To the Editor—Polymyxins are the last resort for the treatment of infections caused by multidrug-resistant bacteria, in particular carbapenem-resistant Enterobacterales (CRE). Resistance to polymyxins used to be due only to chromosomal mutations, but in November 2015, Liu et alReference Liu, Wang and Walsh ¹ described for the first time a colistin resistance mechanism mediated by a new gene (mcr-1) that was present in a transferable plasmid. The mcr-1 has already been described on most continents, being detected in different species and obtained from several sources, including carbapenemase-producing clinical isolates.Reference Skov and Monnet ^{2 ,} Reference Dalmolin, Castro and Mayer ³ Infections due to clinical isolates harboring the mcr-1 and a carbapenem resistance gene is of particular concern because the treatment options would be seriously compromised.Reference Wang, Tian and Zhang ⁴

The aim of this study was to evaluate the prevalence of carbapenemase and mcr-1 genes co-occurring among Enterobacterales clinical isolates in southern Brazil between April 2013 and May 2018.

We evaluated the occurrence of the mcr-1 gene among 4,778 isolates of Enterobacterales with reduced susceptibility to carbapenems obtained from an epidemiologic study in several hospitals in southern Brazil. All isolates were submitted to multiplex real-time polymerase chain reaction with high-resolution melting (RT-PCR-HRM) analysis with primers for bla _KPC, bla _NDM , bla _OXA-48-like, bla _GES, bla _IMP, and bla _VIM and presented positive results for at least 1 of the carbapenemase gene(s) tested.

The presence of the mcr-1 gene was evaluated by pooling 10 isolates together and submitting them to DNA extraction and conventional PCR with specific primers for the mcr-1 gene.Reference Liu, Wang and Walsh ¹ All isolates from a pool with mcr-1 positive result were retested individually by the same conventional PCR to identify the isolate(s) that presented the gene. The amplicons from the individual isolates with positive result in the conventional PCR were submitted to Sanger sequencing and were confirmed as the mcr-1 variant. The minimal inhibitory concentrations (MICs) of several antibiotics were evaluated using broth microdilution method for the individual isolates positive for the mcr-1 gene, and the results were interpreted according to European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines. ⁵

We found only 5 isolates that presented the mcr-1 gene and a carbapenemase gene. All coharboring isolates presented the mcr-1 and the bla _KPC genes. We obtained 2 coharboring isolates (Klebsiella pneumoniae 3111F and Escherichia coli 3431F) in 2014, 1 coharboring isolate (E. coli 5798F) in 2016, and the other 2 coharboring isolates (K. pneumoniae 6701F and E. coli 6699F) in 2018. All 5 isolates were recovered from rectal swabs, with exception of E. coli 6699F, which was recovered from an ascites fluid.

Moreover, 4 isolates presented low-level resistance to colistin (4 mg/L), and 1 isolate (K. pneumoniae 6701F) was susceptible to colistin (0.25 mg/L). All isolates were resistant to ertapenem, meropenem, imipenem, and ciprofloxacin and were susceptible to tigecycline. Susceptibility to aminoglycosides was variable, with most isolates susceptible to gentamicin and intermediate to amikacin (Table 1).

Table 1 Minimal Inhibitory Concentration (MIC)Footnote ^a of Several Antibiotics for Escherichia coli and Klebsiella pneumoniae Coharboring bla _KPC/mcr-1 Genes

^a Performed by broth microdilution.

The prevalence of the mcr-1 gene was very low (0.1%) among carbapenemase-producing Enterobacterales (CPE) in our study. This rate is lower than that reported in Portugal, where 6.69% of the CPE isolates from colonized and infected patients were positive for the mcr-1 gene.Reference Mendes, Novais and Campos ⁶ In Belgium, the prevalence reported was <1% among CRE of human origin.Reference Huang, Bogaerts and Berhin ⁷ These findings demonstrate that the prevalence of mcr-1 with carbapenemase genes is normally very low, although it can differ among countries.

The mcr-1 gene is usually evaluated only among colistin-resistant isolates (MIC>2 mg/L); however, the isolate K. pneumoniae 6701F was susceptible to colistin. Some isolates are mcr-1 positive; nonetheless they are colistin susceptible. One explanation for this is the assumption that the gene might be truncated in isolates positive for the mcr-1 but susceptible to colistin, as already described in Shigella sonnei.Reference Pham Thanh, Thanh Tuyen and Nguyen Thi Nguyen ⁸ Interestingly, the truncated mcr-1 gene could be reactivated after conjugation experiments, resulting in a colistin resistant phenotype.Reference Terveer, Nijhuis and Crobach ⁹ However, a truncated gene is not the only reason to justify the susceptibility; one study reported an E. coli colistin-susceptible mcr-1 positive isolate that presented an intact gene.Reference Liassine, Assouvie and Descombes ¹⁰ Therefore, further studies are needed to elucidate the reason why the mcr-1 gene of the K. pneumoniae 6701F in this study does not promote resistance to colistin. In fact, colistin-susceptible mcr-1 isolates may be contributing to a silent spread of the mcr-1 gene, which may be transferred to multidrug-resistant isolates such as the CPE described in this study.

In conclusion, the current prevalence of mcr-1 is very low, but the detection of 2 isolates in 2018 coharboring bla _KPC/mcr-1 genes is a warning for a possible increase in the prevalence of these isolates in the coming years. Considering the association of mcr-1 with the broad-spectrum resistance mechanisms (eg, carbapenemases), this emergence is of great concern.