Sir, The prevalence of carbapenem-resistant Acinetobacter spp. in human patients is rising. Acquired carbapenem resistance among Acinetobacter spp. is mostly due to class D (OXA) β-lactamases.1 Several OXA-type carbapenemases have been described, including OXA-23. Since then, OXA-23 has been increasingly encountered among Acinetobacter spp. from humans worldwide1 and recently it has also been found in an Acinetobacter species from cattle.2 However, knowledge about carbapenemase-producing Acinetobacter spp. of animal origin remains very limited, making it difficult to assess its impact on public health. In this study, we describe the presence of OXA-23-producing strains belonging to an undescribed Acinetobacter species in the faeces of hospitalized horses. In February 2012, faecal samples were collected from 20 hospitalized horses at the Faculty of Veterinary Medicine, Ghent University, Belgium. A total of 2 g of sample was suspended in 18 mL of buffered peptone water and incubated aerobically overnight at 37°C. Thereafter, a loopful of this enrichment medium was inoculated on MacConkey agar plates (Oxoid Ltd, Basingstoke, UK) containing 1 mg/L imipenem and on a selective Acinetobacter medium without carbapenems.3 Both media were incubated overnight aerobically at 37°C. Two Acinetobacter isolates (H60437 and H60467) were obtained from the MacConkey agar plates containing imipenem and two (H60716 and H60730) from the selective Acinetobacter medium. All isolates were from different horses hospitalized in different stables and were not associated with the illness for which the horses were hospitalized. The horse from which isolate H60437 was obtained had been treated for a wound infection with penicillin [30 mL of procaine penicillin intramuscularly (im) for 5 days]. The horse from which isolate H60716 was obtained had been treated with cefquinome (450 mg im) for 4 days at home and with penicillin (21 mL of procaine penicillin im) for 5 days after surgical drainage of a metacarpal abscess. The two other horses did not receive antimicrobials during their hospitalization. The 16S rRNA gene of these isolates was sequenced and phylogenetic analysis revealed that they all belonged to a separate entity, a not yet formally described Acinetobacter species (Figure S1, available as Supplementary data at JAC Online), showing >98% sequence similarity with each other.4 Genotypic comparison of the four equine isolates was performed by PFGE.4,5 H60437 and H60467 had the same fingerprint pattern, indicating that they were identical, whereas the two other isolates had fingerprint patterns that differed from each other and from the pattern of H60437 and H60467 (data not shown). The MIC of imipenem was determined using the broth microdilution method according to the CLSI guidelines (Table 1).6 The two isolates (H60437 and H60467) obtained from the MacConkey agar plates containing imipenem had an MIC of 16 mg/L, indicating clinical resistance to this antimicrobial agent.6 The MICs of imipenem for the two other isolates, H60716 and H60730, were 0.12 and 0.03 mg/L, respectively, indicating that they were susceptible to imipenem.6 Since colistin is considered to be a last-choice drug for treatment of human patients infected with multidrug-resistant Gram-negative bacteria, the MIC of this antibiotic was also determined using the microdilution method. All four isolates had an MIC of 2 mg/L, indicating clinical susceptibility.6 Results of susceptibility testing for other antimicrobial agents, as determined by the Kirby–Bauer disc diffusion test, are shown in Table 1.6 Characteristics of the Acinetobacter isolates obtained from faeces of hospitalized horses aThe control strains used were Escherichia coli ATCC 25922 (0.06–0.25 mg/L) and Staphylococcus aureus ATCC 29213 (0.015–0.06 mg/L). bAntimicrobial drugs tested were the following: ampicillin (AMP), amoxicillin/clavulanic acid (AMC), ceftiofur (CFT), cefquinome (CFQ), imipenem (IPM), meropenem (MEM), tetracycline (TET), sulphonamides (SUL), trimethoprim (TMP), gentamicin (GEN), neomycin, amikacin, enrofloxacin and colistin. Characteristics of the Acinetobacter isolates obtained from faeces of hospitalized horses aThe control strains used were Escherichia coli ATCC 25922 (0.06–0.25 mg/L) and Staphylococcus aureus ATCC 29213 (0.015–0.06 mg/L). bAntimicrobial drugs tested were the following: ampicillin (AMP), amoxicillin/clavulanic acid (AMC), ceftiofur (CFT), cefquinome (CFQ), imipenem (IPM), meropenem (MEM), tetracycline (TET), sulphonamides (SUL), trimethoprim (TMP), gentamicin (GEN), neomycin, amikacin, enrofloxacin and colistin. The presence of blaOXA-like carbapenemase genes in the isolates was studied by PCR.2 A blaOXA-like gene was found in isolates H60437 and H60467, but was absent in isolates H60716 and H60730. Sequence analysis of the amplicon obtained from isolates H60437 and H60467 confirmed the presence of blaOXA-23 (Table 1). The genetic environment of blaOXA-23 was investigated by PCR mapping and sequencing.2 Upstream of this gene, the insertion sequence ISAba1 was detected in both H60437 and H60467. This insertion sequence, which belongs to the IS4 family,7 contained the promoter sequences [a −35 sequence (TTAGAA) separated by 16 bp from a −10 sequence (TTATTT)] that up-regulate the expression of blaOXA-23, resulting in high-level hydrolysis of carbapenems.7 ISAba1 also possessed two 16 bp inverted repeats and the +1 transcription start was found 60 bp upstream of the start codon (GTG) of the blaOXA-23 gene and located inside this insertion sequence. The genetic features described above are most likely responsible for the imipenem resistance detected in our isolates. BLAST analysis indicated that the genetic environment of blaOXA-23 was part of a Tn2008 transposon. Acinetobacter radioresistens has been identified as the progenitor of blaOXA-23.7 This was based on the identification of this gene on its chromosome and the fact that this species is susceptible to carbapenems because it lacks ISAbaI containing the promoter sequences responsible for up-regulation of blaOXA-23.7 ISAba1 has been reported to originate from Acinetobacter baumannii.7 It mediates mobilization of the blaOXA-23 gene and it most probably plays a role in the spread of this resistance gene to different Acinetobacter species and possibly also to the novel Acinetobacter species isolated here. The blaOXA-23 oxacillinase gene in combination with this insertion sequence has indeed already been reported among different Acinetobacter spp., including A. baumannii, Acinetobacter johnsonii, Acinetobacter pittii and Acinetobacter baylyi from humans1,8,9 and Acinetobacter genomospecies 15TU from cattle.2 The dynamic spread of blaOXA-23 among Acinetobacter isolates may therefore be difficult to control since it is not associated with a single Acinetobacter species and has been demonstrated in bacteria from humans as well as from animals. Our study provides the first description of a novel OXA-23-producing Acinetobacter species isolated from horses. The fact that an Acinetobacter species showing acquired resistance to carbapenems is present in horses is worrying. This may not only complicate treatment of Acinetobacter infections in this animal species,10 but it may also be considered a public health hazard. Indeed, animals are a potential reservoir of resistant bacteria and carbapenems are considered to be last-choice antimicrobials in human patients infected with multidrug-resistant Acinetobacter species.1 This work was supported by internal funding. None to declare. Figure S1 is available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/). We thank Nathalie Van Ryselberghe, Arlette Van de Kerckhove, Leentje Van Simaey and Bart Saerens for their skilled technical assistance.
No takes yet. Share an insight, caveat, or question.
Smet et al. (2012) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: