Bacterial subtyping (or often called "DNA fingerprinting" or "bacterial genotyping") is a group of methods that are commonly used to differentiate bacterial isolates at sub-species or strain levels in epidemiological studies of various pathogens. Our previous work had been focused on development and evaluation of multilocus sequence typing (MLST) methods for short-term epidemiological subtyping of L. monocytogenes. In L. monocytogenes, virulence genes evolve rapidly under strong selective pressures, and are often more polymorphic than housekeeping genes. Analysis of virulence gene sequences therefore improves the discriminatory power (ability to differentiate bacterial strains) and epidemiologic relevance of MLST-based schemes for analyzing L. monocytogenes strains. Unlike L. monocytogenes, enterohaemorrhagic E. coli (EHEC) O157:H7 strains are highly clonal. Although E. coli O157:H7 strains appear to be genetically diverse by pulse-field gel electrophoresis (PFGE) analysis, this type of genetic diversity is largely attributable to numerous insertions and deletions associated with bacteriophages and other mobile genetic elements (e.g. transposons).
	Most of genes with known housekeeping and virulence functions in E. coli O157:H7 are highly conserved (>99.9% identical) at both nucleotide and amino acid sequence levels, therefore, MLST analysis can not provide effective strain differentiation for E. coli O157:H7. Our recent effort was to design a subtyping scheme to target genome-wide single nucleotide polymorphisms (SNPs) to differentiate strains and study the long-term epidemiology of E. coli O157:H7. By using comparative genome sequencing (CGS) microarrays, we analyzed approximately 1,200 genes in ten E. coli O157:H7 genomes and have identified >800 SNPs in many conserved genes that are potentially useful for SNP genotyping. Currently we are developing a high-throughput SNP genotyping assay to analyze more E. coli O157:H7 strains with diverse epidemiological backgrounds. Data generated in this study will be critical for us to better understand the distribution, transmission and epidemiology of E. coli O157:H7 and to develop effective preventative strategies to reduce the incidence of E. coli infections (hemorrhagic colitis). Subtyping data in this study will also help us to understand the evolution of E. coli O157:H7 from the genomic point of view.
Molecular subtyping results along with epidemiological data collected in many studies have revealed significant variations in virulence among different bacterial subpopulations (e.g. serotypes and genetic lineages) within the same bacterial species. For example, in L. monocytogenes, there are 13 different pathogenic serotypes, however, only three serotypes (1/2a, 1/2b and 4b) are responsible for the vast majority (>90%) of foodborne illness, indicating that these serotypes may possess higher virulence potential to infect human and cause disease than other serotypes. In addtion to enhanced virulence, bacterial subpopulations may also differ by host preference. For instance, in E. coli O157:H7, two distinct genetic lineages (or clades) have been identified with apparent host specificity (human or cattle), indicating that some E. coli O157:H7 strains are more adapted to infect and colonize in human hosts whereas others are more adapted to bovine hosts. The genetic mechanisms for these variations in virulence and host preference is largely unknown, yet remains very intriguing.
	The overall hypothesis in our study of bacterial virulence variations is that subpopulation-specific virulence factors do exist and may express differentially during colonization in environmental niches and infection in host species. By using comparative genomics tools, we have identified many serotype and lineage-specific genes (or putative open reading frames) with hypothetical or unknown functions in both L. monocytogenes and E. coli O157:H7 genomes. New DNA microarrays are being designed and used to study the differential gene expression to further characterize these putative virulence factors and predict their functions in virulence and pathogenesis during environmental colonization and human infections. In addition to basic studies in bacterial pathogenesis and epidemiology, our lab have ongoing research projects focusing on: i) study of stress responses of model foodborne pathogens (e.g. Salmonella enterica); ii) development of rapid assays for real-time detection/monitoring of bacterial pathogens in food matrices.