Chart for Fruits and Vegetables in Short-term Transport or St orage. University of California – Division of Agriculture and Natural Resources. Publication 21560. http://postharvest.ucdavis.edu/Pubs/postthermo.shtml Last accessed on July 14, 2005. Todd, E. 1989. C.D. Preliminary Estimates of Costs of Foodborne Disease in the United States. J. Food Prot. 52(8):595-601. Wu, F. M., M.P. Doyle, L.R. Beuchat, J. G. Wells, E.D. Mintz and B. Swaminathan. 2000. Fate of Shigella sonnei on Parsley and Methods of Disinfection. J. Food Prot. 63:568-572. Zaika, L.L. 2001. The Effect of Temperature and Low pH on Survival of Shigella flexneri in Broth. J. Food Prot. 64(8):1162-1165. Zaika, L.L. 2002. Effect of Organic Ac ids and Temperature on Survival of Shigella flexneri in Broth at pH 4. J. Food Prot. 65:1417-1421.

PAGE 70

59 BIOGRAPHICAL SKETCH Dirk Sampath was born in the town of Siparia, Trinidad, in the twin-island Republic of Trinidad and Tobago, West I ndies, on November 15, 1957. He graduated from the University of Florida with his Bach elor of Science in nutritional science, in 1992. He then returned to Trinidad and worked on his parent’s family-owned farm. In 1995 he returned to Orlando, Florida, and then moved to Athens, Georgia. His venture there led him to work in the poultry industry for GoldKist. He then returned to the University of Florida in 2002 to pursue his Ma ster of Science in food science and human nutrition. After graduating, Dirk plans to be employed in the food or food-related industry.


Permanent Link: http://ufdc.ufl.edu/UFE0013128/00001

Material Information

Title: Survival of Inoculated shigella spp. on Tomato and Orange Surfaces
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: UFE0013128:00001

Permanent Link: http://ufdc.ufl.edu/UFE0013128/00001

Material Information

Title: Survival of Inoculated shigella spp. on Tomato and Orange Surfaces
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: UFE0013128:00001


This item has the following downloads:


Full Text












SURVIVAL OF INOCULATED .\/igel// spp. ON TOMATO AND ORANGE
SURFACES















By

DIRK M. SAMPATH


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

Dirk M. Sampath

































To all the members of my family, and friends, who have encouraged me over the years to
strive for better things. And to my nephew and niece, Quinn and Patrina respectively, I
encourage you the same. I hope that as you both move out into the world, you can look
back at the time we were in Trinidad & Tobago with fond memories.















ACKNOWLEDGMENTS

Firstly, I would like to thank my major professor and committee chair, Dr. Keith R.

Schneider, for the opportunity he presented to me in pursuing this master's degree. I also

thank the two other members of my graduate committee, Dr. Renee M. Goodrich and Dr.

Mark A. Ritenour, for all the invaluable help and advice given to me in various aspects of

my project.

I thank all my family who supported me during this time, especially my younger

sister, Martine, my younger brothers, Link, Brett and Verne, and last but not least, my

mother, Cynthia, who, thankfully, is my lone, living parental witness for this, as my

father, Dr. Martin Sampath, flew away on Tuesday, December 5, 1995. I also thank my

elder sister, Sylvanna, who exposed me to a lot of progressive things when we were much

younger (sometime in the last century), and whom I still consider to be much more

civilized than I may ever attain.

I thank all my labmates past and present, for the advice and help they all freely

gave to me whenever I needed it during my tenure here at UF.

My work here at the University of Florida was supported by the USDA-CREES

IFAFS Grant number 00-52102-9637.

















TABLE OF CONTENTS


page

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

L IS T O F T A B L E S ................ ........................................................................ ... v ii

FIG U R E ...... ...................................................................................... ............... .............. ix

A B ST R A C T ................. ...................................................................................... ..... x

CHAPTER

1 INTRODUCTION ............... ................. ........... ................. ... ..... 1

2 LITERA TURE REVIEW .......................................................... ..............6

,. . ..ll. ...................................... ............................................. . 9
Fresh Tomato Production and Handling in Florida..................................................12
Fresh Orange Production and Handling in Florida................. ............................13

3 M ATERIALS AND M ETHOD S ........................................ ......................... 16

Initial Preparation of Rifampicin Stock Solution. .................... ................... .......... 16
G row th Study ............... ................. ........................................... 16
Preparations of Rifampicin-resistant ./'lge,//A ...................................................... 16
Adaptation of organisms to rifampicin in tryptic soy broth (TSB). .............16
Transfer of organisms onto Rif80 tryptic soy agar slants and storage.........17
Procedure for the Grow th Studies ............................................ ............... 18
R ecov ery Stu dy ................................................................... ................. 19
Preparations Prior to the Recovery Study ................................. ............... 19
Acquisition of produce ............... .......................................... .................. 19
Placement of produce prior to inoculation. .............................................19
Preparation of PBS rinsate ............. .............. ......... .. ............. 19
Procedure for the Recovery Study.............. ............. ....... ............... 20
Preparation of organism source inoculum............................... ...............20
Determination of CFU/ml in the source inoculum................ ........... 21
Inoculation onto produce surface ...................................... ............... 21
Recovery of initial inoculum in BPW rinsate .............. ................ 21









Su rv iv al Stu dy ................................................................2 2
Statistical A analyses ............................................................ ........ .... 22

4 R E S U L T S .............................................................................2 3

Growth Study of S. sonnei and S. flexneri ............................................................... 23
Recovery and Survival Studies for Tomatoes and Oranges ......................................24
Recovery Study for Tom atoes ............................................................................. 26
Recovery Study for Oranges......................................................... .............. 27
Su rv iv al Stu dy .........................................................................2 7
T om ato Survival Study .......................... .. ................................. ............... 28
\/V g ell// t sonn ei ............................................... ................. 2 8
1\/ /lg e/l t flexneri ............................................................... ............... 3 1
O range Survival Study ............................................... ............................. 34
.\ i/g e ll/ so n n ei ............................................... ................ 3 4
.\/li /g e/ t flexneri ............................................................... ............... 36

5 DISCUSSION .............. .......... ...... ................. .........39

T om ato Survival Study .......................... .. ................................. ............... 40
g ell t sonnei ........................................ .................40
1/i lg e/l t fl ex n eri ...................................................................................... 4 2
O range Survival Study ............................................... ............................. 43
\ lirg e/' t son n ei ............................................... ................ 4 4
1/ihlg e/l t fl ex n eri ...................................................................................... 4 5

6 C O N C L U SIO N ......... ......................................................................... ........ .. ..... .. 49

LIST OF REFEREN CES ............................................................ .................... 51

B IO G R A PH IC A L SK E TCH ..................................................................... ..................59
















LIST OF TABLES


Table page


4-1. Organism, temperature/relative humidity combinations, and time periods for the
tom ato and orange survival studies. .............................................. ............... 25

4-2. Tomato recovery of .\/nge//At spp. from 0.1% buffer peptone water (BPW) after
inoculum dried (1.0 to 2.5 hr). ............................................................................ 26

4-3. Orange recovery of.\ i/ge//A, spp. from 0.1% buffer peptone water (BPW) after
inoculum dried (1.0 to 2.5 hr). ............................................................................ 27

4-4. Comparison of S. sonnei survival population decline on 'Florida 47' tomatoes at
13C/60%RH and 13C/90%RH conditions. ................................ ..................29

4-5. Comparison of S. sonnei survival population decline on 'Florida 47' tomatoes at
30C/60%RH and 30C/90%RH conditions ................................. ................29

4-6. Cross-comparison of S. sonnei survival population decline on 'Florida 47'
tomatoes at 130C/60%RH and 30C/60%RH conditions. ......................................30

4-7. Cross-comparison of S. sonnei population decline on 'Florida 47' tomatoes at
13C/90%RH and 30C/90%RH conditions .................................... ............... 31

4-8. Comparison of S. flexneri survival population decline on tomatoes at
13C/60%RH and 13C/90%RH conditions .................................... ............... 32

4-9. Comparison of S. flexneri survival population decline on 'Florida 47' tomatoes
at 30C/60%RH and 30C/90%RH conditions................................................. 32

4-10. Cross-comparison of S. flexneri survival population decline on 'Florida 47'
tomatoes at 130C/60%RH and 30C/60%RH conditions ......................................33

4-11. Cross-comparison of S. flexneri survival population decline on 'Florida 47'
tomatoes at 130C/90%RH and 30C/90%RH conditions ......................................34

4-12. Comparison of S. sonnei survival population decline on oranges at 130C/60%RH
and 13C/90% RH conditions .............................................................................34

4-13. Comparison of S. sonnei survival population decline on oranges at 300C/60%RH
and 30C/90% RH conditions .................................. ...................................35









4-14. Cross comparison of S. sonnei survival population decline on oranges at
13C/60%RH and 30C/60%RH conditions .................................... ............... 36

4-15. Cross-comparison of S. sonnei survival population decline on oranges at
13C/90%RH and 30C/90%RH conditions .................................... ............... 36

4-16. Comparison of S. flexneri survival population decline on oranges at
13C/60%RH and 13C/90%RH conditions .................................... ............... 37

4-17. Comparison of S. flexneri survival population decline on oranges at
30C/60%RH and 30C/90%RH conditions .................................... ............... 37

4-18. Cross-comparison of S. flexneri survival population decline on oranges at
13C/60%RH and 30C/60%RH conditions .................................... ............... 38

4-19. Cross-comparison of S. flexneri survival population decline on oranges at
13C/90%RH and 30C/90%RH conditions .................................... ............... 38






















FIGURE

Figure page

4-1. Growth curves of rifampicin-adapted S. sonnei and S. flexneri grown in TSB-
Rif80 at the stationary phase. ..............................................................................24















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

SURVIVAL OF INOCULATED ,h/ngel/A spp. ON TOMATO AND ORANGE
SURFACES

By

Dirk M. Sampath

December 2005

Chair: Keith R. Schneider.
Major Department: Food Science and Human Nutrition

This study examined the survival of .\lhgell,/t sonnei and .\l lge/a flexneri on the

surfaces of Florida-grown tomatoes and oranges at temperature and relative humidity

conditions mimicking those that could be encountered during the growing season, at

harvest, and during postharvest storage. 'Florida 47' tomatoes and navel and 'Valencia'

oranges were used in this study. Inoculated samples were placed in one of four

temperature/relative humidity environments: 13C/60%RH, 13C/90%RH, 30C/60%RH

or 30C/90%RH.

In some instances, the temperature and relative humidity combination had a marked

effect on the survival of inoculated ,.\/ge/ll spp. on produce surfaces. There were no

significant differences in S. sonnei and S. flexneri population reduction on tomato

surfaces held at 130C at 60% or 90%RH. On the tomato surfaces at 300C, S. sonnei

populations declined slower at 60%RH compared to 90%RH. In contrast, there was no

significant difference in population decline when S. flexneri was observed under the same









conditions. At 60%RH, there was no significant difference in population decline of S.

sonnei or S.flexneri at 13 and 30C. At 90%RH, the S. sonnei population declined

significantly slower at 130C than at 30C. No significant difference in population decline

was observed with S. flexneri under this condition.

On orange surfaces stored at 130C, significantly more S. sonnei survived at 60%RH

(indicated by a 1.47 logo CFU population decline) than at 90%RH (4.71 logo CFU

decline). A similar observation was made for S. flexneri under this same condition, with

1.45 logo CFU decline per fruit on Day 7 at 60%RH, and 4.06 logo CFU decline per

fruit, at Day 7. At 30C, the population decline was significantly lower at 60%RH for S.

sonnei (4.24 logo CFU) and S. flexneri (5.66 logo CFU) than at 90%RH (5.32 and 6.41

logo CFU decline per fruit, for S. sonnei and S. flexneri, respectively). At 60%RH,

significantly more S. sonnei survived at 130C than at 300C after 7 days. The same pattern

of behavior was observed for S. flexneri. At 90%RH, survival of both S. sonnei and S.

flexneri was greater at 130C than at 30C. Both S. sonnei and S. flexneri approached the

least limit of detection on the orange surfaces at Day 7, under storage conditions of

130C/90%RH.

The results indicate that higher temperature (300C) and higher relative humidity

(90%) favored the population reduction of S. sonnei on the tomato surfaces. Storage

conditions of 130C and 85-90%RH for 48 hr inhibits the survival ofS. sonnei and S.

flexneri on tomatoes, thereby decreasing the risk of shigellosis foodborne disease. Both

organisms survived better on orange surfaces at 130C and 60%RH. However, oranges

stored at the present conditions and times under which the tomatoes are stored could

probably represent a greater risk vehicle of foodborne shigellosis, compared to tomatoes.














CHAPTER 1
INTRODUCTION

Recent outbreaks of some foodborne illnesses have been traced back to the

consumption of contaminated fresh fruits and vegetables (Beuchat 2002). As health-

conscious individuals seek a healthier lifestyle, one of the choices made in this regard is

to consume more plant-oriented foodstuffs. In the U.S. and other countries, successful

campaigning to this end has inevitably led to the increased consumption of more of the

traditional, and a greater variety of, plant-based foods, especially fresh fruits and

vegetables (Beuchat and Ryu 1997). Along with the increased consumption of these

foods, there has been an increase in the reported incidences of foodborne illnesses

(Beuchat 2002). Since 1997, as a result of a report to President Clinton of the U.S.

entitled, "Food Safety from Farm-to-Table: A National Food Safety Initiative," increased

cooperation between the U.S. Department of Health and Human Services (USHHS) and

the U.S. Department of Agriculture (USDA) has led to combined efforts to protect the

health of the American consumers, with more emphasis placed on monitoring domestic

and imported produce for safety (Food Safety and Inspection Service [FSIS] 2000).

Increased monitoring, in combination with the increasing consumption trend, may lead to

a higher occurrence of reported produce-associated, foodborne illness outbreaks among

the consumer population. When such outbreaks do occur traceback operations,

conducted by the health agencies responsible for such activities, have to be implemented

in order to determine the origin and destinations of the produce implicated. Some of the

produce items for which traceback operations have been conducted as a result of reported









incidences of foodborne illnesses from 1990 to 2000 have been tomatoes, cantaloupes,

scallions, leaf lettuce, raspberries, basil and basil-containing products, various berries,

green onions and parsley (Guzewich and Salsbury 2001). Since the U.S. depends on

seasonal imports for an adequate, continuous supply of perishable produce, not only are

monitoring operations of such imports important, but so also are accurate traceback

information about sources of product, if foodborne illness outbreaks do occur. For

example, inaccurate traceback of a foodborne illness incident in June 1996 by the Texas

Department of Health initially identified strawberries from California as the product

contaminated with Cyclospora, when in fact it was Guatemalan raspberries (Calvin

2003).

Bacteria can attach to the surfaces of many produce items. They may be found in

pores, dents and other surface aberrations where they are more protected from adverse

environmental conditions (Seo and Frank 1999). They can also be found adhering to

cuts, cracks, and perforations in the produce surfaces (Liao and Cooke 2001; Takeuchi

and Frank 2000; Burnett et al. 2000), where they may be further protected. Some

produce are washed in dump tanks using recirculated water. However, proper sanitation

procedures must be followed, since such water treatments may present a potential health

risk as microbes are washed off fruit surfaces and accumulate in the water used for

cleaning. The contaminated water then can contaminate all subsequent fruit passing

through it. In addition to contaminating the surfaces, the contaminated water may also

penetrate into the produce by whatever accessible openings are present on their surfaces.

Bartz (1982) showed that Erwinia carotovora (subspecies carotovora) invasion into

tomatoes could be prevented if their submersion times and depths were less than two









minutes and 17 centimeters, respectively. In a more recent study, Duffy et al. (2005)

reported that parsley submerged for 15 minutes in a peptone-inoculated, 3-strain

Salmonella suspension had higher populations of loosely-attached, strongly-attached, and

internalized Salmonella cells than parsley submerged for 3 min in the same suspension.

Foodborne bacteria have been shown to have a greater potential to survive on fresh-

cut produce than on those with intact peels or rinds, because the bacteria had access to the

nutrients exuded on the cut produce surfaces (Francis et al. 1999). Golden et al. (1993)

found that Salmonella grew well on the interior tissues of watermelon, honeydew and

cantaloupe melons at 230C, such as would be found at a roadside stand. Even in cold

storage, human pathogens such as Escherichia coli 0157:H7 have been shown to survive.

After 34 hr storage at 5C, E. coli 0157:H7 levels of 3.1 and 3.0 CFU/gm, respectively,

on cut watermelons and cantaloupes were unchanged (Del Rosario and Beuchat 1995).

Temperature and relative humidity are the most important environmental characteristics

which affect the populations of the natural microflora and microbial pathogens found on

fresh fruit and vegetable produce (Brackett 1987). These environmental factors may thus

enhance the bacterial survival on the produce and lead to increased risk of foodborne

illness due to contaminated produce.

Bacterial contamination of fresh produce can occur at any time during production,

harvesting and postharvest handling. For example, contamination can occur in the field

by use of contaminated irrigation water, by harvesters and postharvest handlers, in

unclean processing facilities, in food-service establishments, and by consumers

themselves (Beuchat 1996). Under laboratory conditions, simulated commercial washing

and sanitizing procedures typically results in a one to two log reduction in pathogen









numbers, and this reduction is inadequate for microbiological safety (Sapers 2001). In

some commercial applications the actual reduction in pathogen numbers may be lower

thereby compromising produce safety (Sapers 2001). Large-scale growers, wanting to

keep the produce as "fresh" as possible so as to maximize sales, and minimize losses due

to drop in quality and spoilage, try to speed up delivery from the farm to the final outlets.

As a result, some washing and rinsing procedures at these facilities may not be adequate

to remove pathogens on the produce surfaces to safe levels, and may even increase the

microbial load on them if the wash water is contaminated. A study performed by Lang et

al. (2004) found that significantly more pathogens were recovered from the surface

tomatoes dip-inoculated in cell suspensions, which represented washing produce in water

in dump tanks, compared to those recovered when the tomatoes were spot- or spray-

inoculated. More pathogens were recovered 1 hr after application and were more viable

than those recovered after 24 hr after (Lang et al. 2004). Since some foodborne

pathogens have low infective doses, the potential danger to consumers exposed to

inadequately washed produce is apparent. In preventing contamination of produce,

effective procedures must be implemented at the preharvest and postharvest aspects of

production (Cliver 1997).

In 1999, of the 1,040 cases of.\/ligel//t infections causing illnesses that were

reported at FoodNet Surveillance sites, 61% were due to S. sonnei and 29% from S.

flexneri (Centers for Disease Control and Prevention [CDC] 1999). Of the 15% of cases

requiring hospitalization from bacterial infections in general, 12% of those were due to

./lge/lt (CDC 1999). In the U.S., S. sonnei accounted for greater than 75% of

shigellosis cases (Gupta et al. 2004). In its 2000 FoodNet Surveillance Annual Report,









the CDC stated that from 1996 to 2000, the incidence of salmonellosis declined while the

overall incidence of shigellosis increased, with noticeably large increases in California

and Minnesota (CDC 2000).

This study examined the survival of each of the bacterial pathogens, S. sonnei

(ATCC 9290) and S. flexneri (LJH 607) on the surfaces of tomatoes (Florida 47 variety),

and Navel and Valencia oranges. Both tomatoes and oranges are grown commercially in

Florida, with cash receipts in 2003 being $516 and $983 million respectively (National

Agricultural Statistics Service [NASS] 2005a). The temperature and relative humidity

combinations used in the current studies were set to mimic commercial field and

postharvest conditions experienced by tomatoes and oranges in Florida. The objectives

in the study were to establish the growth characteristics of rifampicin-adapted S. sonnei

and S. flexneri in tryptic soy broth (TSB), to determine the recovery after drying of both

1/nge/l/t inoculated around the blossom ends of tomatoes and oranges, and to observe the

survival of.\lnge//A -inoculated tomatoes and oranges at temperature/relative humidity

conditions of 13C/60%RH, 13C/90%RH, 30C/60%RH and 30C/90%RH.














CHAPTER 2
LITERATURE REVIEW

In the U.S. it is estimated that over 76 million illnesses and 5,000 deaths are due to

diseases perpetuated by foodborne organisms (Mead et al. 1999). Foodborne diseases

and conditions deemed nationally reportable are: botulism, brucellosis, cholera,

enterohemorrhagic E. coli (EHEC), post-diarrheal hemolytic uremic syndrome,

listeriosis, salmonellosis, shigellosis, typhoid fever, hepatitis A, cryptosporidiosis,

cyclosporiasis, and trichinellosis (Morbidity and Mortality Weekly Report (MMWR)

2005). Predictions are that from 2000 to 2020, the market share for citrus and apples will

increase by 27% each, grapes 24%, tomatoes 19%, lettuce 24%, other fruit 26%, and

other vegetables 22%, with the least increase of 8% shown by fried potatoes/chip

consumption (Lin et al. 2003). This projected increase in overall produce consumption,

combined with the addition of foreign sources to satisfy year-round domestic demand has

increased the likelihood of reportable foodborne illnesses.

Many food poisoning incidences worldwide have been attributed to contaminated

fresh fruits and vegetables. In 1989-90, Salmonella-contaminated cantaloupe from

Mexico and Central America were responsible for causing illness in an estimated 25,000

people in the U.S. (CDC 1991; Lund and Snowdon 2000). In 1991, this organism was

also responsible for 33 cases of infection via watermelon consumed at a picnic and school

fair in Michigan (Blostein 1993). Frozen Brazilian mamey [apple] imported to the U.S.

in 1998-99 caused 13 cases of Salmonella Typhi food poisoning (Lund and Snowdon

2000). Cantaloupe containing E. coli 0157:H7 caused illness in nine people at an









Oregon restaurant in 1993 (del Rosario and Beuchat 1995). At a Minnesota hotel

restaurant serving fruit salad, coleslaw and tossed salad, Norovirus was responsible for

233 reported cases of foodborne illness in 1982. Norovirus caused 206 cases of food

poisoning in the United Kingdom in 1987 (Lund and Snowdon 2000). In 1992,

Calcivirus in salad eaten at a catered event in Ontario, Canada caused illness in 27 people

(CFSAN-USDA 2001). In England and Wales, almost 6% of intestinal foodborne

disease outbreaks occurring in the 1992-2000 period were linked to fruits, vegetables and

salads (O'Brien et al. 2000). Campylobacterjejuni in salad eaten in British Columbia,

Canada, sickened 330 patrons in a university cafeteria in 1984 (Allen 1985). This same

organism in 1996 caused sickness in 14 people who consumed contaminated lettuce in an

Oklahoma restaurant (CDC 1998). Listeria spp. have been frequently isolated from

vegetables (de Simon et al. 1992; Heisick et al. 1989; Kaneko et al. 1999; Ryu et al.

1992). In 1979, L. monocytogenes contamination of tomatoes, lettuce, and celery likely

caused 25 cases of illness in a Boston hospital, including two fatalities (Ho et al. 1986;

Schelch et al. 1983). In Peru, cabbage compromised with Vibrio cholerae caused 71

deaths in 1991 (Swerdlow et al. 1992). As these incidents show, the affected product, the

product point of origin, and the final venue where the contaminated product was passed

on to the consumer have all been variable over the years.

Juices derived from minimally processed fruits and vegetables, wherever these may

be grown, are also causes for concern with respect to foodborne illnesses. Surface

pathogens on the intact produce can later be passed into the juices when the products are

processed, and may survive if the juices are not pasteurized or improperly handled. Fresh

juices are recognized as causes of foodborne illnesses (Parish 1997). Even the acidic









nature of some juices may not be adequate to kill some pathogens. A study by Zaika

(2002) examined the survival of S. flexneri in the presence of 0.4 M, pH 4 citric and

malic acids at 40C, and found that the pathogen survived for more than 70 days. At 4C,

both S. sonnei and S. flexneri were found to survive in tomato juice for 14 days, and for 8

days at 220C (Bagamboula et al. 2002). Apart from contamination of juices or juice-

derived beverages directly by contaminated produce or processing equipment, equipment

normally deemed clean by current sanitation standards can also be a source of

contamination. Keller et al. (2004) stated that in experiment involving cider production

performed in a small commercial facility, aerobic plate counts (APC) on incoming apples

used in the cider production should correlate linearly with that of the cider APC,

indicating that the counts came from organisms on the apples used. However, they found

no such linear relationship between the apples used and cider made, indicating possible

cross-contamination from the cider-processing equipment (Keller et al. 2004). Workers

with unhygienic practices can contaminate freshly-squeezed juices. Bacteria from such

workers may be transferred directly into the squeezed juices or indirectly from the hands

onto the fruit or vegetable surfaces. Once on the surface, bacteria can be transferred into

the juices during the squeezing operation. In Patiala City, India, Sandeep et al. (2004)

studied the quality of street-vended, freshly-squeezed carrot and Kinnow-Mandarin juices

and found the presence of coagulase-positive Staphylococcus aureus in 30% of the carrot

juice and 12% of the orange juice samples examined. The total fecal coliform counts

(TFCC) and total viable counts (TVC) were about 5 and 6 logo units respectively, for

both juice types (Sandeep et al. 2004).









Zeal for a healthier lifestyle has led some consumers to abstain from

conventionally-grown produce in favor of organically-grown food items. However,

organic produce may not necessarily be safer to eat. In a study performed on a

Minnesota farm, Mukherjee et al. (2004) observed that 8% of fruits and vegetables had

no significant difference in coliform counts between the organically- and conventionally-

grown tomatoes, leafy greens, lettuce, green peppers, cabbage, cucumbers, broccoli,

summer squash, zucchini, bok choi, apples, onions and strawberries. The researchers did

find a six-fold significant increase in E. coli numbers in organically-grown harvested

fruits and vegetables, compared to those conventionally-grown. Even produce garnered

from smaller-scale, home kitchen gardens face similar problems of bacteria-contaminated

water, as those commercially grown on larger scales. In the Campinas Municipality of

Sdo Paulo, Brazil, 19.9% of vegetables examined from kitchen gardens using irrigation

water had fecal coliform counts greater than 200 CFU/g (Simies et al. 2001).

Shigella

.hige/llt species are Gram-negative, facultatively anaerobic, nonsporulating,

nonmotile rods in the family Enterobacteriaceae (Andrews and Jacobson 1998). They are

the causal organisms of shigellosis, also known as bacillary dysentery (Lampel et al.

1999). The four species of.\/lge//At which cause disease, in order of decreasing severity

and serologically typed by their somatic O antigen (Downes and Ito 2001) are S.

dysenteriae (Serogroup A; Shiga toxin production), S. flexneri (Serogroup B), S. boydii

(Serogroup C), and S. sonnei (Serogroup D) (Lampel et al. 1999). Nearly all the

virulence genes lie in a 37-kilobase pair (kbp) cluster within a 180-220-kpb plasmid

(Downes and Ito 2001). By manipulating host-cell phagocytosis, the .h/gel,// enters the

host using its Ipa protein antigens (Downes and Ito 2001). The Ipa proteins are









transported to and concentrated on the bacterial surface by means of the products

expressed from the invasion plasmid gene, ipg, and those responsible for Ipa expression

on the bacterial surface (Downes and Ito 2001). The infection-type of illness presented

by .l/nge//A spp. may be caused by 101-108 colony-forming units (CFU), which may have

an incubation period of 4-6 weeks, with moderate to severe symptoms in those afflicted

lasting days to weeks (Council for Agricultural Science and Technology (CAST) 1994).

Shigella Contamination in Food

.\/nge/ll spp. are not associated with any one food; contamination is mainly due to

mishandling by food handlers with poor personal hygiene (Downes and Ito 2001). The

foods most often associated with .\/nlge/t are potato salads, chicken, shellfish (Downes

and Ito 2001) and raw vegetables (Andrews and Jacobson 1998). In an experiment to

study the survival of S. flexneri in coleslaw, crab salad, carrot salad, cabbage salad and

potato salad at 40C, it was found that the organism survived for 11 days minimum in all

five salads, and was not killed by the low pH or inhibited by the normal flora inherently

present in the salads (Rafi and Lunsford 2002).

Many vegetables have an internal pH of 4.5 or higher, and are able to support

bacterial growth, while many fruits (apples, oranges, tomatoes) have a lower internal pH

that inhibits bacterial growth (De Roever 1998). During cider production, apples have

been found to allow the slow growth of E. coli 0157:H7 because mold growth increased

the pH of the apple flesh (Fisher and Golden 1998). Carter (1989) noted that during

fresh-squeezed orange juice operations, microorganisms transferred from the orange peel

could propagate in the juice under favorable conditions. Microbial contamination of

plant tissue is for the most part associated with the surfaces of fruits and vegetables, and

sound fruits have sterile interiors (De Roever 1998). However, it has been demonstrated









that bacteria on the surface of sound fruits eventually can be internalized (Samish 1963).

Organisms can enter into produce such as tomatoes through naturally occurring openings

and impaired skin surfaces (Bartz and Showalter 1979). Buchanan et al. (1999) found

elevated levels of E. coli 0157:H7 in the inner core region of whole, intact, warm apples

that had been dipped in a cooler temperature suspension of the organism. In a parallel

experiment performed during that same study, when warm apples were submerged in

cold water colored with Red Dye # 40, it was observed that the dye, and by implication

the E. coli, entered into the apples' inner core regions through readily-seen open channels

(Buchanan et al. 1999).

Insects, birds, and dust can act as vectors for plant and human pathogens, especially

after fruits and vegetables have been injured (Beuchat 1996). In one such example,

\/liellt was isolated from ordinary houseflies (Olsten 1998). During harvesting and

handling, citrus fruits can be damaged via plugging, bruises, punctures and splits (Almed

et al. 1973). The quality of the fruit and vegetable thereby diminishes below that

acceptable for safe and palatable consumption, and this quality may decrease faster if

temperature and humidity conditions present during harvesting and packing-house

storage allow and enhance the survival of pathogens like \,/ige/ll

Data from experiments performed on packaged sterile and unsterile vegetables

indicate that .\/ige/ll survived on vegetables tested after 10 days, reaching a 3-7 logo

unit reduction from the initial inoculum (Lampel et al. 1999). It was noted that S. sonnei

grew rapidly on chopped parsley and remained viable after 14 days at 40C (Wu et al.

2000). The survivability of.\ /ge//At may be more pronounced at higher temperatures and

humidities on fruits and vegetables such as oranges and tomatoes. Studies of Salmonella









survivability on harvested tomatoes have been performed mimicking commercial

temperature and relative humidity conditions found during the Spring and Fall/Winter

harvest season, and the survivability of Salmonella serotypes observed on various surface

matrices which match those in Florida packing houses (Allen 2003). ./ngel//t that have

been acid-adapted to reduced pH have been shown to grow more effectively in some

acidic foods than unadapted ./nhlge// Significantly more acid-adapted S. flexneri than

non-adapted ones were recovered from unfermented or fermented porridges made with

corn or corn-cowpea dough (Teteh et al. 2004).

In 1999, of the estimated 448,000 cases of Shigella infections reported in the U.S.,

89,000 were attributed to foodborne sources and 14 cases resulted in death (Mead et al.

1999). Though causing fewer sicknesses each year (approximately 2,500), illness due to

L. monocytogenes results in a much greater rate of mortality (approximately 20%) each

year (Mead et al. 1999). FoodNet sites have reported that in 3,784 cases of.\/nge//at

infections during 1996-1998, S. sonnei accounted for 72%, S. flexneri 22%, S. boydii

0.9%, and S. dysenteriae 0.6%, and that the overall rates of infection were highest

amongst children 1-9 years old (Shiferaw et al. 2000). In January 2000, a multi-state

outbreak of S. sonnei shigellosis gastroenteritis, involving 406 persons, was traced back

to a commercial brand of bean dip consisting of cooked beans, salsa, guacamole, nacho

cheese and sour cream (Kimura et al. 2004).

Fresh Tomato Production and Handling in Florida

In Florida, fresh tomatoes accounted for 40% of U.S. domestic production in 2003

(Economic Research Service [ERS] 2004b). With little change for the last 40 years, the

total tomato production in Florida has been about 40,000 acres, with yields that have

increased steadily during those times (ERS 2004c). This increase can be attributed to









increased utilization of drip irrigation techniques and the development of new tomato

varieties (ERS 2004c). Florida usually competes with Mexico for the early Spring and

Winter U.S. domestic market. The Mexican supply peaks in the winter when Florida is

also the prominent producer during that time (ERS 2004a). Tomatoes are usually planted

in Florida from mid-July to mid-March, and harvesting the crops begin at about mid-

October and ends in June, with temperature and relative humidities averaging 30C

(86F) and 60-90% respectively, during that time (National Agricultural Statistics Service

[NASS] 2005b). Mature-green tomatoes are tomatoes at the point of just changing color

but are still a uniform light green. Mature-green are hand-harvested and then placed in

plastic buckets. The filled buckets are then emptied into pallet bins or the larger

gondolas. Trucks then carry filled bins or gondolas to the packinghouse. The tomatoes

are kept under shade out of direct sunlight in order to allow the tomatoes to cool. The

bins or gondolas are then decanted, allowing the tomatoes to be dumped or flumed into a

dumptank containing heated chlorinated water. From the dumptank, the tomatoes are

rinsed, dried, sized, graded and waxed before being packed in cardboard-type boxes.

After commercial packing, tomatoes are routinely palletized and stored at 120C (Sargent

et al. 2002). Mature, green tomatoes can be stored for 2-5 weeks at 130C and 90-95%RH

(Cantwell 2001). The ideal ripening room conditions used at the commercial level are at

19-21C and 90-95%RH with 50 .il ethylene gas per liter of air (Sargent et al. 2002).

Fresh Orange Production and Handling in Florida

Oranges are usually harvested in Florida from mid-September to June, with Navel

oranges harvested until the end of January, and Valencia oranges from the end of January

to the end of June (NASS 2005a). Temperature and relative humidities average 30C

(86F) and 60-90%RH, respectively, during the orange harvesting time. For short-term









transport and storage (7 days), oranges are stored at 7-10C and 85-90%RH without

chilling injury and decay-causing fungal growth occurring (Thompson et al. 2002).

Florida oranges can also be stored at 1-20C without chilling injury occurring. In the

fields, oranges ready for harvest are mostly picked by hand. The laborers place the

oranges into plastic tubs. When filled, a truck called a "goat," which has a hydraulic

boom, picks up these tubs. The boom then picks up the tubs and dumps them into a

container at the rear of the "goat." The goat then takes its full load and dumps it into a

large tractor trailer. The tractor then takes its loads to packinghouse where, after

dumping and trash removal, the oranges are pre-sized, washed, pre-graded, treated with

fungicide and wax, heated-air dried, graded, box-packed and palletized. Unwashed and

unwaxed oranges may be removed for several reasons at the pre-sizing stage in the

packinghouse operation. They may go into storage for later use. Some packinghouses

have de-greening rooms in which unwashed, unwaxed oranges are temporarily stored and

subjected to ethylene in order to impart a more desirable yellow or orange color to the

fruit. Those oranges destined for the processing plant for orange juice are removed prior

to fungicide and wax applications.

Florida's production of orange juice increased in the 1990's compared to the

previous decade. This was due to a direct response to the harsh freeze weather which

decreased production in the prior decade, resulting in the replanting of the affected freeze

areas in the northern areas of the state, increased planting of trees in the southern areas,

and reorganizing existing groves to accommodate a higher orange tree density (Foreign

Agricultural Service [FAS] 1997). Few oranges were processed in Florida prior to the

1931-32 season, and at the end of that decade most of the output (80%) was still geared









for the fresh market (NASS 2005c). But from 1940 to 1970, there was a steady increase

in the percentage of Florida oranges processed, with most going into the frozen

concentrated orange juice (FCOJ), from around 17% to 90% with a steady state being

maintained around 90% or higher, from 1970 onwards (NASS 2005c).

Not-From-Concentrate (NFC) orange juice consumption has also increased in

recent years, due to it's perceived healthier benefits over FCOJ or reconstituted orange

(FAS 2002). NFC was introduced in the 1950's and its production was much smaller

compared to that in the 1990's (Spreen et al. 2001). In the 1999-2000 orange-growing

season, 50% of the Valencia oranges went into NFC production, compared to the 40%

that went into FCOJ (Spreen et al. 2001).














CHAPTER 3
MATERIALS AND METHODS

Initial Preparation of Rifampicin Stock Solution.

A stock solution of 10,000 ppm rifampicin was made by dissolving 1.0 g of

powdered rifampicin antibiotic (MP Biomedicals; Irvine, CA) in 100 ml HPLC-grade

methanol (Fisher Scientific, Fair Lawn, NJ) at room temperature. This solution was

filter-sterilized using a Nalgene vacuum filter unit (Nalge Nunc International,

Rochester, NY) fitted with a 0.45 itm pore-size filter. This stock solution was tightly

capped, covered in aluminum foil, and stored at lab temperature.

Growth Study

Preparations of Rifampicin-resistant Shigella

Growth study experiments were performed separately for S. sonnei ATCC 9290

(American Type Culture Collection, Manassas, VA) from an isolate obtained by Walter

Reed Army Medical Center and S. flexneri LJH 607 (obtained from Dr. L.J. Harris, UC

Davis). These organisms were adapted to rifampicin prior to conducting the growth

studies. In this manner, the recovered rifampicin-adapted .\/nge//a from all subsequent

experiments could survive and proliferate over any rifampicin-non-adapted organism in

tryptic soy agar (TSA) media fortified with rifampicin. The procedures utilized are

described below.

Adaptation of organisms to rifampicin in tryptic soy broth (TSB).

A cryogenically-frozen bead harboring the S. sonnei or S. flexneri was placed in 10

ml of tryptic soy broth (TSB) (BD-Difco, Sparks, MD) in a sterile, Pyrex test tube. The









prepared tube was placed overnight in a shaker-table incubator at 370C and 30 rpm. After

17-24 hr of incubation, 30 [il of the inoculum was aseptically transferred into 10 ml of

freshly-prepared TSB, using a sterile loop (BD, Sparks, MD) and reincubated. A 10-ml

TSB tube containing 2.5 ppm rifampicin (TSB-Rif2.5) obtained from the stock solution

was inoculated with 30 pl from the second incubation. This TSB-Rif2.5 tube was

incubated for 17-24 hr. Successive inoculations and incubations were performed, each

time increasing the rifampicin concentration as follows: 5, 10, 25, 40, 60, 65, 70, 75 and

80 ppm. Once the maximum concentration of 80 ppm rifampicin (Rif80) was reached,

successive cultures were maintained at this level.

Transfer of organisms onto Rif80 tryptic soy agar slants and storage.

Tryptic soy agar (TSA) (BD-Difco, Sparks, MD) with 80 ppm rifampicin (TSA-

Rif80) was used to make spread plates used for maintenance of rifampicin-resistant

cultures. Plates were inoculated with 0.1 ml of. \/ilge// once the culture had reached the

maximum rifampicin concentration as described above. Spread plates were incubated at

37C. Suitable single colonies were chosen and streaked for isolation onto S. sonnei/

boydii plating media (Biosynth, Zurich, Switzerland), and incubated overnight. This

procedure was repeated two additional times to select for vigorous, representative

cultures before final colonies were streaked onto TSA-Rif80 slants (one colony per slant)

in screw-capped, sterile Pyrex tubes. The slant tubes were incubated, loosely capped,

for approximately 8 hr at 370C, then once growth was confirmed on the slants, capped

tightly and transferred into a 40C cooler for storage and later use.

.\l/ige/t were also adapted to 200 ppm rifampicin by the same step-wise adaptation

to increasing concentrations of rif-added TSB. The colony isolation procedure was









repeated. The appropriate 200 ppm rif-adapted colonies were stored on TSA-Rif200

slants at 40C.

Procedure for the Growth Studies

For each organism separately, S. sonnei ATCC 9290 and S. flexneri LJH607, the

following procedure was performed. The TSA-Rif80 or Rif200 slant containing the

required organism was removed from the 4C cooler. A 30-Cl sterile BactoTM loop (BD-

Difco, Sparks, MD) was used to remove an aliquot of the required rifampicin-adapted

./hgel//l, and inoculate a 10-ml TSB tube with the corresponding rifampicin

concentration (either Rif80 or Rif200). This inoculated tube was allowed to incubate

overnight at 370C and 30 rpm. From this tube, two more successive transfers into 10-ml

TSB tubes and incubations were performed. From the third incubated TSB tube, loop

transfers were carried out into three sterile, 125-ml Pyrex flasks, which contained 100

ml of TSB-Rif80. The flask's spout was covered with sterile aluminum foil, and all three

inoculated flasks were placed in the shaker-table incubator set at 370C and 30 rpm.

From each flask, starting from the third hour of incubation, and each hour

thereafter, a 1.0-ml aliquot of culture was serial-diluted into sterile tubes each containing

9.0 ml of phosphate buffer solution (PBS) (MP Biomedicals, Irvine, CA). From each

dilution tube, 1.0 ml was removed, pour plated onto TSA-Rif80 pour plates and

enumerated after incubation at 370C for 24 hr.

Colony counts from 25 to 250 colony-forming units (CFU) on the pour plates were

recorded. In some instances, more or less CFU were also recorded when no other counts

were available. These outlier counts were clearly noted as not to overly bias the data set.

The CFU/ml were then calculated from these counts and dilutions, the logo CFU/ml for

the three flasks at each time increments was determined and a mean population at each









time was calculated. From these values, a plot of logo mean CFU/ml versus incubation

time was performed. The lag, exponential, and stationary growth phases were plotted for

each organism.

Recovery Study

Preparations Prior to the Recovery Study

Acquisition of produce

The tomatoes and oranges used in this study were commercially produced and

obtained directly from the packer. 'Florida 47' tomatoes were picked green in the field

and packed directly into boxes (field-packed) that were delivered to the laboratory

located in Gainesville, FL. Commercially mature oranges were picked by University

personnel under typical harvesting conditions from research groves located at Lake

Alfred and packed into commercial 4/5 bushel bags. No washed or waxed produce were

used in this experiment. About 4 to 6 hr after acquisition of produce from the packers or

field, the produce items were stored in dry, ventilated containers in a dark, 4C/40%RH

cooler, prior to inoculation.

Placement of produce prior to inoculation.

The required number of tomato or orange samples were removed from the cooler

and placed blossom end up on pre-autoclaved, cooled fiberglass trays at lab temperature.

To ensure dry surfaces, the samples were air-dried on the trays overnight at room

temperature.

Preparation of PBS rinsate

Sterile Stomacher bags (Seward, Thetford, UK) were each aseptically filled with

100 ml of buffered peptone water (BPW) at pH 7.0. (BD-Difco, Sparks, MD.), and the









tops sealed with sterile bag clips (Fisher Scientific, Canada). The sealed bags were stored

in a 4C cooler prior to use.

Procedure for the Recovery Study

For each organism separately, S. sonnei ATCC 9290 and S. flexneri LJH607, the

following procedure was performed:

From the TSA-Rif200 slant containing the target organism, three successive loop

transfers into 10-ml TSB-Rif200 tubes were completed and the tubes were shaker-

incubated at 370C and 30 rpm. A sterile, 125-ml Pyrex flask containing 100 ml of TSB-

Rif200 was inoculated with a loop transfer from the final, incubated 10-ml TSB-Rif200

tube. The flask's spout was loosely covered with sterile aluminum foil, placed in the

shaker-table incubator set at 370C and 30 rpm, then removed from the incubator after the

organism had reached stationary phase. This time had been determined from the

previously described growth studies.

Preparation of organism source inoculum

The stationary phase incubate was aseptically removed from the 125-ml flask,

placed in a sterile, tightly-capped, centrifuge Falcon tube (Fisher Scientific, Pittsburg,

PA). The prepared tube was centrifuged at 4,000 rpm for 10 min in a CentraMP4R

centrifuge (International Equipment Company, Needham Heights, MA). The supernatant

was then decanted and discarded. A 10-ml volume of BPW was added to the centrifuge

tube and the pellet resuspended using a Vortex Genie2 vortex (Scientific Industries,

Inc., Bohemia, NY). This procedure was repeated twice followed by a final resuspension

in BPW.









Determination of CFU/ml in the source inoculum

Triplicate 1.0-ml aliquots of the prepared source inoculum were separately serial-

diluted into sterile tubes each containing 9.0 ml of BPW. From chosen dilution tube, 1.0

ml was removed to make the corresponding TSA-Rif80 pour plate. Pour plates were then

incubated at 370C for 24 hr. For each sample, triplicate replicates were performed and

recorded. The CFU/ml were then calculated from these counts and corresponding

dilutions, and the average CFU/ml computed for each sample.

Inoculation onto produce surface

A repeater pipetter (Brinkmann, Westbury, NY) fitted with the 5-100 pl sterile tip

(Brinkmann, Westbury, NY) and set to deliver 10 pl, was aseptically filled with the

source inoculum, according to the manufacturer's directions. Ten 10 [il drops (100 [il

total) were then aseptically delivered in a circular pattern onto the surface surrounding,

and 1.0-2.0 cm away from, the blossom end of each fruit. Care was taken so as to ensure

that the applied drops did not 'run' together. The delivered inoculum was then allowed to

air-dry for approximately 1-2 hr.

Recovery of initial inoculum in BPW rinsate

After the inoculum had dried (about 1.5-2.0 hr), the samples of each fruit were

aseptically placed into the pre-prepared, BPW rinsate-filled, Stomacher bag and

resealed. Each bag contained one fruit for a total often samples. The microorganisms

were recovered using a 'rub-shake-rub' method (Burnett et al. 2001) and the initial

inoculum level was calculated. A 1.0-ml aliquot was removed from the stomacher bag

and serially diluted into sterile tubes each containing 9.0 ml of BPW. For each dilution

tube, 1.0 ml was removed to make the corresponding TSA-RifS0 pour plate. Triplicate

pour plates were made for each dilution tube. Plates were incubated and counted as in









previous procedures. The CFU/ml were then calculated from these counts and

corresponding dilutions, and the average CFU/100 ml computed for each sample.

Survival Study

Inoculated produce were placed in either a Caron 6030 (Marietta, OH) or Barnstead

International Lab-line E-22560-16D (Dubuque, IA) environmental humidity chamber set

at one of the following temperature/relative humidity combinations: 13C/60%RH;

13C/90%RH; 30C/60%RH; or 30C/90%RH. From time 0 (after drying but prior to

placement within the chamber) and other time increments within the chamber, an

appropriate number of samples were removed, each fruit transferred into a 100 ml, PBS

filled Stomacher bag, then inoculum was recovered using the 'rub-shake-rub' method as

previously described for the recovery study. Serial dilutions, pour platings, incubation

and enumeration also followed those performed in the recovery study.

Statistical Analyses

StatisticaTM (Statsoft, Tulsa, OK) was used to analyze the results. Tukey's Honest

Significant Difference (HSD) test (P < 0.05) was utilized to examine data from survival

studies for significance. The recovery studies for both .\l/nge// organisms were analyzed

using the calculated t-statistic to determine whether the two means of the species'

respective initial source and recovered populations were significantly different (P <

0.05).














CHAPTER 4
RESULTS

Growth Study of S. sonnei and S. flexneri

Growth studies were performed for both S. sonnei and S. flexneri in order to

determine the time from initial incubation when the organisms reached their stationary

phases. Typically, stationary phase microorganism are used in order to consistently

obtain a similar inoculum size. The approximate start time of the stationary phase for

each organism was noted and the CFU/ml exhibited at that phase was recorded. The

/nlge/ll spp were adapted to 80 and 200 ppm rifampicin before growth studies were

performed.

Figure 4-1 shows the stationary phases of the rifampicin-adapted ./nhge/lt

organisms. All the organisms show counts above 8.0 logo units at their respective

stationary phases, which commenced after approximately eight hours incubation. The

logo CFU/ml for each time period were averaged over the course of the 8 to 12-hr

incubation for each of the rifampicin-adapted ./nge//t organisms, and the means used to

determine whether significant differences occurred amongst them. The Tukey's Honest

Significant Difference (HSD) test was applied to those means. \/nge//At sonnei adapted to

200 ppm rifampicin grew to significantly higher titers (8.81 logo CFU/ml) than S. sonnei

adapted to 80 ppm rifampicin (8.54 logo CFU/ml) (P < 0.05). However, S. sonnei

adapted to 200 ppm rifampicin was significantly lower than S. flexneri adapted to 80 ppm

rifampicin (8.98 logo CFU/ml), and not significantly different from S. flexneri adapted










to 200 ppm rifampicin (8.74 logo CFU/ml). The lowest recorded mean was noted for S.

sonnei adapted to 80 ppm rifampicin, and the highest recorded titer was seen with S.

flexneri also adapted to 80 ppm rifampicin.


Growth Curves of
Rifampicin-adapted S. sonnei and S. flexneri
grown in TSB-Rif 80.

9.50


9.00 -*- S. sonnei 200 ppm
S-*-S. sonnei 80 ppm
Loglo CFU/ml
-A- S. flexneri 200 ppm
8.50 _-.-- S. flexneri 80 ppm



8.00
8.0 9.0 10.0 11.0 12.0
Time (hr)


Figure 4-1. Growth curves of rifampicin-adapted S. sonnei and S. flexneri grown in
TSB-Rif80 at the stationary phase.

Recovery and Survival Studies for Tomatoes and Oranges

These studies were performed to examine the survival of.\/ilge//t spp. on the

surfaces of unwashed, unwaxed tomatoes and oranges obtained directly from the field.

For the recovery study, the area surrounding the blossom end of each produce type were

inoculated with the species being tested. After drying for approximately 1-2 hr at room

temperature, .\/ige/l/ were recovered from each fruit, using the 'rub-shake rub' method

utilizing Stomacher" bags filled with 100 ml of BPW. The recovered organisms were

then enumerated by serial dilution followed by plating onto TSA-Rif80 plates incubated

at 37C for 24 hr. The average CFU per tomato and orange was recorded.









The survival study utilized the same inoculation procedure used for the recovery

study. However, in addition to recovering and enumerating the ./nge//At from produce

samples immediately after the drying period, the remaining inoculated samples were

placed in environmental humidity chambers. The humidity chambers were pre-set and

pre-equilibrated at the temperature-humidity parameters under which the survival of the

\/nge//lt species was to be observed, prior to placement of the inoculated produce within.

After specific residence times in the humidity chamber, produce samples were removed

and analyzed. Recovery of.\,h/ge//t into BPW-filled Stomacher bags and enumeration

of the .\lhge//At therein were carried out in same manner as with the recovery studies. The

following time periods in Table 4-1 were used to perform the survival studies.

Table 4-1. Organism, temperature/relative humidity combinations, and time periods for
the tomato and orange survival studies.
Tomato Survival Study
Temperature/Relative Time periods for each
Organism .
OrganisHumidity Combinations experiment
130C/60%RH
.1/Nige/,t sonnei 130C/90%RH
ATCC 9290 30oC/60%RH
30C/90%RH
130C/600%RH 0, 1, 2, 4, 6 hr
13C/60%RH
13oC/90%RH
.\1/lge/llflexneri LJH607 30C/60%RH
30oC/60%RH
30oC/90%RH
Orange Survival Study
Temperature/Relative Time periods for each
Organism .
OrganisHumidity Combinations experiment
130C/60%RH
\/gellt sonnei 130C/90%RH
ATCC 9290 30oC/60%RH
30oC/90%RH
130C/60%RH
13C/90%RH
.\/llge/l flexneri LJH607 30C/60%RH 0, 1, 2, 7 days
30oC/90%RH
30oC/90%RH









For both recovery and survival studies, the CFU recovered per tomato and orange

were log-transformed and plotted against time, and the data interpreted from these. For

the purposes of data analysis, plates with no countable colonies were given the value of

1.0, yielding a lower detection limit of 100 CFU/sample or 2.0 logo units using the

formula logo [100*(CFU+ 1)].

Recovery Study for Tomatoes

Both .\ligel // sonnei and i.igell, iflexneri organisms were recovered off the tomato

surfaces in 0.1% buffered peptone water (BPW) adjusted to pH 7.0. Table 4-2 shows the

logo CFU/ml of.\/nge//a applied onto the tomato surface, the logo CFU/ml recovered,

and the logo unit reduction for each species after an average drying time of

approximately 1.0 to 2.5 hr. Using probability tables, the calculated t-statistic and the

degrees of freedom between the species respective initial source and recovered

populations were used to compare with the tabulated t-value at P < 0.05. For each

Table 4-2. Tomato recovery of .\/nge//A spp. from 0.1% buffer peptone water (BPW)
after inoculum dried (1.0 to 2.5 hr).

S. sonnei (Loglo CFU/ml) + (st. dev) recovered from Tomatoes.

Trial (n=3) logo Initial logo Recovered logo Reduction

8.19a+ 0.22 5.35b+ 0.80 2.83

S. flexneri (Logio CFU/ml) (st. dev) recovered off Tomatoes

Trial (n=3) logo Initial logo Recovered logo Reduction

8.69a 0.12 6.91b+ 0.35 1.78
Values are mean + SD of three replications. Different letters (ab) within rows indicate a significant
difference in microbial counts (P < 0.05).









species and their recovered numbers, superscripts with the same letter above their

respective means are not significantly different. Both species recovered were

significantly lower from their initial inoculum sizes. S. sonnei had a 2.83 logo CFU/ml

reduction in recovery and S. flexneri, 1.78 logo CFU/ml post drying.

Recovery Study for Oranges

Using calculated t-statistic and probability tables as performed for the tomato

recovery study, it was found that the logo CFU recovered from orange surfaces were

significantly lower (P < 0.05) from their inoculum sources for both organisms (Table 3-

4). The drying times for the inoculum on the oranges at room temperature were also

approximately 1.0 to 2.5 hr. Both species recovered were significantly lower from their

initial inoculum sizes post-drying. S. sonnei had a 1.46 logo CFU/ml reduction in

recovery and S. flexneri, 1.37 logo CFU/ml.

Table 4-3. Orange recovery of.\/i/ge//A spp. from 0.1% buffer peptone water (BPW) after
inoculum dried (1.0 to 2.5 hr).

S. sonnei (Loglo CFU/ml) + (st. dev) recovered from Oranges.

Trial (n=3) logo Initial logo Recovered logo Reduction

8.73a 0.01 7.27b+ 0.21 1.46

S. flexneri (Logio CFU/ml) (st. dev) recovered off Oranges

Trial (n=3) logo Initial logo Recovered logo Reduction

8.75a+ 0.13 7.38b 0.15 1.37
Values are mean + SD of three replications. Different letters (ab) within rows indicate a significant
difference in microbial counts (P < 0.05).

Survival Study

For both tomatoes and oranges, survival studies of.\/i/ge/lt spp. on fruit surfaces

were conducted, and the studies deemed most appropriate were used to investigate the









temperature-relative humidity effects on the ./ngel//l organisms. The temperature-

humidity conditions used in the survival study were: 130C/60% relative humidity (RH),

13C/90%RH, 30C/60%RH and 30C/90%RH. Loglo CFU/ml values recovered for each

inoculum were transformed as logo [100*(CFU + 1)] CFU/fruit (tomato or orange).

The data was analyzed using ANOVA statistics. The logo CFU/tomato and orange

recovered at each time point were transformed to reflect the logo CFU decline from the

initial, dried inoculum (0 to 6 hr for the tomatoes and 0 to 7 days for the oranges). The 6-

hr and 7-day end-time limits (for tomatoes and oranges, respectively) were selected since

these sampling times were the first to have no detectable growth. The ANOVA analyses

were constructed to show the relationship of the temperature parameter held constant,

while observing the effects of humidity. Using the same data, cross-comparisons,

pertaining to the relationship of the humidity parameter held constant while observing the

temperature effect, were also computed and described. The ANOVA tables showed

significant differences between and among each of the effects (P < 0.05). Tukey's HSD

method was used to compute probabilities for determining significance between groups.

Tomato Survival Study

Shigella sonnei

For tomatoes inoculated with S. sonnei and stored at 130C at 60% and 90%RH

(Table 4-4), there no significant difference (P < 0.05) in total reduction (CFU/tomato)

was seen between 60 and 90%RH at any of the time periods examined.









Table 4-4. Comparison of S. sonnei survival population decline on 'Florida 47' tomatoes
at 13C/60%RH and 13C/90%RH conditions.

S. sonnei; Temperature: 13C
Logio [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr.
Hour 60% RH 90% RH
0 0.00 0.00
1 0.55 (+ 0.56)a 0.44 (+ 0.24)a
2 0.87 (+ 0.53)a 0.76 (+ 0.25)a
4 0.79 (+ 0.50)a 0.85 (+ 0.23)a
6 0.91 (1 0.50)a 0.93 (+ 0.23)a
Values are mean + SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).

The S. sonnei populations held at 300C and observed at 60% and 90%RH (Table 4-

5) presented significantly different population declines at the 2-hr sampling period (0.62

logo CFU at 60%RH and 4.30 logo CFU at 90%RH). For the 60%RH, the S. sonnei

population decline at 2 hr (0.62 loglo CFU) was not significantly greater than those at the

1-, 4- and 6-hr time points. However, for the S. sonnei population at 90%RH, there was a

significant change in population decline from the 1-hr (1.17 logo CFU) to the 2-hr (4.30

logo CFU) time period. S. sonnei was undetected at 90%RH after 4 and 6 hr. A

population decline of 1.59 logo CFU was indicated after 6 hr at 60%RH.

Table 4-5. Comparison of S. sonnei survival population decline on 'Florida 47' tomatoes
at 30C/60%RH and 30C/90%RH conditions.
S. sonnei; Temperature: 30C
Logio [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr
Hour 60% RH 90% RH
0 0.00 0.00
1 1.18 ( 0.50)a 1.17 (0.22)a
2 0.62 (+ 0.54)a 4.30 (A 0.41)b
4 1.40 (+ 0.52)a 4.62 (+ 0.21)b
6 1.59 ( 0.54)a 4.62 (+ 0.21)b
Values are mean + SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).









For S. sonnei populations observed on tomatoes cross-referenced at 60%RH and

compared atl3 and 30C (Table 4-6), both populations indicated no significant population

decline, either between or within populations, from the 1 to 6-hr time period. The

population decrease at 60%RH and 13C at 1 hr (0.55 logo CFU) was not significantly

different from the decline at 2 hr (0.87 logo CFU), 4 hr (0.79 logo CFU) and 6 hr (0.91

logo CFU). Also, the population decrease at 60%RH and 30C at 1 hr (1.18 logo CFU)

was not significantly different from the decline at 2 hr (0.62 logo CFU), 4 hr (1.40 logo

CFU) and 6 hr (1.59 logo CFU). At the end of the 6-hr period, S. sonnei presented a

population decline of 0.91 logo CFU at 130C and 1.59 logo CFU at 300C, which were

not significantly different.

Table 4-6. Cross-comparison of S. sonnei survival population decline on 'Florida 47'
tomatoes at 130C/60%RH and 30C/60%RH conditions.
S. sonnei; Relative Humidity: 60%
Logio [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr.
Hour 130C 30C
0 0.00 0.00
1 0.55 ( 0.56)a 1.18 ( 0.50)a
2 0.87 (+ 0.53)a 0.62 (+ 0.54)a
4 0.79 ( 0.50)a 1.40 (+ 0.52)a
6 0.91 ( 0.50)a 1.59 ( 0.54)a
Values are mean + SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).

For S. sonnei observed on tomatoes at 90%RH and compared at 13 and 300C (Table

4-7), both populations declined significantly (P < 0.05) after 1 hr. The total decline in

population recorded for the 13 and 300C sample groups were 0.44 and 1.17 logo CFU,

respectively. For the 300C trials, there was a significant difference in total reduction at 2

hr (4.30 logo CFU) compared to the 1 hr (1.17 logo CFU) sampling period. No

significant differences were observed for the remaining time periods when compared to









the 2-hr time period. At 13C, there were no significant population decline for the

remaining time periods compared to the 1-hr time period.

When comparing between the 13 and 30C sample groups at 90%RH, S. sonnei

recovered values were significant different at each time for the entire 6-hr period. At 4

and 6 hr, the 30C populations were below detectable limits. At 6 hr the population

decline recorded at 130C was 0.93 logo CFU, which was significantly different from the

below-detectable-limit counterpart at 300C.

Table 4-7. Cross-comparison of S. sonnei population decline on 'Florida 47' tomatoes at
130C/90%RH and 30oC/90%RH conditions.
S. sonnei; Relative Humidity: 90%
Logio [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr.
Hour 130C 30C
0 0.00 0.00
1 0.44 ( 0.24)a 1.17 ( 0.22)b
2 0.76 (+ 0.25)a 4.30 (A 0.41)b
4 0.85 ( 0.23)a 4.62 ( 0.21)b*
6 0.93 ( 0.23)a 4.62 ( 0.21)b
4.62 log reduction = below detectable limits for this data set.
Values are mean + SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).

Shigellaflexneri

The S. flexneri populations were compared at 130C and observed at 60% and

90%RH (Table 4-8). No significant differences in population decline were observed,

either between or within both populations when all the time points from 1 to 6 hr were

compared (P < 0.05), due to the greater variability of the obtained results.









Table 4-8. Comparison of S. flexneri survival population decline on tomatoes at
13C/60%RH and 13C/90%RH conditions.
S. flexneri; Temperature: 13C
Loglo [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr
Hour 60% RH 90% RH
0 0.00 0.00
1 2.25 ( 1.46)a 0.85 (+ 1.66)a
2 3.55 (+ 1.39)a 1.59 ( 1.00)a
4 3.06 ( 1.51)a 1.30 ( 0.74)a
6 2.49 (+ 1.14)a 1.16 ( 0.89)a
Values are mean + SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).

Within each population at their respective relative humidities, there was no

significant difference in population decline from the 2 to 6-hr period. Between S. sonnei

populations, at both relative humidities, there were no significant differences in

population increases observed at 1 hr (0.63 and 1.00 logo CFU at 60 and 90%RH,

respectively). Additionally, for both relative humidities, no significant differences were

observed in population declines at 2 hr (0.12 and 1.05 logo CFU for 60 and 90%RH,

respectively) or at 6 hr (2.28 and 2.78 logo CFU for 60 and 90%RH, respectively).

There was a significant difference noted at 4 hr, with a population increase of 0.50 logo

CFU seen for the 60%RH sample group and a population decline of 1.57 loglo CFU for

the 90%RH trial.

Table 4-9. Comparison of S. flexneri survival population decline on 'Florida 47'
tomatoes at 300C/60%RH and 30C/90%RH conditions.
S. flexneri; Temperature: 30C
Logio [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr.
Hour 60% RH 90% RH
0 0.00 0.00
1 -0.63 (+ 1.53)a -1.00 (+ 1.06)a
2 0.12 ( 1.57)a 1.05 ( 1.27)a
4 -0.50 ( 0.87)a 1.57 (- 1.04)b
6 2.28 (- 0.84)a 2.78 (- 1.07)a
Values are mean + SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).










There was a significant difference, between S. flexneri groups, in survival

population decline at both temperatures cross-compared at 60%RH, at the 1-hr period

(Table 4-10), with the decline at 13C being the greater (2.25 logo CFU) than at 300C.

The population decline at 2 hr at 13C (3.55 logo CFU) was significantly different than

that at 30C (0.12 logo CFU). At the end of 6 hr, there was no significant difference

between the population declines at 13C (2.49 logo CFU) and 30C (2.28 logo CFU).

Within S. sonnei populations at both temperatures, there were no significant differences

in population declines during the entire 6-hr period.

Table 4-10. Cross-comparison of S. flexneri survival population decline on 'Florida 47'
tomatoes at 130C/60%RH and 30C/60%RH conditions.
S. flexneri; Relative Humidity: 60%
Logio [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr.
Hour 130C 30C
0 0.00 0.00
1 2.25 (+ 1.46)a -0.63 (+ 1.53)b
2 3.55 (+ 1.39)a 0.12 (+ 1.57)b
4 3.06 (+ 1.51)' -0.50 (+ 0.87)b
6 2.49 (+ 1.14)a 2.28 (+ 0.84)a
Values are mean + SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).

For the S. flexneri populations at 90%RH and compared at 130C and 30C (Table 4-

11), there were no significant differences found in population declines between the

population groups at each of the 1- to 6-hr time periods. Although at 1 hr a population

gain of 1.00 logo CFU was detected for the species at 300C, this was not significant when

compared to the decline at the 2-hr time point (1.05 logo CFU). No significant

differences in population decline at 300C were observed between the 2-hr, 4-hr and 6-hr

time periods.









Table 4-11. Cross-comparison of S. flexneri survival population decline on 'Florida 47'
tomatoes at 130C/90%RH and 30C/90%RH conditions.
S. flexneri; Relative Humidity: 90%
Loglo [100*(CFU + 1)] (st. dev) decline / tomato from 0 hr.
Hour 130C 30C
0 0.00 0.00
1 0.85 (+ 1.66)a -1.00 (+ 1.06)a
2 1.59 ( 1.00)a 1.05 (+ 1.27)a
4 1.30 ( 0.74)a 1.57 (+ 1.04)a
6 1.16 (0.89)a 2.78 (+ 1.07)a
Values are mean + SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).

Orange Survival Study

Shigella sonnei

At 130C and 60% or 90%RH (Table 4-12), S. sonnei had no significant difference

in their population declines on Day 1, though after seven days there was a marked

difference, with 1.47 logo CFU and 4.71 logo CFU presented at 60%RH and 90%RH

respectively. Within the population at 60%RH, there were no significant differences

between the decline at Day 2 and Day 7. Within the population at 90%RH, there were no

significant differences between the decline at Day 4 and Day 7. However, reductions

seen at Day 1 were significantly different (2.15 logo CFU) than Day 4 (4.12 logo CFU).

Table 4-12. Comparison of S. sonnei survival population decline on oranges at
130C/60%RH and 130C/90%RH conditions.
S. sonnei; Temperature: 13C
Logio [100*(CFU + 1)] (st. dev) decline / orange from Day 0.
Day 60%RH 90%RH
0 0.00 0.00
1 0.95 ( 0.44)a 2.15 ( 0.86)a
4 ND* 4.12 ( 0.93)
7 1.47 ( 0.40)a 4.71 ( 0.57)b
ND = No data recorded for these times.
Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).









At 30C and 60%RH or 90%RH (Table 4-13), there was a significant difference

between population declines at Day 4, with the less decline (4.24 logo CFU) at 60%RH

and populations below detection limits at 90%RH (5.32 logo CFU). No survivors were

detected for either 60 or 90%RH at Day 7.

Table 4-13. Comparison of S. sonnei survival population decline on oranges at
30C/60%RH and 30C/90%RH conditions.
S. sonnei; Temperature: 30C
Logio [100*(CFU + 1)] (st. dev) decline / orange from Day 0.
Day 60%RH 90%RH
0 0.00 0.00
1 ND* 1.56 ( 0.21)
4 4.24 ( 0.12)a 5.32 ( 0.09)b**
7 5.32 ( 0.09)a 5.32 ( 0.09)a
ND = No data recorded for this time.
*5.32 log reduction = below detectable limits for this data set.
Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).

/ngel//lt sonnei populations on oranges held with 60%RH at 13 or 300C is shown in

Table 4-14. The population decline at Day 1 (0.95 logo CFU) was not significantly

different from the declines at Day 7 (1.47 logo CFU) for the 13C population. For the

30C population, the decline at Day 4 was significantly different than that at Day 7 (5.32

logo CFU), when no organism growth could be detected. At Day 7, both the 13 and

30C sample group had significant population declines, 1.47 logo and 5.32 logo CFU,

respectively. When compared to each other after 7 days, the 30C group had a

significantly greater decline than the one observed at 130C.









Table 4-14. Cross comparison of S. sonnei survival population decline on oranges at
13C/60%RH and 30C/60%RH conditions.
S. sonnei; Relative Humidity: 60%
Loglo [100*(CFU + 1)] (st. dev) decline / orange from Day 0.
Day 130C 30C
0 0.00 0.00
1 0.95 ( 0.44) ND*
4 ND 4.24 ( 0.12)
7 1.47 ( 0.40)a 5.32 ( 0.09)b**
ND = No data recorded for this time.
*5.32 log reduction = below detectable limits for this data set.
Values are mean + SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).

With 90%RH at 300C, S. sonnei populations dropped to undetectable levels on

oranges by Day 4 and after, indicated by a 5.32 logo CFU decline. At 130C at the same

%RH, there was a significant population decline in survivors at Day 4 (4.12 logo CFU)

compared to that at Day 1 (2.15 logo CFU), but no significant difference in decline was

observed between Day 4 and Day 7 (4.71 logo CFU).

Table 4-15. Cross-comparison of S. sonnei survival population decline on oranges at
130C/90%RH and 30oC/90%RH conditions.
S. sonnei; Relative Humidity: 90%
Logio [100*(CFU + 1)] (st. dev) decline / orange from Day 0.
Day 130C 30C
0 0.00 0.00
1 2.15 ( 0.86)a 1.56 ( 0.21)a
4 4.12 ( 0.93)a 5.32 (+ 0.09)b**
7 4.71 (+ 0.57)a 5.32 (+ 0.09)a**
5.32 log reduction = below detectable limits for this data set.
Values are mean + SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).

Shigellaflexneri

At 130C, The S. flexneri population declines were significantly less, at all times, in

60% relative humidity conditions than those at 90%RH (Table 4-16). At 90%RH, there

was no significant differences in population declines at Days 1, 2 and 7. At 60%RH,

there were no further significant differences in population decline between Day 1, 2 or 7









readings. At 60%RH, there was no significant difference in population decline when Day

1 was compared to Day 2 (1.06 logo CFU), but it was significant when compared to Day

7 (1.45 logio CFU)

Table 4-16. Comparison of S. flexneri survival population decline on oranges at
13C/60%RH and 13C/90%RH conditions.
S. flexneri; Temperature: 13C
Logio [100*(CFU + 1)] (st. dev) decline / orange from Day 0.
Day 60%RH 90%RH
0 0.00 0.00
1 0.37 ( 0.36)a 3.07 (- 1.04)b
2 1.06 (+ 1.06)a 3.40 (- 0.87)b
7 1.45 (- 0.38)a 4.06 (0 0.84)b
Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).

For the S. flexneri held at 30C, there were significant differences in population

declines at Day 1 (1.80 logo CFU at 60%RH and 3.82 logo CFU at 90%RH) (Table 4-

17). Day 2 declines were significantly different between relative humidities, with 2.60

logo CFU at 60%RH, and the almost complete extinction represented by a decline of

5.97 logo CFU at 90%RH. The population decline at 60%RH on Day 7 (5.66 logo CFU)

was significantly different to that at Day 2. At 90%RH, no survivors were detected on

Day 7.

Table 4-17. Comparison of S. flexneri survival population decline on oranges at
30C/60%RH and 30C/90%RH conditions.
S. flexneri; Temperature: 30C
Logio [100*(CFU + 1)] (st. dev) decline / orange from Day 0.
Day 60%RH 90%RH
0 0.00 0.00
1 1.80 ( 0.38)a 3.82 (- 0.47)b
2 2.60 ( 1.02)a 5.97 ( 0.56)b
7 5.66 (- 0.96)a 6.41 (- 0.08)a*
6.41 log reduction = below detectable limits for this data set.
Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).









At 60%RH, S. flexneri populations were significantly different at 130C compared to

30C at each time period sampled (Table 4-18). For Day 1 the population decline was

0.37 logo CFU at 13C, while a greater decline of 1.80 logo CFU was noted at 300C.

For Day 2, the population decline was 1.06 logo CFU at 130C, while the greater decline

at 30C was 2.60 logo CFU. For Day 7, the population decline was 1.45 logo CFU at

13C, while the greater decline at 300C was 5.66 logo CFU.

Table 4-18. Cross-comparison of S. flexneri survival population decline on oranges at
13C/60%RH and 30C/60%RH conditions.
S. flexneri; Relative Humidity: 60%
Logio [100*(CFU + 1)] (st. dev) decline / orange from Day 0.
Day 130C 30C
0 0.00 0.00
1 0.37 (+ 0.36)a 1.80 (- 0.38)b
2 1.06 ( 1.06)a 2.60 (- 1.02)b
7 1.45 (+ 0.38)a 5.66 (+ 0.96)b
Values are mean + SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).

At 90%RH (Table 4-19), S. flexneri population reductions were not significantly

different at Day 1 at either temperature. Significant differences were observed at Day 2,

with populations more than 2.5 logo CFU lower at 300C, compared to at 130C. No

survivors were detected by Day 7 at 300C (6.41 loglO CFU), compared to a 4.06 logo

CFU reduction seen at 130C.

Table 4-19. Cross-comparison of S. flexneri survival population decline on oranges at
130C/90%RH and 30oC/90%RH conditions.
S. flexneri; Relative Humidity: 90%
Logio [100*(CFU + 1)] (st. dev) decline / orange from Day 0.
Day 130C 30C
0 0.00 0.00
1 3.07 ( 1.04)a 3.82 ( 0.47)a
2 3.40 ( 0.87)a 5.97 ( 0.56)b
7 4.06 ( 0.84)a 6.41 ( 0.08)b*
6.41 log reduction = below detectable limits for this data set.
Values are mean SD of three replications utilizing 10 samples/rep. Different superscripted letters (ab)
within rows indicate a significant difference in population declines (P < 0.05).














CHAPTER 5
DISCUSSION

Consumers moving towards healthier lifestyles are consuming more fresh fruit and

vegetables. Concomitant with this shift in consumption is an increased probability of

foodborne illness from such produce. Produce growing in the fields or moving to the

final consumer undergoes many levels of pre- and post-production processes that present

opportunities for produce contamination. In response, producers and packer/shippers

have implemented and improved measures to assure consumers of a safe product. The

goal is to prevent contamination in the first place; if contaminated, all foodborne bacteria

cannot be easily removed without injuring the product itself. Fruit-handling equipment

has to be regularly cleaned and sanitized so that they don't become a source of

contamination. In recent years, .\/Ngel// has emerged as an important illness-causing

pathogen that has been identified as a produce contaminant, especially in salad-type

preparations. This study observed the survival behavior of S. sonnei and S. flexneri on

the surfaces of unwashed, unwaxed tomatoes and oranges. The inoculated produce were

subjected to environmental conditions of temperature and relative humidity commonly

experienced while growing in the field, or during postharvest storage and transport.

Growth curves were established in order to identify where and at what time the

stationary phases of both occurred. The stationary phase was chosen as a time in the

growth curve that a constant inoculum level could be harvested. Additionally, organisms

in the stationary phase of their growth cycle are more resilient, thus making for an

improved model for the study of bacterial decline. l/nge/ll sonnei (Rif-200) was found









to be not significantly different from S. flexneri (Rif-200) based on Tukey's Honest

Significant Difference (HSD) test (P < 0.05). Both reached stationary phases after 8 hr

incubation with populations at around 8.5 logo CFU/ml (Figure 4-1).

Tomato Survival Study

Shigella sonnei

For S. sonnei-inoculated tomatoes subjected to 130C and 60% or 90%RH, there

were no significant differences (P < 0.05) between populations, as they declined

irrespective of the differing relative humidities and times. The comparison of S. sonnei

populations maintained at 300C and observed at either 60% or 90%RH showed that the

organism survived better at the lower humidity range. S. sonnei populations declined

quickly and were undetected at 90%RH after 4 hr.

For S. sonnei observed on tomatoes and cross-compared at 60%RH at 13 and 30C,

both populations declined equally over the 6-hr period. With the S. sonnei observed on

tomatoes at 90%RH and compared at 13 and 30C, the 13C population declined slower

over the 6-hr period. Thus at 130C the S. sonnei declined slower (survived better) than

those stored at 30C.

At a storage temperature of 13C, relative humidity ceased to be a factor in

influencing survival of the S. sonnei on the tomato surfaces. When comparing 90%RH

and 60%RH stored at 300C, the survival of S. sonnei, only those tomatoes held at

evaluated temperature and relative humidity significantly declined. Beattie and Lindow

(1999) stated that higher humidity enhances bacterial survival by making more free

moisture available to the organism. Conversely, the results of this study indicated that

relative humidity and higher storage temperatures appeared to take precedence in

influencing metabolic activity in S. sonnei, causing the organism to inactivate faster on









the tomato surface. This is similar to the results obtained by Allen (2003) in which it was

observed that lower quantities of Salmonella were recovered off tomato surfaces held at

30C/80%RH, compared to those recovered at 200C/60%RH or 20C/90%RH.

As indicated in Tables 4-6 and 4-7, the total reduction seen for S. sonnei at 90%RH

for 30C was significantly greater than at 90%RH for 13C, or with either temperature at

60%RH, on the tomato surfaces. The higher temperature and relative humidity,

combined with the lack of nutrients on the tomato surface, could have caused S. sonnei to

use up its energy reserves and so die off quicker. Thus temperature/relative humidity

conditions that would normally be more favorable for growth of S. sonnei in the presence

of suitable nutrients, reduces the organism's ability to survive on the tomato surfaces in

the absence of those nutrients. In experiments conducted on parsley, S. sonnei has been

observed to behave similarly (Wu et al. 2000). Whole parsley incubated at 210C,

inoculated with either an initial 6.19 or 3.23 logo CFU inoculum, increased less than 1.0

logo CFU for both inocula after 1 day, which was followed by a subsequent decline in

numbers after 2 days. However, the organism proliferated on chopped parsley incubated

at 21C, from initial inocula of 6.48 and 3.49 logo CFU/g to 9.20 and 6.32 logo CFU/g,

respectively, in 2 days (Wu et al. 2000). These results support the hypothesis that an

increase in moisture and/or nutrients is crucial for bacterial growth on produce.

Inability to survive at a higher temperature for a longer period was also observed

for other organisms inoculated on other produce surfaces. When Listeria monocytogenes

was inoculated onto unwashed carrots at titers of 2.4 logo CFU (incubated at 5C) and

3.0 logo CFU (incubated at 15C), less than 1.0 logo CFU was recovered after the 18 and

7 day sampling periods, respectively (Beuchat and Brackett 1990). However, the









opposite behavior has also been observed with other organisms. From cantaloupe rind

surfaces initially inoculated with 5.2 logo CFU of a 4-strain cocktail ofE. coli 0157:H7

and incubated at temperature/relative humidity conditions of 5C/93%RH and

25C/93%RH, Del Rosario et al. (1995) recovered less than 1.0 logo CFU (after 8 days)

and 7.1 logo CFU (after 21 days), respectively.

The data for this study suggested that the S. sonnei populations survived equally

well at 130C, irrespective of relative humidity. However in contrast, Wu et al. (2000)

found that S. sonnei (inoculated at a high and low concentrations) on both whole (6.19

and 3.2 logo CFU/g) and chopped (6.5 and 3.5 logo CFU/g) parsley declined during a

14-day storage period at 5C. For the whole, inoculated parsley, the S. sonnei population

significantly declined to approximately 4.1 and <1.0 logo CFU/g, respectively. For the

chopped parsley, recovered levels were 3.9 and <1 logo CFU/g, respectively after 14

days of storage (Wu et al. 2000).

Shigellaflexneri

The S. flexneri populations held at either 13 or 30C and observed at 60% or

90%RH (Tables 4-8 and 4-9) were compared. Neither temperature changes from 13 to

30C nor humidity changes from 60 to 90%RH in any of the temperature/relative

humidity combinations had any significant effect on population decline seen on the

tomato surfaces at the 6-hr sampling period. Some early sampling periods showed

significance, though these appear to be an artifact of sample variation. Another study by

Islam et al. (1993) observed S. flexneri survival on chopped cucumbers. It was found that

at 25C, the S. flexneri population increased from an initial inoculum size of about 5.7

logo CFU to approximately 8.5 logo CFU/g after 6 hr, then declined to 7.2 logo CFU/g

after 72 hr (Islam et al. 1993). At 5C, the S. flexneri population remained steady at its









initial inoculum size of about 5.8 logo CFU/g for 72 hr. It is possible that the same trend

might occur with S. flexneri at the two temperatures used in this study (13 and 30C), if

the experiments were performed with chopped or sliced tomatoes that allow nutrients to

be available, thereby making a more favorable environment for bacterial growth.

In this study, both S. sonnei and S. flexneri were studied individually on the tomato

surfaces. Joy (2005) observed that a ./nge//At cocktail population, comprised of both S.

sonnei and S. flexneri, survived longer on tomato surfaces at fall/winter conditions of

27C/90%RH, than at 270C/60%RH. Joy (2005) also concluded that on wounded tomato

surfaces, the i.\/ge/ll cocktail in combination with Erwinia carotova survived least at

temperature/relative humidity conditions of 27C/60%RH, compared to conditions at

27C/90%RH. Joy (2005) postulated that the soft rot caused by the Erwinia, which was

more notable at 90%RH than 60%RH, may increased the availability of nutrients and

increased the pH of the tomato, which is approximately pH 4.5 (Guo et al. 2001). These

factors may have contributed to the survival and proliferation of the ,\/nge/l/

Orange Survival Study

The most notable observation from this study was the increased survival of.h/ige//at

spp. inoculated on orange surfaces as compared to tomatoes. Survival studies conducted

on tomatoes were conducted over a 6 hr time period, whereas studies utilizing oranges

were conducted over 7 days. It is important to emphasis this point to clarify comparison

between tomatoes and oranges. It is also notable to point out that in all cases, .\/nge/lt

spp. inoculated onto orange surfaces that were stored at higher temperatures, were

eliminated quicker.









Shigella sonnei

At the 13C/60%RH and 13C/90%RH, S. sonnei populations had different survival

behavior. l/ge//At sonnei survived better at 60%RH than at 90%RH (Table 4-12). Under

the same conditions, S. sonnei behaved differently than in the tomato survival study, as

no difference in survival behavior was observed at these parameters. It may be that under

the same conditions of temperature and relative humidity, the orange surfaces exhibit

different characteristics than the tomato surfaces, thereby allowing the S. sonnei to

survive better at the lower humidity. Both surface texture and structure of vegetables

play important roles in the attachment and survival of bacteria on them (Kauze and

Joseph 2001). For the survival of S. sonnei on the orange surface, the organism may have

been better shielded from the environment on the rougher orange surface as compared to

the relatively smooth tomato surface.

Table 4-13 showed that at the storage temperature of 30C, there was significantly

less reduction observed at 60%RH (4.24 logo CFU) on Day 4, compared to the 5.32 logo

CFU reduction seen at 90%RH. Stine et al. (2005) found that the inactivation of S.

sonnei on cantaloupe surfaces was unchanged at 22.70C when the relative humidity was

increased from 47.1% compared to 90.3%. However, it was also found that on bell

pepper surfaces, the S. sonnei inactivation at 24.8C/48.8%RH was significantly lower

(1.16 logo CFU/g) than at 24.80C/86.1%RH (1.48 logo CFU/g). Since bell peppers have

similar surface topography to that of tomatoes, it is reasonable to expect survival

characteristics to be similar.

.\/ngel/t sonnei was found in greater numbers on Day 7 at 130C/60%RH (1.47 logo

CFU reduction) than at 130C/90%RH (4.71 logo CFU reduction) (Table 4-12). No

survivors were detected at Day 7 at 300C/60%RH or 30C/90%RH. The overall survival









at 13C was greater than what was observed for 30C, for both relative humidities tested.

As with the previously mentioned study conducted on pepper surfaces, the lower

temperature favored better survival.

The analysis of S sonnei at 60%RH noted better survival at 130C as compared to

30C (Table 4-14). The higher temperature could have caused the S. sonnei to deplete its

nutrients and stores of energy quicker.

At a 90%RH Ssonnei survived better at 130C than at 30C at 4 days, though

differences seen at the Day 7 time period were not significant (Table 4-15). Higher

temperature may have caused the S. sonnei to deplete its nutrients and stores of energy

quicker, resulting in below detectable levels of the organism after 4 days. The S. sonnei

also declined in the 13C group, though not as quickly, reaching a level statistically

equivalent to the 30C group after 7 days. Despite the high relative humidity that would

intuitively support the survival due to an increased availability to water, .\/ngel/l was

effectively reduced from the orange surface in the 7 day study, though not as quickly as

seen in the tomato trials.

The results indicated that as the temperature increased from 13 to 300C, survival of

S. sonnei on the orange surfaces decreased at 60%RH. When the relative humidity was

increased from 60%RH to 90%RH, the population still showed better survival at 13C

compared to 300C.

Shigellaflexneri

.\l/getll flexneri presented similar behavior as that observed for S. sonnei on

orange surfaces. Table 4-16 indicated that for the 13C group at Day 7, the 1.45 logo

CFU decline for 60%RH was significantly less than the decline for 90%RH (4.06 logo

CFU). Table 4-17 indicated that at 300C at Day 2 there was a significantly greater









decline for the 90%RH group (5.97 logo CFU) compared to the group at 60%RH (2.60

logo CFU). It was noted in Table 4-17 that, although there were no significant

differences in decline between groups at Day 7, for the 90%RH, no survivors were

detected, with the 60%RH group presenting a 5.66 logo CFU decline in survivors.

When relative humidity was increased from 60%RH to 90%RH, as seen with the S.

sonnei tests, the S. flexneri exhibited better survival at 130C, compared to 30C. Table 4-

18 indicated that at 60%RH on Day 7 for the 13C group, there was significantly less

population decline (1.45 logo CFU) compared to the 30C group (5.66 logo CFU). In

Table 4-19, at Day 7, the 13C group showed significantly better survival (4.06 logo

CFU decline) than the 30C group, which was below detectable limits.

The ability for S. flexneri to survive better at lower temperatures has been reported

in previous studies. Tetteh and Beuchat (2003) reported that S. flexneri acid-adapted to

pH 4.5 were seen to survive, in TSB acidified to pH 3.5, for 2 hr at 480C, less than 1 day

at 30C, and 6 days at 40C. The acid-adapted S. flexneri showed a 2.5 logo CFU/ml

decrease when held for 6 days at 40C, compared to a 6.0 logo CFU/ml reduction seen

with the unadapted, control cells (Tetteh and Beuchat 2003). Zaika (2001) found that S.

flexneri also survived better at lower temperatures as pH increased. In brain-heart

infusion broth media at pH 4, the organism was undetected after 5, 15, 23, 85 and 85 days

at incubation temperatures of 37, 28, 19, 12 and 4C, respectively (Zaika 2001). In the

same media at pH 3, the survival characteristics were 1, 7, 9, 16 and 29 days for those

same respective temperatures (Zaila 2001). At pH 2, S. flexneri populations dropped to

undetectable levels after 1 to 3 Days at 190C or lower. However, when held at 370C or

28C, S. flexneri populations dropped to undetectable levels after only 2 and 8 hr,









respectively (Zaika 2001). When the media was prepared at pH 5, populations decreased

by 0.5-1.0 logo CFU/ml after 75 days at 40C, were undetectable after 135 days at 120C,

though populations increased rather than decreased at the higher temperatures studied

(Zaika 2001).

Based on the findings of Zaika (2001) and Tetteh and Beuchat (2003), an

alternative hypothesis for the survival of S. flexneri on the orange surface could have

been the organism's ability to adapt to the acidic orange surface, while the initial

inoculum was drying. This adaptation may have allowed it to survive better at 13C

compared to 30C, once the produce was placed in the humidity chamber. This acid

adaptation may have attributed a cross-protection effect to enable the organism to better

survive at 60%RH compared to 90%RH on the orange surface. In contrast, as there were

no significant differences in survival at both temperatures and relative humidities on the

tomato surface, one may infer that no acid adaptation of S. flexneri occurred here.

Results seen for S. sonnei on the orange surface mimic those seen for S. flexneri,

where survival was enhanced at lower temperatures and relative humidities. The results

also suggest that acid adaptation and cross-protection did not occur, though further

research would be necessary to rule out this possibility. In contrast with the observations

of S. flexneri on the tomato surface, S. sonnei only showed significantly better survival at

30C/60%RH compared to 300C/90%RH, and 13C/90%RH compared to 300C/90%RH.

For both organisms, drier field conditions would allow them to survive better on

the orange surfaces, but when the oranges are transferred to storage conditions of

13C/90%RH, the organisms are less likely to survive. There is less chance that field-

contaminated S sonnei or S. flexneri could survive long enough to reach consumers on









surface-contaminated tomatoes as compared to oranges which can survive for up to 7

days. Typically, short storage time for tomatoes at this temperature and relative humidity

is for 48-hr duration, while transit storage time may be for 1 or 2 weeks. Enough time

would thus elapse for the organisms to be extinguished, and so be of limited threat to

end-consumers. In this study where oranges were held at 130C/90%RH, both organisms

approached the least unit of detection after Day 7 (data not shown). Oranges, if no

degreening time is needed, may be packed within 24 hours of harvest and then spend at

least a few days in the transit/distribution chain. If degreening time is required for better

color, the fruit will be held in Florida at approximately 85F and 95% RH with

approximately 2-5 ppm ethylene for 1 to 3 days (sometime more for grapefruit). This

represents only slightly greater .\/igel/h foodborne threat to end-consumers, than is found

for tomatoes.

Also, one must also consider that tomatoes and oranges are consumed differently.

Usually tomatoes are consumed cut into slices or smaller, bite-sized chunks with the skin

intact, such as would be found in salad preparations. Oranges are not consumed with the

skins, which are discarded prior to eating. In this sense, one may perceive a greater threat

of. /ige//,t foodborne illness arising from the consumption of tomatoes compared to

oranges.














CHAPTER 6
CONCLUSION

The growth curves for each rifampicin-adapted .\/lge//t spp. were established, and

their stationary phases noted in order to consistently obtain an appropriate inoculum size

for each to be used in experiments on the produce surfaces. Recovery of.\/ige//At from

the inoculated tomato or orange surfaces, by the 'rub-shake-rub' method, was performed

to ensure consist inoculum recovery.

The tomato survival study indicated that both S. sonnei and S. flexneri behaved

similarly when compared at 130C and observed at 60 and 90%RH. At 13C, both S.

sonnei and S. flexneri showed no significant difference in population survivability,

irrespective of the relative humidities. However humidity significantly affected survival

of S. sonnei when held at 300C and compared at 60 and 90%RH. .\l/gel//t sonnei

survived longer on the tomato surfaces at 60 than at 90%RH. .\/lgel// flexneri survived

equally well on tomato surfaces at all temperature/relative humidity combinations used in

the study.

The orange survival study indicated that both S. sonnei and S. flexneri survived

longer at 60 than at 90%RH for both temperatures of 13 and 30C. Both organisms also

survived longer at 13 than at 300C when cross-compared at either 60 or 90%RH. This

indicated that lower temperature and lower humidity aided in the survival of the .\/lge//t

on the orange surfaces.

Typical storage conditions used in industry for tomatoes and oranges in are 10 and

13C, and 85 and 90%RH, respectively. The results of the survival study indicated that









the storage temperature/relative humidity and times presently used should hinder or

eliminate, rather than enhance, the survivability of.\//ige//t on the tomatoes. Hence,

tomato-associated ./Nige/la outbreaks are most likely not due to the present storage

conditions used in the industry, but rather due to an outside source of contamination, such

as contamination from retail or food service workers with poor personal hygiene,

inoculating at the point of consumption after the produce has been removed from storage

or the retail consumers who may themselves have unhygienic practices while handling

such produce.
















LIST OF REFERENCES


Allen A.B. 1985. Outbreak of Campylobacteriosis in a Large Educational Institution--
British Columbia. Can. Dis. Weekly Rep. 2:28-30.

Allen, R.L. 2003. A Recovery Study of Salmonella spp. from the Surfaces of Tomatoes
and Packing Line Materials. (Master's thesis, University of Florida, 2003).
http://www.uflib.ufl.edu/etd.html. Last accessed on July 18, 2005.

Almed, E.M., F.G. Martin and R.C. Fluck. 1973. Damaging Stresses to Fresh and
Irradiated Citrus Fruit. J. Food Sci. 38:230-233.

Andrews, W. H. and Jacobson, A. 1998. .\lngle/l Bacteriological Analytical Manual.
8th.ed_http://www.cfsan.fda.gov/-ebam/bam-6.html. Last accessed on September 7,
2005.

Bartz, J.A. 1982. Infiltration of Tomatoes Immersed at Different Temperatures to
Different Depths in Suspensions ofErwinia carotovora subsp. carotovora. Plant
Dis. 66: 302-306.

Bartz, J.A. and R.K. Showalter. 1979. Postharvest Water Intake and Decay of Tomatoes.
Citrus and Vegetable Magazine. 3(44):7 & 28.

Bagamboula, C.F., M. Uyttendaele and J. Debevere. 2002. Acid Tolerance of.\lngell/,
sonnei and .ige/ll, iflexneri. J. Appl. Microbiol. 93:479-486.

Beuchat, L.R. 1996. Pathogenic Microorganisms Associated with Fresh Produce. J. Food
Prot. 59(2):204-216.

Beuchat, L.R. 2002. Ecological Factors Influencing Survival and Growth of Human
Pathogens on Raw Fruits and Vegetables. Microbes and Infection. 4:413-423.

Beuchat, L.R. and J.H. Ryu. 1997. Produce Handling and Processing Practices. Emerg.
Infect. Dis. 3:459-463.

Beuchat, L.R. and R.E. Brackett. 1990. Survival and Growth ofListeria monocytogenes
on Lettuce as Influenced by Shredding, Chlorine Treatment, Modified Atmosphere
Packaging and Temperature. J. Food. Sci. 55(3):755-758, 870.

Beattie, G.A. and S.E. Lindow. 1999. Bacterial Colonization of Leaves: A Spectrum of
Strategies. Phytopathol. 89(5):353-359.









Blostein, J. 1993. An Outbreak of Salmonella javiana Associated with Consumption of
Watermelon. J. Environ Health. 56(1):29-31.

Brackett, R.E. 1987. "Microbiological Consequences of Minimally Processed Fruits and
Vegetables." J. Food Qual. 10:195-206.

Buchanan, R.L., S.G. Edelson, R.L. Miller and G.M. Sapers. 1999. Contamination of
Intact Apples after Immersion in an Aqueous Environment Containing Escherichia
coli 0157:H7. J. Food Prot. 62(5):444-450.

Burnett, A.B. and L.R. Beuchat. 2001. Comparison of Sample Preparation Methods for
Recovering Salmonella from Raw fruits, Vegetables and Herbs. J. Food Prot.
64(10):1459-1465.

Burnett, S.L., J. Chen and L.R. Beuchat. 2000. Attachment of Escherichia coli 0157:H7
to the surfaces an Internal Structures of Apples as Detected by Confocal Scanning
Laser Microscopy. Appl. and Environ. Microbiol. 66:4679-4687.

Calvin, L. 2003. Produce, Food Safety, and International Trade: Response to U.S.
Foodborne Illness Outbreaks Associated with Imported Produce. Chapter 5. In: J.
Buzby (Ed.), International Trade and Food Safety: Economic Trade and Case
Studies. USDA, Economic Research Service, AER-828.
http://www.ers.usda.gov/publications/aer828/. Last accessed on July 8, 2005.

Cantwell, M. 2001. Properties and Recommended Conditions for Storage of Fresh Fruits
and Vegetables. In: Postharvest Technology.
http://postharvest.ucdavis.edu/Produce/Storage/propty.shtml. Last accessed on
May 11, 2005.

Carter, R.D. 1989. Technical Manual: Fresh-Squeezed Florida Orange Juice
Production/Packaging/Distribution. Florida Department of Citrus, Scientific
Research Department, University of Florida, Lake Alfred, FL.

Centers for Disease Control and Prevention [CDC]. 1991. Multi-State Outbreak of
Salmonellapoona Infections United States and Canada, 1991. MMWR. 40:549-
552.

Centers for Disease Control and Prevention [CDC]. 1998. Outbreak of Campylobacter
enteritis Associated with Cross-Contamination of Food--Oklahoma, 1996.
MMWR. 47:129-31.

Centers for Disease Control and Prevention [CDC]. 1999. FoodNet Surveillance Report
for 1999 (Final Report).
http://www.cdc.gov/foodnet/annual/1999/pdf/FoodNet 1999Annual Report.pdf
Last accessed on July 10, 2005.









Centers for Disease Control and Prevention [CDC]. 2000. FoodNet Surveillance Report
for 2000 (Final Report).
http://www.cdc.gov/foodnet/annual/2000/2000finalreport.pdf Last accessed on
July, 10, 2005.

Center for Food Safety and Applied Nutrition U.S. Food and Drug Administration
[CFSAN-USDA]. 2001. http://www.cfsan.fda.gov/-comm/ift3-4o.html. Last
accessed on April 27, 2005.

Cliver, D.O. 1997. Virus Transmission via Food. Food Technologist. 51:71-78.

Council for Agricultural Science and Technology (CAST). 1994. Foodborne Pathogens:
Risks and Consequences. Task Force Report no. 122. September 1994. pp. 12.

De Roever, C. 1998. Review: "Microbiological Safety Evaluations and
Recommendations on Fresh Produce." Food Control. 9:321-347.

de Sim6n, M. C. Tarrag6 and M.D. Ferrer. 1992. Incidence ofListeria monocytogenes in
Fresh Foods in Barcelona (Spain). Int. J. Food Microbiol. 16:153-156.

Del Rosario, B.A. and L.R. Beuchat. 1995. Survival and Growth of Enterohemorrhagic
Escherichia coli 0157:H7 in Cantaloupe and Watermelon. J. Food Prot. 58(1):105-
107.

Downes, F.P. and K. Ito (eds.). 2001. Microbiological Examination of Foods.
Washington: Sheridan Books, Inc. pp. 381-382.

Duffy, E.A., L. Cisneros-Zevallos, A. Castillo, S.D. Pillai, S.C. Rick and G.R. Acuff.
2005. Survival of Salmonella Transformed to Express Green Fluorescent Protein on
Italian Parsley as Affected by Processing and Storage. J. Food Prot. 68(4):687-695.

Economic Research Service [ERS]. 2002. Oranges: The Most Consumed Fruit in
America.
http://www.ers.usda.gov/Briefing/FruitAndTreeNuts/fruitnutpdf/oranges.pdf Last
accessed on July 12, 2005.

Economic Research Service [ERS]. 2004a. Commodity Highlight: Fresh Tomatoes. U.S.
Department of Agriculture.
http://www.ers.usda.gov/Briefing/Vegetables/vegpdf/FrTomatoHigh.pdf. Last
accessed on July 13, 2005.

Economic Research Service [ERS]. 2004b. Impact of Greenhouse Tomatoes on the Fresh
Field Tomato Industry. U.S. Department of Agriculture.
http://www.ers.usda.gov/publications/err2/err2g.pdf Last accessed on July 13,
2005.









Economic Research Service [ERS]. 2004c. Tomatoes: Background. U.S. Department of
Agriculture. http://www.ers.usda.gov/briefing/tomatoes/background.htm. Last
accessed on July 13, 2005.

Foreign Agricultural Service [FAS]-USDA. 1997. World Horticultural Trade and U.S.
Export Opportunities. http://www.fas.usda.gov/htp2/circular/1997/97-
08/aug97cov.htm. Last accessed on July 14, 2005.

Foreign Agricultural Service [FAS]-USDA. 2002. Recent Developments in the World
Orange Juice Trade and the U.S. Competitive Position.
http://www.fas.usda.gov/htp2/circular/2000/00-02/ojspecial.htm. Last accessed on
September 18, 2005.

Fisher, T.L. and D.A. Golden. 1998. Fate of Escherichia coli 0157:H7 in Ground Apples
Used in Cider Production. J. Food Prot. 61(10): 1372-1374.

Food Safety and Inspection Service [FSIS]. 2000.
http://www.fsis.usda.gov/OA/background/2000fsi.htm. Last accessed on July 7,
2005.

Francis, G.A., C. Thomas and D. O'Beirne. 1999. The Microbial Safety of Minimally
Processed Vegetables. Int. J. Food Sci. Technol. 34:1-22.

Golden, D.A., E.J. Rhodehamel and D.A. Kautter. 1993. Growth of Salmonella spp. In
Cantaloupe, Watermelon and Honeydew Melons. J. Food Prot. 56:194-196.

Guo, X., J. Chen, R.E. Brackett and L.R. Beuchat. 2001. Survival of Salmonellae on and
in Tomato Plants from the Time of Inoculation at Flowering and Early Stages of
Fruit Development through Fruit Ripening. Appl. Environ. Microbiol. 67(10):4760-
4765.

Gupta, A., C.S. Polyak, R.D. Bishop, J. Sobel and E.D. Mintz. 2004. Laboratory-
Confirmed Shigellosis in the United States, 1989--2002: Epidemiologic Trends and
Patterns. Clin. Infect. Dis. 38:1372-1377.

Guzewich, J.J. and P. Salsbury. FDA's Role in Traceback Investigations for Produce.
Food Safety Magazine. 2001.

Heisick, J.E., D.E. Wagner, M.L. Nieman and J.T. Peeler. 1989. Listeria spp. on Fresh
Market Produce. Appl. Environ. Microbiol. 55:1925-1927.

Ho, J.L, K.N. Shands, G. Friedland, P. Eckind and D.W. Fraser. 1986. An Outbreak of
Type 4b Listeria monocytogenes Infection Involving Patients from Eight Boston
Hospitals. Arch. Intern. Med. 146:520-523.

Joy, J.A. 2005. Survival of Salmonella and .nhlge//l on Tomatoes in the Presence of the
Soft Rot Pathogen, Erwinia Carotovora. (Master's thesis, University of Florida,
2003).









Islam, M.S., M.K. Hasan and S.I. Khan. 1993. Growth and Survival of ./ngel//tflexneri
in Common Bangladeshi Foods under Various Conditions of Time and
Temperature. Appl. Environ. Microbiol. 59(2):652-654.

Kaneko, K.I., H. Hayashidami, Y. Ohtomo, J. Kosuge, M. Kato, K. Takahashi, Y. Shiraki
and M. Ogawa. 1999. Bacterial Contamination of Ready-to-Eat Foods and Fresh
Products in Retail Shops and Food Factories. J. Food Prot. 62:644-649.

Kauze, A.S. and H. Joseph. 2001. Quantitative Determination of the Role of Lettuce Leaf
Structure on Protecting E. coli 0157:H7 from Chlorine Disinfection. J. Food Prot.
64:147-151.

Keller, S.E., S.J. Chirtel, R.J. Merker, K.T. Taylor, H.L. Tan and A.J. Miller. 2004.
Influence of Fruit Variety, Harvest Technique, Quality Sorting, and Storage on the
Native Microflora of Unpasteurized Apple Cider. J. Food Prot., 67(10):2240-2247.

Kimura, A.C., K. Johnson, M.S. Palumbo, J. Hopkins, J.C. Boase, R. Reporter, M.
Goldoft, K.R. Stefonek, J.A. Farrar, T.J. Van Gilder and D.J. Vuglar. 2004.
Multistate Shigellosis Outbreak and Commercially Prepared Food, United States.
Emerg. Infect. Dis. 10(6): 1147-1149.

Lampel, K.A. 2001. .\/ngIe/I In: Downes, F.P. and K. Ito. (Eds.). Compendium of
Methods for the Microbiological Examination of Foods (4th ed., pp. 381-384).
United States of America: Sheridan Books, Inc.

Lampel, K., R.C. Sandin and S. Formal. 1999. .\l/ge/ll Introduction and Detection by
Classical Cultural Techniques. In: Robinson, R.K., C.A. Batt and P.D. Patel (Eds.).
2000. Encyclopedia of Food Microbiology. 3:2015-2020.

Lang, M.M., L.J. Harris and L.R. Beuchat. 2004. Evaluation of Inoculation Method and
Inoculum Drying Time for Their Effects on Survival and Efficiency of Recovery of
Escherichia coli 0157:H7, Salmonella, and Listeria monocytogenes Inoculated on
the Surface of Tomatoes. J. Food Prot. 67(4):732-741.

Liao, C.K. and P.H. Cooke. 2001. Can. J. Microbiol. 47:25-32.

Lin, B., J.N. Variyam, J. Allshouse and J. Cromartie. 2003. Food and Agricultural
Commodity Consumption in the United States: Looking Ahead to 2020.
USDA/ERS, AER-820. http://www.ers.usda.gov/publications/aib792/aib792-
7/aib792-7.pdf. Last accessed on July 13, 2005.

Lund, B.M. and A.L. Snowdon. 2000. Fresh and Processed Fruits, Chapter 27. In: B.M.
Lund, T.C. Baird-Parker and G.W. Gould (Eds.). The Microbiological Safety and
Quality of Food, Volume I. Gaithersburg (MD): Aspen. pp. 738-758.

Mead, P.S., L. Slutsker, V. Dietz, L.F. McCaig, J.S. Bresee, C. Shapiro, P.M. Griffin and
R.V. Tauxe. 1999. Food-related Illness and Death in the United States. Emerg.
Infect. Dis. 5:607-617.









Morbidity and Mortality Weekly Report (MMWR). 2005. 52(54):1-85.
http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5254al.htm. Last accessed on
September 18, 2005.

Mukherjee, A; D. Speh; E. Dyck and F. Diez-Gonzales. 2004. Preharvest Evaluation of
Coliforms, Escherichia coli, Salmonella, and Escherichia coli 0157:H7 in Organic
and Conventional Produce Grown by Minnesota Farmers. J. Food Prot. 67(5):894-
900.

National Agricultural Statistics Service [NASS]-USDA. 2005a. Florida Agricultural
Facts. http://www.nass.usda.gov/fl/rtocOv.htm.

National Agricultural Statistics Service [NASS]-USDA. 2005b. Weekly Weather and
Crop Bulletin. 92:37.
http://usda.mannlib.cornell.edu/reports/waobr/weather/2005/full/wwcb3705.pdf.
Last accessed on September 20, 2005.

National Agricultural Statistics Service [NASS]-USDA. 2005c. Frozen Concentrated
Orange Juice. In: Trends in U.S. Agriculture.
http://www.usda.gov/nass/pubs/trends/concentratedoj.htm. Last accessed on July
10, 2005.

O'Brien, S., R.T. Mitchell, I.A. Gillespie and G.K. Adak. 2000. The Microbiological
Status of Ready-to-Eat Fruit and Vegetables. Discussion paper ACM/476 of the
Advisory Committee on the Microbiological Safety of Food.
http://www.foodstandards.gov.uk/pdf files/papers/acm476.pdf.

O'Brien, S., R.T. Mitchell, I.A. Gillespie and G.K. Adak. 2000. The Microbiological
Status of Ready-to-Eat Fruit and Vegetables. Discussion paper ACM/510 of the
Advisory Committee on the Microbiological Safety of Food.
http://www.food.gov.uk/multimedia/pdfs/acm510A.pdf.

Olsen, A.R. 1998. Regulatory Action Criteria for Filth and Other Extraneous Materials.
III. Review of Flies and Foodborne Enteric Disease. Reg. Toxicol. Pharmacol.
28:199-211.

Parish, M.E.1997. Public Health and Nonpasteurized Fruit Juices. Crit. Rev. Microbiol.
23:109-119.

Rafi, E. and P. Lunsford. 1997. Survival and Detection of .\/ge//atflexneri in Vegetables
and Commercially Prepared Salads. J. Assoc. of Anal. Chem. Intl. 80:1191-1197.

Ryu, C.-H., S. Igimi, S. Inoue and S. Kumagai. 1992. The Incidence of Listeria Species
in Retail Foods in Japan. Int. J. Food Microbiol. 16:157-160.

Samish, Z., R. Etinger-Tulczynska and M. Bick. 1963. The Microflora Within the Tissue
of Fruits and Vegetables. J. Food Sci. 28:259-266.









Sandeep, M, D. Aggarwal and A. Ganguli. 2004. Microbiological Analysis of Street-
vended Fresh Squeezed Carrot and Kinnow-Mandarin Juices in Patiala City, India.
Internet Journal of Food Safety, V.3, 1-3.
http://www.foodhaccp.com/internetjoumal/ijfsv31.pdf Last accessed on July 27,
2005.

Sapers, G.M. 2001. Efficacy of Washing and Sanitizing Methods. Food Technol.
Biotechnol. 39(4):305-311.

Sargent, Steven A., and C.L. Moretti. 2002. Tomato. In: The Commercial Storage of
Fruits, Vegetables, and Florist & Nursery Crops (3rd ed.).
http://www.ba.ars.usda.gov/hb66/138tomato.pdf. Last accessed on September 18,
2005.

Schelch, W.F., P.M. Lavigne, R.A. Bortolussi, A.C. Allen, E.V. Haldane, A.J. Wort,
A.W. Hightower, S.E. Johnson, S.H. King, E.S. Nicholls and C.V. Broome. 1983.
Epidemic Listeriosis--Evidence for Transmission by Food. N. Engl. J. Med.
308:203-206.

Seo, K.H. and J.F. Frank. 1999. Attachment of Escherichia coli 0157:H7 to Lettuce
Leaf Surface and Bacterial Viability in Response to Chlorine Treatment as
Demonstrated by Using Confocal Scanning Laser Microscopy. J. Food Prot. 62:3-
9.

Shiferaw, B., S. Shallow, G. Kazi, S. Segler, D. Soderlund, T. Van Gilder and the EIP
FoodNet Working Group. 2000. .liigell,/ Then and Now: Comparing Passive
Surveillance for Shigellosis in Five FoodNet Sites, 1996-1998. 2nd International
Conference on Emerging Infectious Diseases. Atlanta, GA.

Simies, M., B. Pisani, E.G.L. Marques, M.A.G. Prandi, M.H. Martini, P.F.T. Chiarini,
J.L.F. Antunes and A.P. Nogueira. 2001. Hygienic-Sanitary Conditions of
Vegetables and Irrigation Water from Kitchen Gardens in the Municipality of
Campinas, SP. Braz. J. Microbiol. 32:331-333.

Spreen, L.T., W. Fernandes, Jr., C. Moreira and R.P. Muraro. 2001. An Economic
Evaluation of Hamlin versus Valencia Orange Production in Florida. Department of
Food and Resource Economics, Florida Cooperative Extension Service, Institute of
Food and Agricultural Sciences, University of Florida, Gainesville, FL.
http://edis.ifas.ufl.edu/BODY_FE300. Last accessed on September 18, 2005.

Stine, S.W., I. Song, C.Y. Choi and C.P. Gerba. 2005. Effect of Relative Humidity on
Preharvest Survival of Bacterial and Viral Pathogens on the Surface of Cantaloupe,
Lettuce, and Bell Peppers. J. Food Prot. 68(7):1352-1358.









Swerdlow, D.L., K. D. Greene, R. V. Tauxe, J. G. Wells, N. H. Bean, A. A. Ries, P. A.
Blake, E. D. Mintz, M. Pollack, M. Rodriguez, E. Tejada, L. Seminario, C.
Ocampo, B. Vertiz, L. Espejo and W. Saldana. 1992. Waterborne Transmission of
Epidemic Cholera in Trujillo, Peru; Lessons for a Continent at Risk. Lancet
340(4):28-32.

Takeuchi, K. and J.F. Frank. 2000. Penetration of Escherichia coli 0157:H7 into Lettuce
Tissues as Affected by Inoculum Size and Temperature and the Effect of Chlorine
Treatment on Cell Viability. J. Food Prot. 63:434-440.

Tetteh, G.L. and L.R. Beuchat. 2003. Survival, Growth, and Inactivation of Acid-Stressed
.\/ge//lltflexneri as Affected by pH and Temperature. Int. J. Food Microbiol. 87(1-
2):131-138.

Tetteh, G.I., S.K. Sefa-Dedeh, R.D. Phillips and L.R. Beuchat. 2004. Survival and
Growth of Acid-adapted and Unadapted .h/i/gel/tflexneri in a Traditional
Fermented Ghanian Weaning Food as Affected by Fortification with Cowpea. Int.
J. Food Microbiol. 90:189-195.

Thompson, J., A. Kader and K. Sylva. 2002. Compatibility Chart for Fruits and
Vegetables in Short-term Transport or Storage. University of California Division
of Agriculture and Natural Resources. Publication 21560.
http://postharvest.ucdavis.edu/Pubs/postthermo.shtml. Last accessed on July 14,
2005.

Todd, E. 1989. C.D. Preliminary Estimates of Costs of Foodborne Disease in the United
States. J. Food Prot. 52(8):595-601.

Wu, F. M., M.P. Doyle, L.R. Beuchat, J.G. Wells, E.D. Mintz and B. Swaminathan.
2000. Fate of .\/lge//At sonnei on Parsley and Methods of Disinfection. J. Food Prot.
63:568-572.

Zaika, L.L. 2001. The Effect of Temperature and Low pH on Survival of .\/nge//htflexneri
in Broth. J. Food Prot. 64(8):1162-1165.

Zaika, L.L. 2002. Effect of Organic Acids and Temperature on Survival of .\/nge//A
flexneri in Broth at pH 4. J. Food Prot. 65:1417-1421.















BIOGRAPHICAL SKETCH

Dirk Sampath was born in the town of Siparia, Trinidad, in the twin-island

Republic of Trinidad and Tobago, West Indies, on November 15, 1957. He graduated

from the University of Florida with his Bachelor of Science in nutritional science, in

1992. He then returned to Trinidad and worked on his parent's family-owned farm. In

1995 he returned to Orlando, Florida, and then moved to Athens, Georgia. His venture

there led him to work in the poultry industry for GoldKist. He then returned to the

University of Florida in 2002 to pursue his Master of Science in food science and human

nutrition. After graduating, Dirk plans to be employed in the food or food-related

industry.