Occurrence and diversity of integrons and beta-lactamase genes among ampicillin-resistant isolates from estuarine waters

Abstract

The aim of the present study was to assess the occurrence and molecular diversity of beta-lactamase genes and integrons among Gram-negative ampicillin-resistant bacteria from Ria de Aveiro. Ampicillin-resistant isolates were selected and subjected to genotyping using REP-PCR. Representatives from each REP pattern were affiliated with the following taxa by sequencing the 16S rRNA gene: Aeromonas caviae, A. hydrophila, A. media, A. molluscorum, A. veronii, A. salmonicida, Aeromonas sp., Pseudomonas putida, Pseudomonas sp., Escherichia coli, Escherichia sp., Shigella sonnei, Shigella sp., Klebsiella pneumoniae, K. oxytoca, Raoultella ornithinolytica, R. terrigena, R. planticola, Citrobacter freundii, Morganella morganii and Enterobacter sp. Isolates affiliated with genera Escherichia or Shigella were identified as Escherichia coli using phenotypic-based tests. PCR was used to assess beta-lactamase encoding sequences (bla(TEM), bla(SHV), bla(CARB), bla(CTX-M), bla(IMP), bla(VIM), bla(CphA/IMIS), bla(OXA-A), bla(OXA-B), bla(OXA-C)), class 1 and class 2 integrases, and integron variable regions. Sequence analysis of PCR products was performed. beta-Lactamase genes were detected in 77.8% of the Enterobacteriaceae and in 10.5% of the Aeromonas. The most frequently detected gene was bla(TEM), followed by bla(SHV,)bla(OXA-B), bla(CphA/IMIS) and bla(CARB). Retrieved sequences shared high homology with previously described beta-lactamases. The intI1 gene was present in 29.6% of the Enterobacteriaceae and in 21% of the Aeromonas isolates. The intI2 gene was present in 4 isolates. A total of 13 cassettes included in 12 different cassette arrays were identified. The most frequently found resistance gene cassettes were aadA variants. Previous investigations based on cultivation-independent approaches revealed higher molecular diversity among beta-lactamase-encoding sequences in this estuary. This fact reinforces the hypothesis that cultivation-dependent approaches may underestimate the prevalence of antibiotic resistance genes in environmental samples and may introduce bias in the recovery of their molecular variants.