Occurrence and Genomic Characterization of Two MCR-1-Producing Escherichia coli Isolates from the Same Mink Farmer

Colistin resistance is a real threat for both human and animal health. The mobile colistin resistance gene mcr has contributed to the persistence and transmission of colistin resistance at the interfaces of animals, humans, and ecosystems. Although mcr genes have usually been recovered from food animals, patients, and healthy humans, transmission of mcr genes at the animal-human interface remains largely unknown. This was the first study to isolate and characterize MCR-producing isolates from mink, as well as to report the coexistence of two different MCR-1 producers in the same farmer. The characterization and analysis of two MCR-1-producing E. coli isolates may have important implications for comprehension of the transmission dynamics of these bacteria. We emphasize the importance of improved multisectorial surveillance of colistin-resistant E. coli in this region.

C olistin is one of the most critically important antimicrobials and is considered a drug of last resort for the treatment of infections caused by multidrug-resistant pathogens. The transmissible colistin resistance gene mcr-1 was first discovered in food animals and humans in 2015 (1). Since then, this gene has been detected in pets, wild birds, and environmental samples from various sources (2). Interestingly, a previous study reported the possibility that extended-spectrum ␤-lactamases (ESBLs), carbapenemase enzymes, and MCR-1 have coexisted since the 1980s (3). The cooccurrence of carbapenemase genes or ESBL genes and mcr-1 in the same isolate is of concern because it has the potential to lead to pan-drug resistance (4).
Previous studies have confirmed the spread of mcr genes mainly via epidemic plasmids (IncI2, IncHI2, IncP, IncX4, IncFI, and IncFIB) of various sizes (58 to 251 kb) (5). The acquisition of the mcr-1 gene has been investigated extensively, and diverse genetic contexts surrounding mcr-1 genes have been discovered (6). However, comprehensive information regarding the prevalence of mcr-1 in isolates from fur-bearing animals and the spread of colistin resistance at the animal-human interface remains limited.
We performed a survey of fecal isolates from two adjacent farms (a fur farm and a household farm) located in Shandong Province, China, in June 2016. A total of 20 samples from fur animals, 10 from chicken, 5 from pigs, and 10 from farmers were collected. One fecal sample was collected from each farmer and animal. Fecal samples were collected using sterile swabs and stored in sterile tubes. Subsequently, samples were placed on ice, transported to the laboratory, and processed within 24 h after collection.
Fecal samples were cultured on MacConkey agar plates supplemented with 2 mg/ liter cefotaxime at 37°C for 24 h under aerobic conditions to recover potential ESBLproducing isolates. The MICs were determined by the agar dilution method and interpreted according to CLSI standards (7), and EUCAST breakpoints for tigecycline, colistin, and polymyxin B were applied. Escherichia coli strain ATCC 25922 was used as a quality control.
MCR-positive isolates were characterized by conjugation experiments to assess their ability to transfer colistin resistance (10). Plasmid sizes were determined using S1-PFGE and Southern blotting methods as previously described (8). The identification of the plasmid incompatibility (Inc) group replicon types was performed by multiplex PCR, as described previously (11). The circular image representing comparisons of multiple plasmids was generated by the BLAST Ring Image Generator (BRIG) (12).
Overall, 20 ESBL-producing E. coli isolates were identified from human and animal fecal samples (Fig. 1). We detected bla CTX-M genes in all isolates, including bla CTX-M-14 (n ϭ 10), bla CTX-M-65 (n ϭ 4), and bla CTX-M-54 (n ϭ 1). The remaining ESBL isolates belonged to the bla CTX-M-55 group (n ϭ 5). We further identified 5 (22.7%) mcr-positive E. coli isolates from the fur farm ( Fig. 1). Among them, 3 were isolated from mink samples, while the remaining 2 were detected in the same farmer. DNA sequencing revealed that four isolates carried the mcr-1 gene, while isolate M4 harbored a new mcr variant. The mcr-1.12 gene, as designated here, carries a missense mutation at position 48 (T¡G), which results in a Phe16-to-Leu substitution. MIC determination revealed that all of the MCR-producing strains exhibited multiresistant phenotypes but showed susceptibility to imipenem, meropenem, amikacin, and tigecycline (Fig. 1).
Previous studies have shown that farm workers have a higher rate of carriage of ESBL/AmpC-producing E. coli than the general population (18). The cooccurrence of two different MCR-1-producing isolates in the same farmer indicates that individuals in direct contact with minks are at potential risk for carrying MCR-1 producers, although the magnitude of this risk remains to be elucidated.
Colistin is widespread used in animals but rarely used on mink breeding farms in China (1,19). Our retrospective questionnaire also revealed that only flavomycin was used for growth promotion on the fur farm, while no history of usage of polymyxins was reported from the farmers. Recently, Wu et al. reported the rapid rise in carriage of the ESBL and mcr-1 genes in E. coli of chicken origin in Shandong Province (20). Previous investigations also found that MCR variants were widespread among samples from farmers (21,22). We hypothesized that the introduction of mcr-1-positive bacteria in fur farms had occurred via the food chain-based dissemination pathway, since the farmer reported the breed history of mink fed with chicken bones. Thus far, only limited reports have described the patterns of antimicrobial use and antimicrobial-resistant bacteria among mink (23,24). The epidemiology of antimicrobial-resistant bacteria among mink and farm workers is largely unknown. Therefore, a "One Health" strategy for preventing the spread of colistin resistance at the human-mink-environment interface is essential.
The major types of mcr-1-carrying plasmids, including IncI2, IncHI2, IncX4, IncP, IncFI, and IncFIB plasmids of various sizes, have been identified in E. coli isolates from diverse hosts (5). In fact, mcr-1-carrying IncHI2/IncI2 plasmids were frequently identified from animal, human, and environmental samples in China (20,25). Our data further confirmed that similar IncHI2 and IncI2 plasmids carrying mcr-1 gene are prevalent in various sources in China.
In summary, we describe the emergence of MCR-1 producers in mink and the coexistence of two different MCR-1-producing E. coli isolates in the same farmer. Notwithstanding the limitations of this work, our findings indicate that mcr-1-carrying IncHI2/IncI2 plasmids are widely disseminated in China. Additional studies involving more fur animals and farm workers are urgently needed to assess the dissemination dynamics of mcr genes.