IncI1/ST3 plasmids contribute to the dissemination of the blaCTX-M-1 gene in Escherichia coli from several animal species in France

Abstract

Sir, In Gram-negative bacteria, extended-spectrum b-lactamases (ESBLs) are widespread enzymes conferring resistance to most b-lactams, including third-generation cephalosporins but with the exception of carbapenems and cephamycins. In animals, the blaCTX-M-1 gene is one of the most frequently reported ESBL-encoding genes; in contrast, in humans the blaCTX-M-15 gene is highly prevalent. Plasmids ascribed to different incompatibility (Inc) groups play a key role in the spread of ESBL-encoding genes. However, certain combinations display epidemiological success, such as the blaCTX-M-15 gene on IncFII plasmids in humans, often associated with the B2-O25b:H4-ST131 E. coli clone. Interestingly, in France, the same blaCTX-M-1-carrying IncI1/ST3 plasmid has been reported in various serovars of Salmonella enterica isolated from humans, poultry and cattle, and in Escherichia coli from healthy poultry. – 5 Here, we investigate eight ESBL E. coli isolates recovered from 2006 to 2010 from non-food-producing animals, a cat, four dogs, two horses (one a pony) and a goat, that were confirmed to harbour the blaCTX-M-1 gene by PCR and sequencing. Those isolates were collected through the RESAPATH network, which carries out surveillance of antimicrobial resistance in animals in France (www.resapath.anses.fr). As determined by disc diffusion according to the CA-SFM guidelines (www.sfm-microbiologie.fr), all strains were resistant to ceftiofur but susceptible to cefoxitin, with a typical double-disc synergy. The blaCTX-M-1 gene was preceded by the ISEcp1 element. Three isolates additionally produced the b-lactamase TEM-1 and one presented an OXA-1 (Table 1). Resistance to non-b-lactams varied depending on the isolates, but resistance to tetracyclines and sulphonamides was constant (Table 1). All isolates were unrelated, as proved by distinct PFGE profiles of BlnI-digested genomic DNA (data not shown), and belonged to the phylogenetic groups A (n1⁄42), B1 (n1⁄42), B2 (n1⁄41) or D (n1⁄43) (Table 1). The B2 isolate did not belong to the B2-O25b:H4-ST131 E. coli clone, which has spread worldwide among humans and was also recovered from livestock and companion animals. Resistance to ceftiofur was transferable by conjugation and the blaCTX-M-1 gene was identified in all recipients, together with a single IncI1 replicon (Table 1). As determined on S1-PFGE gels, IncI1 plasmids ranged between 112 and 120 kb in size (Table 1). Southern blot on S1-PFGE gels with blaCTX-M and IncI1 probes demonstrated that the blaCTX-M-1 gene was carried on the IncI1 plasmid. Subtypes of the IncI1 plasmids were determined using the plasmid multilocus sequence typing (‘pMLST’) scheme and all blaCTX-M-1/IncI1 plasmids were of sequence type ST3. Restriction fragment length polymorphism (RFLP) on PstI-digested plasmid DNA from transconjugants showed that the blaCTX-M-1/IncI1/ST3 plasmids were indistinguishable or highly similar (see Figure S1, available as Supplementary data at JAC Online). Indeed, RFLP profiles of the blaCTX-M-1/IncI1/ST3 plasmids from pets were identical to that found previously in S. enterica in humans, poultry and cattle. –5 Those from horses and the goat were indistinguishable as well, albeit slightly different from those from pets. Southern blot on the RFLP gel revealed the same PstI fragment in all plasmids from pets, also of the same size as that previously found in S. enterica (Figure S1; lanes 9 and 10). – 5 For horses and the goat, Southern blot with a blaCTX-M probe confirmed the differences observed on RFLP gels. Hybridization with a blaTEM probe suggested that the slight differences among blaCTX-M-1/IncI1/ ST3 plasmids were probably due to additional resistance (i.e. blaTEM) genes and not to major variations of the plasmid scaffold (Figure S2, available as Supplementary data at JAC Online). In this study, we report indistinguishable or highly similar blaCTX-M-1/IncI1/ST3 plasmids in different E. coli isolates from a wide range of animal species in France. All animals were unrelated, had different owners and originated from highly distant areas. They were also sampled at various periods of time, from 2006 to 2010. Consequently, we demonstrate the spread of the blaCTX-M-1/IncI1/ST3 plasmid in the animal population in France, irrespective of the E. coli backgrounds and animal species. To our best knowledge, this is also the very first report of an ESBL in a goat. In a previous work, we suggested that blaCTX-M-1/IncI1/ST3 plasmids could have transferred to cattle from poultry, a recognized reservoir of IncI1 plasmids carrying ESBL genes. In fact, this ESBL plasmid may have spread more extensively than previously thought into the animal population. Alternately, IncI1 plasmids, which are highly prevalent in animals, may have acquired the blaCTX-M-1 gene independently within different hosts. Equally worrying is the detection of an ESBL producer in small ruminants, farming of which is relatively spared from excessive antibiotic usage. Interestingly, blaCTX-M-1/IncI1/ST3 plasmids successfully expanded in animals in France, whereas most blaCTX-M-1/IncI1 plasmids reported from food-producing animals in the Netherlands were of the ST7 subgroup. Taken together, the differential expansion among countries of different ESBL plasmid subtypes