Microbial Models of the Effect of Storage Temperature on the Growth and Survival of Vibrio Species in Oysters

NON-TECHNICAL SUMMARY: Vibrio bacteria from both endemic and pollution related sources are widely recognized as among the most significant food safety problems related to seafood in the U.S. according to the National Academy of Science, National Shellfish Sanitation Program, and the Food and Drug Administration (FDA). Recently, the World Health Organization, the United Nation's Food Agricultural Organization and the FDA have published risk assessments for Vibrio vulnificus and V. parahaemolyticus in seafood products. In these assessments, they have identified data gaps that increase the uncertainty of risk assessments, namely a lack of predictive models for the behavior of total, virulent and avirulent strains of V. vulnificus and V. parahaemolyticus in post-harvest oysters. The proposed project will address this need by developing predictive models for growth and survival of naturally occuring and laboratory-challenged V. vulnificus and V. parahaemolyticus in shell oysters collected seasonally from two geographical regions under various commercially-relevant storage conditions. The purpose of this project is to determine the effect of storage temperature on the growth and survival of total, virulent and avirulent V. vulnificus and V. parahaemolyticus in post-harvest oysters, and to develop and validate new predictive models.

<P>
PROGRESS: 2006/12 TO 2007/12 <BR>
OUTPUTS: The overall goal of this project is to develop predictive models for the behavior of Vibrio species in oysters. Oysters were collected from Chesapeake Bay (CB) in Summer 2005 and 2006 and then stored at 5, 10, 20, 25, and 30 degree C (degC) for selected time interval and analyzed by a direct plating method (DPM). The Baranyi D-model was fitted to the total Vp growth and survival data to estimate the parameters of lag phase duration (LPD), the growth/inactivation rate (GR) and the maximum population density (MPD). Total Vp was slowly inactivated at 5 and 10 degC. Maximum GR and MPD were observed at 25 degC. No significant LPD was observed. The bias (Bf) and accuracy (Af) factors were 1.00 and 1.03, respectively. Models developed for the growth and survival of total Vp in CB oysters were validated against a new set of model independent data, where oysters were collected from the Gulf Coast (GC) in Summer (Su), Fall (Fa) and Spring (Sp) and analyzed using the same protocols as the primary model experiments. The Bf and Af for the GRs determined in GC oysters during the Su, Fa and Sp were 1.02 and 1.04; 1.02 and 1.05 and 1.03 and 1.06, respectively. The regression analyses of the GR estimates versus harvest conditions suggested that harvest water conditions had no effect on GR. These results indicate that the models developed for the growth of total Vp in CB oysters are valid for Vp growth in oysters harvested from the GC. Pathogenic Vp was only observed at 15, 20, 25, and 30 degC after certain periods of incubation. No pathogenic Vp were detected at t=0h. Predictive models were not developed for the growth and survival of pathogenic Vp in oysters since the level of pathogenic Vp in oysters was very low and random. Currently, we are applying Most Probable Number (MPN)-qPCR for enumeration of pathogenic Vp. Oyster samples collected from the GC in Su, Fa and Sp were analyzed for detecting total Vv at different temperatures by DPM. However, DPM did not work for enumeration of total Vv. Therefore, oysters were collected seasonally (Sp, Su and Fa 2007) from the GC, stored and analyzed for total Vv using MPN method. Vv was found to be inactivated at 5 and 10 degC. Except the Sp 10 degC samples no LPD or GR 'shoulder' was observed. The maximum GR was observed at 30 degC. The Bf and Af were 0.9998 and 1.045, respectively. GRs were similar over the three seasons but lower than the assumptions made in the FAO/WHO Vv Quantitative risk assessment. One undergraduate student graduated with experience in microbiology and molecular biology in Sp 2007. Two abstracts were published and presented to the International Association for Food Protection (IAFP) Annual Meeting 2007. In addition, one abstract has been accepted for presentation to the American Society for Microbiology (ASM) General meeting 2008 and two abstracts have been submitted for presentation to the IAFP 2008. In year 3, predictive models for the growth and survival of total Vv will be validated and models for pathogenic Vp and Vv in oysters will be developed and validated. The effects of storage conditions on sensory and textural characteristics of oysters will be investigated. <BR>
PARTICIPANTS: Two undergraduate students, one graduate student, one postdoctoral associate and our laboratory specialist worked on this project. In addition to PI and Co-PI, collaborator, Kathy Brohawn, Maryland Department of the Environment and the Maryland Watermen's Association were involved on this project. Two undergraduate students, one graduate student and one postdoctoral associate have been trained in microbiology and molecular biology to conduct research on this project. <BR>
TARGET AUDIENCES: The target audiences are the seafood industry, risk assessors, risk managers, regulators and scientific community. The efforts to reach the audiences include presentation of the research findings at the professional meetings and personal communication. The newly developed predictive models may provide valuable information for refining existing risk assessment models and developing Vp and Vv control plans.
<BR>
<BR>
IMPACT: 2006/12 TO 2007/12 <BR>
Though little information is available about the development of predictive models for the growth and survival of Vp and Vv in oysters, the results of this study will assist risk managers and seafood industry in designing and implementing food safety plans to minimize the risk from these pathogens in seafood consumers. Graduation of one undergraduate student with experience in microbiology and molecular biology as well presentation of research findings to the local, regional, national and international professional meetings will enhance the national and international standing of University of Maryland Eastern Shore's research programs, thereby enhancing student recruiting and the recognition of its microbiological research and outreach programs.

<P>

PROGRESS: 2005/12/15 TO 2006/12/15 <BR>

The overall objective of this project is to develop predictive microbial models for the behavior (growth, survival and/or inactivation) of total, virulent and avirulent Vibrio vulnificus (Vv) and V. parahaemolyticus (Vp) strains in oysters as a function of post-harvest storage temperature. To accomplish this objective a postdoctoral associate and a graduate student have been hired and trained in microbiology and molecular biology to conduct experiments for this project. Experimental protocols were developed and sampling sites were selected in a collaborative effort among PI, Co-PI, collaborators, postdoctoral associate and the graduate student. In addition, PI and postdoctoral associate visited the Co-PI's laboratory to learn the predictive modeling. Oyster samples were collected in summer and fall, 2006 and analyzed for the growth and survival of Vv and Vp. Predictive models for the growth and survival of Vv in Chesapeake Bay oysters and total Vp in Gulf Coast oysters collected in summer 2006, as a function of temperature were developed. Vp was slowly inactivated at 5 degree C at a growth rate (GR) of -0.003 log CFU/h. Maximum GR (1.5 log CFU/h) was observed at 25 and 30 degree C. At 10, 15 and 20 degree C, the GR was 0.014, 0.051 and 0.071, log CFU/h, respectively. The bias and accuracy factors for a square root secondary growth model were 1.00 and 1.02, respectively. Maximum population density (MPD) displayed a peak form, with a maximum of 7.9 log CFU/g at 20 degree C. No significant lag phase duration (LPD) was observed. Vv was slowly inactivated at 5 and 10 degree C at a GR of -0.008 and -0.007 log CFU/h, respectively. The maximum GR (0.029 log CFU/h) was observed at 30 degree C. At 15, 20 and 25 degree C, the GR was 0.005, 0.010, and 0.021 log CFU/h, respectively. A LPD was only observed at 10 degree C (218 h). Secondary model for GR had bias and accuracy factors of 1.00 and 1.00, respectively. These results suggest that growth rate of Vp was faster than that of Vv. Development of predictive models for the growth and survival of Vv in Gulf Coast summer oysters, total Vp in Chesapeake Bay summer oysters as well as Vv and Vp in Gulf Coast fall oysters are underway. Two abstracts have been accepted for presentation to the International Association for Food Protection Annual Meeting. These results will soon be submitted for publication to the Journal of Food Protection and Applied and Environmental Microbiology. In addition, collaborative research has increased among University of Maryland Eastern Shore (UMES), FDA, University of Tasmania (UT), Australia and Maryland Department of Environment (MDE). In years 2 and 3, predictive models for the growth and survival of Vv and Vp in oysters will be developed as a function of season of harvest, geographical harvest area, harvest water salinity, as well as in newly introduced Asian oysters. In addition, models will be validated against a new set of model-independent data. Moreover, the effects of storage conditions on sensory and textural characteristics of oysters will be investigated.
<BR>
<BR>

IMPACT: 2005/12/15 TO 2006/12/15 <BR>

Currently, the scientific community, the World Health Organization (WHO), the United Nation's Food Agricultural Organization (FAO) and the US Food and Drug Administration (FDA) recognize that there is inadequate information about the predictive models for the growth and survival of Vv and Vp in shellfish between harvest and consumption, thereby preventing the design of effective science-based intervention technologies to reduce human exposure to these pathogens. Therefore, these new predictive models will address these potentially important variables, thereby assisting risk assessors and the seafood industry in designing interventions to control human health risks. In addition, the resulting research information will have important impact on reducing the uncertainty in risk assessment and will lead to more accurate estimates of human health risk from the consumption of molluscan shellfish. Training of one graduate student and one postdoctoral associate in microbiology and molecular biology as well as establishment of collaborative research among UMES, FDA, UT and MDE will enhance the national and international standing of UMES research programs, thereby enhancing student recruiting and the recognition of its microbiological research and outreach programs.