<%BANNER%>

Modeling the effects of freezing rates, storage temperatures and times on inactivation of Vibrio vulnificus

Permanent Link: http://ufdc.ufl.edu/UFE0024457/00001

Material Information

Title: Modeling the effects of freezing rates, storage temperatures and times on inactivation of Vibrio vulnificus
Physical Description: 1 online resource (101 p.)
Language: english
Creator: Seminario, Diana
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: freezing, inactivation, modeling, predictive, storage, vibrio, vulnificus
Food Science and Human Nutrition -- Dissertations, Academic -- UF
Genre: Food Science and Human Nutrition thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Vibrio vulnificus (Vv) is commonly found in raw oysters. When ingested, it can cause fatal septicemia in immune-compromised individuals. Since consumers prefer raw oysters, post-harvest treatments that reduce Vv while preserving the raw oyster?s sensory characteristics are important. Freezing can reduce Vv and increase the shelf life of oysters to more than 12 weeks when stored at -20degreeC. The objective of this study was to develop predictive inactivation kinetic models for pure cultures of Vv using freezing temperatures and rates, frozen storage temperatures and times. Vv was diluted in phosphate buffered saline (PBS) to obtain about 10^7 colonies/ml. Three sets of vials were frozen at -10degreeC, -35degreeC, and -80degreeC, and stored at -10degreeC. Survival of Vv was followed after freezing and storage at 0, 3, 6 and 9 days. Two other sets were frozen and stored at -35degreeC and -80degreeC, and survival was followed after freezing and every week for six weeks of frozen storage. For every treatment, time-temperature data was obtained using thermocouples in blank vials. The predictive model was developed by analyzing the inactivation data with first order, Weibull, and Peleg inactivation models. The effects of three freezing temperatures on survival of Vv was not significantly different (alpha= 0.05), and caused an average 1.6 log reduction. After one week, samples stored at -35degreeC and -80degreeC showed 0.06 and 1.10 log reductions from the initial counts, respectively. Samples stored at -10degreeC showed more than 3 log reduction regardless of initial freezing temperature. The combined effect of freezing and one week frozen storage resulted in 1.5, 2.6 and 4.9 log reductions for samples frozen and stored at -80degreeC, -35degreeC and -10degreeC, respectively. Inactivation of Vv may be caused by intracellular formation ice crystals and their subsequent growth (re-crystallization) during frozen storage. Storage temperature was the critical parameter in survival of Vv. A modified Weibull model successfully predicted Vv survival during frozen storage: log10(Nt) = log10(No) - 1.22 - {t / 10 ^ (-1.163 - 0.0466 T) ^ (0.00025 T2 + 0.01524 T + 0.49325)}, where No and Nt are initial and time t survival counts, T is frozen storage temperature.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Diana Seminario.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Rodrick, Gary E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024457:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024457/00001

Material Information

Title: Modeling the effects of freezing rates, storage temperatures and times on inactivation of Vibrio vulnificus
Physical Description: 1 online resource (101 p.)
Language: english
Creator: Seminario, Diana
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: freezing, inactivation, modeling, predictive, storage, vibrio, vulnificus
Food Science and Human Nutrition -- Dissertations, Academic -- UF
Genre: Food Science and Human Nutrition thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Vibrio vulnificus (Vv) is commonly found in raw oysters. When ingested, it can cause fatal septicemia in immune-compromised individuals. Since consumers prefer raw oysters, post-harvest treatments that reduce Vv while preserving the raw oyster?s sensory characteristics are important. Freezing can reduce Vv and increase the shelf life of oysters to more than 12 weeks when stored at -20degreeC. The objective of this study was to develop predictive inactivation kinetic models for pure cultures of Vv using freezing temperatures and rates, frozen storage temperatures and times. Vv was diluted in phosphate buffered saline (PBS) to obtain about 10^7 colonies/ml. Three sets of vials were frozen at -10degreeC, -35degreeC, and -80degreeC, and stored at -10degreeC. Survival of Vv was followed after freezing and storage at 0, 3, 6 and 9 days. Two other sets were frozen and stored at -35degreeC and -80degreeC, and survival was followed after freezing and every week for six weeks of frozen storage. For every treatment, time-temperature data was obtained using thermocouples in blank vials. The predictive model was developed by analyzing the inactivation data with first order, Weibull, and Peleg inactivation models. The effects of three freezing temperatures on survival of Vv was not significantly different (alpha= 0.05), and caused an average 1.6 log reduction. After one week, samples stored at -35degreeC and -80degreeC showed 0.06 and 1.10 log reductions from the initial counts, respectively. Samples stored at -10degreeC showed more than 3 log reduction regardless of initial freezing temperature. The combined effect of freezing and one week frozen storage resulted in 1.5, 2.6 and 4.9 log reductions for samples frozen and stored at -80degreeC, -35degreeC and -10degreeC, respectively. Inactivation of Vv may be caused by intracellular formation ice crystals and their subsequent growth (re-crystallization) during frozen storage. Storage temperature was the critical parameter in survival of Vv. A modified Weibull model successfully predicted Vv survival during frozen storage: log10(Nt) = log10(No) - 1.22 - {t / 10 ^ (-1.163 - 0.0466 T) ^ (0.00025 T2 + 0.01524 T + 0.49325)}, where No and Nt are initial and time t survival counts, T is frozen storage temperature.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Diana Seminario.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Rodrick, Gary E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024457:00001


This item has the following downloads:


Full Text

PAGE 1

1 MODELING THE EFFECTS OF FREEZING RATES, STORAGE TEMPERATURES AND TIMES ON INACTIVATION OF Vibrio vulnificus By DIANA MARIA SEMINARIO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILL MENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 200 9

PAGE 2

2 200 9 Diana Maria Seminario

PAGE 3

3 To my family

PAGE 4

4 ACKNOWLEDGMENTS I would like to express my special gratitude to Dr. M urat Ba laban for his guidance and unconditional support but also for showing his students the true meaning of dedication and hard work To Dr. Gary Rodrick for accepting me as his student half way through my graduate studies and helping me complete this p roject. I would also like to thank Dr. Bruce Welt, Dr. Steve Otwell, Victor Garrido, and Dr. Charlie Sims for their guidance and assistance in this project, and Dr. Ronald Schmidt for his countless advice. T o all my labmates throughout these years for th eir friendship and for making the lab my second home I would like to specially thank Maria and Alberto who went out of their way to help me with my research and data analysis. To my parents, Enrique and Aurelia, who made me who I am, and gave me a strong foundation to grow and face life on my own. To my sisters, Daniela, Analia, and Monica, with whom I have share d all kinds of experiences and each one in their special way have helped me to get where I am. To Roberto for his support and patience, and for being there for me every step of the way. Finally, I would like to thank God and the Virgin Mary without them

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 9 ABSTRACT ................................ ................................ ................................ ................................ ... 11 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 13 2 REVIEW OF THE LITERATURE ................................ ................................ ........................ 16 Background on Vibrio vulnificus ................................ ................................ ............................ 16 Ecology ................................ ................................ ................................ ............................ 16 Types of Infection and Disease ................................ ................................ ....................... 17 Population at Risk ................................ ................................ ................................ ............ 18 Vibrio vulnificus and the Oyster Industry ................................ ................................ ............... 18 Occurrence and Seasonality ................................ ................................ ............................ 18 Illnesses Asso ciated with Oyster Consumption ................................ .............................. 19 Infective Dose ................................ ................................ ................................ .................. 20 Methods for Reducing Vibrio vulnificus in Oysters ................................ ........................ 21 Effect of Freezing on Vibrio vulnificus ................................ ................................ .................. 27 Survival of Microorganisms During Freezing ................................ ................................ 27 Viable but Non Culturable State ................................ ................................ ...................... 29 Cold Adaptation ................................ ................................ ................................ ............... 31 Previous Freezing Studies with Vibrio vulnificus ................................ ........................... 33 Microbial Modeling ................................ ................................ ................................ ................ 34 Classification of Models ................................ ................................ ................................ .. 35 Validation of Models ................................ ................................ ................................ ....... 36 Limitations of Models ................................ ................................ ................................ ..... 36 Objectives ................................ ................................ ................................ ............................... 37 3 MATERIALS AND METHODS ................................ ................................ ............................ 39 Bacterial Strain and Culture Conditions ................................ ................................ ................. 39 Sample Preparation ................................ ................................ ................................ ................. 39 Fre ezing Studies ................................ ................................ ................................ ...................... 40 Temperature Data Acquisition ................................ ................................ ................................ 42 Freezing Rates and Slopes ................................ ................................ ................................ ...... 43 Statistical Analysis ................................ ................................ ................................ .................. 44 Model Development ................................ ................................ ................................ ............... 45 Inactivation During Frozen Storage ................................ ................................ ................ 46

PAGE 6

6 Inactivation During Freezing ................................ ................................ ........................... 50 Analysis of Residuals ................................ ................................ ................................ ...... 51 4 RESULTS AND DISCUSSION ................................ ................................ ............................. 52 Freezing Studies ................................ ................................ ................................ ...................... 52 Injured Cells ................................ ................................ ................................ .................... 52 Thawing Temperatures ................................ ................................ ................................ .... 53 Freezing Temperatures and Rates ................................ ................................ ................... 54 Effect of Storage Temperatures and Times ................................ ................................ ..... 57 Effect of Overall Treatments ................................ ................................ ........................... 62 Inactivation Model ................................ ................................ ................................ .................. 64 Inactivation During Frozen Storage ................................ ................................ ................ 64 Inactivation During Freezing ................................ ................................ ........................... 73 Final Inactivation Model ................................ ................................ ................................ 74 Analysis of Residuals ................................ ................................ ................................ ...... 75 5 CONCLUSIONS AND FURTHER STUDIES ................................ ................................ ...... 78 Conclusions ................................ ................................ ................................ ............................. 78 Freezing Studies ................................ ................................ ................................ .............. 78 Inactivation Model ................................ ................................ ................................ ........... 79 Further Studies ................................ ................................ ................................ ........................ 79 APPENDIX A SURVIV AL DATASETS ................................ ................................ ................................ ....... 82 B STATISTICAL ANALYSIS ................................ ................................ ................................ ... 83 LIST OF REFERENCES ................................ ................................ ................................ ............... 95 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 101

PAGE 7

7 LIST OF TABLES Table page 3 1 Freezing studies experimental design ................................ ................................ ................ 40 4 1 Lengths of freezing phases in minutes ................................ ................................ ............... 55 4 2 Slopes of cooling rates ................................ ................................ ................................ ....... 56 4 3 Overall effect of freezing tem peratures in samples stored at 10C. ................................ 60 4 4 First order isothermal analysis ................................ ................................ ........................... 64 4 5 Weibull isothermal analysis ................................ ................................ ............................... 67 4 6 Weibull non isothermal analysis ................................ ................................ ........................ 70 4 7 Peleg isothermal analysis ................................ ................................ ................................ ... 71 4 8 Inactivation during freezing ................................ ................................ ............................... 74 4 9 Comparison between experimental and predicted values ................................ .................. 76 A 1 Survival datasets fro m Section A ................................ ................................ ....................... 82 A 2 Survival datasets from Section B ................................ ................................ ....................... 82 A 3 Survival datasets from Section C ................................ ................................ ....................... 82 B 1 Statistical analysis of the effect of counting methods on bacterial counts, Sec tion A ....... 83 B 2 Statistical analysis of effect thawing temperature (C) on bacterial counts, Section A .... 84 B 3 Statistical analysis of effect of freezing temperature (C) on log 10 reductions, Section B and C ................................ ................................ ................................ .............................. 85 B 4 Statistical analysis of effect of storage temperature ( 35 and 80C) and time (days) on log 10 reductions, Section B ................................ ................................ ........................... 86 B 5 Statistical analysis of effect of storage temper ature ( 35 and 80C) on log 10 reductions when considering each time temperature a separate treatment, Section B. ..... 88 B 6 Statistical analysis of effect of freezing temperature (C) and stora ge time (days) at 10C on log 10 reductions, Section C ................................ ................................ ................. 89 B 7 Statistical analysis of effect of freezing temperature (C) on log 10 reductions when considering each time temperature a sepa rate treatment, Section C ................................ 90

PAGE 8

8 B 8 Statistical analysis of effect of storage temperature (C) and one week of storage on log 10 reductions, Section B and C ................................ ................................ ...................... 91 B 9 Statistical analysis of cumulative effect of freezing and storage temperature (C) and one week of storage on log 10 reductions, Section B and C ................................ ................ 92 B 10 Stat istical analysis of the temperature dependent parameter Alpha for the Weibull model for inactivation during storage ................................ ................................ ................ 93

PAGE 9

9 LIST OF FIGURES Figure page 3 1 Vial and thermocouple ................................ ................................ ................................ ....... 43 3 2 Example of cooling rate vs. time plot ................................ ................................ ................ 44 4 1 Bacterial counts/ml obtained by direct plating (CF U) and by most probable number (MPN) shown in logarithmic scale ................................ ................................ .................... 52 4 2 Thawing curves ................................ ................................ ................................ .................. 53 4 3 Bacterial counts/ml obtained after thawing at 4C and at 30C, in logarithmic scale ....... 54 4 4 Time temperature curves ................................ ................................ ................................ ... 55 4 5 Cooling rate plots ................................ ................................ ................................ ............... 56 4 6 Log 10 reduction during storage at 35C and 80C ................................ .......................... 58 4 7 Log 10 reduction during storage at 10C ................................ ................................ ............ 60 4 8 Log 10 reduction after 1 week of storage at 10C, 35C, and 80C ................................ 61 4 9 Log 10 reduc tion of overall treatments ................................ ................................ ................ 63 4 10 Log 10 D values vs temperature ................................ ................................ .......................... 64 4 11 First order isothermal inactivation during storage at 80C ................................ .............. 65 4 12 First order isothermal inactivation during storage at 35C ................................ .............. 66 4 13 First order isothermal inactivation during storage at 10C ................................ .............. 66 4 14 values vs temperature ................................ ................................ ................................ ..... 67 4 15 values vs temperature ................................ ................................ .......................... 68 4 16 values vs temperature ................................ ................................ ..................... 68 4 17 Weibull isothermal inactivation during storage at 80C ................................ .................. 69 4 18 Weibull isothermal inactivation during storage at 35C ................................ .................. 70 4 19 Weibull isothermal inactivation during storage at 35C ................................ .................. 70 4 20 N values vs. temperature ................................ ................................ ................................ ... 71

PAGE 10

10 4 21 Peleg isothermal inactivation during storage at 80C. ................................ ..................... 72 4 22 Peleg isothermal inactivation during storage at 35C ................................ ...................... 73 4 23 Peleg isothermal inactivation during storage at 10C ................................ ...................... 73 4 24 Inactivation model and experimental data ................................ ................................ ......... 75 4 25 Re siduals vs. predicted values ................................ ................................ ........................... 77

PAGE 11

11 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science MODE LING THE EFFECTS OF FREEZING RATES, STORAGE TEMPERATURES AND TIMES ON INACTIVATION OF Vibrio vulnificus By Diana Maria Seminario May 2009 Chair: Gary Rodrick Major: Food Science and Human Nutrition Vibrio vulnificus (Vv) is commonly found in raw oyster s. When ingested, it can cause fatal septicemia in immune compromised individuals. Since consumer s prefer raw oysters, post harvest treatments that reduce Vv sensory characteristics are important. Freezing can reduce Vv an d increase th e shelf life of oysters to more than 12 weeks when stored at 20 C The objective of this study was to develop predictive inactivation kinetic models for pure cultures of Vv using freezing temperatures and rates, frozen storage temperatures an d times. Vv was diluted in phosphate buffered saline (PBS) to obtain about 10 7 colonies/ml. Three sets of vials were frozen at 10C, 35C, and 80C, and stored at 10C. Survival of Vv was followed after freezing and storage at 0, 3, 6 and 9 days. Two o ther sets were frozen and stored at 35C and 80C, and survival was followed after freezing and every week for six weeks of frozen storage. For every treatment, time temperature data was obtained using thermocouples in blank vials. The predictive model w as developed by analyzing the inactivation data with first order, Weibull, and Peleg inactivation models.

PAGE 12

12 The effects of three freezing temperatures on survival of Vv was not significantly different ( = 0.05), and caused an average 1.6 log 10 reduction. Af ter one week, samples stored at 35C and 80C showed 0.06 and 1.10 log 10 reductions from the initial counts, respectively. Samples stored at 10C showed more than 3 log 10 reduction regardless of initial freezing temperature. The combined effect of free zing and one week frozen storage resulted in 1.5, 2.6 and 4.9 log 10 reductions for samples frozen and stored at 80C, 35C and 10C, respectively. Inactivation of Vv may be caused by intracellular formation ice crystals and their subsequent growth (re c rystallization) during frozen storage. Storage temperature was the critical parameter in survival of Vv A modified Weibull model successfully predicted Vv survival during frozen storage: log 10 N t = log 10 N o 1.22 {[t / 10 ^ ( 1.163 0.0466 T)] ^ (0.000 25 T2 + 0.01524 T + 0.49325)}, where N o and N t are initial and time t survival counts, T is frozen storage temperature.

PAGE 13

13 CHAPTER 1 INTRODUCTION Vibrio vulnificus is a lactose positive, G ram negative bacillus that is commonly found in warm coastal waters 1.6 and 2.3 % (Kelly 1982 ; Tamplin and others 1982 ). Vibrio vulnificus can cause three clinical symptoms : (1) a severe skin infection (2) gastroenteritis and (3) primary septicemia. Skin infection occurs after exposure o f open wounds to Vibrio vulnificus contaminated seawater, while gastroenteritis and septicemia occur after eating raw or partially cooked shellfish containing the bacteria While disease caused from this microorganism is not very common, primary septicemia is of special concern in immune compromised individuals or individual s with liver disease, since it is fatal in 50% of cases Raw oyster ingestion has been suggested as the primary source of septicemia since Vibrio vulnificus accumulates and can multiply in the tissues of the oyster while the oyster is filter feeding ( Tamplin and others 1982; Kelly and Dinuzzo 1985; Klontz and others 1988; Ruple and Cook 1992; Tamplin and Capers 1992; Shapiro and others 1998; Strom and Paranjpye 2000 ). Due to negative publ icity warning consumers of potential risks related to consuming raw oysters, a 60% decline in oyster consumption occurred between 1992 and 2002 (Coleman 2003). As a result, the oyster industry has strengthened conventional techniques, such as high and low temperature treatments, and created new methods to ensure safety. These techniques are meant to reduce the levels of Vibrio vulnificus and other pathogens and preserve Among the post ha rvest treatments industrially applied, heat pasteurization is the most effective in terms of killing the bacteria, but can change the flavor and texture of the oyster High pressure can also kill Vibrio vulnificus but is an expensive technology. Irradiati on preserves original characteristics and extends shelf life, but its use is not openly accepted by oyster

PAGE 14

14 consumers Several other techniques are being developed, such as allowing the oyster to depurate in water with special chemicals to eliminate the bac teria or addin g bacteriophages to kill the microorganism (Bryan and others 1999; Coleman 2003). Unfortunately, none of these techniques preserve raw oysters longer than 21 days under refrigerated conditions On the other hand, freezing of the oysters can increase their shelf life to more than 12 weeks when stored at (Cook and Ruple 1992). Various studies report a reduction in Vibrio vulnificus counts after freezing. Vibrio vulnificus in suspension media kept at ble cells than did and Ruple 1992 ) Similarly, a different study show ed that Vibrio vulnificus in oysters froz en at days and not detectable after 21 days (Jeong and others 1991). These results ag ree with another study carried out in vacuum packed oysters that showed a significant decrease in Vibrio vulnificus numbers in each successive time interval of frozen storage, from 7 to 70 days (Parker and others 1994). Freezing is a common preservation te chnique for foods in general and oysters in particular. If the kinetics of Vibrio vulnificus destruction using this method can be developed, then the model can be applied by the industry more easily and with less costs compared to other existing or potenti al methods, since many producers are already familiar with freezing technology and have the required equipment. Therefore the specific aim of this study wa s to develop quantitative predictive models regarding the inactivation rate of Vibrio vulnificus by d ifferent freezing rates and temperatures as well as different frozen storage temperatures and times. While the application of the model is desi red for Vibrio vulnificus in oysters, this study w as carried out with pure cultures of the bacteria. The uncertai nty of initial numbers of the bacteria and physical differences between

PAGE 15

15 oysters would be sources of variability that make the development of the model difficult. T o avoid this, identical vials inoculated with known numbers of Vibrio vulnificus w ere used fo r this study

PAGE 16

16 CHAPTER 2 REVIEW OF THE LITERA TURE Background on Vibrio vulnificus First desc ribed by Hollis and others (1976 ), Vibrio vulnificus is a Gram negative, lactose positive, motile, curved rod shaped bacterium with a single polar flagellum. This naturally occurring free living inhabitant of estuarine and marine environments has been associated with filter feeding mollusks such as oysters, clams, scallops and mussels. It accumulates and concentrates on their tissues ( Oliver and others 1982; Klontz and others 1988; Strom and Paranjpye 2000 ) Ecology Vibrio vulnificus proliferates in water temperatures above 17C (Kelly 1982, Tamplin and others 1982). In addition it has also be en found in waters as cold as 7.6C ( Wright and others 1996) al though it was previously reported non detectable in waters below 12.5C ( Kelly 1982) Many studies have shown higher concentrations of Vibrio vulnificus during the warmer months of the year, and lower or undetectable levels during the colder months. Klontz and ot hers (1988 ) found a correlation between water temperature and Vibrio vulnificus concentrations; however the correlation was not significant during the warmer months. A subsequent study found a strong correlation between concentrations of Vibrio vulnificus in oysters and water temperature until the temperature reaches 26C. Above this temperature bacterial loads appear to have no additional increase (Kelly 1982; Tamplin and others 1982; Wright and others 1996; Motes and others 1998). In vitro s tudies have s hown that for a twelve hour incubation period, the optimum temperature for the bacteria is 37C ( Kelly 1982). For incubation periods (6 days) Vibrio

PAGE 17

17 vulnificus temperature range i s from 13 to 22 C and temperatures outside this range cause d a 90% reduction in bacterial numbers (Kaspar and Tamplin 1993) Like temperature, s alinity has reportedly shown to be an influence in the occurrence of Vibrio vulnificus Kelly (1982) reported frequently finding the bacteria in salinities lower than 1.65%, while sites w ith salinities of 1.80% and above often yielded negative for the microorganism. However, Tamplin and others (1982) reported that the bacterium was frequently found in salinities of 1.7%, and even more frequently above 2.3% In vitro studies have show n opti mal growth at salinities of 0 .5 to 2% for 12 hour incubation. F or prolonged periods, salinities of 0.5 to 2.5% showed an increase in Vibrio vulnificus numbers (Kelly 198 2; Kaspar and Tamplin 1993). Types of Infection and Disease Vibrio vulnificus can cau se three different diseases: wound infections, primary septicemia and gastroenteritis Primary septicemia is defined as a systemic illness with fever and shock, where Vibrio vulnificus is isolated from blood or other sterile site s Patients usually suffe r from fever and chills that can be accompanied by abdominal cramps, vomiting, diarrhea, and pain in the extremities. Secondary cutaneous lesions, including cellulitis, bullae, and ecchymosis begin to appear within 24 hours. These lesions can become necrot ic and will require surgical debridement or amputation. About 60% of the patients suffer septicemic shocks, and about 50% suffer mental status changes within a week of the illness. Primary septicemia is fatal in 60 75% of the cases. Wound infections are d ef ined as those where the patient had a previous open wound or incurred one while in contact with contaminated seawater or seafood drippings. Analysis of the wound, blood or other sterile site s shows presence of the bacteria. Symptoms are similar to those of primary septicemia, but differ in timing and severity. Inflammation occurs around the wound

PAGE 18

18 site, and the disease can advance to lesions similar to those of primary septicemia, and finally to septicemic symptoms. Fatality ranges from 20 to 30%. Fi nally, gastroenteritis is characterized by abdominal cramps, vomiting and/or diarrhea, with no evidence of open wound infection, and where the bacterium is isolated only from stool These cases do not require hospitalization, and are largely unreported. The over all fatality rate for Vibrio vulnificus is from 30% to 48% ( Klontz and others 1988; Shapiro and others 1998; Strom and Paranjpye 2000). Population at Risk Existing medical conditions of patients that become infected with Vibrio vulnificus have a large imp act o n the outcome of the disease. Shapiro and others (1998) reported the following preexisting conditions from data obtained from the Center for Disease Control and Prevention (CDC) Gulf Coast Surveillance System : liver disease, alcoholism, diabetes, gast rointestinal surgery, peptic ulcer disease, hematologic disorder, immunodeficiency, malignancy, and renal disease. Among Vibrio vulnificus infections reported in the United States between 1988 and 1996, liver disease or alcoholism w ere present in 86 % of th ose cases that resulted in primary sep ticemia. For that period of time, mortality among those patients was 61% (Shapiro and others 1998 ) Vibrio vulnificus and the Oyster Industry Occurrence and Seasonality Several studies have reported close association o f Vibrio vulnificus with oyster tissues. Kelly and DiNuzzo ( 1985 ) suggested that the oyster acquired the bacteria passively by filtration of contaminated seawater. In fact, relationship s between bacterial concentration in seawater and bacterial concentrati on in oysters have been found by several authors. D uring the warmer months of the year, in spring and summer numbers of Vibrio vulnificus in seawater and oysters have

PAGE 19

19 been found at their highest levels, while the bacterial concentration during colder mont hs is very low and often undetectable for both seawater and oysters Ruple and Cook (1992) found that 40% of the oysters harvested in cool months had Vibrio vulnificus while 100% of the oysters harvested in warm weather had the bacteria, and aroun d 67% of them had counts exceeding 10 5 most probable number (MPN) per gram. MPN is a procedure to estimate population density of viable organisms in a sample by serial dilution, enrichment and plating Occurrence of the bacteria in oysters was favored by high temp eratures and relatively low salinity of the harvesting environment These findings coincide with the incidence of human infections due to Vibrio vulnificus Eighty five percent occurred during warm months of the year (Tamplin and others 1982; Kelly and Din uzzo 1985; Ruple and Cook 1992; Tamplin and Capers 1992) In the United States, Vibrio vulnificus has been isol ated from water and/or oysters in a variety of places. Lower densities were found in Pacific, Canadian and North Atlantic waters, while the Gulf of Mexico, mid Atlantic and Chesapeake Bay presented higher densities (Drake and others 2007). Other regions include Belgium, Denmark, Germany, Spain, Sweden and the Neth e rlands (Feldhusen 2000). Illnesses A ssociated with Oyster C onsumption Primary septi cemia from Vibrio vulnificus probably the leading cause of seafood related deaths in the United States, is largely associated with the consumption of raw oysters Between 1988 and 1996, 96% of the primary septicemia cases were on patients who had consumed raw oysters. In that period of time, 204 of the Vibrio vulnificus infections reported were associated with oyster consumption. Of those, p rimary septicemia accounted for 89% of the infections while gastroenteritis accounted for the remaining 11% (S h apiro and others 1998) Although the number of cases of primary septicemia is small, the mortality rate in patients with pre existing medical conditions is high (Ruple and Cook 1992).

PAGE 20

20 Vibrio vulnificus illnesses associated with oyster consumption have been repor ted from different harvesting sites S h a p iro and others (1998) traced back 49% of primary septicemia cases from raw oyster ingestion between 1988 and 1996 in the United States and found that a large portion of these oysters came from Florida and Louisiana The major harvesting sites were Apalachicola B ay in Florida, Mobile Bay in Al abama, Galveston Bay in Texas a several bays in Louisiana. The strong association between Vibrio vulnificus illnesses and raw oyster consumption compared to other fish and she llfish has been attributed to the high concentration of Vibrio vulnificus in oyster tissues and that oysters are commonly eaten raw (Ruple and Cook 1992; Shapiro and others 1998) Some finfish have been found to posses high densities of Vibrio vulnificus however these species are of less concern since, unlike oysters, they are usually cooked before consumption. However, concern has been raised due to their mobility, which can lead to transportation of the microorganism to previously uncontaminated sites (D ePaola and others 1994). Infective Dose The i nfective dose of Vibrio vulnificus in humans is not known and this may be due to large differences in the amount of raw oyster s consumed and bacterial levels per oyster i n those per son s who get the infection. R eports have shown that patients had ingested between 3 to 48 raw oysters usually 24 hours before the onset of the illness. Median incubation period is 18 hours (Klontz and other s 1988) Jackson and others (1997) reported data that showed illnesses occurri ng when Vibrio vulnificus levels in the oysters reached or exceeded 10 3 /g. When relating bacterial concentration with amount of oysters consumed in each case, patients had consumed more than 10 5 bacteria. Blood isolates showed that Vibrio vulnificus was pr esent at 10 7 colony forming units (CFU) per

PAGE 21

21 m illiliter (ml) of blood or more. Further analysis of samples showed that not all Vibrio vulnificus isolated were virulent and indicated that infections are the result of proliferation of a single pathogenic strai n. In Florida, a n estimated 70,000 raw oyster consumers have pre existing medical conditions, and only 5 10 cases are reported per year. Kaysner and others (1989) reported that for oysters weighting 6 to 9 grams, a total number of log 10 3.68 to 4.99 Vibrio vulnificus CFU per oyster is present. Considering that in average 6 to 12 oysters are consumed per meal, the population at risk is consuming large amounts of the bacteria. The low incidences in illnesses suggest that host specificity is a key element in V ibrio vulnificus disease (Jackson and others 1997). Methods for Reducing Vibrio vulnificus in Oysters Due to negative publicity warning consumers of potential risks related to the consumption of raw oysters, a 60% decline in oyster consumption occurred bet ween 1992 and 2002 (Coleman 2003). As a result, researchers have been trying to find ways to control and reduce Vibrio vulnificus in oysters. A variety of processes have been studied, and some have been found to be more effective than others. However, cons preference for raw oysters is a limiting factor. While the main concern of the oyster industry is the elimination of Vibrio vulnificus the reality is that the consumer not only wants a safe product, but also a raw product Temperature control and re frigeration Temperature control of the oysters between harvest and processing or consumption is of critical importance. Several authors have reported that temperature abuse causes an increase in Vibrio vulnificus while refrigeration can avoid this increas e Hood and others (1983) reported that endogenous Vibrio vulnificus showed a significant increase after 7 days of storage at 8C and 20C, but declined after 14 days. Cook and Ruple (1989) found that bacteria l counts

PAGE 22

22 increased a t 22 and 30C but not at 1 0C. Cook (1994, 1997) reported that when oysters are kept at 18C and ambient temperature (22.8 to 31.1C) for 14 and 30 hours respectively the increase in Vibrio vulnificus wa s significant Oysters kept at 10 and 13C ha d no change in Vibrio numbers for 30 hours. F ailure to refrigerate oysters after harvesting can result in almost 2 log i change in bacterial concentration in artificially contaminated shellstock kept at temperatures as high as 22C, attributing the difference in their results to better enumeration techniques than those used in previous studies After reporting an increase of more than 2 logs after 3 days when oysters were stored at 30C, Kaspar and Ta mplin (1993) suggested that the strain used by Murphy and Oliver (1992) lacked the ability to colonize oysters or to compete with the rest of the flora. Lowering the temperature of the oysters with refrigeration and/or icing after harvest is not only used to prevent growth of Vibrio vulnificus between harvest and processing, but according to some authors, can also be a method of reduction itself if applied for prolonged periods O liver (1981) reported a dramatic reduction in Vibrio vulnificus when cells wer e incubated in oyster broth or whole oysters between 0.5 to 4C, with a log reduction for every 0.5 to 2 hours of storage Cook and Ruple (1992) reported large bacterial reductions when oysters were stored at 4, 0 and 1.9C for 14 days. Schwarz (2000) sho wed that rapid chilling of the oysters through an ice water bath followed by refrigeration reduced Vibrio vulnificus between 89 99% as compared to oysters directly refrigerated. On the other hand, Quevedo and others (2005) reported a minimal reduction in Vibrio vulnificus and an increase in total and fecal bacteria after immediate post harvest ice immersion followed by refrigeration Kaysner and others (1989) reported the survival of the microorganism

PAGE 23

23 on shellstock during a 2 week refrigerated storage and in shucked oysters for 6 weeks. It is generally accepted that during refrigerated storage Vibrio vulnificus levels may even decline, however, refrigerated storage cannot be relied upon to eliminate the bacteria (D rake and others 2007). Thermal processing a nd heat shock Pasteurization and canning of the oysters result in an extension of the shelf life as well as a dramatic reduction of the bacterial population. Pasteurization is commonly used in shucked oysters, which need to be refrigerated. Canned oysters are rarely associated with Vibrio vulnificus illnesses, since the processing requirements exceed the time and temperature required for elimination of most resistant pathogens (Richards 2003). Vibrio vulnificus is a fairly heat sensitive organism, with a D value of 12 seco ( Johnston and Brown 2002) In fact, it has been reported that temperatures above 4 death (Cook and Ruple 1992 ; Ama and other s 1994 ). Although the application of heat to oy sters has been recognized as an effective method, heating is complicated by the deterioration of nutritional and sensory quality (Kim and others 1997). An alternative to high temperature process is cold pasteurization, which results in less physical chang es. Cook and Ruple (1992) found that exposing the oysters to 50C for 5 to 10 minutes is sufficient to eliminate Vibrio vulnificus without imparting a noticeable change in appearance or cooked taste. Andrews and others (2000) reported that t reating oyster s for 10 minutes at 50C resulted in a reduction of Vibrio vulnificus to undetectable levels and sensory panels detected no changes in physical characteristics. However, not all microorganisms were eliminated at this temperature; therefore shelf life would be affec ted by spoilage bacteria

PAGE 24

24 Commercial heat shock is the application of low temperature pasteurization followed by a 1 minute cold water spraying. This process is used to facilitate shucking of oysters. Hesselman and others (1999) found that 1 to 4 minutes at 50C reduces Vibrio vulnificus 1 to 4 logs and aids in the shucking process. Similarly, AmeriPure is a patented process in which f resh oysters are cleaned and then placed in a warm bath followed by cold shock in ice cold bath to reduce Vibrio v ulnificus to undetectable levels (Richards 2003). Relaying and d epuration Relaying is a technique that consists of harvesting shellfish from a contaminated area and replanting them into clean water for an extended period. If the water remains clean and t he environmental conditions remain suitable, the shellfish will naturally purge contaminants (Richards 2003). Relaying is not commonly used to remove Vibrio vulnificus from oysters, however Motes and DePaola (1996) showed in a study performed with oysters harvested during summer months and moved to high salinity offshore waters, that Vibrio vulnificus can be reduced to the levels found in January and February, when Vibrio vulnificus related ill nesses have never been reported. Depuration, or controlled puri fication, consists of placing the shellfish in tanks of disinfected, recirculating or flow through seawater to purge contaminants for several days. To allow wat er to recirculate, it needs to b e disinfected, and several chemicals have been used : chlorines, iodophors, ozone (not legal in US), etc, and also physical treatment of the water with ultra violet (UV) light (most commonly used) (Richards 2003). While chlorines and iodophors are very effective in disinfecting the water, the oysters stop pumping in th eir presence (Richards 2003). A s imilar case happens with diacetyl, which eliminates Vibrio vulnificus but when oysters sense th is chemical they stop pumping. With reduced amounts of diacetyl the oysters will continue to filter feed, however the reduced

PAGE 25

25 a mounts of diacetyl were not effective for killing Vibrio vulnificus (Birkenhauer and Oliver 2003). UV disinfection of the water is effective when the temperature in maintained at 15C, while temperatures above 23C can actually cause an increase in Vibrio vulnificus (Tamplin and Capers 1992). This poses a limitation because maintaining the temperature, as well as salinity and dissolved oxygen for these systems can be complicated and expensive. As with relaying, this ristics, but requires extensive labor and equipment. High Pressure f applying pressures around 200 MPa to inactivate Vibrio vulnificus characteristics. One s tudy reported inactivation of pathogenic vibrio species without the risk of them entering the viable but nonculturable ( VBNC ) state, which is a stress produced dormancy state that results in cells that are not able to grow in conventional growth media (see below). H ow ever the study also reported that cells already in that state were more difficult to eliminate (Berlin and others 1999). Another study related pressure with temperature, reporting that increases in times and temperatures causes more death to th e bacteria. However, cold temperatures ( 2 and 1C) help retain the oyster s fresh like characteristics, so a treatment of 250 MPa or more for less than 4 minutes was recommended to achieve a 5 log or more reduction (Kural and Chen 2007). Although the oyst ers are killed with HPP, a study reports that at 400 MPa at 7C in two 5 minutes pulses does not change the flavor and facilitates shucking (Lopez Caballero and others 2000; Richards 2003). Irradiation Irradiation is the use of ionizing gamma rays from rad ionuclides such as Cobalt 60 or Cesium 137 or the use of high energy electrons and X rays produced by machine sources. This method offers various advantages depending on the dose: conservation of raw characteristics,

PAGE 26

26 survival of the oyster therefore shelf life extension, possibility of treating bulk and packed products, etc. Unfortunately, irradiation of oysters has not been approved by FDA yet, requires expensive facilities, and consumers perceive irradiated product s as unhealthy and hazardous. Vibrio vu lnificus is among the most radiation sensitive bacteri a Dixon and Rodrick (1998) reported that Vibrio vulnificus can be eliminated from oysters at doses less than 0.1 kGy. Other studies show that low doses from 0.5 to 1.0 kGy are effective in reducing or eliminating Vibrio vulnificus from live, fresh shucked or frozen oysters. Panelist s were not able to differentiate between treated and untreated oysters (Kilgen and Hemard 1995, Andrews 2003). Other methods Alternative processes are continuously been test ed for the reduction or elimination of Vibrio vulnificus Borazjani and others (2003) tested ultrasound, ozone and organic acids (lemon juice, citric acid and vinegar) for that purpose in whole and half shell oysters. The treatments were ineffective for wh ole oysters. For half shell oysters, treatments with acids were more effective, however all treatments failed to reduce the bacteria to acceptable levels which is 3 MPN/g oyster. Sun and Oliver (1992) found that several FDA approved compounds were effectiv e in causing lethality on Vibrio vulnificus in vitro, however only diacetyl at a minimum concentration of 0.05% was effective in significantly reducing the bacteria in oysters. Freezing and frozen storage Freezing is a popular technique to reduce Vibrio vu lnificus in raw oysters while preserving its shelf life. Various studies report a reduction in Vibrio vulnificus counts after freezing and/or frozen storage. As this topic is the main concern in this stud y, the next section will cover the different effects of freezing o n Vibrio vulnificus

PAGE 27

27 Effect of Freezing on Vibrio vulnificus Survival of Microorganisms During Freezing Freezing microorganisms in simple aqueous solutions has allowed investigators to study the direct effects of freezing on microorganisms without the interventions of undesired variables, and has simplified isolation and enumeration processes following freezing treatments (Lund 2000) However, the actual mechanism of freeze damage to the cell s has n ot been defined yet, and scientists have different theories as to what is affecting the survival of mi croorganisms during freezing, frozen storage and thawing Factors expected to cause damage to cells during freezing are: low temperatures, format ion of intracellular ice formation of extracellular ice, intracellular solute concentration, and extracellular solute concentration. All of these factors will be affected by the rate of freezing. It must be noticed that damage can also occur during thawin g and will be affected by the rate of thawing as well (Lund 2000). In a review about the freezing of biological systems, Mazur (1970) explained that cells generally undergo supercooling, by remaining unfrozen at 10 or 15C, even when the medium around t hem has ice. Since vapor pressure of supercooled water is higher than that of ice, cells need to equilibrate. For slowly cooled cells or cells with high water permeability, equilibrium will be reached by transferring internal water to external ice, resulti ng in dehydration. H owever, if the cell membrane has low water permeability or the cells are cooled rapidly, equilibrium will be achieved when a substance (colloids, dissolved substances, or clumped water molecules) acts as a nucleus for intracellular ice formation The critical rate that defines slow and fast cooling is specific for each cell and depends on its permeability to water and on the ratio of the cell volume and its surface area ( Karow and Webb 1965; Mazur 1970, 1963)

PAGE 28

28 Regardless of how the equi librium was achieved, cells will be subjected which refers to intracellular and extracellular solute concentration that may cause a precipitation of the solutes if their solubilities are exceeded and result in a change i n pH that can be detrimental to cells Slow ly cooled cells will be exposed to this effect for a longer time (Mazur 1970) Lovelock (1953) showed that the effect of the solutes is not instantaneous and attributed the survival of rapidly frozen cells to this observation. Farrant and Morris (1973) suggested that the concentration of solutes itself does not cause injury, but the added stress of freezing and thawing is what causes death of the cells. Formation of intracellular ice produces small crystals that have high surfa ce energies, which will be reduced by growing or by fusing with other small ice crystals. The rate of this process will be higher in smaller crystals and at higher temperatures, so this process is very important during warming. In fact, a study by Bank and Mazur (1973) found that the size of the crystals is a function of initial crystal size, storage temperature and time, and that crystal growth is detectable in temperatures as low as 45C. The formation of large ice crystals is lethal to cells in most cas es and death seems to occur as a result of the extraction of bound water from vital structures and from proteins ( Kar r ow and Webb 1965; Mazur 1970). Mazur (1970) suggested that the opti mum rate range for cell survival is one that will be slow enough to pr event intracellular ice formation, but fast enough to prevent the cell from being affected by prolonged exposure to concentrated solutes However, for some cells, there is no optimum range In their review about the responses of mesophilic bacteria to cold stress, Panoff and others (1998) list membrane damage and DNA denaturation as possible causes of bacterial death after freezing and thawing besides protein damage Bacterial cells, which are sensitive to freezing,

PAGE 29

29 will usually undergo a slow death rate u pon freezing, and their response can be described as an exponential function (Haines 1938; Panoff and others 1998) A lur and Grec z (1975) reported that DNA fragmentation occurred at higher ra tes after fast freezing than it did after slow freezing. However a fter 24 hour storage the slow frozen cells yielded the same results as fast frozen cells. The authors suggested that death was related to DNA and membrane degradation since DNA is attached to plasma membrane. Viable but Non Culturable State The viable but nonculturable state (VBNC) is a dormancy state in which the bacteria fail to form colonies on routine bacteriological media on which they will normally grow, but are alive and capable of renewed metabolic activity (Oliver 2005). This state is different from cell injury, because even though injured cells cannot be grown on selective media, they can still be grown in nutrient rich media (Drake and others 2007). Enumeration of total cell counts, including VBNC cells can be done by a variety of methods, in cluding acridine orange staining, substrate responsive assay, examination of metabolic activity, or by establishing intact cytoplasmic membrane (Oliver 2005). McDougald and others (1998) present a good review of these methods. VBNC state is a response to e nvironmental stresses which would otherwise be lethal to the bacteria, such as suboptimal pH, cold temperatures, and starvation (Oliver 2005). Vibrio vulnificus maintained at ambient temperature in a nutrient free media remained culturable for more than 30 days, and survive d more than 1500. On the other hand, the bacteria enter ed the VBNC state within 40 days when moved to a storage temperature of 5C, while viability tests showed a concentration of more than 10 6 viable cells per ml (Wolf and Oliver 199 2 O liver 1995). Likewise, cells maintained at 35C and moved to 15C prior to storage at 6C remained culturable, while cells moved directly from 35C to 6C entered the VBNC state (Bryan and other s 1999).

PAGE 30

30 Morphologically VBNC cells have been reported to dec rease in cell size, and the shape changes from rods to cocci (Oliver and others 1991). Metabolic changes occur as well. Linder and Oliver (1989) reported an increase in short chain saturated fatty acids as the incubation temperature decreased, which allows membrane fluidity at lower temperatures. The same effect is achieved with an increase in unsaturated fatty acids as temperature decreases, as was found by Oliver and Colwell (1973). Temperature changes cause a decrease in protein, RNA, and DNA synthesis i n less than 15 minutes; however cold shock proteins are synthesized during the first hour (Oliver 1995, McGovern and Oliver 1995). Resuscitation of the VBNC state has been reported to occur by reversing the factor that induced the state in the first place, like increasing the temperature after cold stress (Nilsson and other s 1991). Oliver and others (1995) inoculated Vibrio vulnificus cells in membrane diffusion chambers in estuarine water during cold months, and observed that the cells entered into the VB NC state. When VBNC cells were placed in estuarine waters during the warmer months, cells appear ed to resuscitate into fully culturable state within 24 hours. Oliver and Bockian (1995) reported that VBNC Vibrio vulnificus cells could also resuscitate in vi vo as they observed when injecting the bacteria in mice. The actual existence of the VBNC state has been a topic of discussion for a long time. Some authors state that some viable cells remain viable in unfavorable conditions, and when they are subjected to optimum conditions replicate and become detectable on the appropriate media, giving the appearance of resuscitation (Bogosian and others 2000). Nevertheless, whether the VBNC state exists with its subsequent resuscitation or not, the main concern lies i n the virulence of cells in this state. Linder and Oliver (1989) suggested a loss of virulence in VBNC cells injected in mice, but the infective dose that was used in their study,

PAGE 31

31 ( 5x10 4 cells) was lower than the 50% lethal dose. In a later study, Oliver and Bockian (1995) injected mice with 10 5 VBNC Vibrio vulnificus cells, showing culturable cells in mice that died. Their results suggested that even though the virulence decreases in the VBNC state, the cells can still cause death. Cold Adaptation McGover n and Oliver (1995) studied the response of Vibrio vulnificus to moderate temperature downshift s focusing o n protein synthesis Since the VBNC state in the bacteria can be induced at 10C and results in a halt in growth and protein synthesis, the authors studied a change in temperature from 23C to 13C. Doubling times for the higher temperature occurred in 3 hours, while it took 13.1 hours to achieve the same amount of cells when stored at 13C, after a little lag. They found that the rate of protein synt hesis after the temperature downshift presented a n immediate and sharp d e cline and that during the cold stress response 40 proteins were synthesized at higher levels. Peak expression w as reached in 15 minutes by 5 of the proteins, in 30 minutes by other 1 0, in 1 hour for another 5, in 2 hours by 13, and in 4 hours for the remaining 7. After studying cultures kept at 35C and 15C, Bryan and others (1999) observed that cultures kept at 15C underwent a decline in cell counts, followed by an increase to the initial level and a maintenance of bacterial counts for 162 hours. Colonies kept at 35C showed a n initial increase in the first 24 hours followed by a 3 log decrease in the remaining 138 hours. These results suggested that Vibrio vulnificus can adapt and remain viable at 15C even though growth i s stopped. The authors compared a culture shifted from 35C to 6C, against one with a 3 hour 15C step in the middle. The la t ter showed a 4.5 times higher viability than the first culture. These results suggest that keeping the cultures at the intermediate temperature of 15C before final

PAGE 32

32 storage at 6C allows the bacteria to improve their tolerance to colder temperatures, possibly by synthesizing cold adaptive protein(s) as opposed to the cold shock proteins fo und in other bacteria. In fact, the protein analysis also suggested that proteins responsible for the adaptive response of the bacteria differ from those found in E. c oli Furthermore, McGovern and Oliver (1995) reported in their study that the protein tha t was not present before the temperature downshift and appear ed increased by a factor of 35 one hour after the temperature change, largely differ in apparent molecular mass and isoelectric point from cold shock proteins found in other bacteria. Bryan and others (1999) reported that w hen protein synthesis was inhibited at 15C before storage at 6C, a rapid decline in viable counts was observed. Cells without protein inhibition showed a 6 log higher viability than the others treated with the same temperatur es Even though a decline occurred at 15C in both groups of cells, the cells capable o f synthesizing proteins at 15C were able to maintain their numbers better when stored at 6C T hese results suggested that survival at 6C was possible after protective cold induced proteins were synthesized. Finally, the authors investigated the effect of the intermediate 15C temperature bef ore freezing, and showed that the one frozen directly from 35C. This result suggest s that submitting the bacteria to refrigeration temperatures may be counter productive for freezing as a post harvest process to reduce Vibrio vulnificus in oysters. Most studies of cold adaptation and cold shock in bacteria have been done by counting CFU on solid media. The evaluation of VBNC cells and their resuscitation will probably modify the current understanding of cold responses in cells (Panoff a nd others 1998 )

PAGE 33

33 Previous Freezing Studies with Vibrio vulnificus Several stu dies have been conducted to evaluate the effect of freezing on Vibrio vulnificus in oysters and in suspension media. These studies report a reduction of the bacteria in oysters and oyster meat. Jeong and others (1991) reported a reduction in the bacteria o f 2.8 MPN/g log units in shellstock oysters after 15 days of storage at 25C. Cook and Ruple (1992) froze shucked oyster meat contaminated with naturall y occurring Vibrio vulnificus and observed large reductions, but the bacteria was still present after 1 2 weeks of storage at 20C Oyster meats were frozen by three different methods: Individually Quick Frozen (IQF) with liquid carbon dioxide at 30C, blast frozen at 20C, and conventionally frozen at 20C. Parker and others (1994) reported significant reductions of 3 to 4 logs in Vibrio vulnificus in oysters frozen at 20C. A portion of the oysters analyzed were vacuum packed, and those showed a significantly greater reduction of the bacteria. Most reductions occurred within the first 7 days, but coun ts continued to decline during storage. However, after 30 days, 2 logs of the bacteria were still detected, and after 70 days, some samples still contained 1 log. Mestey and Rodrick (2003) studied the effects of cryogenic freezing and subsequent storage on Vibrio vulnificus by using and comparing carbon dioxide ( 67C) and liquid nitrogen ( 91C) to freeze whole and half shell oysters. For whole oysters, bacterial levels were undetectable after 14 days for carbon dioxide and after 21 days in most cases for liquid nitrogen. Recoverable numbers for half shell oysters were lower at all storage intervals, and were undetectable after 14 days for both methods, and a few times only after 7 days. Studies with pure cultures provide a better understanding and an ease in processing for freezing studies, due to the fewer variables affecting the outcome of experiments. However, it must be remembered at all times that survival of the bacteria in oysters may be different due to variations in freezing and thawing rates and p rotective effect of the oyster

PAGE 34

34 Cook and Ruple (1992) reported that pure cultures suspended in media and frozen at 20C showed a greater decrease than those stored at 4, 0 and 1.9C, but even after 19 days approximately 1 log of culturable cells c ould be recovered. Bryan and others (1999) reported approximately 4 log reduction in 75 hours when Vibrio vulnificus cultures where frozen at 78C but only a 2 log reduction if the microorganism was previously incubated at 15C. Bang and Drake (2002) stu died the effect of freeze thaw cycles on the survival of cold stressed bacteria. Pure cultures were suspended in dilution media and kept at 5C for 5 days. After the incubation period, cultures were subjected to 5 freeze thaw cycles consisting of freezing and storing at 20C for 24 hours followed by air thawing at 23C for 30 minutes. Bacterial counts declined with freeze thaw cycles, but it was found that cold stress did not significantly improve survival. Contrary to previously mentioned studies, Johnsto n and Brown (2002) reported that freez ing did not have much effect on Vibrio vulnificus showing a maximum of 1.5 log reduction after 24 hours. This result was obtained through Thoma count, which allows observation through microscopy. Staining of cells sho wed very little membrane damage. Microbial Model ing Microbial modeling allows the description and prediction of microbial behavior under specific environmental conditions. These conditions can be intrinsic, like pH, or extrinsic, like temperature or salini ty Microbial responses are tested under controlled conditions and the results are then expressed as a mathematical equation that will allow prediction of untested combinations of conditions. Even though several conditions affect the growth or decline in m icrobial populations, only a few have a relevant inf luence, and it is preferred to use as few variables as possible in the equation. It is assumed that the effect of a factor is independent of whether the microorganism is in a broth or food, as long as oth er relevant factors are equivalent

PAGE 35

35 ( Ross and McMeekin 1994; Whiting 1995) Little has been publish ed regarding the microbial modeling of Vibrio vulnificus Classification of Models There are several ways to classify models The most general classification is to divide them into growth models and inactivation or survival model s Within these two categories, they can be classified as primary, secondary and tertiary level models. Primary level models describe the change in a microbial parameter such as CFU/ ml or toxin formation, over time. Examples of these models are exponential growth rate and first order thermal inactivation. Secondary level models describe the responses of primary models to changes in environmental conditions, such as pH or temperature. The Arrhenius relationship or square root model are some examples. Tertiary level models are gen erated by computer software that apply primary and secondary m ode ls into programs that calculate responses to changing conditions. Primary and secondary level models can also be subdivided according to other criteria. They can be linear or nonlinear, segregated or non segregated, structured or non structured. In addition, the models can be descriptive (empirical) or based on microbiological criteria (mechanistic kinetic) (Whiting 1995) Of special interest for this study are inactivation kinetic s or survival models. The term inactivation is generally used to describe a fast decline in microbial population by an active agent, and survival implies a slower decline (Whiting 1995). These types of models are subdivided into four main groups: linear curves, curves with tailing, curves with shoulders and sigmoidal curves. A linear curve, or log linear, describes a first order reaction in which the rate of microbial redu ction is proportional to the population

PAGE 36

36 A curve with tailing is a linear slope that shows a reduction in the rate of microbial decline, i.e. a change in their slope, usually towards the end. The slope of the tailing can be a smaller or a zero slope. Reaso ns proposed for this rate reduction or survival are that a part of the microbial population may be more resistant or that they are protected by external factors, such as dead cells or inactivation of microbicidal agent. A curve with a shoulder starts with a zero slope which, after a transition, is followed by the slope of the linear curve. There are many reason s as to why a shoulder can occur, such as the need of a cumulative e ffect, or death resulting when the synthesis of a particular component is inacti vated. Curves with shoulders or with tailings are know n as biphasic curves, while curves with both conditions are known as sigmoidal curves (Xiong and others 1999). Validation of Models Two steps must be taken to validate a model on ce it has been built. T he first is to test its accuracy with new dat a and new combinations of variables. This will allow an estimation of the goodness of fit and will show if and where additional data is needed. Complex models tend to be very specific, which can be a limitation when testing new data. The second step is to compare model predictions with microbial responses in actual foods. included in the model. Errors in growth or surviv al should always tend towards faster growth rates or better survival, respectively, to make a conservative prediction (Whiting 1995). Limitations of Models Even i f models are usually accompanied by an estimate of the error, the quality of the model

PAGE 37

37 remains subjective. Transforming values, such as log 10 CFU/ml instead of actual population, or pH values instead of hydrogen ion concentration, can result in more linear responses than reality. It is very important that the model is accompanied by a description of its limitations; specific microorganisms, factors tested and considered in the model, ranges for each of these factors, and combinations of factors. The model user must be aware that using the model outside its limitations ma y not give valid answers. It should also be remembered that even if models must be tested in food and therefore competition between the target microorganisms and the rest of the normal mic robial flora should be considered and tested as well, the normal microbial flora varies from sample to sample (Whiting 1995) Objectives The general objective of this study was to develop quantitative predictive models to describe the inactivation of Vibri o vulnificus by freezing temperatures and frozen storage While the applicati on of the model is de sired for the bacteria in oysters, this study was carried out with pure cultures of the bacteria The uncertainty of initial numbers of the bacteria and physi cal differences between oysters would be sources of variability that make the development of the model difficult. To avoid this, identical vials inoculated with known numbers of Vibrio vulnificus were used for this stud y Specific objectives for this study were: To determine the parameters related to freezing and thawing that have an effect on survival of Vibrio vulnificus and use those that result in significant differences to model the survival of the bacteria. To observe and compare the effects of free zing temperatures and freezing rates on survival of the organism. To observe and compare the effects of frozen storage temperatures and times on the effect of the organism.

PAGE 38

38 To develop a mathematical model to predict the survival of the organism under diffe rent freezing and frozen storage conditions.

PAGE 39

39 CHAPTER 3 MATERIALS AND METHOD S Bacterial Strain and Culture Conditions Vibrio vulnificus ATCC 27562 isolated from blood was used for these studies. The bacteria were maintained in 10ml of sterilized Difco (S parks, MD) Marine Broth 2216 at 37C and 1ml was transferred every two days. A transfer was performed 24 hours prior to experimental use. Sample Preparation The following procedure w as followed for each experiment and describes one set of samples. One mil liliter of a fresh culture was transferred into 100ml of Marine Broth and incubated at 37C for 24 hours. The culture of V ibrio vulnificus was then centrifuged for 20 minutes at relative centrifugal force of 2289 x g and re suspended in Phosphate Buffered Saline (PBS) with Magnesium Chloride ( Hardy Diagnostics Santa Maria, C A ) to a concentration of approximately 10 7 CFU/ml. The concentration of Vibrio vulnificus suspended in PBS was estimated spectrophotometric al l y by adjusting the optical density to 0.45 0 at a wavelength of 600nm Twenty five screw cap vials (Nunc* Internally Threaded CryoTube* 3.6ml vials Fisher Scientific Pittsburgh, P A ) were filled with 3ml of the culture suspension To determine viable plate counts of the culture, a 1ml aliquot was serially diluted on PBS and plated onto duplicate Difco (Sparks, MD) Marine Agar 2216 plates and incubated overnight at 37C C olonies were counted following the p rocedure for enumerating bacteria by conventional plate count method stated in the U.S. Foo d and Drug Administration Bacterio logical Analytical Manual Online ( BAM 2001) This method will be referred throughout this study as direct plating.

PAGE 40

40 Freezing Studies Freezing studies were divided in to three sections. The objective of Section A was primaril y to determine the effect of thawing temperatures on the survival of Vibrio vulnificus and to determine the amount, if any, of injured cells with freezing. Based on the results obtained in Section A, Section s B and C were design ed to observe the effects fr eezing temperatures, frozen storage temperatures and times on the survival of the microorganisms. A summary of treatments in each section is shown on Table 3 1. Table 3 1 Freezing studies experimental d esign Section Freezing T (C) Storage T (C) Storage time (days) Thawing T (C) Counting Method A 10 35 80 No storage No storage 3, 30 DP MPN B 35 80 35 80 0,7,14,21,28,35,42 30 DP C 10 35 80 10 0,6,9,12 30 DP DP = Direct plating. MPN = Most probable number Section A A s et of sample tube s w as placed on a tube rack and frozen at 10C using a still air freezer. O nce samples had reach ed the freezer temperature six samples were removed from the freezer Three samples were thawe d for approximately 10 minutes in a water bath until they reached 3 0C, and th r e e were thawed for approximately 165 minutes in a walk in refrigerator at 3C and then warmed up to 30C in a water bath for approximately 6 minutes. S urvival of Vibrio vulnificus was followed as described below Samples were also frozen at 80 C and at 35 C and thawing protocols described above were followed Every experiment was repeated 3 times. To obtain Vibrio vulnificus survival data viable plate counts were determined by direct plating as described above. In addition, a 3 tube Most Pro bable Number ( MPN ) procedure,

PAGE 41

41 which i s a method to estimate the densit y of microorganisms in a liquid, was used with a non sele ctive nutritionally rich liquid media to allow recovery of injured cells (Ray and Speck 1973) Briefly, 3 replicate tubes with 9m l of 1% Alkaline Peptone Water (APW), pH 8.4, used as next three tubes were inoculated with the subsequent dilution, and so on for the next 3 dilutions. After 18 hours of incubation at 37C, the numbers of turbid tubes i e. positive tubes, per dilution were noted and the population density of the sample was obtained by referring to a 3 tube, 5 dilution MPN table ( BAM 2001 ) The difference in counts obtained by direct plating and by MPN was assumed to show the amount of injured cells. Section B A set of sample tube s was placed on a rack and frozen at 80C using a still air freezer. O nce samples had reached the freezer temperature, three samples were removed fro m the freezer The remaining samples were kept for storage at 80 C. T hree samples were removed every week for six weeks. Each time, samples were thawed at 30C in a water bath for approximately 10 minutes for microbiological analysis. Sample tube s were al so frozen and stored at 35C and the sampling procedure described above was followed. Table 3 1 shows the experimental design for the mentioned studies. Every experiment was repeated 3 times. Viable counts were determined by direct plating. Section C A s et of samples was placed on a rack and frozen at 80C using a still air freezer. Once samples had reached the freezer temperature three samples were removed from the freezer and thawed in a water bath at 30C for approximately 10 minutes for microbiologi cal analysis The remaining samples were move d to a 10C freezer for storage. Once sa mples reached 10C three samples were removed and thawed at 30C for approximately 10 minutes for

PAGE 42

42 microbiological analysis. In the same way, three samples were removed from the fr eezer every three days for up to 12 days and thawed as described above. The same procedure was followed for samples frozen at 35C and 10C, and stored at 10C, with the only difference that samples frozen at 10C remained in the same freez er for storage. Table 3 1 shows the experimental design for the mentioned studies. Every experiment was repeated 3 times. Viable counts were determined by direct plating. Temperature Data Acquisition T hermocouples (TCs type T, Omega Eng ineering Stamford C T ) were placed in blank vials containing only PBS to obtain time temperature data of each experiment Perfluoroalkoxy (PFA) insulated 24 gage TCs were connected to a data acquisition system. The module, OMB DAQ 56 from Omega Engineering (Stamford, CT) wa s connected to a computer with a data acquisition and real time display software developed in our lab oratory Blank vials were identical to those used for samples, but were equipped with a Teflon as the one shown in Figure 3 1 The ring had a centered perforation to place and maintain the TC tip centered in the vial, and 2 side perforations to allow water flow and expansion during freezing and thawing. The cap of the vial was also perforated in the center to allow for the TC to be introduced into the vial. For each freezing treatment, sample vials were placed in the rack leaving a space between tubes to have a uniform heat transfer Four vials with TCs were randomly placed between samples, following the fr ee space arrangement. Additional b lank vials were placed in the corner spaces of the rack, since heat transfer was faster in those positions. Once the rack was set up, the data acquisition was started and the rack was introduced into the corresponding free zer. Prior to each experiment, a TC was placed in the freezer to monitor internal ambient temperature throughout the freezing process.

PAGE 43

43 A) B) Figure 3 1. Vial and thermocouple A) Vial equipped with thermocouple and teflon ring. B) Teflon ring. The slow freezing rates at 10C and at 35C caused supercooling (SC) of samples. Unfortunately, the SC was not consistent between and within experiments: it could last for several minutes and be as low as 10C below the freezing point for one sample, but be completely absent in the next one. To avoid or stop SC, the rack was bumped against the freezer wall. The energy released in the impact would initiate the crystal lization process in the samples stopping the SC of it, and beginning the phase change part of the freezing process. Time temperature data was obtained for freezing and thawing processes, and when moving the samples from one freezer to another. Freez ing Rates and Slopes Every freezing experiment had several time temperature datasets, one from ea ch TC. Freezing and cooling rates were calculated by using Equation 3 1 where T t is the temperature of the sample at any given time, T i is the initial temperature of the sample, and T amb is the temperature of the freezer, i. e the ambient temperature. The resulting data was plotted against time and curves similar to the one s hown in Figure 3 2 were obtained (3 1)

PAGE 44

44 Figure 3 2 shows the three stages of the freezing process: first cooling or prefreezing, phase change or freezing, and second cooling or reduction to storage temperature. Slopes and R 2 values for first and second cooling of experiments from Section B and C were calc ulated using linear regressions, and l engths of the freezing phases mentioned above were also determined. These values, obtained from individual TCs, were averaged and standard deviations were calculated. This process was repeated for all experiment al replicates at a given temperature. Values resulting from each of these replicates were averaged an d standard deviations were calculated. Figure 3 2 Example of cooling rate vs time plot Statistical Analysis Statistical analysis of the data consisted of performing an analysis of variance (ANOVA) e test to evaluate the difference in survival means between treatments using SAS 9.1 0.05. Seven aspects were analyzed regarding microbiological data: Mean comparison of Vibrio vulnificus counts obtained by direct plating and MPN to deter mine the presence of injured cells using data from Section A of the experimental procedure.

PAGE 45

45 Mean comparison of bacterial counts obtained by direct plating for samples thawed at 4C and at 30C to determine the effect of thawing temperatures on survival of Vibrio vulnificus using data from Section A of the experimental procedure. Mean comparison of logarithmic reductions (log reductions) immediately after freezing at 10C, 35C and 80C, to observe the effect of freezing on survival of Vibrio vulnificus D ata from sections B and C of the experimental procedure were used Mean comparison of log reductions when storing Vibrio vulnificus at 80C and at 35C, and throughout the storage period, using data from Section B of the experimental procedure. Each combination of storage temperature and time was considered a treatment. The objective was to observe the individual and combined effects of storage temperature and time. Mean comparison of logarithmic reductions of Vibrio vulnificus frozen at 80C, 35C, and 10C, and stored at 10C, and throughout the storage period, using data from Section C of the experimental procedure. Each combination of storage temperature and time was considered a treatment. The objective was to observe the individual and combin ed effects of storage temperature and time. Mean comparison of log reductions after 6 days of storage at 10C and after 7 days of storage at 35C and 80C, to observe the effect of freezing and storage temperature on survival of Vibrio vulnificus Data from Section B and C of the experimental procedure were used. Mean c omparison of overall log reductions of Vibrio vulnificus after 6 days of storage at 10C and 7 days of storage at 35C and 80C to observe the cumulative effects of freezing temperatur e, storage temperature and change of storage temperature. Data from Section B and C of the experimental procedure were used. In addition an ANOVA ( = 0.05) was also used to determine if freezing rates obtained by all freezing temperatures were statistically different. Finally, the same statistical procedure was used to compare different parameters throughout the model development, to verify t hat assu mptions o n which the model was based were accurate. Model Development Statistical analysis of the survival data showed that variables that had an effect in the survival of the Vibrio vulnificus were frozen storage temperature and frozen storage time. Usin g Cancalc Analyze Software ( Engineering and Cybersolutions Inc., Gainesville, FL ) storage data collected from each freezing experiment in Section B and C were analyzed by fitting into

PAGE 46

46 existing inactivation kinetic models: first order, Weibull and Peleg. O nce the inactivation during storage was accurately described, constants were determined to predict inactivation that occurred during freezing. Inactivation During Frozen Storage Each dataset consisted of survival data throughout frozen storage time. Since this section of the modeling only analyzed inactivation during storage, time 0 of storage was considered the initial Vibrio vulnificus population. Based on the results from the statistical analysis the following assumptions were tested regarding inactiva tion of Vibrio vulnificus during storage : Assumption 1: The rate of Vibrio vulnificus inactivation depends on the storage temperature. Assumption 2: If survival of Vibrio vulnificus depends on ly on the storage temperature, then model parameters that are a function of temperature should be statistically the same at the same storage temperature Assumption 1 was tested by analyzing the isothermal kinetics of inactivation of samples stored at the same tempe rature as they were frozen. Therefore, three treatment s were analyzed: frozen at 80 C and stored at 80 C (Fr 80/St 8 0) frozen at 35C and stored at 35C (Fr 35/St 35), and frozen at 10C and stored at 10C (Fr 10/St 10). Each of these treatments had three datasets corresponding to each of the triplicat es. Assumption 2 was tested by perf orming an ANOVA determine if the temperature dependent parameters for same storage temperature of each model were statistically the same regardless of the freezing temperature. Three treatments were analyzed: frozen at 80C and stored at 1 0C (Fr 80/St 10), frozen at 35 C and stored at 10C (Fr 35/St 10 ), and frozen at 10C and stored at 10C (Fr 10/St 10). Each of these treatments had three datasets corresponding to each of the triplicates.

PAGE 47

47 First o rder k inetics First order isothermal analysis of survival data for each treatment yielded D and k values both functions of temperature. The k value represents the rate of inactivation of the microorganism while the D value, or Decima l Reduction Time, is the non logarithmic reciprocal of the k value and represents the time required to reduce the microbial population by a factor of 10. In this case, D values are given in days. This behavior is represented by Equation 3 2, where N o is th e original number of microorganisms and N t is the num ber of microorganisms at time t. (3 2) D values for isothermal datasets were used to determine the z value, which represents the temperature difference necessary to change the D value by one log 10 cycle. In this case, units for z value were degrees Celsius. D values for each triplicate of Fr 80/St 80, Fr 35/St 35, and Fr 10/St 10 were plotted against their correspondent sto r age tempe rature, i. e. 80C, 35C, and 10 C. Since the program does not accept repeated temperature values, the temperatures used were: 80.001, 80.002, and 80.003 instead of 80C; 35.001, 35.002, and 35.003 instead of 35C; and 10.001, 10.002, and 10.003 instead of 10C. The z value is the negative reciprocal of the slope Once the D values and the z value were determined, equations for first order micr obial inactivation (Equation 3 3 and 3 4 ) were used to predict numbers of survivors at any given time The reference temperature was s elected by comparing standard deviations of D values for each storage temperature, and selecting the one with the least variation. ( 3 3 ) ( 3 4 )

PAGE 48

48 For E quation 3 3 N t is the num ber of Vibrio vulnificus at time t in days, N o is the initial number of Vibrio vulnificus during storage, and D T is the D value in days at any given temperature T obtained through Equation 3 4 D Tref is the D value at the reference temperature T ref and z is the z value in degrees Celsius. To verify that the survival of Vibrio vulnificus depends only on the storage temperature (Assumption 2), D values for datasets stored at 10C were determined and compared by e test. Weibull k inetics While first order inactivation kinetics assumes a linear survival curve, most survival curves appear to be nonlinear. Weibull kinetics models survival through the use of two parameters, (Equation 3 6) is a function of temperature described by Equation 3 5. The parameter refers to the shape of t he survival curve, where upward (van Boekel 2002) (3 5) (3 6 ) and values for the three isothermal datasets Fr 80/St 80, Fr 35/St 35, and Fr 10/St 10, values we re plotted against their correspond ing storage temperature 80C, 35C, and 10C, to determine the a and b values. Since the program does not accept repeated temperature values, the temperatures used were: 80.001, 80.002, and 80.003 instead of 80C; 35.001, 35.002, and 35.003 instead of 35C; and 10.00 1, 10.002, and 10.003 instead of 10C. The a value corresponds to the intercept of the straight line, and the b value is the absolute value of the slope as shown in Equation 3 6

PAGE 49

49 Unlike the temperature dependent parameter value. This value is obtained by comparing values per temperature, and selecting an average of the set that presents the least variation. a and b values were determined, Equation 3 7 for Weibull microbial inactivation was used to predict numbers of survivors at any given time, by plugging in Equation 3 6. ( 3 7 ) After the model was tested it became evident t value is selected as described above, for this case, value would better be described as a function of temperature. Therefore, after removing outlier values, quadratic and linear equations were tested for improvement of exper imental data fit To verify that the survival of Vibrio vulnificus depends only on the stora ge temperature values for datasets stored at 10C were compared by performing an ANOVA Peleg k inetics Peleg iso thermal kinetics are described by Equation 3 dependent parameter described by Equation 3 9 The n parameter refers to the shape of the survival curve, where n<1 describes a concave upward curve, n>1 a concave downward curve, a (3 8) (3 9)

PAGE 50

50 After obtaining the n and values for the three isothermal datasets, Fr 80/St 80, Fr 35/St 35, and Fr 10/St 10, the correspondent stor age temperature were plotted against [ln(e 1)] to determine the survival parameters of the microorganism, i.e. the critical temperature (T c ) and k value Since the program does not accept repeated temperature values, the temperatures used were: 80.001, 80.002, and 80.003 instead of 80C; 35.001, 35.002, and 35.003 instead of 35C; and 10.001, 10.002, and 10.003 instead of 10C. The k value corresponds to the slope of the straight line and T c which is the critical temperature at which the inact ivation intensifies, to the intercept of the straight line with the temperature axis. Unlike the temperature dependent parameter curve, therefore the Peleg model uses only one n value. This value is obtained by comparing all Once the T c and k values were determined, Eq uation 3 10 for Peleg microbial inactivation was used to predict numbe rs of survivors at any given time, by plugging in Equation 3 9 (3 10) After the model was tested it became evident that while traditionally the n value is selected as described above, for this case, n would better be desc ribed as a function of temperature. Therefore, after removing outlier values, a linear equation w as tested for improvement of experimental data fit To verify that the survival of Vibrio vulnificus depends only on the stora ge temperature (Assumption 2), values for datasets stored at 10C were compared by performing an ANOVA Inactivation During F reezing Once the model for Vibrio vulnificus inactivation during storage was selected, it was necessary to add the inactivatio n during freezing to the model

PAGE 51

51 The average and standa rd deviation of the log reduction in survivors during f reezing at all temperatures was found, using data from S ections B and C of the experimental design. The standard deviation was subtracted from the mean log reduction. This way, inactivation of Vibrio vulnificus due to freezing would be accounted for in a conservative way. The number obtained represents the inactivation during freezing. Analysis of Residuals Values predicted by the model were subtract ed from the experimental values to determine the residuals (Equation 3 12). R esidual values were plotted against predicted values to observe the accuracy of the model. Residuals = log(experimental counts) log(predicted counts) (3 12)

PAGE 52

52 CHAPTER 4 R ESULTS AND DISCUSSIO N Freezing Studies Injured Cells Two methods were used to obtain viable counts of Vibrio vulnificus in Section A of the freezing studies: direct plating and most probable number (MPN). The enrich ment media of the MPN procedure allow s in jured bacteria to recover and become viable, while direct plating will only show bacteria capable of forming colonies at the moment of plating. Therefore, by subtracting direct plate counts from MPN counts, the number of injured bacteria can be obtained. D ata from Section A of the freezing studies showed th at bacterial counts after freezing obtained by the two methods were not significantly different ( = 0.05) Overall mean s of counts were 7.01x10 6 colonies/ml and 6.07x10 6 MPN/ml, obtained by direct plating and MPN respectively. Figure 4 1 shows the statistical analysis, with no significant differences ( = 0.05) within any of the three freezing tempera tures and thawing methods Figure 4 1. Bact erial counts/ml obtained by direct plating (CFU) and by most probable number (MPN) shown in logarithmic scale Bars show standard deviations.

PAGE 53

53 These results suggest that exposi ng Vibrio vulnificus to freezing tre atments does not result in injured cells; cells will either die or retain their ab ility to form colonies. Whether inactive cells are really dead or they are in the VBNC state needs to be further investigated. Thawing Temperatures Samples in Section A of th e experiments were thawed at 30C and at 4C t o observe the effect of thawing temperatures on the survival of Vibrio vulnificus after freezing. On average, thawing at 30C until water bath temperature was reached (30C) lasted 10 minutes. On the other hand slow t hawi ng lasted on average 166 minutes until refrigerator temperature was reached (2 to 4C). Samples were then warmed up to 30C, which on average lasted 7 minutes. As examples, Figure 4 2 A shows a plot of thawing at 3C followed by warming to 30C and Figure 4 2B shows thawing at 30C. In both cases, samples were thawed from 35C. A) B) Figure 4 2. Thawing curves. A) I n refrigerator temperature and warming up to 30C in a water bath. B) In a w ater bath at 30C.

PAGE 54

54 Overall m icrobial counts obtained by the two thawing methods were not found to be significantly different ( = 0.05). The mean count for thawing at 4C was 8.09x10 6 CFU/ml and for thawing at 30C was 5.93x10 6 CFU/ml. Figure 4 4 shows the statistical analysis, where no significant differences were found between the two thawing temperatures at any of the freezing temperatures. These results show that thawing temperature s, and therefore thawing rates do not have a significant role in the survival of Vibrio vulnificus Figure 4 3 Ba cterial counts/ml obtained after thawing at 4C and at 30C in logarithmic scale. Bars show standard deviations Freezing Temperatures and Rates All sections of the experimental procedure involve d freezing at 10C, 35C, and 80C. However, since the main goal of Section A was to observe the presence of injured cells and the effect o f thawing temperatures, only Section B and C were used to analyze the freezing temperatures and rates. Two freezing parameters were of main interest: first cooling and phase change. First cooling may define a possible adaptation to cold; slow cooling rate s may allow Vibrio vulnificus to synthesize cold adaptation proteins. The length of phase change, on the other hand, will define

PAGE 55

55 the formation of intracellular or extracellular ice, as well as the initial size of the ice crystals. Figure 4 4 shows a sample time temperature curves for vials frozen at 10C and at 80C. A) B) Figure 4 4 Time temperature curves. A) Samples frozen at 80C. B) S amples frozen at 10 C Table 4 1 shows a summary of the aver age length of each freezing phase per freezing temperature. It can be observed that the total freezing time at 10C was roughly 2 times the freezing time at 35C and 4 times the freezing time at 80C. Tabl e 4 1 Lengths of freezing phases in min ute s F reezing temperature 80C 35C 10C Length S T D Length ST D Length ST D First cooling (min) 3.03 0.21 7.50 0.32 20.72 4.26 Phase change (min) 7.90 0.33 27.09 2.38 87.98 18.71 Second cooling (min) 26.15 2.57 38.91 8.28 41.03 3.44 Total (min) 37.09 2.71 73.50 9.79 149.72 20.16 STD= Standard deviation

PAGE 56

56 By p lot ting the cooling rates (Equation 3 1) of first and second cooling for each time temperature curve, slopes and R 2 were determined Averages for each slope per freezing temperature are sh own in Table 4 2. Statistical analysis showed that the slopes of first cooling at different temperatures are different from each other. Figure 4 5 shows a sample cooling rates plots for vials frozen at 10C and at 80C. Table 4 2. Slopes of cooling rate s Freezing temperature 80C 35C 10C Slope STD Slope STD Slope STD First cooling (min 1 ) 0.093 0.006 0.082 0.004 0.063 0.002 R 2 first cooling (min 1 ) 0.983 0.006 0.993 0.005 0.997 0.003 Second cooling (min 1 ) 0.192 0.030 0.099 0 .015 0.099 0.052 R 2 second cooling (min 1 ) 0.993 0.005 0.991 0.003 0.949 0.052 STD= Standard deviation A) B) Figure 4 5 Cooling rate plots. A) Samples frozen at 80C. B) S amples frozen a t 10 C

PAGE 57

57 Survival data showed a decrease in bacterial counts during freezing. Statistical analysis of the log 10 reductions at the three different temperatures showed no significant differences ( = 0.05) Mean log 10 reductions during freezing were 1.5 3 at 80C, 1.77 at 35C, and 1.56 at 10C with an overall mean of 1.63 and standard deviation of 0.41 These results show that while there is a reduction of Vibrio vulnificus during freezing, th e temperature of freezing does not have an effect on the survival. In addition, it can be assumed that the freezing rate does not have an effect on the survival either. T he mechanism of cell damage during freezing should be possible to determine through o bservation with electron microscopy, which was not performed for this study. However, based on literature available, it can be assumed that damage occurred by formation of intracellular or extracellular ice crystals and the subsequent dehydration and/or pu ncture of the cells. Bryan and others (1999) reported better survival of Vibrio vulnificus after freezing due to ada ptation to cold temperatures after a three hour incubation at 15C Even though this study did not include incubation periods, first coolin g rate and length varied with each freezing temperature resulting in different times of exposure to cold temperatures Freezing at 10C allowed Vibrio vulnificus to be exposed to cold temperatures for approximately 20 minutes before reaching sub zero tem perature, while freezing at 80C only allowed for 3 minutes of exposure. Since log 10 reductions obtained by both temperatures w ere statistically the same, it can be assumed that the cells were not able to adapt to cold within 20 minutes. Effect of Storag e Temperatures and Times Storage at 80C and at 35C Section B of the experimental procedure consisted on freezing samples at 80C and at 35C followed by storage at the same temperature Samples were removed immediately after freezing and every week for six weeks to follow survival of the bacteri a. Statistical analysis of

PAGE 58

58 log reductions show ed that the re wa s an interaction between time and temperature. Significant differences es, with a mean log reduction of 0.089 for samples stored at 80C and of 1.262 for samples stored at 35C. Note that log reductions during frozen storage were calculated by considering bacterial counts after freezing, i.e. first data point of storage, as the initial bacterial population. = 0.05) were found through the six weeks of storage at 80C. Samples stored at 35C were shown to be statistically different from each other throughout time, and from those stored at 80 C, as Figure 4 6 shows Figure 4 6 Log 10 reduction during storage at 35C and 80C. Bars show standard deviations. These results show that storage temperature has a significant effect o n the inactivation rate of Vibrio v ulnificus A possible explana tion for this response to different storage temperatures might be in the presence of intracellular ice crystals. As described by Bank and Mazur (1973) t he size of the crystals is a function of initial crystal siz e, storage tem perature and storage time. During freezing at both temperatures, ice crystals were formed inside the cells. Ice crystals formed at the freezing rate corresponding to 80 C were theoretically smaller than those formed at the freezing rate corresponding to 35 C. B ased the on analysis of survival of Vibrio vulnificus d d d d d d d c b b a a a

PAGE 59

59 frozen at different temperatures, it can be assumed that they caused equal amount of damage to cells However, during storage at 80 C ice crystal growth is minimal, while at 35C is already det ectable ( Bank and Mazur 1973) Therefore, intracellular ice crystal growth during storage at 35C may have cause d inactivation of Vibrio vulnificus by damaging membranes. As mentioned in the review of the literature, intracellular ice formation has been a ttributed to rapid freezing rates, while cell dehydration and solutes effects has been attributed to slow freezing rates. Solute effects should cause most inactivation during freezing, where unfrozen cells are exposed to changes in pH and precipitation of solutes However, freezing at 35C and 80C caused the same amount of reduction. Only during storage they behaved differently. These results suggest that what ultimately causes inactivation of Vibrio vulnificus is intracellular ice crystal growth and no t solute effects. Nevertheless, this is only circumstantial evidence; o bservation through electron microscopy is needed to confirm the above mentioned scenarios. Storage at 10 C Section C of the experimental procedure consisted of keeping samples frozen a t 10C, 35C, and 80C stored at 10C. Bacterial counts of the s amples were determined after freezing and after change of colder temperatures to 10C. Additional samples were removed from the 10C freezer every 3 days for up to 9 days to follow survi val of the bacteria. For this section, initial bacterial count were considered as follows: for samples frozen at 10C, initial storage population was the one obtained immediately after freezing; for samples frozen at 35C and 80C the initial storage po pulation was the one obtained after samples were transferred to storage at 10C. Statistical analysis of log reductions showed that the re was no interaction between the freezing temperature and storage time. Mean log reduction throughout time was differen t at every data point. Mean log reduction at days 3 6 and 9 were 2.45, 3.23, and 4.14 respectively

PAGE 60

60 In regards to the effect of freezing temperature on samples stored at 10C s ignificant differences Table 4 3. Table 4 3. Overall effect of freezing temperatures in samples stored at 10C. Freezing temperature 80C 35C 10C Mean log 10 reduction 2.51 ab 2.20 b 2.66 a Means with the same letter are not significantly Data corresponding to the twelfth day was removed since it was available in only one replicate of one treatment (Fr 10C/St 10C set 1). In a similar way, data for the ninth day of Fr 35C/St 10C was not available for two sets (2 and 3), so it was completed by repeating the previous microbial count. Since it is evident that Vibrio vulnificus counts do not increase with time, it was a conservative choice to use the data from the previous data point to have a better statistical analysis. For the complete original dataset, please refer to Appendix 1. Individual data points were analyzed by considering each freezing temperature and time as a separate treatment, as shown in Figure 4 7 Significant found throughout time between treatments, same case at the sixth day. A t the ninth day it was found that samples frozen at 35C were significantly diffe 10C and at 80C. Figure 4 7 Log 10 reduction during storage at 10C. Bars show standard deviations. a a b b b b bc c c d

PAGE 61

61 These results suggest that regardless of the freezing temperature or freezing rate samples stored at the same temperature will have similar rates of inactivation. Just as with the analysis of Section B, it can be assumed that formation of intracellular ice crystals and their subsequent growth is what causes death of cells. One week of s torage To compare the effects of the three storage temperatures on survival of Vibrio vulnificus it was necessary to select one point in time. Therefore, seven days of storage was selected for Section B of the ex perimental procedure, and six days for Secti on C. While this comparison is not completely accurate, it allows a more complete and valuable comparison of effect of storage temperature. Statistical analysis of log reductions during storage showed significant differences between samples stored at the thr difference among survival of samples stored at 10C and frozen at different temp eratures, as shown in Figure 4 8 Figure 4 8 Log 10 reduction after 1 wee k of storage at 10C, 35C, and 80C. Bars show standard deviations. a a c b a Freezing T Storage T

PAGE 62

62 These results confirm that storage temperature is the variable that defines the rate of inactivation of Vibrio vulnificus during 1 week of storage Circumstantial evidence suggests tha t death of cells may be caused by growth of intracellular ice crystals. It is possible that regardless of the initial crystal size, the combination of initial size and growth causes cell injury, with subsequent death. During fast freezing, small ice cryst als are formed, and in warmer temperatures they grow fast er During slow freezing that is still fast enough to allow intracellular ice formation, larger ice crystals form. However their rate of growth at warmer temperatures is slower than that of smaller c rystals. Nevertheless, their original size plus their growth will probably cause the same amount of damage than smaller ice crystals cause Following these assumptions, freezing at 80C results in the formation of small intracellular ice crystals that gr ow very fast at 10C. On the other hand, freezing at 10C results in large r intracellular ice crystals that grow very slow at 10C, but they still grow. Since their initial size was big, when they gr o w, they cause the same damage caused by the fast grow ing small ice crystals. Freezing at 35C lies somewhere in the middle, with medium ice crystals that grow at a medium rate, causing, once again the same amount of damage. E ffect of Overall T reatments To compare the overall effect of the five treatments o n survival of Vibrio vulnificus a point in time was selected. Just like the previous section, seven days of storage was selected for Section B of the experimental procedure, and six days for Section C. While this comparison is not completely accurate, it a llows a more complete and valuable comparison of the overall effect of treatments combining reductions during freezing, storage, and change of storage temperature immediately after f reezing, as shown in Figure 4 9

PAGE 63

63 Fi gure 4 9 Log 10 r eduction of overall treatments. Bars show standard deviations. As shown in Figure 4 9 the move of the samples from freezers at 80C and 35C to a storage temperature of 10C causes an additional 0.75 average log reduction. This added r eduction makes the overall log reduction of Fr 35C/St 10C and Fr 80C/St 10C statistically 10C/St 10C. These results show that while the freezing temperature causes a reduction in survival of Vibrio vulnificus what ultimately determines the rate of inactivation is the storage temp erature. However, the added effect of a change of temperature in storage has a small added effect on the overall reduction. However it must be taken into consideration that for this study the change in temperature was from a colder temperature to a warmer temperature. This causes an increase in crystal growth, which leads to increase in bacterial death. It is very possible that a change to a colder temperature will not cause an additional reduction, or at least that the reduction will not have a significant effect in the overall outcome of the treatment. This should be tested in future experiments. Freezing T Storage T b a a c d

PAGE 64

64 Inactivation Model Inactivation During Frozen Storage First order kinetics First order isothermal analysis of survival data during storage for each triplicate of each treatment yielded D and k values, as shown in Table 4 4 A plot of log D values vs temperature (Figure 4 10 ) had a slope whose reciproca l resulted in a z value of 29.45 C Fr 80/St 80 set 3 resulted in an outlier, therefore was removed from the analy sis. Table 4 4. First order isothermal analysis Freezing Temperature (C) Storage Temperature (C) Set number D value (days) k value ( days 1 ) 80C 80C 1 596.81 0.0039 2 502.08 0.0046 3 155.94 0.0148 35C 35C 1 29.84 0.0978 2 29.19 0.07 89 3 23.54 0.0772 10C 10C 1 2.33 0.9888 2 2.25 1.1081 3 2.08 1.0249 *Outlier Figure 4 1 0 Log 10 D values vs t emperature

PAGE 65

65 Aside from the z value, it was necessary to determine the reference temperature (T ref ) and reference D value (D Tref ) to use f irst o rder E quation s 3 3 and 3 4. A comparison of standard deviations of D values for each storage temperature showed that D values corresponding to 10C had the least variation, therefore T ref was 10C and the av erage D value resulted in D Tref of 2.22 days. Using T ref and D Tref Equation 4 1, or more specifically Equation 4 2, was used to determine D T for each storage temperature. Knowing the initial number of Vibrio vulnificus during storage, predicted values for survival at each storage temperature were determined at each time t by plugging D T into Equation 4 3 where N f is Vibrio vulnificus numbers after freezing (4 1) (4 2) (4 3) Figure 4 11 First order isothermal inactivation during storage at 80C. D=528.94 days.

PAGE 66

66 Figure 4 12 First order isothermal inactivation during storage at 35C. D=15.66 days. Figure 4 13 First order isothermal inactiva tion during storage at 10C. D=2.22 days. While the first order equation did a fairly good job of describing inactivation during storage at 80C, as shown in Figure 4 11, it became evident that inactivation during storage at 35C and 10 did not follow first order inactivation kinetics; the curve shows an upward concavity (Figure 4 12 and 4 13).

PAGE 67

67 Weibull kinetics Weibull isothermal analysis of survival data during storage for each triplicate of each treatment yielded and values, as shown in Table 4 5 A plot of log values vs temperature (Figure 4 1 4 ) showed that Fr 80/St 80 sets 1 and 2 were outliers, therefore those values were removed from the dataset. values resulted in a= 1.163 and b=0.0466. Table 4 5 Weibull isothermal analysis Freezing Temperature (C) Storage Temperature (C) Set number value value 80C 80C 1 9.27 x10 4 0.242 2 1.03 x10 123 0.008 3 221.534 0.876 35C 35C 1 3.246 0.239 2 5.741 0.285 3 5.548 0.276 10C 10C 1 0.204 0.406 2 0.032 0.203 3 0.501 0.489 *Outlier s Figure 4 1 4 values vs t emperature 35 =0.3796 values showed less variability at the mentioned temperature However, upon values

PAGE 68

68 would be best described by relating it to temperature through a linear or a quadratic equation (Figures 4 1 5 and 4 1 6 ). By testing these two equations, it was determined that the quadratic equation was more adequate than the linear one. Figure 4 1 5 values vs t emperature Figure 4 16 values vs t emperature Using a and b Equation 4 4 was used to determine T for T for each temperature was determined using Equation 4 5. Knowing the initial number of Vibrio

PAGE 69

69 vulnificus during storage, predicted values for survival at each storage temperature were determined at each time t by plugging in Equ ations 4 4 and 4 5 into Equation 4 6 ,, where N f is the bacterial numbers after freezing (4 4 ) (4 5 ) (4 6) The predictions for inactivation during stora ge obtained through Weibull kinetics resulted fairly accurate as shown in Figures 4 1 7 4 1 8 and 4 19 Next, assumption 2 was tested: if survival of Vibrio vulnificus depends only on the storage temperature, then model parameters that are a function of t emperature should be statistically the values from the three experiments stored at survival of the Vibrio vulnifi cus only depends on storage temperature. Figure 4 17 Weibull isothermal inactivation during storage at

PAGE 70

70 Figure 4 18 Weibull isothermal in activation during storage at 35 2.94 30 Figure 4 19 Weibull isothermal inactivation during storage at Table 4 6 Weibull non isothermal analysis Freezing Temperature (C) Storage Temperature (C) value 80C 10C 0.2102 a 35C 10C 0.2613 a 10C 10C 0.2458 a

PAGE 71

71 Peleg kinetics Peleg isothermal analysis of survival data during storage for each triplicate of each and n valu es, as shown in Table 4 values resulted in k=0.0275 and T c = 34.52. Fr 80/St 80 set 3 resulted in an outlier, therefore was removed from the analysis. Table 4 7. Peleg isothermal analysis Freezing Temperature (C) Storage Temperature (C) Set number value n value 80C 80C 1 0.18481 0.24176 2 0.09778 0.00821 3 0.00883 0.87584 35C 35C 1 0.75475 0.23895 2 0.60811 0.28463 3 0.62315 0.27603 10C 10C 1 1.90462 0.4058 2 2.00812 0.20271 3 1.4025 0.48917 *Outli er n was found to be best described by n 10 =1.7718, since n values showed less variability at the mentioned temperature. However, upon testing the model, it became evident that n values would be best described by relating it to temperature through a linea r equation (Figures 4 2 0 ). Figure 4 20 N values vs t emperature

PAGE 72

72 Using k and T c Equation 4 7 was used to determine T for each storage temperature. n T for each temperature was determined using Equation 4 8 Knowing the initial number of Vibrio vulnificus during storage, predicted values for survival at each storage temperature were determined at each time t by plugging in Equations 4 7 and 4 8 into Equation 4 9 where N f is the bacterial counts after freezing (4 7) (4 8 ) (4 9 ) While Peleg kinetics did a fairly g ood job of describing inactivation during storage at 35C, as shown in Figure 4 2 2 it became evident that inactivation prediction d uring storage at 10C and 80C was inadequate. In fact, Peleg predicts that with time, storage at 80C will cause an inc rease in the bacterial population. Obviously, this is false (Figure s 4 21 and 4 2 3 ). Figure 4 21 Peleg isothermal inactivation during storage at 80C. 0.10

PAGE 73

73 Figure 4 2 2 Peleg isoth ermal inactivation d uring storage at Figure 4 23. Peleg isothermal in activation during storage at 10 C. Inactivation During Freezing Table 4 8 shows log 10 of initial counts before freezi ng (N o) log 10 of counts after freezing (N f ) and the corresponding log 10 reductions. Average log 10 reduction was calculated along with the standard deviation. To obtain a conservative constant for describing the inactivation of Vibrio vulnificus during fr eezing the standard deviation was subtracted from the average log 10

PAGE 74

74 reduction (Equations 4 10, 4 11, 4 12) Note that it was previously determined statistically that inactivation due to freezing is not significantly different among the three temperatures. Table 4 8 Inactivation during freezing Freezing Temperature (C) Storage Temperature (C) Log 10 N o Log 10 N f Log 10 reductions 80C 80C 8.03 6.36 1.66 8.03 6.95 1.08 7.58 5.93 1.65 80C 10C 7.58 5.93 1.65 7.94 6.36 1.58 7.87 6.32 1.55 35 C 35 C 7.64 6.24 1.40 7.94 6.40 1.53 7.81 6.38 1.44 35C 10 C 7.81 6.38 1.44 7.80 5.24 2.56 7.51 5.28 2.24 10C 10C 8.23 7.31 0.91 7.45 5.69 1.76 7.76 5.75 2.00 Average log 10 reduction = 1.63 (4 10) Standard deviatio n = 0.41 (4 11) Average log 10 reduction Standard deviation = 1.22 (4 12) Therefore, inactivation during freezing can be described through Equation 4 13, regardless of the freezing temperature. (4 13) Final Inactivation Model Combining the inactivation kinetics during storage as described by the Weibull model (Equation 4 6) and the inactivation during freezing (Equation 4 13), the final inactivation model is described by Equation 4 14. Figure 4 2 4 shows the e xperimental values from the five

PAGE 75

75 treatments (Fr 80/St 80, Fr 80/St 10, Fr 35/St 35, Fr 35/St 10, Fr 10/St 10) and the values predicted by Equation 4 14. (4 14) Figure 4 24. Inactivation m odel and experimental d ata Analysis of Residuals Table 4 9 shows log 10 experimental values and log 10 predicted values as well as the residuals values Note that time 0 days corresponds to the values after freezing. A plot of residuals against predicted values is shown in Figure 4 25. Residual values above 0 result when the prediction of surviving microorganisms was lower than the experimental value, which is undesirable for a n inactivation model. Analysis of residuals showed that all residual values are below 0.20 indicating that whenever inactivation is Before freezing

PAGE 76

76 represent when the prediction of surviving microorganism was greater tha n the experimental value. Figure 4 25 shows se veral occasions in which the inactivation was underestimated, however this results in a more conservative approach which is desirable fo r inactivation models. The lowest residual value was 1.50. Table 4 9 Comparison between experimental and predicted va lues Freezing Temperature (C) Storage Temperature (C) Storage Time (days) Experimental values (log 10 ) Predicted values (log 10 ) Residual 80C 80C 0 6.60 6.70 0.10 7 6.47 6.67 0.20 14 6.59 6.65 0.06 21 6.45 6.62 0.17 28 6.51 6.60 0.09 35 6.53 6.57 0.05 42 6.44 6.55 0.11 80C 10C 0 5.80 6.60 0.80 3 3.19 3.91 0.72 6 2.30 3.14 0.84 9 1.12 2.58 1.46 35C 35C 0 6.35 6.59 0.25 7 5.27 5.33 0.06 14 5.01 5.08 0.07 21 4.91 4.91 0.01 28 4.62 4.77 0.15 35 4.74 4.66 0.08 42 4.66 4.56 0.10 35C 10C 0 5.01 6.51 1.50 3 2.74 3.82 1.08 6 2.12 3.05 0.92 9 2.34 2.49 0.15 10C 10C 0 6.86 6.71 0.15 3 3.80 4.02 0.22 6 3.16 3.24 0.08 9 2.85 2.69 0.16 12 1.30 2.24 0.94

PAGE 77

77 Figure 4 25 Residuals vs predicted values

PAGE 78

78 CHAPTER 5 CONCLUSIONS AND FURTHER STUDIES Conclusions Freezing Studies There are many variables affecting the survival of Vibrio vulnificus during freezing. In these experiments, seve ral factors have been studied by submitting pure cultures of the bacteria to freezing, thawing, and frozen storage treatments B y comparing two bacterial counting methods, it was determined that submitting Vibrio vulnificus to freezing conditions does not result in injured cells; cells will either die or retain their ability to form colonies. However, there is a possibility that the bacteria entered the VBNC state which needs to be further investigated. When submitting Vibrio vulnificus to freezing treatm ents there is a reduction in bacterial survival, regardl ess of the freezing temperature. This suggests that the temperature of freezing does not have an effect on the survival. Likewise, different thawing temperatures showed no significant differences betw een bacterial counts Whether the reduction occurs during freezing or thawing remains unclear, but it can be suggested that occurs due to intracellular ice formation. Analysis of storage data revealed that storage temperature has a significant effect on t he inactivation rate and will ultimately define the survival of Vibrio vulnificus Storage of the bacteria at different temperatures revealed warmer temperatures are most lethal to the bacteria C older temperatures show slower inactivation rates. In addit ion, when storing Vibrio vulnificus frozen at different temperatures in the same storage temperature, similar inactivation rates were observed. These results suggest that temperatures that allow a faster rate of crystal growth during storage are more letha l for Vibrio vulnificus However this circumstantial evidence needs to be confirmed by electron microscopy.

PAGE 79

79 Finally, an analysis of the overall effect of the freezing treatments on Vibrio vulnificus showed that the reduction caused by changing frozen sampl es to a different temperature for storage cause d an additional reduction in survival. However, only a change from colder to warmer frozen storage was studied. Nevertheless, these results suggest that the crystal growth caused while samples were warming up caused the increased lethality of the treatments. Inactivation Model As recognized by several authors, first order kinetics was not suited for describing the reduction of all bacterial populations, and certainly not Vibrio vulnificus during frozen storage. Peleg inactivation kinetics also failed to predict adequately the inactivation of Vibrio vulnificus regardless of trying to fit the non temperature dependent n value to temperature. The model developed to describe the inactivation kinetics of Vibrio vuln ificus during freezing and frozen storage was found to be a combination of a constant for describing the reduction during freezing, and an adaptation of the Weibull model for describing the reduction during frozen storage. Unlike conventional Weibull kinet ics, this study showed that, in case of Vibrio vulnificus a function of temperature. In our case described through a quadratic equation. Further Studies Like many other studies, these results answer some questions regarding the inactivation of Vibrio vulnificus with freezing and frozen storage, but also raise several others. While the existence of a VBNC state in Vibrio vulnificus may always be debatable, it is very important to examine that possibility in future studies. Oliver and Bockian (1995) reported that VBNC cells were still capable of causing illness and death, therefore two things should be taken into consideration. The first one is to determine if Vibrio vulnificus enters the VBNC state during freezing o r frozen storage. The second one is to determine if cells already in the VBNC

PAGE 80

80 state are inactivated by the mentioned treatments. If one of these possibilities is true, then they must be included in the inactivation model presented in this work. This study presents circumstantial evidence of Vibrio vulnificus inactivation due to intracellular ice crystal formation and growth. Observation of cells after freezing and at different points of the storage will confirm these claims. Transmission Electron Microscopy (TEM) will allow a detailed analysis of cell membrane and organelles, as well as the identification and measurement of intracellular ice crystals. The initial ice crystal size is mainly defined by the freezing rate, or the length of the phase change step in the freezing process, not by the freezing temperature. By submitting sample vials to different freezing rates at the same freezing temperature, the role of these two variables in the inactivation of Vibrio vulnificus can be isolated. Survival data and T EM observation will allow a better understanding of the mechanism of damage during the freezing process. As discussed by many authors, Vibrio vulnificus is able to adapt to cold temperatures when submitted to temperatures around 15C for a few hours (McGo vern and Oliver 1995 ; Bryan and others 1999) This adaptation can be elucidated by submitting the bacteria to freezing treatments and frozen storage temperatures. Through TEM analysis, t he physical response of the bacteria and their ability to s urvive thes e treatments can be observed. If the reason for inactivation Vibrio vulnificus during freezing and frozen storage is in fact intracellular ice crystal growth, then cold adapted bacteria cap able of a better survival should show decreased intracellular forma tion. Regarding the model development, models should be validated under conditions different from the ones used to build them. While the model presented in this study accurately describes the inactivation of Vibrio vulnificus as shown by the residual anal ysis, an appropriate validation should be conducted.

PAGE 81

81 Finally, the model presented in this work was developed with vials of pure cultures. This model should be tested and adapted to describe the inactivation of Vibrio vulnificus in oysters, considering add itional variables, such as size and shape of the oysters.

PAGE 82

82 APPENDIX A SURVIVAL DATASETS Section A Table A 1 Survival datasets from Section A Freezing Temperature Initial Counts by direct plating (CFU/ml) Thawing at 30C Thawing at 4C Direct plating (CFU/ml) Most probable number (MPN/ml) Direct plating (CFU/ml) Most probable number (MPN/ml) 80 1.55E+08 1.75E+07 8.93E+06 5.27E+07 1.25E+07 1.07E+07 1.24E+07 4.38E+07 6.88E+06 1.25E+07 1.60E+08 8.28E+06 3.63E+06 1.46E+06 2.30E+06 35 6.90E+07 3.42E +06 6.97E+06 1.16E+06 2.03E+06 5.30E+07 3.00E+05 6.53E+05 2.45E+05 1.84E+05 1.27E+08 7.98E+05 1.38E+06 1.32E+06 1.16E+06 10 5.30E+07 2.38E+06 2.15E+06 4.68E+06 4.60E+06 1.69E+08 2.38E+06 2.41E+06 1.83E+06 4.05E+06 1.13E+08 5.93E+06 9.10E+03 2.55E +06 3.60E+03 Section B Table A 2 Survival datasets from Section B Fr 80/St 80 Fr 35/St 35 Control Set 1 Set 2 Set 3 Set 1 Set 2 Set 3 Set 1 Set 2 Set 3 Initial 1.06 x1 0 8 1.07 x1 0 8 3.80 x1 0 7 4.35 x1 0 7 8.65 x1 0 7 6.50 x1 0 7 6.15 x1 0 7 5.65 x1 0 7 1.68 x1 0 8 0 days 2.31 x1 0 6 8.83 x1 0 6 8.50 x1 0 5 1.74 x1 0 6 2.54 x1 0 6 2.38 x1 0 6 5.25 x1 0 7 5.18 x1 0 7 7 days 1.88 x1 0 6 5.97 x1 0 6 1.00 x1 0 6 1.07 x1 0 5 2.19 x1 0 5 2.33 x1 0 5 3.53 x1 0 7 4.60 x1 0 7 4.05 x1 0 7 14 days 2.67 x1 0 6 8.18 x1 0 6 7.08 x1 0 5 6.90 x1 0 4 1.53 x1 0 5 8.32 x1 0 4 1.59 x1 0 7 1.42 x1 0 7 1.22 x1 0 7 21 days 1.74 x1 0 6 6.03 x1 0 6 6.43 x1 0 5 5.32 x1 0 4 8.89 x1 0 4 1.04 x1 0 5 5.73 x1 0 6 3.50 x1 0 6 1.66 x1 0 6 28 days 1.48 x1 0 6 7.70 x1 0 6 4.97 x1 0 5 2.90 x1 0 4 3.85 x1 0 4 5.78 x1 0 4 2.82 x1 0 6 1.76 x1 0 6 7.48 x1 05 35 days 2.10 x1 0 6 7.47 x1 0 6 5.18 x1 0 5 3.32 x1 0 4 7.28 x1 0 4 5.77 x1 0 4 2.06 x1 0 6 1.76 x1 0 6 1.17 x1 0 6 42 days 2.03 x1 0 6 5.75 x1 0 6 5.65 x1 0 5 4.84 x1 0 4 4.40 x1 0 4 2.12 x1 0 6 9.73 x1 0 5 1.53 x1 0 6 Section C Table A 3 Survival datasets from Section C Fr 80/St 10 Fr 35/St 10 Fr 10/St 10 Set 1 Set 2 Set 3 Set 1 Set 2 Set 3 Set 1 Set 2 Set 3 Initial 3.80 x10 7 8.70x10 7 7.40x10 7 6.50x10 7 6.35x10 7 3.25x10 7 1.68x10 8 2.82x10 7 5.70x10 7 After freezing 8.50x10 5 2.28x10 6 2.10x10 6 2.38x10 6 1.73x10 5 1.89x10 5 2.06x10 7 4.94x10 5 5.67x10 5 0 days 3.32x10 4 1.63x10 6 2.21x10 5 1.65x10 5 7.98x10 4 6.55x10 4 3 days 2.50x10 2 2.09x10 3 2.31x10 3 1.00x10 3 4.40x10 2 2.20x10 2 1.51x10 4 1.53x10 3 2.18x10 3 6 days 5.00x10 1 2.80x10 2 2.70x10 2 3.48x10 2 2.00x10 1 3.00x10 1 3.43x10 3 6.40x10 2 2.75x10 2 9 days 0 1.00x10 1 3.00x10 1 2.20x10 2 2.10x10 3 0 4.00x10 1 12 days 2.00x10 1

PAGE 83

83 APPENDI X B STATISTICAL ANALYSIS Freezing Studies Table B 1 Statistical analysis of the effect of counting methods on bacterial counts Section A A) ANOVA. B) M eans Method 1 : d irect plating Method 2 : MPN A) B)

PAGE 84

84 Table B 2 Statistical analysis of effec t thawing temperature (C) on bacterial counts Section A A) ANOVA B) Means by freezing temperature. C) Means by thawing temperature A) B)

PAGE 85

85 C) Table B 3 Statistical analysis of effect of freezing temperature (C) on log 10 reductions Section B an d C A) ANOVA. B) Means A)

PAGE 86

86 B) Table B 4 Statistical analysis of effect of storage temperature ( 35 and 80 C) and time (days) on log 10 reductions Section B A) ANOVA. B) Means by time in days. C) Means by temperature. A)

PAGE 87

87 B) C)

PAGE 88

88 Table B 5 Stati stical analysis of effect of storage temperature ( 35 and 80 C) on log 10 reductions when considering each time temperature a separate treatment, Section B. A) ANOVA. B) Means. A=Fr 80/St 80 C C=Fr 35/St 35 C ; numbers represent storage time in days. A) B)

PAGE 89

89 Table B 6 Statistical analysis of effect of freezing temperature (C) and storage time (days) at 10C on log 10 reductions, Section C. A) ANOVA. B) Means by time in days. C) Means by temperature. A) B)

PAGE 90

90 C) Table B 7 Statistical analysis o f effect of freezing temperature (C) on log 10 reductions when considering each time temperature a separate treatment Section C A) ANOVA. B) Means B=Fr 80/St 1 0C D=Fr 35/St 10 C E=Fr 10/St 10C ; numbers represent storage time in days. A)

PAGE 91

91 B) Table B 8 Statistical analysis of effect of storage temperature (C) and one week of storage on log 10 reductions Section B and C. A) ANOVA. B) Means. A)

PAGE 92

92 B) Table B 9 Statistical analysis of cumulative effect of freezing and storage temperature ( C) and one week of storage on log 10 reductions, Section B and C. A) ANOVA. B) Means. A= Fr 80/St 80C, B=Fr 80/St 10C, C=Fr 35/St 35C, D=Fr 35/St 10C, E=Fr 10/St 10C; numbers represent storage time in days; change to storage at 10C. A)

PAGE 93

93 B) Inacti vation Model Table B 10 Statistical analysis of the temperature dependent parameter Alpha for the Weibull model for inactivation during storage A) ANOVA. B)Means. A)

PAGE 94

94 B)

PAGE 95

95 LIST OF REFERENCES Alur MD, Grecz N. 1975. Mechanism of injury of Escherich ia coli by freezing and thawing. Biochem BiophysRes Commun 62:308 312. Ama AA, Hamdy MK, Toledo RT. 1994. Effects of heating, pH and thermoradiation on inactivation of Vibrio vulnificus Food Miccrobiol 11:215 227. Andrews LS. 2003. Low dose gamma irradiat ion to reduce pathogenic vibrios in live oysters ( Crassostrea virginica ). J Aquatic Food Prod Tech 12(3):71 82. Andrews LS, Park DL, Chen YP. 2000. Low temperature pasteurization to reduce the risk of vibrio infections from raw shell stock oysters. Food Addit and Contam 19(7):787 791. [BAM] Bacteriological Analytical Manual. 2001. United States Food and Drug Administration. Rockville, Md. Available from : http://www.cfsan.fda.gov/~ ebam/bam toc.html. Accessed Feb 25 2009. Bang W, Drake MA. 2002. Resistance of cold and starvation stressed Vibrio vulnificus to heat and freeze thaw exposure. J Food Prot 65(6):975 980. Bank H, Mazur P. 1973. Visualization of freezing damage. J Cell Bi ol 57:729 742. Berlin DL, Herson DS, Hicks DT, Hoover DG. 1999. Response of pathogenic vibrio species to high hydrostatic pressure. Appl Environ Microbiol 65(6):2776 2780. Birkenhauer JM, Oliver JD. 2003. Use of diacetyl to reduce the load of Vibrio vulnif icus in the eastern oyster. J Food Prot 66(1):38 43. peroxide sensitive culturable cells of Vibrio vulnificus gives the appearance of resuscitation from a viable but non culturable state. J Bacteriol 182(18):5070 5075. Borazjani A, Andrews LS, Veal CD. 2003. Novel nonthermal methods to reduce Vibrio vulnificus in raw oysters. J Food Safety. 23(2003): 179 187. Bryan PJ, Steffan RJ, DePaola A, Foster JW, Bej AK. 1999. Adaptive response to cold temperatures in Vibrio vulnificus Curr Microbiol 38:168 175. Coleman E. 2003. The Gulf Oyster Industry, Seizing a better future. Louisiana Sea Grant College Program for the National Sea Grant College Program. Cook DW, Rupl e AD. 1989. Indicator bacteria and Vibrionaceae multiplication in post harvest shellstock oyster. J Food Prot. 52:343 349. Cook DW, Ruple AD. 1992. Cold storage and mild heat treatment as processing aids to reduce the numbers of Vibrio vulnificus in raw oy sters. J Food Prot 55(12):985 989.

PAGE 96

96 Cook DW. 1994. Effect of time and temperature on multiplication of Vibrio vulnificus in postharvest gulf coast shellstock oysters. Appl. Environ. Microbiol 60(9):3483 3484. Cook DW. 1997. Refrigeration of shellstock: cond itions which minimize the outgrowth of Vibrio vulnificus J Food Prot 60(4):349 352. DePaola A, Capers GM, Alexander D. 1994. Densities of Vibrio vulnificus in the intestines of fish from the U.S. Gulf Coast. App Environ Microbiol 60(3):984 988. Dixon D, R odrick GE. 1998. Effect of gamma radiation on shellstock oysters: Extension of shelf life and reduction in bacterial numbers with particular reference to Vibrio vulnficus In: Combination Processes for Food Irradiation. IAEA PressVienna, Austria. 97 110. Drake SL, DePaola A, Jaykus LA. 2007. An overview of Vibrio vulnificus and Vibrio parahaemolyticus Compr Rev Food Sci F 6:120 144. Farrant J, Morris GJ. 1973. Thermal shock and dilution shock as causes of freezing injury. Cryobiol 10:134 140. Feldhusen F 2000. The role of seafood in bacterial foodborne disease. Microbes Infect 2:1651 1660. Haines RB. 1938. The effect of freezing bacteria. Proc R Soc B 124:451 463. Hesselman DM, Motes ML, Lewis JP. 1999. Effects of a commercial heat shock process on Vibri o vulnificus in the American oyster, Crassostrea virginica harvested from the Gulf Coast. J Food Prot 62(11):1266 1269. Hollis DG Weaver RE, Baker CN, Thornsberry C. 1976. Halophilic Vibrio species isolated from blood cultures. J C lin M icrobiol 3(4):425 431. Hood MA, Ness G, Rodrick G, Blake N. 1983. Effects of storage on microbial loads of two commercially important shellfish species, Crassostrea virginica and Mercenaria compechiensis Appl. Environ. Microbiol 45:1221 1228. Jackson JK, Murphree RL, Tampl in ML. 1997. Evidence that mortality from Vibrio vulnificus infection results from single strains among heterogeneous populations in shellfish. J C lin M icrobiol 35(8):2098 2101. Jeong Y, Birbari W, Rodrick GE. 1991. Evaluation of freezing treatment on Vibr io present in shellstock oysters and clams. Proc Trop Subtrop Fish Technol Soc Am 353 355. Johnston MD, Brown MH. 2002. An investigation into the changed physiological state of Vibrio bacteria as a survival mechanism in response to cold temperatures and s tudies on their sensitivity to heating and freezing. J Appl Microbiol 92:1066 1077. Karrow AM, Webb WR. 1965. Tissue freezing: A theory for injury and survival. Cryobiol 2(3):99 108.

PAGE 97

97 Kaspar CW, Tamplin ML. 1993. Effect of temperature and salinity on the s urvival of Vibrio vulnificus in seawater and shellfish. App Environ Microbiol 59(8):2425 2429. Kaysner CA, Tamplin ML, Wekell MM, Stott RF, Colburn KG. 1989. Survival of Vibrio vulnificus in shellstock and shucked oysters ( Crassostrea gigas and Crassostrea virginica ) and effects of isolation medium on recovery. App Environ Microbiol 55(12):3072 3079. Kelly MT. 1982. Effect of temperature and salinity on Vibrio (beneckea) vulnificus occurrence in a gulf coast environment. App Environ Microbiol 44(4):820 824. Kelly MT, Dinuzzo A. 1985. Uptake and clearance of Vibrio vulnificus from Gulf Coast oysters ( Crassostrea virginica ). App Environ Microbiol 50(6):1548 1549. Kilgen MB, Hemard MT. 1995. Evaluation of commercial irradiation and other processing methods for Vibrio vulnificus control in Louisiana Oysters. Presented in 20 th Annual Meeting of Seafood Science and Technology. Humacao, Puerto Rico. Kim CM, Jeong KC, Rhee JH, Choi SH. 1997. Thermal death times of opaque and translucent morphotypes of Vibrio vulnific us Appl Environ Microbiol 63:3308 3310. Klontz KC, Lieb S, Schreiber M, Janowski HT, Baldy LM, Gunn RA. 1988. Syndromes of Vibrio vulnificus infections: clinical and epidemiological features in Florida cases, 1981 1987. Ann Intern Med 109:318 323. Kural A G, Chen H. 2007. Conditions of reduction of Vibrio vulnificus in oysters through high hydrostatic pressure treatment. Int J Food Microbiol 122:180 187. Linder K, Oliver JD. 1989. Membrane fatty acid and virulence changes in the viable but nonculturable sta te of Vibrio vulnificus Appl Environ Microbiol 55(11):2837 2842. Lopez Caballero ME, Perez Mateos M, Montero P, Borderias AJ. 2000. Oyster preservation by high pressure treatment. J Food Prot 63(2): 196 201. Lovelock JE. 1953. The haemolysis of human red blood cells by freezing and thawing. Biochim Biophys Acta 10:414 426. Lund BM. 2000. Freezing. In: Lund BM, Baird Parker TC, Gould GW, editors. Microbiological Safety and Quality of Food. Vol 1. Gaithersburg: Maryland. p 122 145. Mallett JC, Beghian LE, Me tcalf TG, Kaylor JD. 1991. Potential of irradiation technology for improved shellfish sanitation. J Food Safety 11(4):231 245. Mazur P. 1963. Kinetics of water loss from cells at subzero temperatures and likelihood of intracellular freezing. J Gen Physiol 47:347 369. Mazur P. 1970. The freezing of biological systems. Science. 168(3934):939 949.

PAGE 98

98 McDougald D, Rice SA, Weichart D, Kjelleberg S. 1998. Nonculturability: adaptation or debilitation? FEMS Microbiol Ecol 25:1 9. McGovern VP, Oliver JD. 1995. Inducti on of cold responsive proteins in Vibrio vulnificus J Bacteriol 177(14):4131 4133. Mestey D, Rodrick GE. 2003. A comparison of cryogenic freezing techniques and their usefulness in reduction of Vibrio vulnificus in retail oysters. In: Villalba A, Reguera B, Romalde JL, Beiras R, editors. Molluscan Shellfish Safety. Santiago de Compostela, Spain: Consellera de Pesca e Asuntos Martimos da Xunta de Galicia and Intergovernmental Oceanographic Commission of UNESCO. p 467 474. Motes M, DePaola A. 1996. Offshor e suspension relaying to reduce levels of Vibrio vulnificus in oysters ( Crassostrea virginica ). Appl Environ Microbiol 62(10):3875 3877. Motes ML, DePaola A, Cook DW, Veazey JE, Hunsucker JC, Garthright WE, Blodgett RJ, Chirtel SJ. 1998. Influence of water temperature and salinity on Vibrio vulnificus in Northern Gulf and Atlantic Coast oysters ( Crassostrea virginica ). App Environ Microbiol 64(4):1459 1465. Murphy SK, Oliver JD. 1992. Effects of temperature abuse on survival of Vibrio vulnificus in oysters. Appl Environ Microbiol 58(9):2771 2775. Nilsson L, Oliver JD, Kjelleberg S. 1991. Resuscitation of Vibrio vulnificus from the viable but nonculturable state. J Bacteriol 173(16):5054 5059. Oliver JD. 1981. Lethal cold stress of Vibrio vulnificus oysters. Appl Environ Microbiol 41(3):710 717. Oliver JD. 1995. The viable but non culturable state in the human pathogen Vibrio vulnificus FEMS Microbiol Lett 133:203 208. Oliver JD. 2005. The viable but nonculturable state in bacteria. J Microbiol 43(S):93 100. Oliver JD, Bockian R. 1995. In vivo resuscitation, and virulence towards mice, of viable but nonculturable cells of Vibrio vulnificus Appl Environ Microbiol 61(7):2620 2623. Oliver JD, Colwell RR. 1973. Extractable lipids of gram negative marine bacteria: fatty acid composition. Int J Syst Bacteriol 23:442 458. Oliver JD, Hite F, McDougald D, Andon NL, Simpson LM. 1995. Entry into, and resuscitation from, the viable but nonculturable state by Vibrio vulnificus in an estuarine environment. Appl Environ Micr obiol 61(7):2624 2630. Oliver JD, Nilsson L, Kjelleberg S. 1991. The formation of nonculturable cells of Vibrio vulnificus and its relationship to the starvation state. Appl Environ Microbiol 57(9):2640 2644.

PAGE 99

99 Oliver JD, Warner RA, Cleland DR. 1982. Distrib ution of Vibrio vulnificus and other lactose fermenting vibrios in the marine environment. Appl Environ Microbiol 45(3):985 998. Panoff JM, Thammavongs B, Gueguen M, Boutibonnes P. 1998. Cold stress responses in mesophilic bacteria. Cryobiol 36:75 83. Park er RW, Maurer EM, Childers AB, Lewis DH. 1994. Effect of frozen storage and vacuum packaging on survival of Vibrio vulnificus in Gulf Coast oysters ( Crassostrea virginica ). J Food Prot 57(7):604 606. Peleg M. 2003. Microbial survival curves: interpretation s, mathematical modeling, and utilization. Comments Theor Biol 8:357 87. Quevedo AC, Smith JG, Rodrick GE, Wright AC. 2005. Ice immersion as a postharvest treatment of oysters for the reduction of Vibrio vulnificus J Food Prot 68(6):1192 1197. Ray B, Spec k ML. 1973. Enumeration of Escherichia coli in frozen samples after recovery from injury. Appli Microbiol 25(4):499 503. Richards GP. 2003. The evolution of molluscan shellfish safety. In: Villalba A, Reguera B, Romalde JL, Beiras R, editors. Molluscan Sh ellfish Safety. Santiago de Compostela, Spain: Consellera de Pesca e Asuntos Martimos da Xunta de Galicia and Intergovernmental Oceanographic Commission of UNESCO. 221 245. Ross T, McMeekin TA. 1994. Predictive microbiology. Int J Food Microbiol 23:241 2 64. Ruple AD, Cook DW. 1992. Vibrio vulnificus and indicator bacteria in shellstock and commercially processed oysters from the Gulf Coast. J Food Protect 55(9):667 671. Schwarz JR. 2000. Rapid chilling of oyster shellstock: a post harvest process to reduc e Vibrio vulnificus Presented in 25th Annual Meeting of Seafood Science and Technology. Longboat Key, Florida. Shapiro RL, Altekruse S, Hutwagner L, Bishop R, Hammond R, Wilson S, Ray B, Thompson S, Tauxe RV, Griffin PM. 1998. The role of Gulf Coast oyst ers harvested in warmer months in Vibrio vulnificus infections in the United States. J Infect Dis 178:752 759. Strom MS, Paranjpye RN. 2000. Epidemiology and pathogenesis of Vibrio vulnificus Microbes Infect 2:177 188. Sun Y, Oliver JD.1994. Effects of GR AS compounds on natural Vibrio vulnificus populations in oysters. J Food Prot 57:921 923. Tamplin M, Rodrick GE, Blake NJ, Cuba T. 1982. Isolation and Characterization of Vibrio vulnificus from Two Florida Estuaries. App Environ Microbiol 44(6):1466 1470.

PAGE 100

100 Tamplin ML, and Capers GM. 1992. Persistence of Vibrio vulnificus in tissues of Gulf Coast oysters, Crassostrea virgnica exposed to seawater disinfected with UV light. App Environ Microbiol 58(5):1506 1510. van Boekel MAJS.2002. On the use of the Weibull model to describe thermal inactivation of microbial vegetative cells. Intl J Food Microbiol 74:139 159. Whiting RC. 1995. Microbial modeling in foods. Crit Rev Food Sci 35(6):467 494. Wolf P, Oliver JD. 1992. Temperature effects on the viable but noncultu rable state of Vibrio vulnificus FEMS Microbiol Ecol 101:33 39. Wright AC, Hill RT, Johnson JA, Roghman MC, Colwell RR, Morris JG Jr. 1996. Distribution of Vibrio vulnificus in the Cheasepeake Bay. Appl. Environ. Microbiol 62(2):717 724. Xiong R, Xie G, E dmondson AE, Sheard MA. 1999. A mathematical model for bacterial inactivation. Int J Food Microbiol 46:45 55.

PAGE 101

BIOGRAPHICAL SKETCH Diana Maria Seminario was born on 1982 in Guayaquil, Ecuador After graduating from high s cho ol with honors, she e ntered the Escuela Superior Politecni ca del Litoral (ESPOL) to pursue a degree in f ood e ngineering with a minor in food product d evelopment In her sophomore year she joined Productos Elaborados Bolivar S.A, a banana processing company, as Quality Control Assistant and eventually became HACCP Coordinator. F ollowing her graduation in 2006, Diana was offered an assistantship in the University of Florida to pursue her Master of S cience degree in f ood s cience, under the supervision of Dr. Murat Balaban. On Ja nuary of 200 8 Dr. Gary Rodrick was selected as her new graduate advisor after Dr. Balaban accepted a position at the University of Alaska. I n the summer of 20 07, Diana obtained an internship in the Quality Assurance and Food Safety Department of one of th e During her stay at the University of Florida, Diana represented the university in four occasions as part of the Dairy Judging Team and the Food Science College Bowl Team. In addition she served as treasurer for the Gator Chapter of the Florida Association for Food Protection.