1 USE OF POST HARVEST PROCESSES TO REDUCE Vibrio parahaemolyticus IN RAW OYSTERS: BLAST FREEZI NG AND HEAT SHOCK By LEANN HELDT-WIEAND MANLEY A THESIS PRESENTED TO THE GADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGEE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008
2 Leann Heldt-Wieand Manley
3 To my husband James I dedicate this work. Without his encouragement, advice, and love I would not have run this course.
4 ACKNOWLEDGEMENTS My husband Jam es Manley has been the kind and guiding motivation behind this achievement. I never imagined that I would be writing an acknowledgement for a thesis, and because of him, his encouragement and academic advice throughout this arduous endeavor, it is finished. This degree also belongs to my mother, Katherine Wieand Budman, who showed me by example that a masters degree is within my reach. Her pursuit of academic achievement was with many obstacles, all of which she overcame, setting a perfect model for me to follow. I also thank my late Grandmother, Emma Boyer Wiea nd, who insistently pushed me to pursue higher education. Her prayers and encouragement were not fruitless and because of her, the idea of even pursuing a masters degr ee entered my thoughts. I wish to thank my major prof essor, Dr. Steven Otwell, dire ctor of this project: and lab managers Charlene Burke, Melissa Evans and Vi ctor Garrido for their assistance, support and patience during my masters program. With much admiration, I wish to send gr atitude to Tommy Ward, Dacky Ward, Smokey, Webbs Seafood and to all of the oystermen who are the backbone of this industry! Thanks are also due to the members of my supervisory committee, Dr. Shirley Baker and Dr. Anita Wright, for their advi ce and assistance during my research. I also appreciate the support and help from my lab mates Jennette, Mi lan, Mike H, Mike M, Don, Delores and Jeffrey who made the work enjoyable.
5 TABLE OF CONTENTS page ACKNOWLEDGEMENTS.............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES.........................................................................................................................8 ABSTRACT...................................................................................................................................10 CHAP TER 1 MICROBRIAL PATHOGENS IN OYSTERS...................................................................... 20 Characteristics of Vibrio parahaemolyticus ...........................................................................23 Epidemiology and Pathogenesis...................................................................................... 24 Prevalence........................................................................................................................28 Physiology.......................................................................................................................29 Detection and Enumeration Methods.............................................................................. 30 Real time PCR Detection................................................................................................. 31 Post-Harvest Processes....................................................................................................34 Research Hypothesis and Objectives...................................................................................... 37 2 MATERIALS AND METHODS........................................................................................... 39 Oyster Harvesting and Handling............................................................................................ 39 Oyster Thermal Intervention and Transportation to PHP .......................................................40 Blast Freezing..................................................................................................................42 Heat Shock.......................................................................................................................42 Analytical Methods for V. parahaemolyticus Enum eration................................................... 42 Confirmatory MPN Molecular tlh P robe Protocol.......................................................... 44 PCR Real Time Detection Methods................................................................................ 44 Statistical Analysis.......................................................................................................... 46 3 RESULTS...............................................................................................................................51 Initial Screening......................................................................................................................51 Thermal Intervention........................................................................................................... ...51 Post Harvest Processes...........................................................................................................52 PCR Analysis..........................................................................................................................53 4 DISCUSSION AND CONCLUSION.................................................................................... 59 Initial Screening and Thermal Intervention............................................................................ 63 Preliminary Work Involving Bacterial Competition....................................................... 64 Thermal Intervention Issues............................................................................................ 65
6 PCR Analysis..........................................................................................................................66 Conclusions.............................................................................................................................68 LIST OF REFERENCES...............................................................................................................70 BIOGRAPHICAL SKETCH.........................................................................................................78
7 LIST OF TABLES Table page 4-1 Temperature and salinity measurements of Apalachicola Bay (11 m ile area, station 350)....................................................................................................................................544-2 Mean log MPN g-1 levels of V. parahaemolyticus in Apalachicola Bay for the initial screening and the subsequent thermal intervention (Pre-PHP) for all trials. 3 samples per trial, per level, to obtain the means. SD = standard deviation.....................................554-3 Heat shock post harvest proce ss validated at day 7 by reducing V. parahaemolyticus levels by >3.52 log. Neither the blast freezing post harvest process nor the control reduced V. parahaemolyticus > 2.4 log. All units are in mean log MPN g-1. SD = standard deviation..............................................................................................................564-4 One-tailed T test for V. parahaemolyticus reduction per treatment, trial and interval. Ho = 3.51 log reduction; Ha 3.52. P value must be < .05 for significance* ( .05).....56
8 LIST OF FIGURES Figure page 1-1 Florida oysterman tonging for oysters. (P icture courtesy of Victor Garrido). ............... 193-1 Oyster handling and enumeration per trial three trials were performed for each PHP method and control............................................................................................................ 483-2 Microbial analysis per sample of V. parahaemolyticus at initial screening and after thermal intervention........................................................................................................... 493-3 Enumeration steps per sample for the cont rols and post harvest treated samples after -20C storage for 7 and 14 days.........................................................................................504-1 Map of Apalachicola Bay FDACS illustra ting their water testi ng stations. Testing station 350 is the station where the water sa linity and temperature was sampled, as it is the closest to the two harvesti ng areas used in this project............................................ 544-2 Mean levels of V. parahaemolyticus ; Pre-PHP (thermal intervention), PHP day 7 and PHP day 14 for heat shock, blast and the controls. The heat shock post harvest process validated ( 3.52 reduction) at day 7.................................................................... 554-3 Comparison of the mean reduction in V. parahaemolyticus of the heat shock samples between the FDA tlh Probe, BAX real time PCR, and the Cepheid SmartCyclerIITM real time PCR.................................................................................................................. ...574-4 Ellab TracksenseTM time (hours) and temperature (Cel sius) data of the blast freezing and heat shock post harvest processes............................................................................... 58
9 LIST OF ABBREVIATIONS AP probe Alkaline Phosphate probe APW Alkaline Peptone Water ASW Artificial Sea Water BAM Bacteriological Analytical Manual CFU Colony Forming Unit CPS Capsular Polysaccharide CT Cholera Toxin FDA Food and Drug Administration FDACS Florida Department of Ag riculture and Consumer Services LPS Lipopolysaccharide LSD Least Significant Difference mCPC modified Cell obiose-Polymixin B-Colistin MPN Most Probable Number PBS Phosphate Buffered Saline PCR Polymerase Chain Reaction PHP Post Harvest Process PHT Post Harvest Treatment ppt Parts per thousand TCBS Thiosulfate Citrate Bile salts Sucrose T1N3 Tryptone 10%, NaCl 3% VBNC Viable But Non-Culturable YSI Yellow Springs Instruments
10 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 USE OF POST HARVEST PROCESSES TO REDUCE Vibrio parahaemolyticus IN RAW OYSTERS: BLAST FREEZI NG AND HEAT SHOCK By Leann Heldt-Wieand Manley August 2008 Chair: Name W. Steven Otwell Major: Food Science and Human Nutrition In addition to the discomforts of human illne ss that can result from the consumption of raw oysters with elevated levels of Vibrio parahaemolyticus, such events result in large economic losses due to closed oyster harvesting ar eas and reduced sales of oysters, ultimately leading to significant adverse consequences for the industry that needs additional and optional controls. The problems are caused by V. parahaemolyticus, which is a Gram negative, halophilic, facultative anaerobe that has a probable infective dose of 106 and is relatively resistant to reduction in oysters. This study investigated blas t freezing and heat shock as unique post harvest processes (PHP) that could be used to reduce or eliminate V. parahaemolyticus in oysters destined for raw consumption. The goal was to reduce V. parahaemolyticus by 3.52 logs. Based on previous validation studies invol ving various post harvest processes to reduce Vibrio vulnificus it was reasoned that blast freezing or heat shocking could be used as an effective post harvest process to reduce V. parahaemolyticus. The blast freezing process involved product exposure to cold air (-29C) blown at 25 m.p.h for 8 hours followed by continual storage at 20C. The heat shock pr ocess involved immersion of raw oysters into 65C water for 5 minutes immediately followed by an ice slush immersion, then continual storage at 20C. Our hypotheses were that bl ast freezing (-29C forced air) and heat shock (5
11 min. at 65C), will individually cause 3.52 log reduction of V. parahaemolyticus in raw oysters, C. virginica to yield a safer product for raw cons umption. Likewise, it was hypothesized that there will be no significant di fference between the specified FDA tlh probe results and the multiplex real time PCR results in enumeration of V. parahaemolyticus in PHP oysters. All tests were conducted with oysters, Crassostrea virginica harvested between July 16 and September 10 in Apalachicola Bay, Florid a. The raw oysters were enumerated for V. parahaemolyticus ; directly upon harvesting (i nitial screening), after thermal intervention (to normalize bacterial loads), and af ter 7 and 14 days frozen (-20 C) storage following each post harvest process. Three trials were completed at different times for each PHP using 8 bushels of oysters per trial. The specified FDA protocols (MPN and Direct Plating) were used to enumerate the V. parahaemolyticus for the environmental and thermally intervened levels. The specified FDA MPN (most probable number) protocol was used to enumerate all the post harvest processing levels. Additionally, a multiplex real ti me PCR assay was performed and compared to the tlh probe results in one heat shock trial to determine the accuracy of this PCR enumeration for V. parahaemolyticus The heat shock process provided > 3.52 l og reduction evident after 7 days. The blast freezing process, which exceeded -32C, only reduced V. parahaemolyticus levels ~2 log by day 14 which did not reach the expected 3.52 log reduction necessary for the FDA validation. Continued storage time (210 days) for the blast process only imparted an additional reduction of 0.5 log. Additionally, the tlh confirmation of V. parahaemolyticus levels of the heat shock samples from the one trial did not differ significantly from a multiplex real time PCR assay. This real time PCR enumeration was a tlh, tdh, trh multiplex assay performed with a SmartCyclerIITM (Cepheid).
12 CHAPTER 1 OYSTERS AS A FOOD The first evidence for oysters used as a food in North Am erica is provided by the Lenape Indians of the present day New York. The Lena pe valued oysters as a food, they harvested oysters for food and used their shells for a vari ety of everyday needs. Archaeologists have found and preserved middens, piles of oyster shells, and their heights represent a crude estimate of the Lenapes consumption of oysters. The middens, ofte n as high as four feet are noted as markers of pre-European settlements at the mouth of the Hudson River. The Lenape offered Henry Hudson his first taste of these bivalves in the ear ly 1600s. So it is as early as the 1500s that there is evidence of oyster consumpti on in the United States (Kurlansky, 2006). Opening or shucking the oyster surely was a c hore for the Lenape, as it is a difficult task today with the tools used in commerce. An oys ter can exert ~22 pounds of pressure with the abductor muscle to hold its shells tightly closed (Churchill, 1920) The Lenapes tool of choice is assumed to be flint rock, as they origina lly did not have metal shucking knives until the European settlers arrived (Kurlansky, 2006). The Europeans provided the knife and the Lenape taught them how to prepare the oyster; other than eati ng them raw, the Lenape prepared oysters by wrapping them in seaweed and cooking them over a fire until they opened (Kurlansky, 2006), possibly the first effort at post harvest processing. At first, oysters were not shi pped; those who did not reside by the water did not eat oysters. In August of 1807, Robert Fulton made his first run with the Clermont the first steamboat, from Manhattan to Albany. It was not long after this initial voyage that many other major cities were receiving passengers, goods, and produce from the Manhattan area by steamboat. In this same year, enormous quantities of oysters were sh ipped to upstate New York and to Europe (Kurlansky, 2006). Between 1828 and 1861, 64 steamboats listed Apalachicola, Florida as their
13 home port with more than twice that number active on the Apalachicola river system (Chapel, 2007). Consumer access led to a greater demand for oysters. By the mid 1800s, oyster harvesting constituted the most valuable fishery in the United States. According to Parks (1985), Americans in the late 1800's were enveloped in a "great oyster craze," noting: "No evening of pleasure was complete without oys ters; no host worthy of the name failed to serve 'the luscious bivalves,' to his guests. In every town there were oyster parlors, oyster cellars, oyster saloons, and oyster bars, houses, stalls, and lunchrooms." Oysters and oyster meats alike were sold from house to house in cities by street peddlers (Freeman, 1989; Kurlansky, 2006). By the post Civil War era oysters were considered such a valuable commodity that oyster wars ensued. Hired gun boats were employed by private companies to pr otect private oyster beds with some states employing an oyster navy to keep order (King, 2004). The annual yield in the United States in 1910 was ~ 30,000,000 bushels, with a monetary return to the fishermen of $15,000,000 (Churchill, 1920). The growing demand for oysters was fed by their nutritional value. In a Turtox News reprint, Curtis Newcombe from the Virginia Fish eries Laboratory stated, For reasons of health, people need to be taught to eat oy sters, just as they are taught to eat calcium-rich milk or eat vitamin-rich vegetables. This is true, for one pound of oysters provides an adult approximate daily amounts of 12% energy, 50% protein, 26% calcium, 40% phosphorous, > 184% of iron and 110% iodine (Newcombe, 1946). And not to forget th e claim that oysters ar e an aphrodisiac, Dr. George Fisher at Barry University in South Florida announced in March, 2005 at an American Chemical Society meeting that he had found tw o compounds in oysters, D-aspartic acid and Nmethyl-D-aspartate, that were effective in releasing the sexual horm ones testosterone and
14 estrogen. This sounds promising, but through numerous medical literature searches, this was the only finding that appeared at all valid for this claim. Currently, nearly 99% of the oys ters harvested in the United States come from the Atlantic and Gulf coasts (NOAA). Since the 1990s Louisian a has produced 42% of the total US harvest (King, 2004), while the Chesapeake produced a mere 2% (King, 2004). Specifically, C. virginica is an important resource in the Gulf of Mexico where production represents 50% of the U.S. total harvest (FDA CFSAN, 2005). The total catch re ported for this species to FAO for 1999 was 145,733 tons. The countries with the largest catches in 1999 were USA (98,892 tons) and Mexico (43,285 tons). Combined global aquaculture and cap ture production of C. virginica is ~ 200,600 tons (FAO, 2000). Oysters are members of the family Ostrea cea, class Bivalvia, in the phylum Mollusca. The three main species of oysters consumed in the United States are the native Eastern oyster, Crassostrea virginica ; the native Olympia oyster, Ostreola conchaphila (ranging from Alaska to Mexico); and the Pacific oyster, Crassostrea gigas (the basis of a large industry from Alaska to Mexico, although mainly in Washi ngton state) (MacKenzie, 1996). C. virginica, the oyster species used in this study, is the most consumed species of oyster in the US. Otherwise known as the American Oyster, these bivalves live as sess ile organisms, cemented to the substrate on their left valve (Cunningham, 1885). This left shell be comes misshapen due to the attachment and grows slightly larger and more cupped. The asymme trical nature of their shells, and cupped left shell, serves as a means to contain their liquor (or blood/hemolymph) (Brooks, 1891). This asymmetrical cupped left shell also affords the oy ster industry a simple and cost effective means to keep oysters alive during shipping (Churchill, 1920).
15 Oysters ecological niche includes estaurine and sound environments with salt levels between 0.5 and 1.5%, but they are intolerant of prolonged exposure to fresh water or marine salinities (Galtsoff, 1964). For instance, when sali nity is high, oysters are susceptible to oyster drills (a snail), crabs, an d a protozoan parasite called Perkinsus marinus or more commonly called dermo (MacKenzie, 1989). Oyster drills al one are capable of kil ling 85 percent of the young oysters on a reef. They are f ound in shallow areas of tidal to subtidal zones (between 0.575 meters), and prefer a firm substrate such as pilings, hard rock bottoms, and substrates hardened with the oyster shells of previous generations (Galtsoff, 1964). These epi-benthic filter feeder s, ingest a variety of algae, bacteria, and small detrital particles. Fecal and pseudofecal material is important in sediment production and deposition, providing sites for remineralizing bacterial action and a food source for deposit feeders (Churchill, 1920; Galtsoff, 1964). The oyster eat s by opening their shell slightly and filtering food from the surrounding water through their gills with cilia action (Churchill, 1920). Oysters convert aquatic microbes and nutrien ts into an eatable meat as they filter the water (Brooks, 1891). Under ideal conditions an oyster can filter up to 34 liters of water per hour (Galtsoff, 1964) and what they selectively f ilter becomes what we consume. The fact that oysters are filter-feeding organi sms presents a health problem to those who enjoy raw oysters. As an oyster filter feeds it accumulates any bacteria, algae, viruses and contaminants in their intestinal tract. Oysters do have a met hod of selective feeding, but the detection of the selected part icles still is not well understood. The labial palps serve as a mechanism of a selectivity that sorts particulat es as they pass through the palps and enter the mouth (Churchill, 1920; Brooks, 1891).
16 Oysters do have a basic immune system to wa rd off pathogens. Their immune system is composed of hematocytes that phagocytose path ogens, or form cysts around larger infectious agents such as parasites. They also have hum oral immunity that includes lectins, which are composed of carbohydrates that tag a pathogen fo r removal by the hematocytes or agranulocytes (Genthner et al ., 1999). In Apalachicola Bay it was found th at oysters immune systems become more active in the warmer months when th e bacterial levels are elevated (Fisher et al ., 1996). Conversely, in the Chesapeake Bay it was found that oysters immunity (hemolymph lysozyme level) is higher in the winter months (Chu and La Peyre, 1989). One reason for the summer months heightened immunity is that possibly on e or more of the four phases of phagocytosis (attraction, attachment, internaliz ation, and intracellular degradatio n) is more active because the oysters are at a higher me tabolic rate (Genthner et al ., 1999). But, these are suggestions, and evidence for these rates of imm unity are not yet clear. Because bacteria and viruses are ubiquitous in an oysters environment they are a problem in raw shellfish. This is the problem of oysters as a food. More specifically, it is a problem for those who enjoy raw oysters on the half shell. Although the safe oyster proverb only eat raw oysters during the mo nths with an r originated as a warning about English oyster larvae (Mackenzie, 1996), it serves a useful purpo se in the U.S. by encouraging consumption of oysters during the cooler months when they are microbiologically safer to consume. This saying holds true to some extent, but is no guarantee for safe consumption. Consumers who wish to enjoy raw shellfish create a challenge to harves ters and distributors who aim to provide a safe seafood product. Consumer safety is greatly increased if raw oysters receive some type of post harvest process (PHP) to reduce the loads of potential pathoge ns (FDA CFSAN, 2005). Most Vibrio
17 infections arise from the consumption of raw or undercooked shellfish (FDA CFSAN, 2005). Effective and affordable post harvest processes ar e needed to assist the industry and ensure the safety of the consumer. Post harvest processi ng equipment is extremely expensive in both the initial cost and its high maintenance requirements; regular testing is needed to insure the apparatus is operating to specification and mainta ining the FDA required low levels of bacteria in the end product. Initial attempts at post harvest processing be gan with the first shucking and packaging of oysters that started in Connecticut in the ea rly 1800s, where families would hand-shuck the oysters at home. Then the meats were taken to a packing house and packed in little wooden kegs or in metal tins surrounded with ice. Oyster s were first canned in Baltimore in the 1820s (Churchill 1920). By the late 1800s the oysters were shucked and packaged at the packing houses (MacKenzie, 1996). Washing these oysters was accomplished by blowers, which were 200 gallon freshwater tanks fitted with tubing by whic h air could be blown in to stir up the meats to remove the mud and debris off of the meats before packing (Churchill, 1920). Although it was not until 1961 that the earliest re ported work on the effect of heat on the microbiological quality of sh ellfish occurred (Dombroski et al ., 1999), the industry had employed heat in post harvest for a century. As early as 1858 Louis Mu rray introduced oyster scalding. Although this process wa s designed to facilitate openi ng the oyster, now we are aware of the food safety implications. This initial effort has become known as heat shocking or heat treatments. Henry Evans invented a proce ss that consists of placing the oysters in cars of iron framework, 6 to 8 feet [1.8 to 2.4 m] long, and holding about 20 bushels of unshucked oysters, and the cars are run on a track from the wharf to a steam-tight box, ranging from 15 to 20 feet [4.6 to 6.1 m] long, and fitted with a ppliances for admitting the steam at any desired pressure, and a door at each end of the box permitting the
18 entry of the car, and then so arranged that the doors can be closed, thus making a practically air-tight compar tment. The steam is turned on for about 15 minutes, the chest is then opened and the cars run into the shucking shed, where employees, each provided with a knife, are able to separate very easily the oysters from the shell. After they are steamed and shucked they are washed in cold wa ter and sent to the fillers' table. Here they are placed in cans, weighe d and hermetically sealed. The cans are then put into a cylindrical basket and lowered into the process kettle, in which they are steamed to a su fficient degree to kill all germs of fermentation. After coming from the process kettle, they are cooled in a large vat of cold water and then tr ansferred to the la beling and packing department. The total cost of ha ndling a bushel of oysters in the Baltimore canneries has been estimated at 29 cents, while the average price during recent years of a bushel of oysters for the canning trade has been about 55 cents (Hunt 1903)." As early as 1885, in Apalachicola, Florida, John G. Ruge and his brother established the Ruge Brothers Canning Company and they became Flor idas first successful pasteurized commercial oyster packers (Chapel, 2008). Another substan tial contribution for safer oyster storage and shipping came out of Apalachicola, Florida in 1851; John Gorrie wa s granted the patent for the first ice machine (Chapel, 2008). This invention had a huge impact on food storage and safety as a whole; ice did not have to be shipped from the Northern regions. Public perception of oyster safety has long been an issue. In 1906, in the heat of concern over food safety, the U.S. Congress passed several "Pure Food Laws" (Churchill, 1920) which completely changed oyster handling, packing, and shipping methods. He alth officials and journalists vented a flood of cr iticism about the lack of cleanlin ess of oyster beds and industry handling practices (MacKenzie, 1996). Cases of typhoid poisoning and ga strointestinal trouble were blamed on the consumption of oysters simply if the patient had admitted eating them. Even today, some consumers are simply unaware of their own compromised health status when consuming raw oysters despite the requi red posted warnings in raw bars: There is an associated risk with co nsuming raw oysters. If you have chronic illness of the liver, stomach or blood, or ha ve immune disorders, you are at greater
19 risk of serious illness for raw oysters, and should eat oysters fully cooked. If unsure of your risk, consult a physician (Florida Administrative Code). Subsequently, their sickness has led to media hype and ultimately concerns over whether raw oysters are actually a safe food (Stephenson, 1994). Even a few cases of V. parahaemolyticus can cause substantial economic losses to the shellfish industry. The combined efforts of the FDA, the ISSC (Interstate Shellfish and Sanitation Commission) and members of the shellfish industry have developed guidelines to validate a post harvest process for the reduction of microbial pat hogens in shellfish. In order to validate a post harvest process for V. parahaemolyticus, a reduction from 10,000 to < 3 mean log MPN g -1 (a 3.52 log reduction) must be obtained (ISSC, 2007). Consumers of shellfish, especially immunocompromised individuals, are better served as more post harvest processes are validated. Figure 1-1. Florida oysterman tonging for oysters (Picture courtesy of Victor Garrido).
20 CHAPTER 2 MICROBRIAL PATHOGENS IN OYSTERS The m embers of the microbial family Vibrionaceae are characterized as Gram negative rods that are halophilic and facultatively anae robic. There are more than 30 species of Vibrio and at least 12 of them are associated with foodborne illnesses (Chakraborty et al ., 1997). The main Vibrio pathogens associated with the consumpti on of oysters, through documented occurrence of illness, are Vibrio vulnificus Vibrio cholerae, and Vibrio parahaemolyticus. They are ubiquitous in estuarine waters where oysters thrive, and pose more of a threat as the waters warm during the summer months (> 20C water temperature). Thes e bacteria are also present in the winter months, possibly just in a VBNC stat e (viable but non-culturable) (Baffone et al ., 2006). During 2005-2006, an 18-month study carried out in the waters of the Adriatic Sea detected these three Vibrio with PCR methods, but not with culturebased methods indicating a VBNC state. There are three bio-types of pathogenic V. vulnificus a naturally occurring estaurine bacterium which can be isolated in molluscan shellfish. The CDC first identified V. vulnificus in 1976 (Oliver and Kaper, 2001). Biotype 1 is al most exclusively known to infect humans (gastroenteritis, primary septicemia and wound inf ections) and thus poses th e greatest concern to the seafood industry. This biotype causes a number of different types of infe ctions in humans and it is known to have high phenotypic variation (Johnston et al, 1986; Hlady, 1997; Oliver and Kaper 2001). The factor once thought to be most a ssociated with the cause of shock and death in humans infected with V. vulnificus is the cells LPS (lipopol ysaccharide), or endotoxin (McPherson et al ., 1991). Now studies have shown a posit ive relationship between the increased expression of CPS (capsular polysacchari de) and virulence (Chatzidaki-Livanis et al ., 2006; Yoshida et al ., 1985). It has been found that V. vulnificus with a translucent phenotype form more biofilm (Joseph et al. 2004); however, the opaque morphology is found to be responsible
21 for disease (Chatzidaki-Livanis et al ., 2006). It is theorized that the expression of CPS may increase as V. vulnificus enters the host and a morphotype ch ange may occur (Chatzidaki-Livanis et al ., 2006). V. vulnificus is associated with severe and freque ntly fatal infections in immunosuppressed people (Dombroski et al ., 1999). From 1981 through 1992, 125 persons with V. vulnificus infections were reported to the Florida Department of Health and Rehabilit ative Services of those 44 died (MMWR, 1993). The most severe human disease syndrome caused by V. vulnificus is primary septicemia, associated with consumption of raw oysters. Severe wound infections can result if compromised dermal tissue comes into contact with V. vulnificus in saline coastal water. Despite the fact that septic V. vulnificus infections are relatively rare, significant attention is focused on monitoring and controlling this foodborne illness. The infective dose for V. vulnificus in oyster meat shown to induce septicemia in immunocompromised people is estimated to be 100 bacteria/g (FDA CFSAN, 2005). It is well established that post harvest processes are very e ffective at reducing V. vulnificus in raw oysters, as both heat shock and blast fr eezing processes have been FDA validated (V. Garrido unpublished). The time required to reduce viable V vulnificus cells 1 log (D value) is 1.3 min. .41 min. at 46C and 48 C respectively. At 50C, V. vulnificus dies so quickly that death curves cannot be attained (Dombroski et al ., 1999). V. cholerae is the bacterium that causes cholera. V. Cholerae is unique among Vibrios in the fact that it can grow agar without added Na Cl (CDC, 1993). Importa nt distinctions within V. cholera are made based on whethe r or not a strain of V. cholera can produce cholera toxin (CT). Additionally distinctions are made based on two serogroups, O1 and O139, associated with epidemics of Cholera, although not all of the stra ins of those serotypes cause human disease, or
22 produce CT. The basis for serotyping V. cholera is the LPS somatic antigen. The O1 serogroup has long been associated with epidemic and pandemic cholera; while the O139 Bengal serogroup was first recognized during an epidemic in eastern India and Bangladesh in 1993 (CDC, 1993). Humans with acute cholera excrete up to 108 V. cholerae per g of feces, and the symptom of explosive diarrhea is a mechanis m for the bacteria to find its wa y back into the water supply. The incubation period ranges from 3 hours to 5 days (Levine et al., 1981). If a patient with cholera is not treated with antibiotics, they can excrete the bacteria for up to tw o weeks even after the symptoms stop (Bennish, 1994). Protein phosphorylati on in the infected leads to increased Clsecretion and this results in decreased NaCl absorption and ultimately water that is not reabsorbed by the GI tract is excreted as explos ive diarrhea. Transmission of cholera is primarily the fecal-oral route, and often through contaminated water. So me of the major foods associated with transmission of O1 are crabs, shrimp, raw fish and oysters. In the US there were only 7 cases of serotype O1 from Gulf Coast seaf ood between 1996-2005; the majority of these cases were due to foreign travel (FDA CFSAN, 2005). Infectious doses of V. cholerae needed to consistently cause dia rrhea in volunteers is 1011 CFU; if the stomach pH is buffered with antacids then that dose is reduced to 104 cfu. (Levine et al., 1981). Two Vibrios which compete with V. parahaemolyticus in enumeration methods are Vibrio alginolyticus and Vibrio mimicus. V. alginolyticus grows into numerous bright yellow colonies on TCBS (Thiosulfate Citrate Bile salts Sucr ose). At more concentrated dilutions (10 -1 and 10 2), V. alginolyticus can outcompete V. parahaemolyticus on TCBS, rendering itself more than a simple annoyance while attempting to enumerate V. parahaemolyticus This competition mainly occurs in environmental samples during the warmer summer months. G.C. Fletcher et al., (1985) group found that there remains a need to develop an enrichment medium that will effectively
23 eliminate V. alginolyticus without adversely affecting th e survival and growth of V. parahaemolyticus. The other competitor during V. parahaemolyticus enumeration is V. mimicus which presents as blue-greenish colonies on T CBS. This bacterium is sometimes accidentally mistaken as V. parahaemolyticus and dotted during transfer from the TCBS to the T1N3 (10% Tryptone, 3% NaCl) plates. It has been demonstrated that TCBS cannot differentiate V. parahaemolyticus from some strains of V. vulnificus and Vibrio mimicus (Su and Liu, 2007). Unconfirmed species of competitive bacteria gr owing as bright yellow colonies on TCBS complicated the start of this study in early July 2007 as it outcompeted V. parahaemolyticus for resources; in addition, blue green competitiv e colonies presented on TCBS consistently throughout the study. Characteristics of Vibrio parahaemolyticus V. parahaemolyticus is a Gram negative, motile, halophilic, non-spore forming facultative anaerobe that lives in oysters. It follows that the ecological niche of V. parahaemolyticus is one of an estuarine organism: me sophile range 8.3C 45.3C, (Miles et al ., 1997) optimal 35C 37C; pH range 4.8 11, op timal 7.5 8.6; NaCl range 0.5 10%, optimal 3%. V. parahaemolyticus in the Gulf thrives in the warmth of the summer months. A study by Gooch (2000) shows that when oysters are exposed to ambient air temperature of 26C (78.8F) V. parahaemolyticus levels increases by 1.9 logs CFU g -1 after 10 hours in June, and increases by 2.4 logs cfu g -1 after 10 hours in July (Gooch, 2000). A more conservative figure for their regeneration ability is an increase of 13 26 fo ld in oysters exposed to 26C for > 24 hour period (Kaufmann et al ., 2002). One problem for the oyste r industry is the increase in Vibrio parahaemolyticus loads upon harvesting. Once removed from the water, bacterial loads in oysters can increase (Cook et al ., 1999; Gooch, 2000). A study by Cook et al ., (1999) shows that the ambient air temperature of the oysters is not significantly increase d upon harvesting, and
24 exposure to air, the increase in bacterial loads appears to be caused by th e fact that the oysters have stopped filter feeding (Cabello et al ., 2005). Epidemiology and Pathogenesis V. parahaemolyticus was first discovered in Japan in 1950 when it caused a m ajor outbreak of 272 illnesses and 20 deaths associated w ith sardines (FDA CFSAN, 2005). The first documented outbreak in the US occurred in Ma ryland in 1971; the food medium was steamed crabs. There were twelve other outbreaks of V. parahaemolyticus between 1969 and 1972 involving oysters. In Washington State during 1990 there were 30 confirmed cases resulting from eating raw oysters. The largest outbreak wa s in Galveston Bay, Texas during 1998 resulting in 416 confirmed cases due to the O3:K6 tdh + serotype (MMWR, 2006b). V. parahaemolyticus guidelines vary throughout the wo rld. In the United States, the 2005 FDA risk assessment of V. parahaemolyticus states that there is a 0.001% chance of a healthy person suffering symptoms if exposed to a 104 CFU dosage. The infective dose is estimated to be 106 CFU. If that dose is increased to 108 CFU, then the probability of illness increases to 50% (FDA CFSAN, 2005). Japans limits are similar in that <102 V. parahaemolyticus MPN/g is permissible in seafood for raw consumption; the te mperature of Japans seafood is required to be maintained at < 10C throughout its distribution and storage (FAO/WHO, 2003). Additionally, after harvesting and during seafood preparation, fish and shellfish are washed with disinfected seawater or potable water. Denmark exercises some import c ontrols for seafood from non-EU countries, examining about 50% of ready-to-eat seafood for V.parahaemolyticus and other Vibrio species, and sporadically testi ng raw and frozen seafood as well. Denmark also sets a limit on V. parahaemolyticus at < 102. Several European co untries reject raw seaf ood if any level of V. parahaemolyticus species are detected (FAO/WHO, 2003).
25 Any, healthy or immunocompromised, person who consumes an infectious dose of V. parahaemolyticus can become infected, ultimately deve loping symptoms of gastroenteritis. Symptoms of a V. parahaemolyticus infection include diarrhea (occasionally bloody), nausea, vomiting, fever and chills with a duration of 1 3 days (FDA CFSAN, 2005). The incubation time is between 4 96 hours; once the bacteria attach to the intestinal trac t infection starts and symptoms appear. This disease is self-limiti ng and best treated by administering fluids. Antibiotics are usually unnecessary, but in ve ry severe cases tetracycline, ampicillin or ciprofloxacin are used (MMW R, 2006b). For the immunocompromised, and those with underlying medical conditions, there is a great er risk, and a greater probability that gastroenteritis may develop into septicemia a nd potentially death (FDA CFSAN, 2005). The atrisk population in the U.S. is estimated by the Ce nter for Science in the Public Interest (CSPI) to be at around 60 million in 1997, and this popula tion includes those who have: peptic ulcer disease, diabetes, liver disease, hematological disease, immunodeficiency, alcoholism, gastric surgery, heart disease, cancer, renal disease and transplant recipien ts (FDA CFSAN, 2005). In order to identify V. parahaemolyticus with nucleic acid methods there are three genes typically targeted: tlh, tdh and trh. The thermolabile hemolysin ( tlh) gene is the targeted gene for general identification of V. parahaemolyticus. A pathogenic virulence factor for this bacterium is the tdh (thermostable direct hemolysin) ge ne. The gene has low occurrence in V. parahaemolyticus populations, but causes most of the health consequences in those infected with V. parahaemolyticus (Miyamoto et al ., 1969). This tdh gene occurs in over 90% of the clinical strains isolated worldwide (Baffone et al ., 2006). It is believed that tdh+ strains multiply more efficiently in human intestines and tdhstrains multiply more efficiently in the environment (Gooch, 2000). It is known that most strains of V. parahaemolyticus sampled from the
26 environment or seafood are not pathogenic, as the ratio of tdh to tdhis ~ 1:10000 (.01%) (Su and Liu, 2007). This isolate is rarely found in environmental strains or in food (Cook et al ., 2002b). In the Gulf Coast region the FDA states the occurrence of the tdh gene to be at .02% and, considerably higher (.2%) in the Northwes t Region. Another route of infection is through compromised epidermis; if this bacterium has access to an open wound, infection will occur (FDA CFSAN, 2005). The second virulence gene targeted is the trh (thermostable direct related hemolysin) gene. This gene ( trh ) produces a hemolysin toxin that was first shown to be lethal in mice at high concentrations by Sochard and Colwe ll in 1977 and at lower concentrations it was shown to cause diarrhea (Yeung and Boor 2004). Trh and tdh are similar in that they have 70% nucleotide sequence identity (Nishibuchi et al ., 1989). Sochard and Colwell (1977) also demonstrated that trh is heat labile to 60C for 10 minutes which was an identifying factor in its discovery, as tdh is heat stable. Interestingly, recent studies have shown ev idence of horizontal gene transfer of tdh among the Vibrios ; as other Vibrio species than V. parahaemolyticus reacted positively. In a study by Bej et al. Vibrio hollisae, Vibrio mimicus and V. cholera non 01 reacted positively with a tdh probe (Bej et al ., 1999). Baffone et al showed that a V. vulnificus sample produced amplicon with tdh primers and also that V. alginolyticus strains produced amplicon with trh primers (Baffone et al ., 2006; Nordstrom et al. 2007). Strains of V. parahaemolyticus are typically serot yped on the basis of the O (somatic) and K (capsule) antigens. There are as many as 13 O groups and 71 K types that can be identified by commercial antisera (Yeung and Boor 2004). Recently, the O4 strains have been associated with gastroenteritis cases (DePaola et al., 2000; Nolan et al., 1984). A surveillance study in Calcutta, India demonstrated that there is a lack of general association be tween illnesses and serotypes.
27 Multiple diverse serovars were identified from various infected patients (Okuda et al., 1997). In 1996, however, the infection rate of V. parahaemolyticus increased in India along with the association of the O3:K6 serotype isolated from the sick (Okuda et al ., 1997; Ramamurthy and Nair, 2005). These serotype characteristics were shown to be nearly identical to trh negative and tdh positive strains identified in India and Southeast Asia. Since it takes a very low number of pathogenic cells to make a human ill, th is strain is difficult to enumerate Additionally, other non-pathogenic strains mask th e presence of pathogenic cells. O4:K68 and O1:KUT are two strains which have been reported as genetically similar to O3:K6 isolates, as they share similarly arbitrarily primed PCR finge rprints (Yeung and Boor, 2004). Virulence capability of a strain can be altere d through the transfer of genetic elements both horizontally and vertically betw een bacterial populat ions (Waldor and Mekalanos, 1996). There is evidence that the hemolysin gene is mobile among bacterial populations. Tdh has been demonstrated to exist on both plasmid and chromosomal DNA (Nishibuchi and Kaper, 1985). PCR results have shown that trh and tdh primers produce amplicon in samples of V. alginolyticus and V. vulnificus respectively (Baffone et al ., 2006). Additionally, tdh was shown to produce amplicon in Vibrio hollisae (Bej et al ., 1999). DNA hybridization using the tdh molecular probe revealed positive hybridization of nonparahaemolyticus strains of Vibrios such as V. hollisae, V. mimicus and V. cholerae non-01 (Nishibuchi and Ka per, 1985) (Nishibuchi et al ., 1986). It is generally accepted that tdh and trh are virulence genes for V. parahaemolyticus alone. These virulence factors have been identified for V parahaemolyticus although an in depth understanding of this bacterias ability to cause disease remains largely unknown (Yeung and Boor 2004).
28 Prevalence V. parahaemolyticus is an em erging pathogen. In 2006, there were 154 laboratory confirmed cases of Vibrio food borne infections in th e United States (MMWR, 2006b). Of those 154, 64% were confirmed as V. parahaemolyticus and 12% were confirmed as V. vulnificus. Subtyping of V. parahaemolyticus isolates from these cases indi cated that 18 of the 23 tested were serotype O4:K12, which is unr elated to the pandemic strain, O3 :K6, that was first identified in Asia in 1996 and later emerged in the United States in 1998 (Daniels et al., 2000). The outbreak in Galveston Bay, Texas in 1998 with 416 cases (98 culture confirmed) did contain the O3:K6 strain (DePaola et al ., 2000; FDA CFSAN, 2005). It is difficult to establish an accura te count of illnesses resulting from V. parahaemolyticus Not all states require that V. parahaemolyticus infections be reporte d to the state health department (FDA CFSAN, 2005). Studies suggest that approximately 20 V. parahaemolyticus illnesses are estimated to exist for each labor atory-confirmed case reported to the CDC (Mead et al ., 1999). The CDC collaborates with the Gulf Coast states of Alabama, Florida, Louisiana, and Texas to monitor Vibrio infections from that region. This mo nitoring results in about 30-40 cases of V. parahaemolyticus infections reported each year. FoodNet (foodborne diseases active surveillance network) is a program of the CDC which also tracks V. parahaemolyticus in regions outside the Gulf Coast. In 1997, the incidence of diagnosed V. parahaemolyticus infection recorded by FoodNet was .25/100,000 ( MMWR, 2006a). The incidence of Vibrio infections in 2006 (mostly, but not limited to raw seafood and oyste rs) has increased to the highest level since FoodNet began conducting surveillance (MMWR, 2006a). Since this pathogen causes only mild diarrh ea in a healthy person, many cases are not reported to physicians. In order to determine a confirmed case of V. parahaemolyticus, it must be
29 isolated from an infected patient s stool. Then a probable case is defined as gastroenteritis in a person who can be epidemiologically linke d to a confirmed case (FDA CFSAN, 2005). Physiology Physiological responses to cool er temperatures m ay afford V. parahaemolyticus an increased resistance to freezing post harv est processes. In a study by Johnston et al ., (2002) at 4C and 12 days of starvation the V. parahaemolyticus cells entered a VBNC state (when bacterial cells cannot be cultured on media yet retain viabil ity) although they appeared to be metabolically active and maintained their membrane integrity. Another change noted by Johnstons group is that their init ial rod shape became coccoid at 4C, and that the cells showed signs of membrane blebbing. Blebbing is explai ned as modifications (bumps) in the outer membrane of the bacteria that ar e frequently involved in resist ant mechanisms to biocides. Once cells change form (rod to coccoid) they are not detectable with current culture-based methods and it is suggested that this change minimizes cell maintenance (increasing survival) as it increases the surface to volume ratio (Johnston et al., 2002; Jiang and Chai, 1996). Johnstons group shows that freezing at -2 0C does not inactivate these Vibrio at all. Worth noting is that the initial levels in Johnstons study were 109, which is more than 10,000 times the natural levels found in oysters. Johnston s group shows that the V. parahaemolyticus clumps together with filaments and form biofilms. In addition to their resistance to cold temperatures, V.parahaemolyticus were found to have a D value (time required at a specific temperature required to reduce a species of bacteria 1 log) of 1.75 min at 55C (Johnston et al. 2002). This D value is greatly contrasted to that of V. vulnificus which dies so quickly at 50C it is impossible to determine a death curve. In addition to entering a VBNC state, anothe r explanation for the fr eezing resistance of V. parahaemolyticus may be due to a sigma factor of RNA polymerase (rpoS gene) that controls
30 adaptive responses to stress. The rpoS gene has previously been f ound to significantly impair the ability of bacteria to survive environmental stre ss (low pH, and extreme temperatures), but in a study by Vasudevan et al., (2006) the mutant with the rpoS gene removed from V. parahaemolyticus was significantly less able to survive in 4C and 18C than the wild type. The medium for this study was ASW (alkaline sea wate r) and fish homogenate and it was carried out for 21 days at 4C and 56 days at 18C (Vasudevan et al ., 2006). Detection and Enumeration Methods In the present study, tw o FDA specified and ISSC endorsed BAM (Bacteriological Analytical Manual) enumerati on protocols were used. Both the MPN (Most Probable Number) and Direct Plating protocols were used to pe rform both the initial screening and temperature abused enumerations (Kaysner and DePaola, 2004) The post harvest process enumerations were performed using only the MPN protocol. The MPN protocol has an enrichment stage and a confirmatory stage. The enrichment stage first calls for a 1:1 dilution of the homoge nate with PBS (phosphate buffered saline); once blended it is diluted with PBS in serial dilutio ns. From each PBS dilution, 1 mL is aliquoted into 9 mL of APW (alkaline peptone water) for enrichment of the V. parahaemolyticus This enrichment step is not considered a diluti on, but rather an environment for dilution of V. parahaemolyticus injured cells to rejuvenate after the homogenation of the oyster meat. This is important as the bacteria are in an oyster medium and if wrongl y assumed to be dead, they can rejuvenate exponentially given optimal conditions rendering the final oyster product high in possibly pathogenic bacterial loads. After the APW enrichment, T CBS (thiosulfate citrate bile salts sucrose) agar is used as a selective medium; it supplies the V. parahaemolyticus with sucrose as an energy source while the bile salts select out Gram positive competitive bacteria. The positive appearing, green, colonies that grow on TCBS are then dotted onto the final
31 selective media, T1N3, which contains 10% tryptone and 3% salt content. At this point the confirmatory step starts when the f ilters are lifted directly from the T1N3 plates to hybridize with the (AP) alkaline phosphatase-labeled tlh probe which yields total V. parahaemolyticus levels in a sample. (thermolabile hemolysin) (Kaysner and DePaola, 2004). The Direct Plating protocol reports results in enumeration of colony forming units (CFU) and is much less than MPN. Homogenate and PBS dilutions are aliquoted and spread onto T1N3 agar (Kaysner and DePaola, 2004). The Specifi ed FDA confirmation step for both presumptive methods (MPN and Direct plating) includes the tlh probe and it was used to enumerate total V. parahaemolyticus levels in this project (K aysner and DePaola, 2004). Real time PCR Detection PCR (polymerase chain reaction) in the field of bacterial id entification, is relatively new, only 25 years old. In 1983, Kary B. Mullis was driving through Ca lifornia one night, contemplating how to use DNA polymerase with o ligonucleotide primers in order to identify a given nucleotide at a given position in a comp lex DNA molecule, when he was struck by the concept of PCR. His inspiration for developing PCR was necessity, he needed to amplify a DNA sample he was studying. At that time DNA was am plified by cellular repr oduction; to carry out research in that manner is labori ous at best. During this drive he contemplated the idea of making unlimited DNA copies from a single copy of DN A, and called the method "Polymerase Chain Reaction" (PCR). It was only a few months be fore he conducted the first successful PCR experiment (Saiki et al ., 1985). Ten years after his drive in Ca lifornia, he was awarded the Nobel Prize in Chemistry in Stockholm for this discovery. PCR was first published in 1985, with Klenow polymerase used as the elongation enzyme (Saiki et al ., 1985). Due to the heat instability of th e Klenow polymerase, new enzyme had to be added for every new cycle, and the maximum lim it of the product length was 400 bp (base pairs).
32 Then in 1988, the first report using DNA polymerase from Thermophilus aquaticus ( Taq polymerase) was published (Saiki et al ., 1988). This polymerase improved the value of PCR immensely due to its heat stab ility. The introduction of the automatic programmable heating block in the same report also removed the labori ous need for three different water baths for the procedure. PCR is now accepted for many different bacterial identification protocols, but it is still not an accepted ISSC protocol for enumeration of V. parahaemolyticus in shellfish. PCR is a useful tool for enumeration and identification of bacteria by quantifying the amplicons, or target genes. Real time PCR be gins by heat-denaturation (95C) of a DNA sample into single strands. The primers, complementary to the 3' ends of the target DNA segment of interest, anneal to their specifi c region on the denatured DNA when the temperature is lowered to 50C. The probes contain a repor ter dye and a quencher dye. A probe binds to their sequence on the target DNA just after a primer. The Taq polymerase then attaches to the DNA strand behind the primer and binds dNTPs elongating the amplicon; during this process the probes are cleaved by the nuclease activity of the Taq separating the reporter and quencher, causing the reporter dye fluorescence to increase. Fluorescence will only occur if the target gene is present in the sample as the primers and probes bind specif ically to their complementary sequence. When synthesis is complete in one cycl e, the whole mixture is then reh eated to 95C to melt the newly formed DNA duplexes. After the te mperature is lowered again, anot her cycle of synthesis takes place. Repeated cycles of melting (heating to 95C) and synthesis (cooling to 59C) quickly amplify the sequence of interest At each cycle, the number of copies of the sequence between the primer sites is doubled; th erefore, the desired sequence increases exponentiallyabout a million-fold after 20 cycles. It is called real time because the quantities of amplicon fluoresces
33 and are measured with every cycle and there is no need for post-PCR processing with gel electrophoresis (Pierce, 2003). The PCR platform used in this study is the SmartCycler IITM from Cepheid (Sunnyvale, CA). The multiplex assay used in this study targets three genes in V. parahaemolyticus; thermolabile hemolysin ( tlh ), thermostable direct hemolysin ( tdh ) and thermostable related hemolysin ( trh ). It also includes a novel internal amplification control (IAC) that controls for false negatives (Nordstrom et al ., 2007), although the IAC cannot c ontrol for all false negative possibilities; a bad primer or probe (due to incorrect diluting, poor storage, etc.) could also cause a false negative. In that case, if, for instance a tlh primer was bad and V. parahaemolyticus was present, but came up negative, then the resulting negative control c ould alert the researcher of the problem. Industry needs an accurate high-throughput enum eration method to assist them in meeting the FDA requirement that 25% of all Florida she llstock must receive some type of validated PHP. This FDA mandate requires extensive amounts of enumerations as th ey must validate their process and continue with monthly verificati ons. The objective of in cluding real time PCR analysis of the heat shock samples from this study is to analyze its accuracy compared to the molecular AP tlh and tdh probe results. One benefit of the study will be to increase real time PCR data evidence favoring acceptance as an official ISSC method of V. parahaemolyticus enumeration. The presently accepted ISSC mo lecular AP probe confirmation enumeration method for V. parahaemolyticus is laborious, taking up to 4 days to complete. The real time PCR method would take one 8 hour day to enumerate up to 46 samples from extraction to enumeration. Real time PCR reagents are expens ive, costing approximately $5.00/sample to run,
34 but the savings in reduction of labor costs and th e high cost of the AP probe would outweigh the costs of the PCR reagents. Post-Harvest Processes The two post harvest processes used in this study, blast freezing and heat shock, are not new to the seafood industry. The par ticu lar blast freezer used in this study has been in service for 30 years (Ward, 2007). Blast freezing to -29C (20F) is a mechanical freezing process that completely freezes the oyster meat in approximate ly 2 hours. The oysters are placed fully intact, onto rolling racks that are pushed into the walk-in freezer. Cold air, at a temperature of C, is fan blown, or blasted, into the freezer at 25 m.p.h. quickly reducing the temperature of all the products in the freezer. This type of freezing does have oscillations between -29C through ~ -21C throughout the 8 hour blast process. Blas t freezing differs from traditional mechanical freezing in that mechanical freez ing does not use forced air and can take days for items to completely freeze solid.. The blast method ensu res uniformity and expediency in the freezing process of the oyster meat. The oysters are treate d in the blast freezer for 8 hours, then removed and immediately stored at -20C. There are slig ht oscillations in temper ature during the freezing because the unit has to de-ice as necessary (Figure 4-4). The heat shock post harvest process, aimed at assisting in the shucking of the oyster, utilizes brief exposure to hot water. This proce ss also reduces pathogens in the oyster. Using hot water to reduce pathogens is not a novel proc ess either; scalding oysters has been used successfully since 1858 (MacKenzie, 1996). In the heat shock post harvest process, oyster shellstock is subjected to 65C (150F) for 5 minutes, then imme diately cooled in an ice slush for 10 minutes. Hesselman et al., (1999) finds that the interactio n of the heating and the quick chilling is responsible for the significant reduction in bacteria in the oysters. Once chilled, the oysters are immediately placed in an onsite mech anical freezer and stored at C. The 65C
35 water in the heat shock PHP is changed daily wh ich requires 5 hours total (1 hour to refill and 4 hours to reheat); the water supply is chlorinate d and tested every 6 months to meet coliform standards from "Standard Methods for the Exam ination of Water and Waste Water." Webbs Seafood processes approximately 400 bags of oyste rs (60 pounds/bag) daily. In order to reduce V. parahaemolyticus 1 log (1 D value) requires exposure to 52C for 1.3 1.6 minutes (Andrews, 2000). Careful application of this heat shock PH P is critical because studies have shown the oyster texture and taste are negatively affected due to protein degradation at temperatures above 52.5C (Andrews et al ., 2003a). Alternative methods that ar e either presently or pote ntially practiced to reduce V. parahaemolyticus are depuration, high pressure processing, and irradiation. Depuration is a series of moving live oysters from a contaminat ed environment to one of clean water, or by filtering the water they are residing in. A study by Cabello et al. (2005) shows nearly all fluorescent tagged V. parahaemolyticus (VpGFP) was eliminated from oysters depurated for 70 hours. Depuration is done in order to allow the oyster to flush out c ontaminates from their system by filtering clean water th rough their system naturally. Ozone, UV, chlorine and a variety of water cleansing processes are used to treat th e water in attempts to clean it for intake, and ultimately flushing, of the oyster to allow it to release its microbe contaminates. In as early as 1911, depuration experiments were being performe d as a potential method of reducing pathogens in oysters (Canzonier, 1991). In these early experiments it was found that if oysters were depurated for only two days a good elimination of the indicator bacteria was found (Canzonier, 1991). The first commercial depuration processing started in 1921 in New York (Canzonier, 1991). A form of depuration, and possibly the firs t kind of value-added attempt with oysters, used in some states such as New Jersey, was ca lled floating. With floating the oysters were
36 tied to docks allowed to float in cleaner water and less saline waters. They were allowed to pump water through their gills for 24 hours before they were removed for packaging. Oysters that were allowed to float yielded more meats and had a be tter appearance as they were allowed to purge sediment and microorganisms from their meat (MacKenzie, 1996). Since the shallower is less saline, water naturally moves into the oyste r, plumping up the meats and adding volume and value to their harvest. High pressure processing invol ves compressing oysters in wate r under pressure to reduce waterborne pathogens. In a study demonstrating the effectiveness of this process, inoculated oysters showed a 6 log reduction of V.parahaemolyticus at 241mPa (34,953psi) for 11 minutes (Koo et al ., 2006). Another study using 300 mPa (43,511 psi) reduced V. parahaemolyticus > 5 logs in 180 seconds (Su and Liu, 2007). Koo et al (2006) finds V. parahaemolyticus to be much more resistant to reduction with high pressure processing than V. vulnificus Oysters essentially shuck themselves around 40,000 psi, so one of the benefits of the high pressure process is the pre-shucking of the oyster. The resulting dead oysters must be banded together after processing in order to maintain the liquor; this makes it easy for the consumer to check if the product has been treated as the oyster would be pre-shucke d in the package. High pressure processing is considered a no heat pasteurization method (or cold pasteurization). It requir es only a 5F of heat transfer and therefore the oyster does not suffer from taste or texture changes (Koo et al ., 2006). Irradiation involves subjecting oysters to low doses of radiation from Cobalt in order to extend their shelf-life. In a study demonstrating the effect s of irradiation on oysters for bacterial reduction, it was shown that V. parahaemolyticus levels reduce by 6 logs when the oyster is exposed to 1 kGy. This, too, is a non-thermal process, which reduces the potential quality loss due to high heat exposures. Also, it was shown that even irradiation exposures as high as 3 kGy
37 did not change the sensory attributes of the oyste r meat (Flick, 2007). With a O3:K6 strain with an initial V. parahaemolyticus load of 104 CFU/g it was found that a dose of 1.0-1.5kGy caused a reduction to non-detectable levels with a high oyster survival rate In addition, a sensory analysis was performed with 146 volunteers, and there was no significant difference in quality noted from irradiated and non irradiated oysters (Andrews et al., 2003b; Su and Liu, 2007). Research Hypothesis and Objectives In an effort to develop processi ng controls to reduce or elim inate V. parahaemolyticus in raw oysters, this research was intended to add ev idence as to the effectiveness of post harvest processes with blast freezing and heat shock to yield safe oysters for raw consumption. Following FDA specified microbial confirma tory MPN and direct plating protocols, V. parahaemolyticus were enumerated at harvesting time (initial screening), immediately after thermal intervention (Pre-PHP), and then again af ter 7 and 14 days in frozen storage following the post harvest processes. One objective of th is study was to determine which post harvest processing method was more effective--heat shock or blast freezing--at reducing V. parahaemolyticus from an FDA specified mini mum level of 10,000 bacteria g -1. This project is part of a larger validation program that will establish whethe r these two post harvest processes can be approved for commercial use to provide a 3.52 log reduction in V. parahaemolyticus levels found in oysters. Both blast freezing and heat shock processing methods were previously validated for V. vulnificus in 2005 and 2006 respectively (V. Garrido unpublished). The other objective was to compare the conf irmatory MPN enumeration of V. parahaemolyticus from heat shock samples to numbers obtained using a multiplex real time PCR assay with the SmartCyclerIITM (Cepheid). Our hypotheses were that blast freezing and heat shock, would individually cause a 3.52 log reduction of V. parahaemolyticus in raw oysters ( Crassostrea virginica ) to yield a safer
38 product for raw consumption. In addition, there would be no significant difference between the required confirmatory MPN results and a new V. parahaemolyticus multiplex assay with the SmartCyclerIITM real time PCR platform in enumeration of V. parahaemolyticus in PHP oysters.
39 CHAPTER 3 MATERIALS AND METHODS The overall approach was to determ ine the e ffectiveness of two post harvest processes, blast freezing and heat shock, to reduce V. parahaemolyticus by 3.52 logs in raw oysters. Oyster Harvesting and Handling In each of th e three trials for each PHP, 8 bushels of oysters were collected from Apalachicola Bay as assisted by Wards Seafood. The oysters were harvested by hand tonging (Figure 1-1) from two of Wards private l eases, nos. 609 and 572, during the summer of 2007. Oysters for the first trial were harvested on 7/16/07, the second on 8/20/07, and the third on 9/10/07. The water temperatures a nd salinity levels of Apalachicola Bay during this study were measured by the FDACS (Florida Department of Agriculture and Consumer Services) lab based in Apalachicola. Temperature and salinity measur ements are taken regularly in Apalachicola Bay as part of regulatory guidelines; hence, the da tes of salinity and temperature measurements do not exactly coincide with the co mmencement of each trial, but they do coincide for the months of July, August and September 2007. The temperature and salinity measurements were taken with a YSI (Yellow Springs Instruments 6820V2) probe during mid-morning hours. The oysters were harvested then immediately washed with Apalachicola Bay water and held at an ambient temperatur e of ~85F (shade) in burlap bags at Wards Seafood for ~30 minutes. This holding stage is common in commerc ial practice. They were then transported in burlap bags (15 minutes) directly to the Univer sity of Floridas Oyster Industry Lab in Apalachicola for initial screening by microbi al analyses to determine environmental V. parahaemolyticus levels. Eight bushels of oysters were used for each trial for each PHP. For each trial, 36 oysters were removed randomly from the 8 bushels for initial screening of V. parahaemolyticus using MPN enumeration (Figure 3-1) (Kaysner and DePaola, 2004).
40 Oyster Thermal Intervention and Transportation to PHP The norm al environmental level of V. parahaemolyticus in Gulf Coast oysters is approximately 100 CFU g -1 in the summer months (Cook et a l., 2002a; FDA CFSAN, 2005). Thermal intervention was necessary to elevate these low initial levels of V. parahaemolyticus to a minimum of 104 MPN g -1 (10,000 bacteria/g of oyster meat) in accordance with process validation protocols established by the FDA (FDA, 2005). The pathogenic level of 10,000 MPN g -1 has been demonstrated to cause illness in test animals (FDA CFSAN, 2005), and is assumed potentially problematic for human consumption. In order to validate a post harvest processes for V. parahaemolyticus in raw oysters, a potentially pa thogenic worse case scenario (104 MPN g -1), must be reduced by 3.52 logs (FDA, 2005). For the thermal intervention, th e 8 bushels of oysters were divided into two groups and placed in a Precision 30M incubator at 26C fo r 18 20 hours. Tracer time-temperature probes were inserted into two oysters. For insertion, a hole was drilled into th e shell and the probe was inserted in order to determine the temperatur e of the oyster meat th roughout the incubational thermal intervention; due to the extreme temper atures of the post harvest processes (-29C and 65C), Ellab TracksenseIITM time-temperature probes were used. After the thermal intervention, 72 oyster s were removed randomly for microbial enumeration with both confirmatory MPN and dir ect plating analyses to determine if the FDA minimum level of 10,000 MPN g -1 was attained (Figure 3-2). This sampling period following thermal intervention was designated as Pre-PHP in the statistical analys is of the data. All remaining oysters from thermal intervention were transported (5 minutes) to commercial coolers (4C) to simulate changes that oysters experience during typical handling schedules after harvest. Igloo 25 gallon coolers were used for both gro ups of oysters for stor age and transport. The oysters were held (4C) for 5 hour s prior to separation into two equal groups for their assigned
41 PHP. During each PHP, an Ellab TracksenseIITM time temperature recorder was inserted, as previously described above, in two oysters fr om each group in order to record temperature profiles through each PHP. The Tracer time temp erature recorder is not intended for extreme temperatures such as freezing and 50C, hence the switch to Ellab. Th e heat shock group of oysters were transported (1 hour) to Webbs Seafood in Panama City, Florida, in their respective Igloo coolers. The blast freezing group was take n out of the coolers at the same time and placed onto aluminum trays and rolled into Wards blast freezer. A complete flow chart of the oyster handling during each tr ial is shown in Figure 3-1. A portion (72 oysters) of the main group of oysters in each trial was set aside to serve as a control. The control group in each trial was separated from the main group of oysters after thermal intervention treatment. The control group was held in a walk-in cooler at a constant temperature of 4C (refrigerat ion temperature) throughout the study. The rationale for this holding temperature is that the c ontrols were not to receive any PHP treatment including frozen storage at (-20C). The cont rol group serves as a measure of consequence for V. parahaemolyticus growth when a PHP is not used. The control group was maintained for 14 days at 4C, which is consistent with state of Fl orida requirements for the maximum allowed storage time in refrigeration (FDACS, 5L-1.003). This expiration date is stamped on a mandatory harvesting tag and attached to freshly harvested oysters. All oyster retailers are required to keep this tag on hand for 90 days after receiving a shipment of oysters. Within 24 hours following each PHP, all oysters were transported back to the University of Florida, Gainesville in their re spective coolers (in a frozen st ate -20C) (4 hour s) the following day. Upon arrival at the University of Florid a, the two PHP groups were placed in a -20C freezer (Brown, Salisbury, N.C.) and the control group was placed in a 4C cooler. Enumeration
42 was conducted on each of the PHP samples and the control (n = 18) at the designated 7 and 14 day intervals. Blast Freezing Blast freezin g is not a new process and this pa rticular blast freezer used in this study has been in service for 30 years (Ward, 2007). Blast freezing to -29C (20 F) is a mechanical freezing process that freezes the oyster meat in approximately ~4 hours. The oysters were placed, fully intact, onto rolling racks that were pu shed into the walk-in freezer. Cold air, at a temperature of C was fan blow n, or blasted, into the freezer across the oysters at 25 m.p.h., which reduced the temperature of the oysters fr eezer. This type of freezing does oscillate between -29C and ~ -21C throughout the 8 hour bl ast process in order to de-ice. The oysters were treated in the blast freezer for 8 hours then to complete this process the oysters were removed and immediately stored at -20C. Heat Shock The heat shock post harvest process, utilizing hot water to reduce pat hogens, is not a novel process either; scalding oysters has been used successfully fo r 150 years (MacKenzie, 1996). In the heat sho ck post harvest process, oyster shellstock were placed on a conveyor belt that immersed them in a 65C (150F) water bath for 5 minutes, then they were immediately cooled in an ice slush for 10 minutes. Once chille d, the oysters were immediately placed in a mechanical freezer and for storage at C to co mplete this post harvest process. To reduce V. parahaemolyticus 1 log (1 D value) requires exposure to 52C for 1.3 1.6 minutes in ASW (Andrews et al ., 2000). Analytical Methods for V. parahaemolyticus Enumeration Figure 3-2 is a flow chart representing the FDA specified MP N and direct plating (CFU) protocols used in this study on raw oysters for both the initial and thermal intervened
43 enumerations (Kaysner and DePaola, 2004). The APW (alkaline peptone water Difco), TCBS (thiosulfate citrate bile salts sucrose Difco) and the T1N3 (Tryptone 10%, NaCl 3% Difco ) media were all prepared at the Oyster Industry Lab (NELAP certified lab, DOH ID # E71992). Figure 3-3 provides a schematic for the post harvest enumeration steps. These figures differ in the lack of a direct plating step for Figure 3-3 as we ll as different MPN dilutions. For each trial, 36 oysters were removed for mi crobial enumeration for the initial screening at the time of harvest, and then again after th ermal intervention. Twelve oysters represent one sample as each enumeration was done in triplica te. They were shucked, homogenated (together) for 90 seconds (sterile Waring blenders), diluted 1:1 in PBS (phosphate buffered saline) and then serially diluted (Eppendo rf pipettes) in sterile PBS cups out to 10-6. For the MPN method, 1mL was aliquoted from the PBS serial dilution to the corresponding APW (alkaline peptone water) dilution and the APW tubes were incubated (Fishe r Scientific Isotemp) overnight at 37C. The APW tubes were noted for turbidit y (bacterial presence) and then used to streak TCBS plates. The TCBS plates were incubated overnight at 37C and then observed the following day for round, opaque, sticky green colonies, 2 to 3 mm in diameter which are indicative of V. parahaemolyticus For the direct plating method (Fi gure 3-2) two 0.20 g aliquots from the homogenate were weighed and spread onto duplicate 10-1 T1N3 plates. Following the serial dilutions, two 100L aliquots from the 10-1 PBS cups were spread onto the duplicate T1N3 10-2 plates, and so on to the 10-6 as per the protocol. The T1N3 plates were incubated overnight at 37C. This method was followed for the initial sc reening and the thermally intervened samples only as a back-up method in case of competitive b acteria posing difficulty in retrieving accurate data with the MPN method. For the post harvest enumerations (Figure 3-3), one difference was that the MPN dilution tubes were taken out to the 10-4, with 5 tubes for the 10-1 and 10-2 dilutions
44 and 3 tubes for 10-3 and 10-4 dilutions. The other difference is there was no direct plating protocol performed for post harvest enumerations. Confirmatory MPN Molecular tlh Probe Protocol Following the presum ptive steps of the MPN protocol, the tlh probe confirmatory MPN steps started with presumptive green colonies from the TCBS plates being dotted onto duplicate T1N3 plates in rows (up to 5 dots per row). Controls (V. parahaemolyticus, V. vulnificus and V. cholera ) are dotted directly onto the T1N3 plates in addition to the tw o control strips (previously dotted filter strips with known positive coloni es) included in the hybridization protocol. The extra controls were intended as insurance as they are developed separately to a certain point in the protocol, and then are joined with the vali dation filters. These additional controls help to pinpoint the cause of a failure in the protocol if something goes awry (controls do not develop). The tlh (5Xaa agc gga tta tgc aga agc act g 3) and tdh (5Xgg ttc tat tcc aag taa aat gta ttt g 3) probes were purchased through DNA Technology, Denmark. The dotted and incubated T1N3 plates were lifted and the FDA tlh AP probe V. parahaemolyticus hybridization protocol was followed to attain confirmatory MPN enumeration for all samples (Kaysner and DePaola, 2004). PCR Real Time Detection Methods The real tim e PCR thermal cycling was c onducted using the Cepheid SmartCycler IITM. Trial 1 heat shock dilutions of the thermally intervened (Pre-PHP), day 7 and day 14 samples from this study were extracted and used for this PCR analysis following a V. parahaemolyticus multiplex assay (Nordstrom et al ., 2007). In this assay the pathogenic genes, trh and tdh and the nonpathogenic gene tlh were targeted. A ll of the primers --tlh, tdh, trh, IAC (internal amplification control) and the tlh and IAC probes were prepared by Integrated DNA technologies (IDT, Coralville, IA). The IAC DN A was prepared at the Gulf Coast Seafood Lab in Dauphin Island, Alabama. The tdh and trh probes were prepared by Applied Biosystems and
45 the Platinum Taq polymerase, 50 mM MgCl, PCR amplification buffer and dNTPs were prepared by Invitrogen (Carlsba d CA). This is a hot start Taq and does not become active until it reaches 95C; this reduces the chance of denaturation through freezing, thawing and mishandling. This assay has a quantitative ability as the CT (cycle threshold) values can be compared to a standard curve; or, as in this project, the samples tested were previously direct plated so the CT values can be compared to that data for a CFU value as well as the MPN from the APW tubes. The PCR analysis was performed at the Da uphin Island Gulf Coast Research Lab in Dauphin Island, Alabama. There were five SmartCycler IITM thermacyclers daisy chained together in two sets; two to one computer and three to another. A ll of the extractions were stored at -20C for ~2 months and were analyzed on the same day with the assistance of Jessica Nordstrom and Dr. A. DePaola. Each SmartCycler IITM was set to stage 1, heating to 95C for 60 seconds for the initial denaturation and activation of the hot start Taq polymerase. Stage 2 consisted of heating to 95C for 5 seconds and then cooling to 59C for 45 seconds. This second stage was set to repeat 45 times. The fluorescent threshold was set to 15 for greater sensitivity of the CT values. The FTTC25 dye set was selected due to the repo rters used on the probes (FAM, TET, Texas Red and Cy5) and total volume of the tubes, 25L. Other than the IAC control for false negatives, water and V. parahaemolyticus strain TX 1029 served as a pos itive control that contained all three gene markers. Since these samples were al l processed on the same day one large master mix was created and aliquoted accordingly. The SmartCycler IITM results were compared to the AP tlh probe data and to the BAX Dupont real time PCR tlh data.
46 The BAX real time assay data used for comparison purposes is a Vibrio multiplex ( V. cholerae, V. vulnificus and V. parahaemolyticus ). This platform is aimed at industry where efficient, high through-put is essential. For DNA extraction, 5 L from overnight enriched MPN tubes was aliquoted into a cluste r tube containing 200L of lysate solution and promptly capped. The tray of cluster tubes was inserted into a 37C Dupont heating block for 20 minutes. The tray was then placed into another 95C Dupont heating block for 10 minutes. Next the tray was placed into a Dupont cooling rack for 5 minutes and the samples were frozen (-20C) for later (~1 month) analysis. Upon starting the analys is, the samples were thawed and the Dupont reaction tubes containing the reaction tablets (all inclusive for reagents, primers, etc) were added to a 96well, -20C tray. Us ing Duponts multichannel pipetter, 30L of extraction product was added to each reaction tube. The reaction tube s are then placed into a 2mL, 96-well tray and spun for 1 minute at 4C. The reaction tubes we re then placed into the BAX reaction tray analyzed for 40 cycles set on the Vibrio spp assay setting. Statistical Analysis Bacterial numbers were converted to log10 values for statistical analysis. For each trial there were 3 samples (12 oysters = 1 sample) fo r the thermally intervened (Pre-PHP) treatment, and 3 samples each for the control, and each PHP (heat shock and blast freezing). This totals 21 samples for each trial. There were 3 trials performed, resulting in a total population of 63 samples for the entire project. Statistical analysis for the response ( 3.52 mean log MPN g -1 reduction of V. parahaemolyticus ) was analyzed at a significance level of = 0.05 by a one tailed t-test (Minitab) and the corresponding P values were noted for significance (P < 0.05). The hypothesis entered was Ho = 3.51 log reduction; Ha 3.52 log reduction. If the P value was < 0.05 ( 0.05) then that treatment was significant and the null hypot hesis was rejected, meaning the reduction was 3.52 log.
47 For the V. parahaemolyticus enumeration comparison between the tlh probe and the two real time PCR methods, a total of 24 samples we re analyzed a 2 way ANOVA (Excel) statistical test was performed. In this comparison only the heat shock trial 1 samples were analyzed which included three samples at Pre-PHP, and 3 samples at each time interval (days 7 and 14) for each enumeration method. There was no data for day 14 for the BAX Q7 as it validated at day 7 with a > 3.52 log reduction.
48 Figure 3-1. Oyster handling and enumeration per trial three trials were performed for each PHP method and control. Oyster samples harvested from Apalachicola Bay; 8 bushels of oysters are washed and transported (15 minutes) to the Oyster Industry Lab. Initial enumeration for Vp using Direct plating and MPN protocols. n = 36 oysters Thermal Intervention performed on all oysters. Incubated for 18 20 H @ 26C. Temp. probes inserted. Enumeration performed using both direct plating and MPN protocols to assure 10000 g-1 present. n=36 oysters. Designated Pre-PHP. Oysters placed in refrigeration (4C) for 5 hours. HEAT SHOCK: HS group of oysters transferred for heat shock PHP (1H). The next day they are shipped back to UF frozen (4H) and stored at -20C. Temperature probe inserted. BLAST FREEZE : Blast group of oysters transferred for PHP. The next day they are shipped back to UF frozen (4H) and stored at -20C. Temperature probe inserted. PHP oysters held at 20C, control at 4C. Oysters from each treatment are randomly selected and enumerated at day 7 and at day 14 with the confirmatory MPN protocol. CONTROL: Control oysters held at 4C throughout the project. Oysters are randomly selected and enumerated at day 7 and at day 14 with the MPN protocol
49 Figure 3-2. Microbial an alysis per sample of V. parahaemolyticus at initial screening and after thermal intervention. PBS 80mL PBS 99mL PBS 99mL PBS 99mL 20 mL 11 mL11 mL11 mL11 mL 1 mL (0.1g) 100 L 1 mL (0.01g) 1 mL (0.001g) 1 mL (0.0001g) PBS 99mL 1 mL (0.00001g) Incubate O.N. 37C. Incubate O.N. 37C PBS 99mL 1 mL (0.000001g) Step 3B: extract from trial 1 Pre-PHP MPN tubes for PCR (freeze). Step 3A : Streak PBS on TCBS for isolation Step 2: 1 mL of PBS into APW (9 mL) Step 1A: Homogenize (90 sec) oyster meats (10-12 oysters) + PBS (1:1) Step 4B: DNA tlh Probe Hybridization Incubate O.N. 37C 100 L 100 L 100 L 100 L QPCR FDA multiplex for Vp Step 6: multiplex PCR for trial 1 Pre-PHP. T1N3 x2 11 mL TCBS X 2 Step 4A : Replica dotting onto T1N3 Step 5: DNA tlh Probe Hybridization .20 g 1 sample = 12 Oysters Step 1B: Direct Plating PBS to T1N3 duplicate plates. Incubate O.N. 37C.
50 Figure 3-3. Enumeration steps per sample for th e controls and post harvest treated samples after -20C storage for 7 and 14 days. PBS 80mL PBS 99mL PBS 99mL PBS 99mL 20 mL 11 mL11 mL11 mL 1 mL (0.1g) 1 mL (0.01g) 1 mL (0.001g) 1 mL (0.0001g) Incubate O.N. 37C Incubate O.N. 37C Step 3B: extractions of heat shock trial 1 MPN tubes for PCR (freeze). Step 3A : Streak PBS on TCBS for isolation Step 2: 1 mL PBS into APW (9 m L) Step 1: Homogenize (90 sec) oyster meats (10-12 oysters)+ PBS (1:1). Incubate O.N. 37C QPCR FDA multiplex for Vp Step 6: multiplex PCR for heat shock samples @ D 7 & 14. T1N3 x2 TCBS X 2 Step 4 : Replica dotting onto T1N3 1 sample = 12 oysters Step 5: DNA tlh Probe Hybridization
51 CHAPTER 4 RESULTS Initial Screening Apalachicola Bay water temperatures and sali nity levels were taken into account during this study. These abiotic water measurements are recorded regularly (Yellow Springs Instruments 6820V2 probe) in Apalachicola Bay by the FDACS lab at tes ting station 350 (Figure 4-1; Courtesy of William Davis, FDACS). Since these salinity and temperature measurements were taken by another lab, the dates do not exac tly correspond to the ha rvesting dates of this study but they are taken during th e same months in the same ar ea of Apalachicola Bay (Table 41). Only the salinity leve ls reveal a large change during this study. In these two instances salinity decreased during the second week of July and th e second week of August below 25 ppt, but such fluctuations are not atypical and the initial screening analyses re vealed a typical environmental level of V. parahaemolyticus in Apalachicola Bay. During th e months of July, August and September, 2007, the mean log MPN level of V. parahaemolyticus in Apalachicola Bay oysters was ~ 102 (Table 4-2). Thermal Intervention All trials received the same parameters of thermal intervention 18 20 H at 26C. The average incubator temperature for all three trials in this study was 24.89C .17 C. During thermal intervention V. parahaemolyticus reached a 104 g-1 mean log MPN of whole oyster homogenate as required by the FDA validation protocol (Table 4-2). The levels for all three trials ranged from 4.0 to 5.9 MPN (Table 42). Interestingly, the levels of V. parahaemolyticus attained with thermal intervention were not influenced by the initial levels at the time of harvest.
52Post Harvest Processes A one tailed t-test (Minitab) showed that the reduction of V. parahaemolyticus (MPN g-1) comparing all treatments (blast, heat s hock and controls) are only significant ( 3.52 log reduction) with the heat shock post harves t process (P value = 0.03) (Table 4-4). This blast freezing rate reduced V. parahaemolyticus ~ 2 log in 14 days, without any significant reduction (Table 4-3; Ta ble 4-4). This process is signifi cantly (P = 0.99) less effective than the heat shock post ha rvest process at reducing V. parahaemolyticus in raw oysters with a standard deviation of 0.1 (Table 4-3; 4-4). Additional data, outside this experimental design and not reported reveals only a ~ 0.5 log reduction of V. parahaemolyticus after 210 days of continuous frozen (-20C) storage following blast freezing. The average freeze rate for oysters during the blast freezing proce ss in this study was a decrea se of 0.84C/min. The Ellab temperature probe data for both post harvest pr ocesses in this study ar e shown in Figure 4-4. Heat shock reduced the V. parahaemolyticus significantly (P = 0.03) with an average standard deviation of 0.3. Overall, the heat shock process reduced the V. parahaemolyticus 3.52 log in 7 days (Figure 4-2; Table 4-3) and validated accord ing to regulatory requirements. Although the heat shock process wa s not significantly different between day 7 and day 14, this lends to the consideration that the reduction most likely took place in the first few days after the process. This lack of significance in reducti on between day 7 and day 14 was expected as the heat shock process reduced V. parahaemolyticus so low that it would be difficult for it to reduce significantly any further from such a low MPN. The average freeze rate for the oysters during the subsequent freezing in the heat s hock process was 0.51C/min. (Figure 4-4). The controls, kept in refrige ration (4C) throughout this pr oject, showed that common refrigerated storage can reduce V. parahaemolyticus levels > 1 log (Figure 4-2), Gooch et al., (2002) had similar results in a refrigeration study of V. parahaemolyticus in raw oysters reaching
53 a .8 log reduction over 14 days (Gooch et al., 2002). Considering th e 5.13 mean log MPN g -1 initial load of V. parahaemolyticus in this study > 1 log is a re asonable reduction. The typical Gulf Coast levels of V. parahaemolyticus in the summer months are 102 MPN g -1 (Cook et al., 2002a), so these results shows that refrigeration could be an effective way to keep raw oysters safe for 14 days without subjecting them to extreme heat or cold. PCR Analysis The levels of V. parahaemolyticus detected by the Cepheid real time PCR and the AP tlh probe were very similar (Figure 4-3). The trial 1 heat shock post harvest processed samples were analyzed using a real time PCR meth od and the ISSC accepted molecular AP tlh probe protocol. A statistical comparison using a tw o way analysis of variance (Ex cel) indicated that there was no significant difference (P = 0.569) in the compared results between these enumeration methods for the one trial with heat shock samp les. The real time PCR SmartCycler II TM was found to be as accurate as the AP tlh probe method. Both day 7 and day 14 of the trial 1 heat shock processed samples yielded no positives (MPN g-1 <0.18) as did the AP tlh probe. For the pre-PHP samples, there was a standard deviation of 0.16 (Figure 4-3). These results re veal that real time PCR using the V. parahaemolyticus multiplex assay is at least as accurate as the ISSC accepted AP tlh probe protocol. PCR should be considered by the ISSC (Interstate Shellfish Sa nitation Committee) as a protocol for V. parahaemolyticus analysis in shellfish. Worth noting is that there no tdh+ samples in the PCR results and this is comparable to additional data accrued during the trial as a tdh AP probe was utilized for environmental data collection. For this entire study, based on the tdh AP probe, there were no tdh+ samples found in the oysters harvested from Apalachicola Bay. In addition, data accrued using the BAX Dupont real time PCR platform of the same trial using heat shock samples allowed for a further accuracy comparison (Figure 4-3). These
54 results were analyzed side by side with the tlh probe and the Cepheid SmartCyclerIITM results with no significant difference (p = 0.56) and a 0. 13 standard deviation. Note that day 14 for the BAX was not done as that trial valida ted at day 7 (Figure 4-3). Figure 4-1. Map of Apalachicola Bay FDACS illustr ating their water test ing stations. Testing station 350 is the sampling stat ion used in this study. Table 4-1. Temperature and salinity measurements of Apalachicola Bay (11 mile area, station 350). Date Depth (m) S. Salinty (ppt) S. temp(C) B. Salinity B. Temp 7/9/2007 4 39.0 29.4 39.0 29.3 7/16/2007 5 35.5 29.1 35.3 29.1 8/6/2007 6 32.8 29.9 33.5 29.5 8/29/2007 5 24.6 30.1 24.5 30.2 9/27/2007 5 25.2 27.4 25.1 27.5 Both the surface (S) and bottom (B) temperatures are presented for the months of July, August and September, 2007. Courtesy of the FDAC S lab in Apalachicola (William Davis).
55 Table 4-2. Mean log MPN g-1 levels of V. parahaemolyticus in Apalachicola Bay for the initial screening and the subsequent thermal intervention (Pre-PHP) for all trials. 3 samples per trial, per level, to obtain th e means. SD = standard deviation. 0 1 2 3 4 5 6Control Heat Shock BlastLOG (MPN/g) Pre-PHP PHP D7 PHP D14 Figure 4-2. Mean levels of V. parahaemolyticus; Pre-PHP (thermal intervention), PHP day 7 and PHP day 14 for heat shock, blast a nd the controls. The heat shock post harvest process validated ( 3.52 reduction) at day 7. Trials Initial levels SD Thermal intervention levels SD 1 2.0 .3 4.0 .0 2 1.8 .8 5.9 .1 3 1.7 .7 5.3 .0
56 Table 4-3. Heat shock post harvest process vali dated at day 7 by reducing V. parahaemolyticus levels by >3.52 log. Neither the blast freezing post harvest process nor the control reduced V. parahaemolyticus > 2.4 log. A ll units are in mean log MPN g-1. SD = standard deviation1 *largest reduction for blast 2.4 mean log MPN gTable 4-4. One-tailed T test for V. parahaemolyticus reduction per treatment, trial and interval. Ho = 3.51 log reduction; Ha 3.52. P value must be < .05 for significance* ( .05). All units in mean log MPN g-1 Trial Treatment Day 7 reduction P Treatment Day 14 reduction P Control 1.52 0.93 Control 0.04 0.93 1 Blast 1.79 0.99 Blast 1.78 0.99 Heat Shock 3.66 0.12 Heat Shock 3.97 0.02 Control 1.87 0.93 Control 1.87 0.93 2 Blast 1.87 0.99 Blast 1.82 0.99 Heat Shock 4.75 0.05 Heat Shock 4.88 0.03 Control 1.33 0.93 Control 1.33 0.93 3 Blast 2.13 0.99 Blast 2.45 0.99 Heat Shock 5.05 <0.00 Heat Shock 4.81 0.01 significance is that there was a 3.52 mean log MPN g-1 reduction of Vp Trial Pre-PHP Treatment Day 7 SD Treatment Day 14 SD Control 2.5 0.9 Control 4.0 0.0 1 4 Blast 2.2 0.1 Blast 2.3 0.5 Heat Shock 0.4 0.1 Heat Shock 0.1 0.2 Control 4.0 0.0 Control 4.0 0.0 2 5.9 Blast 4.0 0.0 Blast 4.0 0.0 Heat Shock 1.1 0.5 Heat Shock 1.0 0.6 Control 4.0 0.0 Control 4.0 0.0 3 5.3 Blast 3.2 0.1 Blast 2.9 0.2 Heat Shock 0.3 0.0 Heat Shock 0.5 0.5
57 0 1 2 3 4 5 6Pre-PHP PHP D7 PHP D14LOG (MPN/g) tlh probe BAX Cepheid nd Figure 4-3. Comparison of the mean reduction in V. parahaemolyticus of the heat shock samples between the FDA tlh Probe, BAX real time PCR, and the Cepheid SmartCyclerIITM real time PCR.
58 -32.40 -22.40 -12.40 -2.40 7.60 17.60 27.60 37.60 47.60 57.60 C 00:00 08:00 16:00 24:00 32:00 Figure 4-4. Ellab TracksenseTM time (hours) and temperature (Cel sius) data of the blast freezing and heat shock post harvest processes. Green Line Heat Shock temp data Black Line Blast temp data Heat Shock Trans p ortation Blast freezer c y clin g
59 CHAPTER 5 DISCUSSION AND CONCLUSION The heat shock post harvest process validated in 7 days by reducing V. parahaemolyticus levels > 3.52 mean log MPN g-1. Heat shock PHP (65C) efficiently reduced V. parahaemolyticus from potentially pathogenic levels (~105 mean log MPN g-1) to ~0.5 mean log MPN g-1. The blast freezing (-29C) post harvest process only reduced V. parahaemolyticus levels ~ 2 log by day 14. The control (4C) had a ~ 1 log reduction in 14 da ys. In the comparison of V. vulnificus to V. parahaemolyticus reduction with these two post harvest processes, V. parahaemolyticus clearly demonstrated more ability to resist freezing. Additionally, this study shows that real time PCR, using the V. parahaemolyticus multiplex assay, is as accurate as the ISSC accepted AP tlh probe protocol as there was no significant difference between the tlh probe protocol and the real time PCR platforms. The post harvest processes used in this project were also validated in 2005 (blast) and in 2006 (heat shock) for V. vulnificus in raw oysters (V. Garrido unpublished). The heat shock process validated for both V. vulnificus and V. parahaemolyticus by reducing them from potentially pathogenic levels (~105 mean log MPN g-1) in 7 days. Most likely this PHP achieved validation reduction levels at day 1, but there is no FDA validation requirement to enumerate at that time interval hence no evidence for this suggestion. The blast freezing process of V. vulnificus in raw oysters (2005) validated by attaining a reduction of > 3.52 mean log MPN g -1 at day 56 (V. Garrido unpublished). Conversely, in this present study blast freezing (whi ch exceeded -32C) only reduced V. parahaemolyticus levels by ~ 2.5 log at day 210, clearly showin g its freezing resistance abil ity. This post harvest process does not appear effective at reducing V. parahaemolyticus in a time efficient manner and it is
60 costly for the industry to hold the product at -20C for such an extended period of time (> 6 months). A study by Ching Lin et al., (2003) shows that when V. parahaemolyticus is exposed to a cold shock exposure (15C and 20C for 2 and 4 hour intervals) before a freezing treatment it becomes more resistant to reduction. The V. parahaemolyticus cells in TSB (tryptic soy broth) cold shocked at 15C for 2 and 4 hour intervals then treated at 5C for 4 days increased in population. Similarly, in this presen t study, the oysters were also held at a cold temperature (4C) for 5 hours before receiving the blast freezing pr ocess (-29C). The 4C 5 hour exposure of the oysters in our study is the typical temperatur e exposure of naturally harvested commercial oysters; oyster harvesters are requi red to place oysters on ice (10C) if out of the water for more than 10 hours (>84F water temperature) (FDA C FSAN, 2005). Typically af ter freshly harvested oysters are landed, they are immediately washed and placed in refrigeration and held until processing or shipment to a retailer. Possibly, th e 5 hours in refrigeration in our study enhanced a resistance ability in V. parahaemolyticus allowing it to resist reduction with the blast freezing process. A study by Muntada-Garriga et al., (1995) shows that freezing effectively reduces V. parahaemolyticus cells frozen in homogenate An initial level of 107 V. parahaemolyticus was reduced by 6 logs after 18 weeks of storage at -24C. This is an extraordinary reduction in V. parahaemolyticus compared to the results of our study. A dditionally, at initial levels of only 104, V. parahaemolyticus were reduced down to non-detectable le vels in the same time period. It must be considered that V. parahaemolyticus was frozen in homogenate which differs from the oyster industry in which the whole oyster is frozen, as in our st udy. The medium used may be as important as a cold shock trea tment prior to freezing; the en richment broth (TSB); the non-
61 enrichment (ASW), or a whole food product pr ovides different niches for survival of V. parahaemolyticus. Furthermore, the results from Muntada-Garriga et al., still allows the suggestion that cold shocking before freezing does play a role in enabling V. parahaemolyticus to resist freezing. Johnston et al., (2002) finds that the V. parahaemolyticus clumps together with filaments and forms biofilms that could enhance resist ance to which assists in resisting freezing. Freezing of V. parahaemolyticus cells in ASW at -20C for 24 hour s had no effective in reducing V. parahaemolyticus as there was < 1 log reduction from a 10 9 CFG g-1 starting point. V. parahaemolyticus shows similar resistance in our study wh en frozen in the whole oyster, which is what the consumer will be ingesting. A study of raw oysters in frozen st orage examining the reduction of V. parahaemolyticus, and is similar to our blast freezing results in that the V. parahaemolyticus was reduced > 3 log at 23C for 3 months (Chae, 2007). This study ut ilized inoculated Pa cific Oysters to 105 CFU g-1, and is similar to our study in that the whole oyste r was used and the initial levels (our thermally intervened numbers) are similar. No cold shock treatment was applied and the samples were enumerated at intervals of one mont h, three months and four months. The blast freezing storage (-20C) and control (4C) re sults of our study are similar to that of a study by Vasudevan et al., (2002) in which inoculated fish fillets with V. parahaemolyticus initial loads of 104 showed a significant ~ 1 log reduction af ter the product was held at 4C for 9 days, similarly in our study the controls reduced V. parahaemolyticus from a ~105 initial load down ~1.5 logs after 7 days. Vasudevan et al., (2002) also shows that freezing (-18C) of fish fillets (a whole food product medium) inoculated with V. parahaemolyticus results in a significant reduction (~ 2 log) af ter 5 days as did our present study for the blast freezing PHP.
62 Similarly, a study by Wong et al., (1994), using fish and shrimp homogenates inoculated to 104 with V. parahaemolyticus finds that when held at 4C for 6 days there is a 1 log reduction; while -30C storage for 3 days resulted in a 3 log reduction. Both of th ese studies reflect our findings. The heat shock post harvest process is fast, re latively easy to perform, aids in shucking, and is cost effective as little freezing time is needed prior to product availability for the consumer. In this project, the heat shock PHP reduced V. parahaemolyticus ~4.5 log in only 7 days. A possible oversight with th is study is that an enumeration of the heat shock PHP samples was not performed at any interval prior to the 7 day analysis. Since this was an FDA validation no specific enumeration intervals are recommended; t could be that the heat shock PHP reduced the V. parahaemolyticus immediately following the process. Al so, since after the 7 day analysis there was no significant difference in reducti on by day 14, the reduction most likely took place immediately following the heat shock process. One consideration that might be perceived as a drawback is that the hot water in which the oyste rs are immersed must be tested regularly for pathogenic bacteria this would not be necessary with a freeze only process. The data collected in this project is similar to that of other heat reduction studies done on V. parahaemolyticus. In Andrews et al., (2003a) oysters inoculated with a more resilient strain of V. parahaemolyticus, O3:K6, were heat shocked at 65C a nd similar reductions were observed. The heat shock PHP, also known as blanching or low temperature pasteurization, was used to reduce V. parahaemolyticus in a study by Andrews et al. (2000) in live oysters resembling the heat shock results of our project. In Andrews study, the live oyste rs were allowed to take up the V. parahaemolyticus naturally in inoculated tanks. On ce the inoculation was completed, sample oysters were removed, shucked, homogenized and analyzed for initial loads that reached 105. The remaining oysters were pasteurized at 65C in increments of 5 minutes. At 10 minutes the
63V. parahaemolyticus levels fell to non-detectable levels. Al so, spoilage bacteria were reduced 2 3 logs, lengthening the shelflife of the oysters (Andrews et al., 2000). Because Vibrio multiply rapidly, even low levels of V. parahaemolyticus in harvested products can rapidly increase to in fectious levels if not rapidly refrigerated after harvest and maintained at proper temperatures during tr ansport, processing, and storage (i.e., <50F [<10C]) (FDA CFSAN, 2005). The control temperature of 4C reduced V. parahaemolyticus ~ 1 log similar to other studies. Cook et al., found significantly higher (10 100 fold) levels of V. vulnificus and parahaemolyticus in the oysters at market than at the time of harvest, but encouragingly the consistently low temperatures of storage at restaurants resulted in a decrease of Vibrio (Cook et al., 2002a). Interestingly, the C temperature of the blast PHP did not significantly reduce V. parahaemolyticus compared to the 4C temperature of the control. The reduction at 4C shows that oysters can be kept safely for 14 days, but it is not successf ul for long term storage as the MPN was increasing by ~.5 log by day 14. Gooch et al., (2002) had similar results in a refrigeration study of V. parahaemolyticus in raw oysters reachi ng a 0.8 log reduction over 14 days (Gooch et al., 2002). Considering th e initial load of V. parahaemolyticus in this study of 5.13 mean log MPN, a > 1 log is an acceptable reduction. Initial Screening and Thermal Intervention The initial levels of V. parahaemolyticus in the Apalachicola Bay were within typical limits (Cook, 2002; FDA CFSAN, 2005), averaging ~ 102 for this study. Normally, if all other conditions are constant, the initial load of bacteria in a sample affects the exponential growth of the population. This study found that the thermal in tervention did not boost the growth of the initial levels of V. parahaemolyticus in each trial as expected. For example, in the first trial the initial levels attained 2 mean log MPN g -1 and, after thermal intervention for 18 20 hours, the
64 levels only increased to 4 mean log MPN g -1. Conversely, in the s econd trial the initial screening reached 1.8 mean log MPN g -1 and after thermal intervention the V. parahaemolyticus population grew to an excep tional 5.9 mean log MPN g -1. This is almost a 2 log difference in population growth with nearly the same initial screening level. During the pre-validation thermal intervention enumerations, and once during the validation, competitive bacterium presented and alternative measures were employed. Additional data was collected using the tdh AP probe FDA protocol. It is worth noting that there was no tdh+ V. parahaemolyticus samples found in Apalachicola Bay during this validation study. Both the tdh probe protocol and the re al time PCR samples noted no tdh+ samples in either the initial, environmental levels, or th e thermally intervened (artificially elevated) 105 MPN g-1 samples. Preliminary Work Involving Bacterial Competition Competing bacteria presented difficulties with MPN enumerations after thermal intervention at the more concentrated dilutions (101 and 102) during the July 2007 pre-validation microbial analyses. This competition was noted as V. parahaemolyticus was negative TCBS and came up negative at the lower (101 and 102) dilutions, but at higher (103 and 104) dilutions V. parahaemolyticus tested positive. The competitive bacteria made it difficult to retrieve accurate MPN data, and a direct plating method was empl oyed as a back-up method for the following prevalidation run. The data from the direct plating was also difficult to interpret accurately. Cook et al., (2002b) reported similar difficu lties using the direct platin g method; background bacteria produced confluent lawns or crowded plates. In one instance they employed a two-step direct plating strategy. Since this competition mainly pr esented itself during th ese pre-validation runs in our study (only observed once during the actual validation) it was not investigated further. More studies are needed to a ssist in enumeration methods of thermal intervention samples.
65 The competitive bacteria presented itself as neon yellow colonies commandeering the resources of the TCBS agar and it was assumed to be V. alginolyticus, though no presumptive testing was performed. Fl etcher (1985) studying V. parahaemolyticus in oysters, had a similar enumeration problem caused by yellow colonies covering the TCBS plates at the more concentrated dilutions. They also incorporated direct plating as an alternative enumeration method, but it was found to be less accurate at lower numbers of bacteria. These competitive yellow colonies were identified as V. alginolyticus. In a review by Su and Liu (2007), it was stated that TCBS cannot differentiate V. parahaemolyticus from some strains of V. vulnificus and Vibrio mimicus. Cooks group found similar confusion in MPN enumeration in a 1998 1999 survey, with V. parahaemolyticus negative at the 101 and 102 dilutions but positive at the 103 and 104 dilutions. In that study, three coll aborating labs attained th e same results on the same samples, showing that technician error was not a cause for false negatives. For enumeration purposes Cooks group used an unconventional approach of only accepting the highest, all positive, dilution yielding the target organism. In the third trial of this present study we also had to choose this option of selecting the highest, all positive, dilution in order to ascertain an accurate MPN due to competing bacteria. As V. alginolyticus is implicated as an often annoying competitor for V. parahaemolyticus, it is interesting that in a 2005 2006 study of Vibrio in the Adriatic Sea, V. alginolyticus was found to be the predominate species. Out of 96 samples V. cholera was 1/96, V. vulnificus was 3/96, V. parahaemolyticus was 10/96 and V. alginolyticus was 24/96. More importantly, Baffone et al.,(2006) revealed V. alginolyticus strains produced amplicon with a trh primer. Thermal Intervention Issues On one trial, V. parahaemolyticus levels failed to a ttain the FDA required 104 MPN g -1 even after thermal in tervention. The average MPN of th e initial screening levels or
66 environmental levels for this failed trial was 20 MPN g -1, which helps to explain the inability to reach 104 MPN g -1. If V. parahaemolyticus is at low levels in the oyster meat initially, then thermal intervention does not appear to produce 104 MPN g -1 levels. The use of thermal intervention to artificially raise levels of V. parahaemolyticus is not an easy endeavor. It adds an additional 24 hours to th e already lengthy (3 day) enumeration protocol and it is not always successful, wasting time a nd labor costs as well as the media and other consumables. It is not a foolpr oof method to attain required 104 MPN g -1 levels of bacteria. Failure is not discovered until many resources and much time have been invested. Also a lab must have a large incubator to perform this treatment. Another setback is th at some of the oysters may die during this process and have to be culled from the population; in some experiments this scenario may detrimentally reduce the sample size to insufficient numbers. Often during the thermal intervention step competitive bacteria commandeered the TCBS plates. Bacterial competition was present in the initial screeni ng; however, they did not cause enumeration difficulties until after the thermal intervention. Though there are many drawbacks to incubation as a means of increasing bacterial loads in a sample, a positive aspect is that it is much less labor intensive and more cost effective than having th e oysters filtering inoculat ed water in aquarium tanks. PCR Analysis These results reveal that real time PCR, using the V. parahaemolyticus multiplex assay, is as accurate as the ISSC accepted AP tlh probe protocol; the ISSC is the governing body that dictates which bacterial enumer ation protocols are accepted for the shellfish industry. Previous studies have demonstrated the effica cy of real time PCR to enumerate V. parahaemolyticus (Nordstrom et al., 2007; Cook et al., 2002b; Sechi et al., 2000). The compared results of
67 Cepheid real time PCR to the AP tlh and tdh probe revealed strong similarities. PCR is not yet an accepted ISSC protocol for V. parahaemolyticus analysis in shellfish. This Cepheid real time PCR multiplex assay is novel in its IAC development, which controls for false negatives. The assay wa s developed by the Dauphin Island Gulf Coast Research Lab in Alabama and Nordstrom et al., (2007) utilized their assay in determining accuracy analyzing 117 samples isolated from clinical, environmental and food sources for V. parahaemolyticus tlh, trh and tdh genes. With standard curves they demonstrated that each target gene can be detected down to 1 CFU mL -1 on a SmartCyclerIITM from Cepheid. Also they reported amplification efficiencies of 84%, 94% and 95% for tlh, tdh, and trh respectively (Nordstrom et al., 2007). Bej et al., (1999) did a similar study of 111 samples isolated from clinical, environmental and food sources using a similar V. parahaemolyticus PCR multiplex assay (tlh, trh and tdh). The platform for this V. parahaemolyticus multiplex assay was a Perkin Elmer thermacycler model 480, and it was not real time. They completed additional tests for the PCR positive trh and tdh samples by retesting them on Wagatsuma agar and showed an absolute correlation for the PCR pos itive hemolytic positive strains (Bej et al., 1999). In 2007, Wright et al., did a PCR study of V. vulnificus in oyster homogenates befo re and after nitrogen immersion post harvest process. They compar ed real time PCR using the SmartCyclerIITM (Cepheid) platform to the ISSC accepted AP tlh probe protocol. The real time PCR-MPN results for all samples were comparable (R2 =.97 Pearsons) and the study concludes that real time PCR is a sensitive and cost-effectiv e alternative to the standard tlh probe methods (Wright et al., 2007). High through-put, accurate enumeration met hods are needed to assist the industry in maintaining a safe, healthy product for consumers of raw oysters. The presently accepted ISSC
68V. parahaemolyticus enumeration protocol, the AP tlh probe, takes up to 4 da ys to complete; it is an arduous and time consuming protocol. As th is study has shown, real time PCR is a method that is equally as accurate, and more time efficient, for enumerating V. parahaemolyticus than the accepted AP tlh probe protocol. The real time PCR en umeration method can be completed in one day; the drawbacks are that the reagents and th e equipment (SmartCyclerIITM) are expensive, including a suggested $3500/yr maintenance package cost. Conclusions The heat shock PHP is an effective process to ensure the safety of those who consume raw oysters. This is the first study to validate ( 3.52 reduction) the heat shock post harvest process for V. parahaemolyticus in raw oysters, and it did so in 7 days. A validation allows a post harvest processor to state on their tags that their product is post harvest processed, hence reduced in V. parahaemolyticus level. This process affords the industr y a larger market as some states and many large-scale retailers only offer PHP oysters to the consumers. In this study, heat shock PHP reduced V. parahaemolyticus from potentially pathogenic levels of ~105 mean log MPN g-1 down to 0.5 mean log MPN g-1. Conversely, the blast freezing post harvest process, which exceeded -32C, reduced V. parahaemolyticus only ~ 2 log over 14 days; this clearl y demonstrated the freezing resistance ability of V. parahaemolyticus when only an additional 0.5 log re duction was attained at day 210. The controls demonstrated that refrig eration (4C) can i nhibit growth of V. parahaemolyticus as it only showed a ~ 1 log reduction in raw oysters that were thermally abused. If oysters are harvested properly handled within suggested ISSC harvesting and handling guidelines refrigeration maintains a safe raw pr oduct if consumed within 14 days. Validation protocols proved to be troubl esome during this study. As the Apalachicola Bay oyster V. parahaemolyticus levels were typical (102 MPN g-1) for the summer months,
69 thermal intervention (an FDA approved method of elevating bacterial levels to104 MPN g-1) was employed. Consequently, the thermal interventi on method also elevated competing bacteria which produced enumeration difficulties with the MPN and direct pl ating methods. Improved validation protocols are needed in order to assist and encourage industry to validate more post harvest processes in order to meet the FDA mandat e that 25% of Florida shellstock must receive a PHP. Additionally, the real time PCR findings present data that shou ld assist the adoption of an ISSC accepted high-throughput real time PCR protocol for the enumeration of V. parahaemolyticus. If accepted as a protocol real time PCR will substa ntially shorten the present tlh probe protocol by 3 4 days, saving tim e and resources for the oyster industry.
70 LIST OF REFERENCES Andrews L. S., Park D. L., Chen Y.-P. (2000) Lo w temperature pasteurization to reduce the risk of Vibrio infections from raw shell-stock oysters. Food Add Cont 19:787-791. Andrews L.S., DeBlanc S., Veal C.D., Park D.L. (2003a) Response of Vibrio parahaemolyticus 03:K6 to a hot water/cold shock pasteurization process. Food Additives and Contaminates.20:331-334. Andrews, L., Jahncke, M., Mallikarjunan, K. (2003b) Low dose gamma irradiation to reduce pathogenic Vibrio in live oysters. J Aqua Food Prod Technol 12: 71-82. Baffone, W., Tarsi, R., Pane, L., Campana, R., Repetto, B., Mariottini, G., Pruzzo, C. (2006) Detection of free-li ving and plankton-bound Vibrios in coastal waters of the Adriatic Sea (Italy) and study of their pathoge nicity-associated properties. Env Micriobiol 8: 12991305. Bartlett, John.(1919) (Swift, Jonathan. Polite C onversation: Dialogue ii) Familiar Quotations, 10th ed. Bej, A. K., D. P. Patterson, C. W. Brasher, M. C. L. Vickery, D. D. Jones, and C.A. Kaysner. (1999) Detection of total a nd hemolysin-producing Vibrio parahaemolyticus in shellfish using multiplex PCR amplification of tl, tdh and trh. J Microbiol Meth 36:215-225. Bennish, M. L. (1994) Cholera: pathophysiology, clin ical features, and treatment, p. 229. In I. K. Wachsmuth, P. A. Blake, and Olsvik (ed.), Vibrio cholerae and cholera: molecular to global pe rspectives. ASM Press, Washington, D.C. Brooks, William, K. (1891) The Oyster. Johns Hopkins University Press. Butler, Phillip (1949) Gametogenesis in the oyster under conditions of depressed salinity. Chesapeake Shellfish Investigations. Biol bull 96:263-269. Cabello, A.E., Espejo, R.T., and Romero, J. (2005) Tracing Vibrio parahaemolyticus in oysters (Tiostrea chilensis) using a Green Fluor escent Protein tag. J Exp Mar Biol Ecol 327:157166 Canzonier, Walter. (1991) Historical perspective on commercia l depuration of shellfish. In W.S. Otwell, Rodrick, G.E., Martin, R.E. (ed.), Molluscan Shellfish Depuration. CRC Press. Inc., Boca Raton, Fla. Centers for Disease Control. (1993) Imported cholera associat ed with a newly described toxigenic Vibrio cholerae O139 strainCalifornia, 1993. Morbid. Mortal. Weekly Rep. 42:501. Chae, Minjung. (2007) Low temperature post harvest processing at reducing Vibrio parahaemolyticus and Vibrio vulnificus in raw oysters. Masters Thesis; Oregon State University.
71 Chakraborty, S., Nair, G., Shinoda, S. (1997) Pathogenic Vibrios in the natural aquatic environment. Rev Environ Health 12: 63-80. Chapel, G.L., (2007) Apalachicola Historical Society. http://www.visitfranklincounty.com/bibliogaphy.htm Chatzidaki-Livanis, M., Hubbard, M.A., Gor don, K., Harwood, V.J., and W right, A.C. (2006) Genetic distinctions among clini cal and environmental strains of Vibrio vulnificus. Appl Environ Micriobiol 72:6136-6141. Churchhill, E.P. (1920) The oyster and the oyste r industry of the Atlantic and Gulf coasts. Bureau of Fisheries Document No. 890. Washington Government Printing Office. Chu, F.E., and La Peyre, J.F. (1989) Effect of environmental factors a nd parasites on hemolymph lysozyme and protein in American Oysters (Crassostrea virginica). J Invertebr Pathol 54:224-232. Cook D, Bowers J, DePaola A. (2002b) Density of total and pathogenic tdh+ Vibrio parahaemolyticus in Atlantic and Gulf Coast molluscan shellfish at harvest. J Food Prot 65:1873-1880. Cook, D., OLeary, P., Hunsucker, J., Sloan, E., Bowers, J., Blodgett, R., DePaola, A. (2002a) Vibrio vulnificus and Vibrio parahaemolyticus in U.S. retail shell oysters: A national survey from June 1998 to July 1999. J. Food Prot. 65:79-87. Cunningham, J.T. (1885) The resting positions of oysters. Nature 22:597 Daniels NA, Ray B, and Easton A,. (2000) Emergence of a new Vibrio parahaemolyticus serotype in raw oysters: a prevention quandary. JAMA 284:1541. DePaola, A., Kaysner, C., Bowers, J., Cook, D. (2000) Environmental Investigations of Vibrio parahaemolyticus in Oysters after Outbreaks in Washington, Texas, and New York (1997 and 1998). App Environ Microbiol 66:4649-4654. Dombroski, C.S., Jaykus, L., Geen, D.(1999) Occurrence and control of Vibrio vulnificus in shellfish.. J Aqua food prod technol 8:11-25. FAO/WHO. (2003) Joint FAO/WHO food st andards progamme codex committee on food hygiene; Thirty-fifth Session. Orlando, U.S.A., 27 January 1 February 2003. FAO. (2000) Species page. Crassostrea virginica. http://www.fao.org/fishery/species/2669. Accessed online 12/7/07. FDA. (2005) Guide for the Control of Mo lluscan Shellfish. IV Guidance Documents Chapter IV. Naturally Occurring Pathoge ns. 04 Validation/Verification Interim Guidance.
72 FDA CFSAN. (2003) National Shellfish Sanitation Program. Guide for the Control of Molluscan Shellfish. Guidance Documents. Chapte r IV. Naturally Occurring Pathogens. FDA CFSAN. (2005) Quantitative risk assessment on the public health impact of pathogenic Vibrio parahaemolyticus in raw oysters. FDACS, (Florida Department of Ag riculture and Consumer Services). In The comprehensive shellfish control code. Chapter 5L-1.007:12. Fisher, George. (2005) Oysters as an aphrodisiac (D-aspartic aci d, and N-methyl-D-aspartate). Presentation, American Chemical Society, San Diego California. Fletcher, G.C. (1985) The potential food poisoning hazard of Vibrio parahaemolyticus in New Zealand Pacific Oysters. J Intl Microbiol 19:495-505. Flick, G. (2007) Global Aquacu lture Advocate. March/April 2007. Galtsoff, P.S. (1964) The American Oyster Crassostrea virginica Gmelin. Fish, Bull. U.S. Bur. Comm. Fish. 64:185:-218. Genthner, F., Volety, A.K., Oliver L., Fisher, W. (1999) Factors in fluencing in vitro killing of bacteria hematocytes of the Easter Oysters (Crassostrea virginica). App. Environ. Microbiol 65: 3015-3020. Gooch, Janet Ann. (2000) Factors affecting quantitative and qual itative changes in Vibrio parahaemolyticus and Vibrio vulnificus populations in oysters. Di ssertation; Mississippi State University. Gooch, J., DePaola, A., Bowers, J., Marsha ll, D. (2002) Growth and survival of Vibrio parahaemolyticus in postharvest American Oysters. J. Food Prot 65:970-974. Hadly, W.G. (1997) Vibrio infecti ons associated with raw oyster consumption in Florida, 1981 1994). J. Food Prot 60:335-357. In Oliver, J.D. (2006) Vibrio vulnificus. F.L.Thompson, B. Austin, J.G. Swings. The Biology of Vibrios. ASM Press. Washington D.C. pp. 349366. Hesselman, D.M., Motes, M.L., Lewis, J.P. (1999 ) Effects of commercial heat-shock process on Vibrio vulnificus in the American Oyster, Crassostrea virginica, harvested from the Gulf coast. J Food Prot 62: 1266-1269. ISSC Interstate Sanitation and Shellfish Commission. (2007) Task Force II Report: 2007 Biennial Meeting. Jiang, X. and Chai, T. (1996) Survival of Vibrio parahaemolyticus under starvation conditions and subsequent resuscitation of viable, nonculturable cells. App Environ Micriobiol 62:1300-1305.
73 Johnston, J. M., Becker, S.F. and McFarland, L.M. (1986) Gastroenteritis in patients with stool isolations of Vibrio vulnificus. Am. J. Med. 80:336-338. In Oliver, J.D. (2006) Vibrio vulnificus. F.L.Thompson, B. Austin, J.G. Swings. The Biology of Vibrios. ASM Press. Washington D.C. pp. 349-366. Johnston, M.D.. Brown, M.H. (2002) An investigation into the changed p hysiological state of Vibrio bacteria as a survival mechanism in response to cold temperatures and studies on their sensitivity to heating and freezing. J Appl Micro 92:1066-1077. Joseph, L., Wright, A.C. (2004) Expression of Vibrio vulnificus capsular polysaccharide inhibits biofilm formation. J bacterial 186:889-93. Kaufman GE, Myers ML, Pass CL, Bej AK, Ka ysner CA. (2002) Molecular analysis of Vibrio parahaemolyticus isolated from human patients and shellfish during US Pacific northwest Outbreaks. L Appl Microbiol 34:155-161. Kaysner, C.A., and DePaola, A. (2004) Vibrio cholera, V parahaemo lyticus, Vibrio vulnificus and other Vibrios spp. Bacteriological analytical manual online, 8th ed. U.S. FDA, Center for Food Safety and Applied Nutrition. www.cfsan.fda.gov/~ebam/bam-9.html. King, Jam ie L., McGaw, Kay. (2004) Oyster Restoration Series. Status of the U.S. Oyster Resource.15. http://noaa.chesapeakeb ay.net/docs/H abitatConnect Vol5no1NOV2004.pdf. Habitat connections. 5(1). Landings. http://www.st.nmfs.gov/st1/commercial/index.html. Koo, J., Jahncke, M.L., Reno, P.W ., Hu, X., Mallikarjunan, P. 2006. Inactivation of Vibrio parahaemolyticus and Vibrio vulnificus in phosphate-buffered saline and in inoculated whole oysters by high-pressure processing. J of Food Prot.69:596-601. Kurlansky, Mark. (2006) The Bi g Oyster. Ballantine Books. Levine, M. M., R. E. Black, M. L. Clements, D. R. Nalin, L. Cisneros, and R. A. Finkelstein. 1981. Volunteer studies in development of v accines against cholera and enterotoxigenic Escherichia coli: a review, p. 443. In T. Holme, J. Holmgren, M. H. Merson, and R. Mollby (ed.), Acute enteric infections in children. New prospects for treatment and prevention. Elsevier/North-Holland Biomedical Press, Amsterdam. Lin, Ching. Yu, Roch-Chui. Chou, Che ng-Chung. (2004) Susceptibility of Vibrio parahaemolyticus to various environmental stresses after cold shock treatment. Intl J Food Micro 92: 207-215. MacKenzie, Clyde L., Jr. (1989) A guide for enhancing estuarine molluscan shellfisheries. Mar Fisheries Rev. summer. MacKenzie, Clyde L., Jr. (1996) History of oystering in the Unite d States and Canada, featuring the eight greatest oyster estuaries. Mar. Fish. Res. 58:1-78.
74 McCarthy et al., S.A. McCarthy, A. DePaol a, D.W. Cook, C.A. Kaysner and W .E. Hill. (1999) Evaluation of alkaline phosphataseand digoxi genin-labeled probes for detection of the thermolabile hemolysin (tlh) gene of Vibrio parahaemolyticus, Lett. Appl. Microbiol. 28:66. McPherson, V.L., Watts, J.A. Simpson, L.M. and O liver, J.D. (1991) Physio logical effects of the lipopolysaccharide of Vibrio vulnificus on mice and rats. Micriobios 67:272-273. In Oliver, J.D. (2006) Vibrio vulnificus. F.L.Thompson, B. Austin, J.G. Swings. The Biology of Vibrios. ASM Press. Washington D.C. pp. 349-366. Mead, P.S, Slutsker, L, and Diet z, V. (1999) Food-related illness and death in the United States. Emerg Infect Dis 5:607-625. Miles, D.W., Ross Thomas, Olley June and McMeekin Thomas A. (1997) Development and evaluation of a predictive model for the eff ect of temperature and water activity on the gowth rate of Vibrio parahaemolyticus. Intl J Food Micro 38:133-142. Miyamoto, Y., Kato, T., Obra, S., Akiyama, S., Takizawa, K., Yamai, S. (1969) In vitro characteristics of Vibrio parahaemolyticus : its close correlation with human pathogenicity. J. Bacteriol 100:1147-1149. MMWR. (1993) Vibrio vulnificus Infections Associated with Raw Oyster Consumption -Florida, 1981-1992. 42:405-407. MMWR. (2006a) Vibrio parahaemolyticus Infections Associated with Consumption of Raw Shellfish --Three States. 55:854-856. MMWR.(2006b) Preliminary FoodNet Data on the Incidence of Infection with Pathogens Transmitted Commonly Through Food --10 States, 2006. 56: 336-339. Muntada Garriga JM, Rodriguez-Jerez JJ, Lopez-Sabater EI, Mora-V entura MT. (1995) Effect of chill and freezing temperatures on survival of Vibrio parahaemolyticus inoculated in homogenates of oyster meat. L. appl micro 20:225-27. Newcombe, C.L. (1946) Oysters. In Turt ox News. Virginia Fisheries Laboratory, Williamsburg,Va. 24(8). Nishibuchi, M., and Kaper, J. B. (1985) Nucelotide sequence of the thermostable hemolysin of Vibrio parahaemolyticus. J Bacteriol 162: 558. Nishibuchi, M., Hill, W.E., Zon, G., Payne W.I., Kaper, J. B. (1986) Synthetic oligiodeoxyribonucleotide probes to detect Kanagawa phenomenon-positive Vibrio parahaemolyticus. J. Clin. Microbiol. 23: 1091-1095. Nishibuchi, M., T. Taniguchi, T. Misawa, V. Khaeomanee-Iam, T. Honda, and T. Miwatani. (1989) Cloning and nucleotide sequence of the gene (trh) encoding the hemolysin related to the thermostable direct hemolysin of Vibrio parahaemolyticus. Infec Immun 57:2691 2697.
75 Nishibuchi, M., A. Fasano, R. G. Russell, and J. B. Kaper. (1992) Enterotoxigenicity of Vibrio parahaemolyticus with and without genes encoding thermostable direct hemolysin. Infect. Immun. 60:3539. Blackstone, G., Nordstrom, J.L., Vickery, M ., Bowen, M., Meyer, R.F., and DePaola, A. (2003) Detection of pathogenic Vibrio parahaemolyticus in oyster enrichments. J Microbiol Meth 53: 149-155. Nordstrom, J.L., Vickery, M., Blackst one, G., Murray, S., and DePaola, A. (2007) Development of a Multiplex Real time PCR Assay with an Internal Amplification Control for the Detection of Total and Pathogenic Vibrio parahaemolyticus Bacteria in Oysters. Appl Environ Microbiol 73: 5840. Okuda, J. M. Ishibashi, E. Hayakawa, T. Ni shino, Y. Takeda, A.K. Mukhopadhyay, S. Gag, S.K. Bhattacharya, G.B. Nair and M. Nishibuchi (1997) Emergence of a unique O3:K6 clone of Vibrio parahaemolyticus in Calcutta, India, and isola tion of strains from the same clonal goup from southeast Asian travelers arriving in Japan, J. Clin. Microbiol. 35: 3150. Oliver, J.D. (2006) Vibrio vulnificus. F.L.Thompson, B. Austin, J.G. Swings. The Biology of Vibrios. ASM Press. Washington D.C. pp. 349-366. Oliver, J.D. and Kaper, J.B. (2001) Vibrios. In Doyle, M.P., Beuchat, L.R., Montville, T.J. Food Microbiology: Fundamentals and Frontiers ASM Press. Washington D.C. pp. 228-264. In Oliver, J.D. (2006) Vibrio vulnificus. F.L.Thompson, B. Austin, J.G. Swings. The Biology of Vibrios. ASM Press. Washington D.C. pp. 349-366. Oliver, J.D. and Kaper, J.B. (1997) Vibrios. In Doyle, M.P., Beuchat, L.R., Montville, T.J. Food Microbiology: Fundamentals and Frontiers ASM Press. Washington D.C. pp. 228-264 Pierce, Dan. (2003) Real time PCR: The Taqman method. Davidson College. http://www.bio.davidson.edu/Courses/Molbio /MolStudents/spring2003 /Pierce/realtim epc r.htm. Accessed 12/20/07. Ramamurthy, T., and Balakrish Nair, G. (2005) Vibrio parahaemolyticus: The threat of another Vibrio acquiring pandemic potential. 11. National Institute of Cholera and Enteric Diseases, Calcutta, India and International Ce ntre of Diarrhea Disease Research, Dhaka, Bangladesh. Saiki RK, Scharf S, Faloona F, Mullis KB, Horn GT, Erlich HA, Arnheim N. (1985) Enzymatic amplification of beta-globin genomic sequences and restriction site an alyses for diagnosis of sickle cell anemia. Science. 230:1350-1354. Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Hi guchi R, Horn GT, Mullis KB, Erlich HA. (1988) Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science. 239:487-491.
76 Sechi, L.A., Dupre, I., Deriu, A., Fadda, G., Zanetti,S. (2000) Distribution of Vibrio cholerae virulence genes among different Vibrio species isolated in Sardinia, Italy. J. Microbiol 88: 475-481. Sochard, M.R. and Colwell, R.R. (1977) Toxin isolation from a Kanagawa-phenomenon negative strain of Vibrio parahaemolyticus. Microbiol. Immunol. 21:243-54. Stephenson, Frank. (1994) Toward Safer Oyst ers: A gene test for a troubled industry. Research in Review. Su, Yi-Cheng and Liu, Chengchu. (2007) Vibrio parahaemolyticus: A concern of seafood safety. Food Microbiol 24: 549-558. Takahashi, A., Kenjyo, and N., Imur a, K. (2000) Clsecretion in colonic epithelia l cells induced by the Vibrio parahaemolyticus hemolytic toxin related to thermostable direct hemolysin. Infect. Immun 68:5435-5438. Taniguchi, H., R. Hirano, S. Kubomura, K. Higashi, and Y. Mizuguchi. (1986) Comparison of the nucleotide sequences of the genes for th e thermostable direct hemolysin and the thermolabile hemolysin for Vibrio parahaemolyticus. Microb Pathog 1:425. Vasudevan P. Marek P, Daigle S, Hoagland T, Venkitanarayanan KS. ( 2002) Effect of chilling and freezing on survival of Vibrio parahaemolyticus on fish fillets. J Food Safety 22:209-217. Vasudevan, P., Venkitanarayanan, K. (2006) Role of rpoS gene in the survival of Vibrio parahaemolyticus in artificial seawater and fish homogenate. J Food Prot 69: 438-442. Waldor, M.K., and Mekalanos, J.J. (1996) Ly sogenic conversion by a filamentous phage encoding Cholera toxin. Science 272: 1910-1914. Ward, Dacky. (2007) Personal communi cation. Apalachicola, Florida. Ward, L. N., and A. K. Bej. (2006) Detection of Vibrio parahaemolyticus in shellfish by use of multiplexed real time PCR with TaqMan fluorescent probes. Appl Environ. Microbiol 72:2031. Wilson, I. G. (1997) Inhibition and facilitation of nucleic acid amplification. Appl Enviro. Microbiol. 63:3741. Wong H-C, Chen L-L, Yu C-M. (1994) Occurrence of Vibrios in frozen seafoods and survival of psychrotropic Vibrio cholera in broth and shrimp homoge nate at low temperatures. J Food Prot 58: 263-267. Wong H-C, Chen L-L, Yu C-M. (2003) Analysis of the envelope proteins of heat-shocked Vibrio parahaemolyticus cells by immunoblotting and biotin-labeling methods. Microbiol Immunol 47: 313-319.
77 Wright, A.C., Garrido, V., Debeux, G., Farrell-Eva ns, M., Mudbidri, A., Otwell, W.S. (2007) Evaluation of postharvest-processed oysters by using PCR-based most-probable-number enumeration of Vibrio vulnificus bacteria. Appl Environ Microbiol 73:7477-7481. Yeung, P.S.M., Boor, K. (2004) Epidemiology, pat hogenesis and prevention of foodborne Vibrio parahaemolyticus infections. Foodb Path Dis 1:74-88.
78 BIOGRAPHICAL SKETCH Leann Heldt-W ieand Manley grew up in nor theastern Pennsylvania, where she dropped out of high school to pursue a ca reer in the thoroughbred horse racing industry. While working as an exercise rider on three continents, she managed to earn an A.A. degree from Central Florida Community College and a Bachelor of scie nce from the University of Florida. After teaching AP sciences for 6 years at Dunnellon High School and Forest High School, she returned to graduate school. She is currently the underg raduate core lab coordinator for the Howard Hughes Medical Institute grant located in Shands Hospital.