N.H. Sea Grant Research Project Progress Report Today s date: May 17, 2010 Project number: R/CE-137 Project title: Microbial Interactions Influencing the Emergence of Pathogenic Vibrios in Oysters Project initiation date: 2/1/2008 Principal investigator: Whistler, CA; Cooper VC Affiliation: University of New Hampshire Associate investigator(s) and affiliation(s): Technicians, industry partners, collaborators, etc. (specify which) and affiliations: Technician Randi Desy; University of New Hampshire Brief project overview/abstract: Within the last two decades the number of gastroenteritis outbreaks due to Vibrio parahaemolyticus and Vibrio vulnificus has risen steeply worldwide. Consumption of raw or undercooked shellfish represents a common mechanism for infection by these vibrios. Due in part to changing patterns in human land and resource use, and variability in seasonal extremes that may represent larger changes in climate, these emergent pathogens are a public health concern in north temperate regions not previously at risk (McLaughlin et al., 2005; Fitzpatrick, 2002; DePaola et al., 2000; Abbott et al., 1989). Public health and shellfish program managers are now challenged to better understand ecological dynamics that drive shellfish-borne Vibrio illnesses. One major question is whether changing estuarine conditions simply affect total abundance of all bacteria, including potential Vibrio pathogens, or whether certain physical or biological conditions actually may selectively amplify minority pathogenic strains. Because pathogenic biovars are typically rare and are usually not differentiated from total bacterial counts, we require a more nuanced perspective of factors influencing estuarine microbial ecology, including the role and relevance of other microbes residing in shellfish. Furthermore, the ability of bacteria to exchange genes with neighboring populations highlights the potential for spread of virulence determinants among nonpathogens. Thus, this proposal focuses on how physical factors, such as temperature, salinity, and dissolved nutrients, affect vibrios within the context of other resident microbes that may favor or antagonize potential pathogens. This proposal also will delineate in detail the genetic relationships between pathogenic and nonpathogenic Vibrio isolates and quantify the extent of their past and future potential genetic exchange. Both of these objectives will aid development of sensitive and accurate typing methods for pathogenic Vibrios in susceptible ecosystems. More generally, by examining the interplay of
physical conditions and microbial community interactions on the emergence of pathogenic biovars, we will be better equipped to predict broader infectious threats and protect human health. Objectives: 1. Map the population genetic structure and virulence potential of V. vulnificus and V. parahaemolyticus and determine the extent of recombination between strains using multi-locus sequence typing (MLST). 2. Determine the correlated effects of the resident microbial community on the abundance and distribution of Vibrio populations in oysters. Research findings/accomplishments/progress to date: Objective 1: Map the population structure of V. parahaemolyticus and V. vulnificus. From our now three seasons of sampling (June-Dec 2007, and May-Dec 2008 and 2009) we have recovered 264 isolates of V. parahaemolyticus, 36 isolates of V. vulnificus, and 36 isolates of Vibrio cholerae. As expected, no pathogenic biovar genes were detected via multiplex PCR and molecular typing in isolates, indicating these are environmental and not clinical type strains. Characterization of a subset of V. parahaemolyticus isolates for virulence phenotypes revealed that many are as cytotoxic to human colon cells as clinical isolates indicating they do harbor pathogenic characteristics. In contrast to clinical reference isolates which coordinately regulate the production of these putative virulence traits in response to human body temperature (specifically 37 not 28 ), even when the environmental strains express virulence traits they do not do so in a body temperature dependent manner. This suggests temperature regulation may be an indicator of pathogenic potential where known virulence markers are currently absent. Ongoing work is investigating whether horizontal virulence factors (specifically the hemolysin tdh and a type three secretion system harbored on the same pathogenicity island in clinical strains) is expressed in a temperature-dependent manner in environmental GBE isolates. If so, this would indicate that temperature regulation is present in these environmental strains and should virulence genes be acquired through horizontal gene transfer the strains could be converted to human pathogens. A manuscript on this work is currently in preparation. Preliminary analysis of the collection indicates a direct and significant correlation of total populations of V. parahameolyticus with temperature, and an inverse correlation with dissolved oxygen suggesting these are covariates. The seasonality in distribution that follows a temperature gradient is consistent with historical data. The low level of V. vulnificus detection is unusual compared to historical data, and may indicate a shift in population structure, perhaps due to an increase in salinity in the estuary, although the relationship was not significant. A meeting proceeding on this work is in review, and a second manuscript is currently under preparation and near submission for this work. Due to the low level of V. vulnificus recovery as compared to past years, we have focused our MLST analysis on V. parahaemolyticus. We have completed locus sequencing for 9 gene loci on 134 of the 264 isolates, and performed preliminary MLST analysis with this data set. The population structure is hyper-dispersed with no clonality
detected, although within the collection there were two complexes of related strains detected. However, the hyperdespersion is likely an artifact of our collection and processing plan which was devised to avoid collection of clones from the same sample. For our final season (May-Dec. 2009) we revised our collection scheme to allow inclusion of additional strains from the same sample which could represent clones from within a population, but not resulting from enrichment following post-harvest processing. We have completed of our collection as planned, and are in the process of completing MLST analysis on the remaining 130 strains. Although we had not included analysis of V. cholera isolates in our original proposal, our fortuitous collection of these isolates (the first in New Hampshire history) provided a unique opportunity for characterization of the endemic population consistent with this study. We have notably and unequivocally confirmed the identity by molecular typing of the 36 isolates of V. cholerae. of the strains harbor any of the known virulence factors for cholera. The first strains were collected following a rainfall event in 2008 that led to a decrease in salinity and an increase in organic matter, not unlike what is observed in regions hit with monsoons where cholerae outbreaks are problematic. We have completed extensive biochemical and phenotypic analysis of these strains, and completed a 5 gene MLST analysis of the population. The GBE population consists of two distinct types, including a larger population of non-01 type strains which are undergoing horizontal gene transfer and a small population containing a clonal complex that appears 01-like based on sequence identity. There is no evidence that the two populations have undergone recombination with each other. Isolates of both populations were collected at both sites, and in both years indicating they are endemic and not transient. There is no significant correlation with environmental conditions for the detection of these strains. A manuscript is currently under preparation for this work. Objective 2: Determine the correlated effects of the resident microbial population on abundance and distribution of Vibrios. We have completed sample processing for this objective, and are in the sequencing stage of the oyster metagenome. For this analysis, we have chosen to compare two unique sites within the estuary: a site that has historically had higher Vibrio loads (Oyster River) and a site that has had lower Vibrio loads (Nanni Island). These two sites are highly similar in temperature and salinity but differ in dissolved oxygen and an indicator of organic contamination at Oyster River. From each of these two sites we collected 11 oyster samples processed individually, along with overlying water for each of two years (August 2008 and August 2009). Preliminary MPN data indicated that none of the oysters in 2008 contained Vibrios and few in 2009 contained Vibrios. More sensitive Q-RT-PCR methods of detection were applied to these samples to identify oysters with low, medium, and high CFUs from each site, allowing correlation of endemic population with Vibrio loads. The 16s metagentic amplicons (combined V6 and V7 region) for each oyster have been individually tagged by Roche titanium MIDs allowing association of each sequence with each of the 40 source oysters providing a greater flexibility in data analysis. We anticipate sequencing to be complete by June, and data analysis to progress through August allowing some limited hypothesis testing in August at the completion of this project. Impacts to date: We have identified Vibrio cholerae from the GBE, and established a population structure map for Vibrio parahaemolyticus. We have also determined that Vibrio vulnificus
populations are in decline compared to historical data. No pathogenic strains have been detected. Patents: Technology transfer (has a private company utilized this research successfully?): Related grants and contracts (other grants and contracts that funded this research or that were obtained as a result of this research): We obtained an NIH R03 (PI Cooper) to further investigate cytotoxicity of V. parahaemolyticus isolates, to add 3 virulence loci to the MLST scheme, and to perform the MLST analysis of V. cholerae isolates. Problems encountered: Our greatest problem is with the variability of detection by culture based methods including multiple tube fermentation MPN. For that reason, we have invested significant time in developing and optimizing protocols for non-culture detection of Vibrios by Q-RT- PCR. Until we have optimized this method, we were not able to proceed with the metagenomic/ 16s library study. The delay in completing this work has led to an opportunity to develop the new method and also include a second season sample (replicate) in our library which will certainly strengthen the resulting data. We also have encountered difficulties in balancing our goal of obtaining enough isolates with avoiding collection and sequencing of clones as a result of our collection and enrichment methods. However, we adjusted our collection scheme for our final season to obtain multiple isolates from within a sample, allowing us to determine whether hyperdispersion is an artifact of our method or representative of the actual population. Publications to date (please attach PDF if applicable): Presentations to date, with published abstract citation if applicable: American Society of Microbiology, 108th General Meeting, Boston MA, June 1-5, 2008 1. Evaluation of pathogenic potential of Vibrio parahaemolyticus found in New Hampshire s Great Bay Estuary Jenny Mahoney, Nicole Lefebvre, Steve Jones, Cheryl Whistler 2. Identification of Vibrio species found in oysters and water from the Great Bay Estuary Megan Striplin, Jenny Mahoney, Steve Jones, Cheryl Whistler Students associated with project: 1. Megan Striplin (M.S. Microbiology 2010); Thesis title: Distribution and population structure of Vibrio species in the Great Bay estuary of New Hampshire 2. Jenny Mahoney (Ph.D. Microbiology 2011); Dissertation title: A comparative transcriptomics approach to understanding host-specific lifestyles in the genus Vibrio 3. Brian Schuster (M.S. Microbiology 2010); Thesis title: Host-microbe and microbemicrobe interactions that influence niche specialization by Vibrios
4. Jong Yu (M.S. Microbiology 2010); Thesis title: Evaluation of post-harvest processing methods for the remediation of human pathogenic Vibrios in oysters 5. Crystal N. Ellis (Ph.D. 2012) 6. Tucker Noyes, Microbiology (2011) 7. Anna Tyzick, Medical Laboratory Sciences (2010) 8. Nicole Lefebvre, Biology (2008) 9. Freddie Ta, Microbiology (2010) 10. Matt Gerding, Microbiology (2012)