Epidemiologic Study of Vibrio vulnificus Infections by Using Variable Number Tandem Repeats

A 3-year environmental and clinical Vibrio vulnificus survey using simple-sequence repeats typing shows that V. vulnificus biotype 3 constitutes ≈21% of the bacterium population in tested aquaculture ponds as opposed to ≈86% of clinical cases. Simple-sequence repeats proved to be a useful epidemiologic tool, providing information on the environmental source of the pathogen.


The Study
A total of 414 V. vulnificus isolates were studied, including a reference panel of 32 strains previously studied (7). A total of 360 environmental V. vulnificus isolates were successfully retrieved from September 2004 through October 2006 from artificial fish ponds and stores in the western Galilee region of Israel (from 21 samplings), and 22 clinical isolates were retrieved from nearby hospitals during matching years (Tables 1, 2). Fish samples were collected and gills and fins/scales were pooled from ≈10 Tilapia spp.
(300-400 g). Each sample was incubated in 0.5 L modified alkaline peptone water with 4% NaCl, pH 6.9, at 37 o C for 16 to 18 h. Samples were diluted in saline and streaked on thiosulfate-citrate-bile-salts-sucrose (TCBS) agar. Suspected colonies were further grown on chromogenic agar (CHROMagar Microbiology, Paris, France), and validated by amplification of V. vulnificus-specific gene vvh (7). All V. vulnificus colonies were green on TCBS agar. Notably, not all bacterial isolates showed the expected turquoisecolonies on CHROMagar but rather pale white colonies. The latter colonies were further identified as BT3. All other isolates, which were BT1, showed the expected turquoisecolony phenotype. No BT2 isolates were found. However, 6 previously studied isolates (7) showed the expected turquoise-colony.
SSRs were used to genetically characterize 254 clinical and environmental V. vulnificus isolates (Table 1), including 32 previously studied isolates (7). DNA extraction, PCR and primers, SSR sizing, and statistical analysis were conducted as previously described (7). Capillary electrophoresis was performed by using a 3130 Genetic Analyzer and analyzed with GeneMapper-v4.0 (Applied-Biosystems Inc., Foster City, CA, USA). Two to 34 alleles were detected at the 12 SSR loci among the isolates. Environmental isolates were selected from 21 samples with an average 8.5 isolates per enrichment. We removed 71 isolates that had identical SSR genotypes and originated from the same enrichment from the analysis because they were probably clones. Thus, 183 isolates were discriminated to 170 SSR types. SSR variation data was used to calculate genetic relationships among isolates. A genetic distance matrix was generated followed by cluster analysis (7). The resulting dendrogram ( Figure 1) showed clear separation between BT3 isolates and the others (average genetic distance of 0.825 ± 0.101). Genetic distances among BT3 isolates were rather low (average 0.369 ± 0.174) relative to high genetic distances (average 0.804 ± 0.149) found among isolates of the other biotypes, in accordance with our previous analysis of 32 isolates (7). The new studied isolates showed a variety of SSR genotypes and were spread throughout the dendrogram ( Figure 1). Further analysis using eBURST (12) showed similar grouping results (data not shown) (7).
Differentiation of SSR alleles results at locus VV0401 into environmental types (E-types, >12 repeats), and clinical types (C-types, <10 repeats) was tested (13). Of the clinical isolates, 44 isolates (98%) exhibited the expected Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 15 C-type allele/repeat but only 69 isolates (33%) of the environmental isolates showed the E-type allele/repeat, rejecting the null hypothesis (p<0.0001), using Pearson χ 2 test. Notably, 97% of 110 BT3 isolates (clinical/environmental) showed the C-type allele/repeat, in contrast to BT1 isolates (72 isolates, 2%). If, C-type allele/repeat at VV0401 is an indication of potential pathogenicity of V. vulnificus strains, then our results further support the high virulence of BT3. However, additional studies are needed to confirm the relationship of this locus to pathogenicity.
Three clinical BT3 isolates exhibited identical SSR genotypes and 2 clinical BT3 isolates had a genotype related to 5 environmental isolates sampled on related dates from nearby regional areas (Figure 1, panel B). One clinical  (7). §All clinical isolates are associated with fish.  (14) or as viable cells in sediment that can serve as a shelter for some subpopulations (15). The latter scenario is more probable in artificial fish ponds because water circulation is high throughout the growth period and pond sediment remains untouched. A third clinical isolate was analyzed earlier and showed an epidemiologic connection: v254 isolate was obtained from an injured woman (injured by a fish) in December 2003. Analysis of microbial flora on the fish (found in the woman's freezer) identified a V. vulnificus BT3 isolate, v232. These 2 isolates showed identical SSR genotypes, confirming the fish as the origin of infection ( Figure 1,  To strengthen our typing results, we compared the epidemiologic SSR results to those of PFGE in 12 representative BT3 isolates. PFGE was performed and analyzed as described previously (8). Results for PFGE were generally similar to results for SSR ( Figure 2). PFGE patterns were similar (>85%) between isolates. Identical PFGE patterns and SSR genotypes were seen in isolates v239 (clinical) and VVyb95, VVyb194 (environmental), as were isolates v232 (fish) and v254 (clinical). Identical PFGE patterns and single-locus variants in SSR genotypes were seen in clinical isolate v237 and the environmental isolates VVyb89 and VVyb1. Notably, VVyb1, which was isolated a year earlier, differentiated from VVyb89 by another single repeat in an additional single locus, confirming higher resolution of SSR method. Finally, identical SSR genotypes and PFGE patterns that differed by 1 band were seen in v240 and VVyb132.

Conclusions
The developed isolation and enrichment procedures obtained large numbers of BT3 and BT1 from the environment. Results showed that although BT3 makes up only ≈21% of the V. vulnificus isolates from fish, BT3 accounts for ≈86% of the clinical cases and thus could imply high pathogenicity for this group (4). Genetic analysis of this large survey confirms the distinctness (clonality) (5) of BT3 and the high resolution power of the SSR (7).
SSR genotyping of V. vulnificus was used to determine the genetic relatedness between clinical and environmental isolates and identify the source of contamination. SSR  can serve as an epidemiologic tool to indicate the infection source of pathogens such as V. vulnificus, and can potentially provide knowledge for preventive steps in terms of public health. loci. PFGE profiles were compared by using the Dice coefficient followed by unweighted pair group method with arithmetic mean clustering (tolerance, 1.0%). Scale bars represent pattern similarity (% for PFGE and genetic distance for SSR).
All material published in Emerging Infectious Diseases is in the public domain and may be used and reprinted without special permission; proper citation, however, is required.
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the Centers for Disease Control and Prevention or the institutions with which the authors are affiliated.