Gene encoding virulence markers among Escherichia coli isolates from diarrhoeic stool samples and river sources in rural Venda communities of South Africa

  • CL Obi
  • E Green
  • PO Bessong
  • B De Villiers
  • AA Hoosen
  • EO Igumbor
  • N Potgieter
Keywords: Escherichia coli, Virulence markers, Water, Stool, Venda, South Africa


River water sources and diarrhoeic stools of residents in the Venda Region, Limpopo Province of South Africa were analysed for the prevalence of Escherichia coli (E. coli) and the presence of virulence genes among the isolates. A control group of 100 nondiarrhoeic stool samples was included. Escherichia coli was isolated and identified by standard cultural and biochemical methods. Pathogenicity of environmental and human isolates was determined by amplification of genes associated with virulence of E. coli, using specific primers.

Of a total of 228 water and river sediment samples screened, E. coli was recovered from 200 (87.7%), and 135 (67.5%) of these had one or more genes associated with pathogenicity. The highest frequency of isolation of pathogenic strains was found in Ritavi River water and sediment (80.6%), followed by Lotanyanda River (76.9%), and the least (45.8%) in Nzhelele River 2. Escherichia coli was recovered from all of the 252 diarrhoeic stools tested (100%), and 119 (47.28%) of these had one or more genes associated with pathogenicity. The frequency of isolation of potential pathogenic E. coli from humans was highly significant (t = 6.3; pdย•0.01) in comparison to water isolates. Cytotoxic necrotizing Factor 1 (cnf1) and cytotoxic necrotising Factor 2 (cnf2) coding for necrotoxigenic E. coli (NEC); bundle-forming pilus (bfpA) and enteropathogenic attachment and effacement (eaeA) coding for enteropathogenic E. coli (EPEC), occurred in 35% and 34% respectively of river isolates. Heat-stable (ST) and heat-labile (LT) toxin genes coding for enterotoxigenic (ETEC) and Shiga-like toxin 1 (Stx1) and Shiga-like toxin 2 (Stx2) coding for Shiga-like toxin-producing E. coli (STEC) were not encountered in the river isolates. Isolates from stool samples had 21.8% and 12.6% of EPEC and NEC strains respectively; while enterotoxigenic (ETEC), Shiga-like toxin-producing (STEC) and enteroaggregative E. coli (EAEC) had a prevalence of 5%, 5.8% and 5.8% respectively. One human isolate possessed stx2 and eaeA indicating E. coliO157: H7. No genes associated with pathogenicity were observed in human non-diarrhoeic stool isolates. Results have revealed a possibility of a recycling of pathogenic E. coli strains, particularly the EPEC and NEC strains, between the water sources and the local population.

Water SA Vol.30(1): 37-42

Journal Identifiers

eISSN: 0378-4738