Complete sequence of an IncFII plasmid harbouring the colistin resistance genemcr-1isolated from Belgian pig farms

Sir, The monumental increase in antibiotic resistance among important bacterial pathogens, driven by inappropriate and appropriate use of ineffective drugs, is currently recognized as one of the most pressing threats to human health by the WHO.1 In particular, the last decade has seen a significant rise in infections caused by MDR and XDR Gram-negative pathogens such as Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa and Acinetobacter baumannii.2,3 Antibiotics of the polymyxin group such as colistin are sometimes the only drugs to which these bacteria show susceptibility and, therefore, reports of emergence of a plasmid-mediated mcr-1-encoded mechanism of resistance to colistin have been especially alarming.4 We have previously screened rectal swabs collected during 2011–12 in Belgium from post-weaning diarrhoea in pigs and found a 13.2% (7/53) prevalence.5 In this study, we carried out a detailed analysis of the seven mcr-1-harbouring strains (colistin resistance defined by Sensititre >4 mg/L) and also sequenced the plasmid harbouring the gene in one porcine strain. Briefly, colistin MICs were first reconfirmed using a macro broth dilution method and found to be 8 and 16 mg/L for all seven strains. Screening for ESBLs by PCR6 and phenotypic tests (double disc diffusion and Etests) did not identify any strain as an ESBL producer. MLST7 showed that the seven strains belonged to different STs: ST10 (n = 2), ST100 (n = 2), ST90 (n = 1) and novel STs (n = 2). Of these, one (allelic profile: 10-27-5-10-172-9-2) was a single allele variant of ST100 (allelic profile: 10-27-5-10-12-9-2), with an SNP in the malate dehydrogenase-encoding mdh at position 172, and the other (allelic profile: 10-11-4-8-266-8-2) was as a single allele variant of ST10 (allelic profile: 10-11-4-8-8-8-2), also showing differences in mdh at position 266. Geoburst analysis showed that these two novel STs might have evolved from ST100 and ST10, respectively. Of these, E. coli ST10 have previously been associated with human infections8 and KP81, one of the two ST10 strains, was selected for plasmid sequencing. Plasmid DNA was extracted (PureLink HiPure Plasmid Miniprep Kit; Invitrogen, Carlsbad, CA, USA) followed by Nextera XT sample and library preparation and sequencing (MiSeq, v2, 2 × 150 bp; Illumina). The plasmid sequence was de novo assembled (SPAdes v3.6.2), annotated (BASys) and compared with the published mcr-1 gene-carrying plasmid pHNSHP45 (accession number KP347127) using CLC Genomics Workbench v7.5.1 (CLC bio, Denmark). Incompatibility type, pMLST and ISs were defined using PlasmidFinder, pMLST database (https://cge.cbs.dtu.dk/services/) and ISfinder tools (https://www-is.biotoul.fr/).9 Integron types were identified and annotated using BLAST search against the INTEGRALL database (http://integrall.bio.ua.pt/) and the Repository of Antibiotic Resistance Cassettes (http://rac.aihi.mq.edu.au/rac/). Plasmid pKP81-BE (91 049 bp; Figure 1a) encoded 45 ORFs, showed a GC content of 44.9% and belonged to IncFII incompatibility type. The pMLST type was F2:A−B and the serotype was O50/O2:H32. The plasmid backbone showed only 4% similarity to that of pHNSHP45 (accession number KP347127),4 which included the mcr-1 gene and its adjoining IS. Also, pKP81-BE showed 58.52% similarity to plasmid pKH457-3-BE (accession number KU353730), an IncP-type plasmid carrying mcr-1 isolated from E. coli from neonatal diarrhoea in a calf in the Wallonia region of Belgium.5 In contrast to pKH457-3-BE, the mcr-1-harbouring ISApI1 mobile element on pKP81-BE was not truncated and included the transposase-encoding tnpA gene. The pKP81-BE backbone was 99% similar to Salmonella enterica subsp. enterica serovar Typhimurium strain plasmid pHK0653 (accession number KT334335) with 78% query coverage. In addition to mcr-1, pKP81-BE harboured a sul3-containing class 1 integron, In640, previously shown to be associated with Salmonella and E. coli of animal origin.10 This integron showed the presence of genes encoding resistance to trimethoprim (dfrA12), aminoglycosides (aadA1a and aadA2), sulphonamides (sul3) and phenicols (cmlA1) (Figure 1b). Other insertion elements identified were TnAs1, TnAs2, TnAs3, ISPa38, ISPa40 (Tn3), IS26, IS15DIV, IS15DII and IS15 (IS6). (a) Genetic organization and structure of mcr-1-harbouring pKP81-BE. The image was generated using http://stothard.afns.ualberta.ca/cgview_server/. (b) Resistance genes harboured on the In640 integron. Interestingly, both plasmids, porcine and bovine, isolated from the northern and southern parts of Belgium, were present in E. coli belonging to ST10. This ST, although commonly encountered as an antimicrobial-susceptible, low-virulence human intestinal colonizer, has also been associated with human infections and ESBL production.8 However, none of the mcr-1-harbouring E. coli in this study was an ESBL producer based on phenotypic testing. Further screening of all scaffolds derived from the plasmid sequencing identified an IncH12 plasmid along with the mcr-1-harbouring IncFII plasmid. IncFII-type plasmids have been important in spreading resistance genes such as blaNDM-1 and blaCTX-M-15 globally,11 so we chose publicly available IncFII plasmid sequences from E. coli for a comparative analysis. We selected the pGUE-NDM (accession number NC_019089) blaNDM-1-harbouring plasmid from E. coli ST131, the ST reported to be carrying the mcr-1 gene in Europe and China4,12 and pXZ (accession number JF927996) from China and pHK23a (accession number JQ432559) also a pig isolate from China. However, comparative analysis showed only <10% similarly restricted to genetic regions associated with plasmid replication and transfer (data not shown). To our knowledge, this is the first report of the association of mcr-1 with an IncFII plasmid. The fact that we identified two different genetic associations of mcr-1 within a small geographic region (Belgium) further underscores the high promiscuity of the mobile element harbouring mcr-1. The sequence of plasmid pKP81-BE was deposited in GenBank (accession number KU994859). B. B. X. is supported by University of Antwerp Research Funds (BOF-DOCPRO 2012-27450). The study is supported by European Union Horizon 2020 Research and Innovation Programme: Compare [‘COllaborative Management Platform for detection and Analyses of (Re-) emerging and foodborne outbreaks in Europe: Grant No. 643476]. The sponsor had no role in study design, data collection, data analysis, data interpretation or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication. None to declare.