mcr-1 Gene in Latin America: How Is It Disseminated Among Humans, Animals, and the Environment?

1 Programa de Pós Graduação em Microbiologia Agrícola e Do Ambiente, Universidade Federal Do Rio Grande Do Sul, Porto Alegre, Brazil, 2 Laboratório de Microbiologia Aplicada, Instituto de Ciências Básicas da Saúde, Universidade Federal Do Rio Grande Do Sul, Porto Alegre, Brazil, 3 Faculdade de Saúde, Departamento de Farmácia, Universidade de Brasília (UnB), Brasília, Brazil, 4 Laboratório de Pesquisa em Resistência Bacteriana (LABRESIS), Hospital de Clínicas de Porto Alegre, Porto


INTRODUCTION
In the last decade, polymyxins have been reintroduced in the therapeutic arsenal to treat severe infections by carbapenem-resistant Enterobacterales. At that time, reports of polymyxin resistance were all due to chromosomal mutations (1). These mechanisms included (i) modifications of the lipopolysaccharides (LPSs) moiety via the addition of cationic groups; (ii) mutations that lead to the loss of the LPS; (iii) porin mutations and overexpression of efflux pump systems; (iv) overproduction of capsular polysaccharide (CPS) in some Gram-negative bacteria (GNB) that hide the polymyxin-binding sites and the release of CPS-trapping polymyxins; and (v) enzymatic inactivation of polymyxins (2). Although some chromosomal resistance mechanisms have been studied since the 1960's, it was in the late 1990's, after the reintroduction of polymyxins in the therapeutic arsenal, that this problem became more important (3). In fact, this information is supported by the first report of colistin resistance among Acinetobacter baumannii clinical isolates from the Czech Republic in 1999 and Klebsiella pneumoniae from Athens in 2004 (4).
However, in 2015, the mcr-1 gene, associated with IncI2-type plasmid, was identified in Escherichia coli resistant to colistin obtained from food animals and humans in China (1). This finding promoted a great concern in the international scientific community since the last therapeutic option to treat serious infections by multidrug-resistant GNB could be exhausted. With the horizontal transfer, the rapid spread of the mcr-1 gene would be inevitable.
The mcr-1 gene carried by different plasmid types has already been identified in all five continents from different sources and hosts (1,5). Surprisingly, Shen and colleagues, in a retrospective study, characterized the early occurrence of the mcr-1 gene in chicken isolates from 1980's (6).
The polymyxin resistance rate associated to plasmid, as mcr-1, is also low in humans (∼1%) (4). On the other hand, according to a large US surveillance study, the association between mcr-1 and other antibiotic resistance genes, such as extended-spectrum β-lactamase (ESBL) and carbapenemases, may reach 32% of prevalence in K. pneumoniae (11). Regarding the mortality associated with infections caused by colistinresistant isolates in humans, the rate is variable, and it is higher in critically ill patients (30-37%) including those previously exposed to colistin (4). The mortality rate may reach 100% in patients with nosocomial infections caused by pan-drugresistant K. pneumoniae.
It is important to emphasize that the prevalence of mcr-1 gene is higher among production animals, mainly in pig and chicken isolates (5). The data show colistin resistance rates of ∼70% in E. coli isolates from China and ∼90% among Enterobacterales in some European countries (8). So, these data corroborate with the scientific evidence that the worldwide spread of the mcr-1 gene is mainly associated with the large amounts of colistin use in production animals, and its emergence is a particular threat to public health as colistin is considered the last-resort antimicrobial for treatment of severe human infections, and its use in livestock production contributes to emerging resistance globally (1).

mcr-1 IN LATIN AMERICA
In Latin America, a systematic review analysis showed that the prevalence of mcr-1 gene is higher in isolates from animals (8.7%) than in food (5.4%) and humans (2.0%) (12). To the best of our knowledge, the first reports of mcr-1 gene in Latin America dated from July and October 2012 when this gene was identified in E. coli isolates from two inpatients in different hospitals in Argentina ( Table 1) (13). Patients presented neurological disease and diabetes, and the mcr-1-positive isolates were obtained from blood and urine, respectively. In this study, the authors evaluated the presence of the mcr-1 gene in 87 colistin-resistant clinical human isolates from 2008 to 2016 (28 E. coli, 19 K. pneumoniae, 36 of other members of the Enterobacterales, and 4 non-fermenter Gram-negative rods), and nine isolates of E. coli were mcr-1 positive. These isolates were associated with human infections, mainly in males, and the average age of the patients was 68.5 years. All mcr-1-positive E. coli isolates were genetically unrelated as determined by pulsed-field gel electrophoresis, and the resistance mechanism was horizontally transferable by conjugation (13). Still, in 2012, other studies reported mcr-1 harboring E. coli recovered from Kelp Guls in Argentina (14) and from swine in Brazil (Table 1) (15).
It is important to consider that Brazil is the fourth largest pork producer and exporter and the largest chicken meat exporter in the world, which could contribute to the high prevalence of the mcr-1 gene in this country (25). As in other countries, the colistin was extensively used in Brazil as a growth promoter for many years. In 2016, the government published restrictions on the use of colistin in animal production (1,26), which came into force in 2018. However, the use of colistin to treat or prevent infections in veterinary medicine including animal productions is still allowed.
E. coli is the most common species harboring the mcr-1 gene in Latin America countries. However, many other Enterobacterales members such as K. pneumoniae, Salmonella spp., Citrobacter spp., and Enterobacter spp. were also reported as positive for the mcr-1 gene (17,27). In addition to mcr-1, other variants of the gene were reported rarely in Latin America, such as mcr-3, mcr-5, mcr-7, and mcr-9 (28-30).

GENETIC CONTEXT AND DISSEMINATION
OF mcr-1 GENE E. coli isolates harboring mcr-1 gene belong to different sequence types (STs) (31, 32) ( Table 1), indicating that the dissemination of the mcr-1 gene is associated with different clonal strains (1). Loayza-Villa and colleagues investigated the relationship between an E. coli carrying mcr-1 recovered from the gastrointestinal tract of a boy and an mcr-1-positive E. coli from fecal samples and rectal/cloacal swabs from his domestic animals. E. coli strains from domestic animals and from the boy were different; however, all plasmids harboring the mcr-1 gene shared 90% nucleotide identity and a highly conserved backbone, supporting the idea of horizontal dissemination of the mcr-1 gene (32).
In Latin America, the E. coli belonging to CC10 clonal complex, known as the largest human clonal complex, was the most reported in previous studies, including the ST744 and ST10 (1,17,22,33). E. coli CC10 strains are widely disseminated among humans, animals, meat products, and environmental sources (34,35) and are designated as multidrug-resistant strains carrying frequently ESBL, among others (5,31).
The role of ISApl1 in the mobilization of the mcr-1 gene was demonstrated in vitro by transposition. It was suggested that the recombination events associated with mobilization of the mcr-1 gene were initially mediated by two copies of ISApl1 from an unknown progenitor to a plasmid and subsequently transferred to Enterobacterales (41).
Besides that, according to Snesrud et al., the presence of a single or two copies of ISApl1 indicates a recent acquisition of the mcr-1 gene, whereas the absence of this insertion sequence could be correlated with the adaptation of the mcr-1 gene to a new host (41).
The regulation mechanism of mcr-1 gene expression is complex and remains unknown. In general, the gene expression is controlled by its promoter and the corresponding activators and/or inhibitors. Zhang et al. suspect that genes encoding activators and/or inhibitors in the host chromosome may affect the expression of the mcr-1 gene found on plasmids IncX4 and other plasmids. They may vary expressively in unlike genetic backgrounds of the different strains and/or mcr-1harboring plasmids, despite that their promoters are remarkably similar (42).
Although the mobility and dissemination of the mcr-1 gene are associated with ISApl1 and the pap2 gene in most plasmid types (43), the genetic context of the IncX4 plasmid type, in Latin America, is different. This context is characterized by lacking the ISApl1, but it preserves the pap2 sequence and a hypothetical protein (hp) around the mcr-1 gene (26,44). What would be the explanation for that? Snesrud et al. analyzed the genetic environment of the mcr-1 gene associated or not with ISApl1 and concluded that the target site duplications generated by ISApl1 transposition are present even in lack of the ISApl1. This result suggests that the mechanism to mobilize the mcr-1 gene is the same as that observed in other plasmids, and after that, the loss of the insertion sequence by recombination events in IncX4 occurs (45).
Furthermore, the IS26 mobile element upstream to the mcr-1 gene has been also associated with IncX4 plasmid types in Brazil, but there are no other reports in Latin America (26,46) ( Table 1). This Insertion Sequence (IS) plays an important role in the dissemination and evolution of the antimicrobial resistance genes on plasmids, including colistin resistance genes (1).

DISCUSSION
In veterinary medicine, colistin is mainly administrated in pigs and poultry production, for prophylaxis or treatment. The spread of colistin resistance may lead to treatment failure, as well as increase the pathogen transmission reach with quality and economic loss in production animals.
Strong scientific evidence indicates that the mcr-1 gene might have originated from animals because (i) colistin has been used extensively for decades in veterinary practices; (ii) mcr-1 gene was largely identified in several animals and animal food products; (iii) the identification of the mcr-1 gene in E. coli isolate recovered before 1980 in China suggests that the emergence of this gene may be linked to the use of colistin as a growth promoter in the poultry industry; and (iv) genetic features of mcr-1 gene associated with ISApl1 were first identified in Actinobacillus pleuropneumoniae, a common animal pathogen (43), which could be involved in recombination events leading to the mobilization of the mcr-1 cassette.
Finally, a recent study has demonstrated that when colistin is banned from use in animal feed, there was a significant decrease of the mcr-1 gene prevalence in most sources, including pig farms, food, and environment samples (47). Given that the production animals can be a reservoir for mcr-1 gene and its dissemination can occur by food and environment, all countries should apply surveillance, monitoring, and restrictive measures to polymyxins use. In Latin America, Brazil, and Argentina (1) have already banned the use of colistin as a growth promoter, but the impact of this measure has not been evaluated yet.
The problem of antimicrobial resistance is related to the use and abuse of antibiotics in humans, animals, and the environment. Besides that, the mcr-1 gene is disseminated mainly by E. coli clones, with a high capacity to survive in different ecological niches, some of them with pandemic and epidemic potential. So, it seems clear that the One Health approach should be adopted to integrate veterinary and human medicine to address antimicrobial resistance.

AUTHOR CONTRIBUTIONS
SAML, TVD, and AFM: conception of the opinion, collected data, and wrote the paper. ALB and AFM: reviewed and edited. All authors contributed to the article and approved the submitted version.