Molecular analysis of Salmonella serotypes isolated from Prince Edward County, VA waterways via sequential PCR analyses Timothy M. Smith, Jr. and David W. Buckalew Department of Biological and Environmental Sciences Longwood University Farmville, VA Introduction: The Genus Salmonella contains gram-negative, non-sporing, facultative, rod-shaped bacteria. There are two species of Salmonella: S. bongori and S. enterica. Currently, there are a combined 2,579 recognized serovars. It is a topic of debate as to how many of the 2,579 serovars are pathogenic to humans.  Some cite that all serovars can cause salmonellosis while others claim fewer Broiler houses containing chickens are a major source of Salmonella as are most wild avian and reptile species In the case that a sanitation strategy fails at a broiler house, it would be beneficial for the farmer to identify the specific serovar to gain a better understanding of its source and thus be aware of critical points of intervention. Currently, most methods of serotyping are expensive and require a high degree of training to interpret and understand Hong et al. (2008) developed an antigenic key based on a multiplex polymerase chain reaction (MPCR) that utilizes primers specific to the O, H1, and H2 gene sequences for antigens of pathogenic Salmonella isolated from chicken broiler houses in order to serotype clinically relevant serovars. This study utilizes Hong et al.’s (2008) serotyping method to examine the type specificity of Salmonella serovars isolated from local waterways near Farmville, Virginia. More specifically, this study will focus on allelotyping the four major clinically relevant serovars: Typhimurium, Enteritidis, Hadar, and Heidelberg. Materials and Methods: Salmonella Isolation Water samples were collected from three locations in Prince Edward (PE) and Cumberland Counties of Virginia: Appomattox River (APP2), Saylor’s Creek (SAY5), and Green Creek (GRE16). In the laboratory, the samples were assayed via membrane filtration. One milliliter (mL) of sample was diluted with sterile, buffered water and filtered through a 0.45 um pore size filter membrane. The filter membrane was transferred to a 50 mm petri plate containing 1.5 mL of tetrathionate enrichment broth enhanced with novobiocin.. The filter membrane was then transferred to a 50 mm petri plate containing 1.5 mL of sterile Brilliant Green Bile (BGB) broth and incubated for 24 hours at 35 o C. At 24 hours, the plates were examined for all colony forming units (CFU) presumed Salmonella spp. based upon colonial phenotype (i.e. color and morphology - Figure 1). Results: Thirty-two isolates were chosen as representative colonies following incubation on BGB. All isolates were surveyed for Gram staining reaction and cell morphology, of which all isolates tested correctly as being Gram negative and rod-shaped. All 32 of the isolates presented a K/A response on the TSI media which is typical for Salmonella. Additionally, all of the 32 isolates tested positive for the Salmonella Antibody agglutination test, serologically confirming the isolate as Salmonella spp. (See Table 2). PCR Primer Choices For further confirmation that the samples were indeed Salmonella spp. a DNA primer specific for the stn gene which codes for the Salmonella enterotoxin was used. To develop the specific primers needed to identify portions of sequences that consistently differed between groups, the serogroup-specific wba operon (related to the O antigen), fliC allele (related to H1 antigen), and fliB allele (related to H2 antigen) were compared amongst serogroups. Primers for this study were synthesized by Integrated DNA Technologies®, Inc. (US). See Table 1 for a list of chosen primer sequences of alleles for each of the antigens. Confirmatory Testing Representative CFU’s of differing colonial phenotype were photographed and then aseptically transferred to TSI agar slants for further diagnostic testing and for later serological confirmation. TSI agar slants were incubated at 35 o C for 48 hrs. Aseptically obtained samples from the TSI tubes were then subjected to serologic confirmation using Oxoid Rapid Salmonella Antibody Beads™ (See Figure 2). Extraction of Genomic DNA (gDNA) from Isolates Genomic DNA of the isolate was then extracted using the DNeasy Blood and Tissue Kit™ according to the manufacturer’s instructions Figure 1. Filtered Membrane labeled (+) for Salmonella spp. and (-) for others PCR and Gel Electrophoresis Due to logistical conflicts regarding the different primers, a single MPCR with all three major types of primers was not possible. Instead, an isolate first underwent PCR containing the O antigen primers to determine if one of the chosen O antigens was present. Once the O antigen allele was established, if it matched one of the four serovars for the putative O antigen allele then a second allelotyping was performed using the chosen H1 primer set. Again, if the second allelotyping matched one of the four serovars above then a third and final PCR using a selected H2 primer set was completed. The visualization of amplicons present after the PCR’s were made via gel electrophoresis. Figure 2. Agglutination results courtesy of Oxoid™ website An example of (+) agglutination from this experiment Results (cont’d): After completion of the PCR protocol for the O, H1, and H2 (if applicable) antigens, the antigenic formulae were then compared to those listed in the World Health Organization’s 2007 “Antigenic Formulae of the Salmonella Serovars” publication for serotyping analysis (see Table 3). Courtesy of Montana State Biophysics Research website Discussion This study adapted Hong et al.’s (2008) serotyping method, which was initially designed for typing Salmonella from broiler houses, to environmental water samples to gain a perspective of the various serovars that may be found in local PE County waterways. Traditional serotyping of Salmonella spp. involves the use of over 200 non-standardized antisera that is both costly and time-consuming. In Hong et al.’s (2008) experiment, the specificity of this MPCR scheme is a strength considering the parameters of their setting. However, in a study such as this one with samples containing a wide assortment of unknown serovars, compounded by a potentially large number of variant specimens originating from the environmental sample, the specificity now becomes a handicap to the serotyping process For monophasic serovars (the serovars that only contain an H1 antigen allele with an absence of an H2 antigen allele) this serotyping method is less effective because it does not include all possible H2 antigen allele possibilities. It remains unclear whether a particular isolate is indeed a monophasic serovar or whether the isolate is actually a biphasic serovar with an H2 antigen allele heretofore untested. Future studies should examine alternative methods of serotyping environmental isolates with better cost effectiveness than the traditional antisera approach and with the greater precision of Hong et al.’s (2008) MPCR scheme. LONGWOOD UNIVERSITY Department of Biological and Environmental Sciences Table 1: Forward (F) and reverse (R) primer sequences used in PCR reactions – Adapted from Hong et al. (2008) Table 2. TSI and confirmatory testing results from all isolates Table 4. Results from PCR and serovar analyses. A “+” denotes that the expected amplicon was visualized via gel electrophoresis. A “-” denotes that no band was visualized, and “*” denotes that an amplicon(s) was present other than the expected. 24 of the 32 isolates that tested positive for the stn gene (confirmation as Salmonella spp.), 15 of the 24 isolates were typed to S. enterica subsp. enterica (or Group 1 Salmonella – of primary importance as human pathogens). 13 of the 15 isolates typing to Group 1 Salmonella were identified to a selection of our targeted pathogenic serovars. Isolates 18, 21, 24, 26, 27, 29, and 31 amplified the stn region (e.g., were positive for being Group 1 Salmonella spp.) but did not exhibit any of the 5 O antigen alleles that were tested in this study. Isolates 1, 2, 8, 13, 15, 17, and 30 did not amplify any of the H2 allele regions that were used in this study. Courtesy of Montana State Biophysics Research website Acknowledgments: This work was supported by: 1) Cook-Cole College of Arts and Sciences, 2) Cormier Honors College, and 3) Department of Biological and Environmental Sciences, Longwood University. We would also like to thank Dr.’s Amorette Barber and Dale Beach for technical advice during the molecular trials and for agreeing to serve as committee members for T. Smith’s Longwood Senior Honors Thesis Defense