Non-O1/O139, no problem? Fish as reservoirs and vectors of Vibrio cholerae

Vibrio cholerae exhibits over 200 serotypes, with two; O1 and O139 responsible for seven pandemics since 1817. Two critical virulence factors common to both ‘O1’ serotypes are genes coding for the production of the cholera cytotoxin (ctx) leading to secretory diarrhea and the toxin coregulated pilus (tcp) that facilitates colonisation of the small intestine. However, ‘non-O1/O139’ V. cholerae serotypes lacking tcp and ctx encoding genes can still cause gastrointestinal disease through possession of other virulence factors such as heat stable enterotoxin (stn), type III secretion systems (TTSS), and haemolysin toxin (hlyA). Identification of environmental V. cholerae populations is important to ascertain the risk posed to humans that may come into contact with the bacterium.

V. cholerae is a ubiquitous aquatic vibrio and has been found in association with copepods and chironomids acting as environmental reservoirs and vectors facilitating transmission to humans. Outbreaks of cholera have been associated with the consumption of raw, dried and salted fish (Forssman et al, 2007), however, precious little research has considered the importance of fish as an environmental reservoir and vector of V. cholerae. Senderovich et al (2010) surveyed and characterised V. cholerae found within the gastrointestinal tract of various freshwater and marine fish species in Israel.

Fish were obtained from fishermen selling ‘fresh fish’ for human consumption from lakes, fish ponds, rivers, and the Mediterranean sea within Israel (n=110). Middle or lower intestinal contents were commonly directly streaked out onto TCBS agar or rarely enriched in broth culture prior to streaking. V. cholerae suspect colonies were subcultured onto LB agar and subject to oxidase and string tests. The identity of suspect isolates was confirmed by multiplex PCR assay of a V. cholerae specific outer membrane protein (ompW) and the cholera cytotoxin (ctxA). Determination of O1/O139 serotype was achieved by slide agglutination with specific antisera. The presence of additional toxin genes was assayed for all strains. Chitinase activity was determined through a chitin degradation assay.

The majority of the fish species isolated from freshwater habitats (10 out of 14 species; 71 %, n=26) were positive for ‘non-O1/O139’ V. cholerae, compared to only 1 out of 44 species (2.3 %, n=1) of marine species sampled. 50 strains of V. cholerae were isolated from the gastrointestinal tract of the fish species in this study. None of the isolates possessed the genes to produce cholera cytotoxin (ctxA), toxin coregulated pilus (tcp), or non-O1 heat stable enterotoxin (stn/sto). However, all strains possessed the gene toxR, a regulon of the cholera cytotoxin and toxin coregulated pilus, and hapA, responsible for soluble haemagglutinin/protease production. 32 % of the strains were positive for the type III secretion system (TTSS) and almost all of these strains possessed the gene hylA encoding for haemolysin toxin. There was no correlation between fish species and the genotype of the isolated V. cholerae strain. Enumeration of V. cholerae was achieved for two freshwater fish species Sarotherodon galilaeus (St Peter’s fish), and Mugil cephalus (flathead grey mullet) by calculating the colony forming units per gram of intestinal content when cultured on TCBS agar overnight at 37 °C. St Peter’s fish intestine contained 4.8 x 10³ cfu V. cholerae per gram of intestinal contents, while the flathead grey mullet contained 1.4 x 10² cfu.

All V. cholerae isolates possessed the ability to degrade chitin, suggesting a possible benefit to the host through having a ‘commensal’ population of V. cholerae in its gastrointestinal tract, particularly for species consuming a diet high in chitin such as copepod/insect prey. It would be interesting to see if a high chitin diet correlated with an increase in V. cholerae in the G.I. tract.  

This study is the first of its kind to identify fish as reservoirs of V. cholerae. Indeed, the high levels –ca 5 x 10³ cfu- of V. cholerae isolated from St. Peter’s fish is indicative of its prevalence in the aquatic environment. Although all of the strains (n=50) isolated in this study were ‘non-O1/O139’, they may still cause gastrointestinal disease in humans through production of virulence factors. Furthermore, ‘non O1/O139’ serotypes inhabit the same environmental niche as their more pathogenic brothers, and may acquire ctx and tcp genes via horizontal gene transfer. There is therefore a need to be cautious when consuming fish from species known to harbour V. cholerae, to avoid eating raw, dried, or salted fish and ensure fish is thoroughly cooked before consumption.

The propensity for dissemination of V. cholerae by migratory fish species is a novel concept that may become increasingly topical as climate change leads to increased incidence of vibrio diseases such as cholera in subtropical and temperate regions e.g. increased incidence of cholera in the Baltic Sea. 

Main Reference:
Senderovich, Y., Izhaki, I., & Halpern, M. (2010). Fish as reservoirs and vectors of Vibrio cholerae. PLoS One, 5(1), e8607.

Additional Reference:
Forssman B, Mannes T, Musto J, Baumann B, Frei U, et al. (2007) Vibrio cholerae O1 El Tor cluster in Sydney linked to imported whitebait. MJA 187: 345–347.