Show full item record

Author: Drake, Stephenie Lynn
Advisors: Dr. Lee-Ann Jaykus, Committee Chair
Abstract: The Vibrionaceae are environmentally ubiquitous to estuarine waters. Two species in particular, V. vulnificus and V. parahaemolyticus, are important human pathogens that are transmitted by the consumption of contaminated molluscan shellfish. Molluscan shellfish harvesting guidelines stipulate refrigeration within a defined time based on harvest season. Nonetheless, even the more stringent summer guidelines allow for extended storage of oysters at ambient temperatures, providing an opportunity for bacterial proliferation. Recent risk assessments have been based on limited data sets; accordingly, the purpose of this study was to address some of these data gaps in the V. vulnificus and V. parahaemolyticus risk assessments. The objectives of this study were to (i) quantify the levels of total estuarine bacteria, total Vibrio spp., and specific levels of non-pathogenic and pathogenic V. vulnificus and V. parahaemolyticus over the harvest period; (ii) determine if length of harvest time affects the levels of V. vulnificus; and (iii) determine if survival rates under environmental stress conditions differ by strain for both V. vulnificus and V. parahaemolyticus. Oyster and water samples were harvested seasonally from 3 U.S. Gulf Coast sites over 2 years. Environmental parameters were monitored during harvesting. Both surface and bottom water samples (1 L) were taken at the beginning of harvesting and at the end of harvesting. Oyster samples (15 specimens for each time point) were taken at 0, 2.5, 5.0, 7.5, and 10 hrs intervals after being held at ambient temperature during harvesting. Samples were processed for fecal coliforms, Salmonella, total Vibrio spp., aerobic plate count, and total estuarine bacteria using the appropriate media. For enumeration of total V. parahaemolyticus, pathogenic V. parahaemolyticus, and V. vulnificus was done using colony lift hybridization (tlh, tdh+ and/or trh+, and vvhA gene targets, respectively). MPN methods were also used to obtain estimates of pathogenic V. parahaemolyticus (tdh+ and/or trh+) counts. Representative V. parahaemolyticus and V. vulnificus isolates were subjected to phenotyping; V. vulnificus isolates were also subjected to genotyping. Different statistical analysis including the application of generalized linear mixed models (GLMMs), Pearson’s correlations, and analysis of variance (ANOVA) were used to establish relationships where appropriate. The first manuscript describes the field study which examined the distribution and variation in the levels of V. parahaemolyticus and V. vulnificus (both total and pathogenic strains) in shellfish and their overlay waters, and established the effect of seasonal and environmental/ecological factors on these distributions. The best estimate of growth for V. parahaemolyticus (tlh) in oysters based on water temperature was 0.054 log10 per degree C, while the best estimate for growth for V. vulnificus (vvhA) in oysters based on water temperature was 0.068 log10 per degree C. Estimated relationships between tlh and vvhA growth rates and air temperature were consistent with the risk assessments. No statistical relationship could be established between pathogenic V. parahaemolyticus (tdh+ and/or trh+) growth rates and air temperatures. In the second study, we specifically looked at the effect of extended boat deck storage of commercially harvested oysters held at ambient air temperature on the levels of V. vulnificus. During summer, increases in V. vulnificus were as high as 1.4 log10 (CFU/g) after 10 h storage at ambient air temperatures; for spring/fall oysters, a 1.0 log10 increase was observed; and in winter, increases were <0.5 log10. Statistically significant (p<0.05) correlations between ambient air temperature and increases in V. vulnificus counts were noted. Summer V. vulnificus isolates contained a higher percentage (52%) of genotype B strains, which are more often associated with clinical illnesses which was statistically significant. Spring/fall isolates were more often genotype A (80%), associated with environmental origin. In the third study, we examined the relationship of phenotypic characteristics (D-mannitol fermentation) and V. vulnificus genotypes. For a total of 469 V. vulnificus isolates, there was a high degree of concordance between 16SrRNA type and vcg variant type (92-94%) and between either genotyping method and mannitol fermentation (84-91%). Fermentation of D-mannitol may be an appropriate initial screen prior to the application of more labor intensive V. vulnificus genotyping methods. In the fourth study, strains of V. parahaemolyticus and V. vulnificus with different virulence genes/markers were evaluated for their survival under physiological stress conditions. Although there were no statistically significant differences in the recovery of V. parahaemolyticus by strain type under cold stress, there were statistically significant differences in D-values when comparing different V. parahaemolyticus strains subjected to starvation. In some instances, culture media and conditions statistically impacted V. parahaemolyticus (starvation) and V. vulnificus (cold) recovery, extending D-values by 1-3 days when media with pyruvate supplementation and aerobic incubation were used. For V. vulnificus, there were no statistically significant differences in recovery when comparing genotypes under conditions of both starvation and cold stress. In the final study, we looked an alternative method for monitoring oyster temperature during refrigerated cool-down. No statistically significant differences in oyster cooling rates were observed when comparing thermocouple versus button data logger data, irrespective of location in the commercial sacks (top, middle, and bottom) or temperature monitor location (internal vs. external). Taken together, this research provides information which can be used to fill key data gaps in the current risk assessments for V. vulnificus and V. parahaemolyticus. These data can be used in future iterations of the risk assessments to help better predict risk and inform policy as risk managers seek to reduce the disease burden associated with these important foodborne pathogens.
Date: 2008-12-02
Degree: PhD
Discipline: Food Science

Files in this item

Files Size Format View
etd.pdf 1.842Mb PDF View/Open

This item appears in the following Collection(s)

Show full item record