1. Introduction
The use of antibiotics in humans and animals represents the cornerstone of modern medicine. Since their discovery, antibiotics have revolutionized the treatment of bacterial infections around the world. However, bacteria have evolved mechanisms to become resistant to these agents, and a steady increase in this phenomenon has made antimicrobial resistance one of the major health problems affecting humankind.
In the case of Gram-negative bacteria, infections caused by multi-drug resistant-
Enterobacterales,
Pseudomonas aeruginosa and
Acinetobacter sp, are on the rise [
1]. This reality has led to the use of old antibiotics, such as colistin, as a last-resort antimicrobial against these pathogens. Colistin, a polypeptide antibiotic also known as polymyxin E, was discovered in 1949 and was largely used in the 1950s; however, due to its neuro- and nephro-toxicity, it was abandoned in the 1980s in favor of other antibiotics [
2]. Colistin exerts its antibacterial activity against Gram-negative organisms by electrostatically binding the negatively charged phosphate groups of lipid A in the bacterial lipopolysaccharide (LPS), disrupting the integrity of the bacterial membrane [
3].
Currently, antibiotic resistance against colistin is not as worrisome as that against other antimicrobials; however, due to the overuse and misuse of this antibiotic among humans and animals, colistin resistance is on the rise, as described in several reports from different nations [
4,
5,
6]. Resistance against this agent is attributed to the modification of LPS, through cationic substitutions, which reduces the net negative charge of this molecule, impeding the binding of the antibiotic [
7]. Colistin resistance was originally described as being chromosomally encoded; however, in 2015, Liu et al described the first plasmid-mediated resistant gene against this antibiotic,
mcr-1, in
E. coli [
8]. Shortly after its discovery,
mcr-1 was identified in over 20 different countries and, to date, nine additional
mcr-like genes have been described [
9]. Furthermore,
mcr genes have also been identified in
Pseudomonas aeruginosa and
Acinetobacter sp [
10,
11].
With the global spread of
mcr genes among multi-drug resistant bacterial strains and their further horizontal transmission among bacteria, the effectiveness of colistin is under serious jeopardy. An additional factor that might facilitate the spread of colistin resistance is the likelihood of acquiring this plasmid-mediated resistant trait in multi-resistant
E. coli clones such as O25b-ST131, the most common multidrug-resistant high-risk clone associated with extra-intestinal
E. coli infections around the world [
12]. This clone commonly carries resistance genes to antibiotics frequently prescribed in general practice, such as cephalosporins, which are mainly mediated by the
blaCTX-M gene, as well as quinolones, and can easily be transmitted through the consumption of food [
13]. These antibiotics can co-select for colistin-resistant strains and, thus, contribute to the spread of resistance against this agent. ST131
E. coli clones carrying colistin resistance genes have already been reported in environmental and clinical samples [
14,
15].
As the surveillance of antibiotic resistance and research is a key component of the global action plan against antimicrobial resistance [
16], the purpose of the current study was to determine the prevalence of the plasmid-mediated colistin resistance genes
mcr -1 to
mcr -5 among colistin and multi-drug resistant
Enterobacterales,
Pseudomonas aeruginosa and
Acinetobacter sp strains isolated from patients in a tertiary hospital in the city of Toluca, Mexico. Additionally, we sought to assess whether plasmid-mediated colistin-resistant
E. coli strains isolated from these patients belonged to the ST131 clone and co-expressed
blaCTX-M genes.
4. Discussion
Due to the increasing rates of resistance among Gram-negative bacilli against commonly used antibiotics, mainly in bacterial strains isolated form hospitalized patients, colistin has emerged as one of the last-resort antimicrobials in the treatment of these infections. Unfortunately, resistance against this agent is on the rise, particularly in Asia, which is mainly attributed to its overuse in veterinary medicine [
26].
One of the main objectives of the current study was to determine the prevalence of colistin resistance among multi-resistant Enterobacterales, P. aeruginosa and Acinetobacter sp strains isolated from patients in a tertiary hospital in Mexico. In total, 241 strains were included in the study, of which 12 (5.0%) were found to be resistant to colistin according to the CLSI guidelines. Of the 12 colistin-resistant strains, 4 were P. aeruginosa, with a colistin-resistance prevalence of 7.5%, (4/53); 4 were E. coli, with a prevalence of 2.5% (4/162); 3 were K. pneumoniae, with a prevalence of 15.0% (3/20) and one, E. cloacae, with a prevalence of 50.0% (1/2). Although K. pneumoniae presented one of the highest prevalence levels of colistin resistance, no PDR strains were found, unlike P. aeruginosa, for which one strain was found to be resistant to all tested antibiotics. In this study, no Acinetobacter sp isolates were found to be colistin-resistant.
The antibiotic resistance profiles of the colistin-resistant strains found in this study are shown in
Table 3. As all 241 strains included in the current study were multi-resistant isolates, antibiotic resistance to different types among them was high. In the case of
P. aeruginosa, colistin-resistant strains presented an extremely high prevalence of resistance against carbapenems (100%), ceftazidime (100%), ciprofloxacin (100%) and amikacin (100%) and low resistance to tigecycline (25%). For
Enterobacterales, higher rates of resistance were found against ciprofloxacin (100.0%), trimethoprim/sulfamethoxazole (87.5%), cephalosporines (62.5%), and ceftazidime (62.0%); no resistance was found against amikacin. Finally, resistance against at least one of the carbapenems was found in one
E. coli (1/4) and one
K. pneumoniae (1/3) strain.
The prevalence of colistin resistance in this study is in agreement with reports in different geographical areas of the world among muti-resistant strains isolated from hospitalized patients [
10,
27,
28], but higher than the 1.26% worldwide prevalence reported by Dadashi et al [
29]. In Mexico, reports on the prevalence of colistin-resistant strains are scarce; however, in a recent report by the Red Temática de Investigación y Vigilancia de la Farmacorresistencia (INVIFAR network), that included clinical strains isolated from different parts of the country, the colistin resistance prevalence among
K. pneumoniae strains was 17.8% [
30], a rate similar to that found in the current study among isolates of this bacterial species (3/20 = 15.0%). Unlike the INVIFAR report, where no resistance to colistin was found among
E. coli strains, we found that 2.5% of
E. coli strains included in the current study presented this resistance trait. These results confirm that colistin resistance among clinical isolates in our country is already a reality and should be carefully monitored.
As the world has witnessed an increase in resistance against colistin, this phenomenon has been mainly attributed to the emergence and dissemination of plasmid-mediated (
mcr -1 to 10) genes among bacteria. The first
mcr gene (
mcr -1) was originally described in 2015 in China [
8], and in Mexico,
mcr -1 was first identified in 2019 in an
E. coli strain isolated from a fecal sample from a child [
31]. The presence of this plasmid-mediated gene in our country was recently confirmed in
K. pneumoniae strains isolated from different clinical samples [
30]. In the present study, only two
E. coli strains, both isolated from female patients with urinary tract infections that required hospitalization, were found to carry the
mcr -1 gene. No other
mcr gene was found in the strains analyzed and no other bacterial species in addition to
E. coli were found to carry plasmid-mediated colistin resistance genes. In this study, plasmid-mediated colistin resistance genes were not found in
P. aeruginosa. However, as colistin is the last-resource antibiotic against strains of this organism resistant to carbapenems, reports of
mcr genes among multi-resistant strains of this species have started to emerged [
32,
33] and horizontal transmission of
mcr -1 genes has been shown to occur from
P. aeruginosa to other bacterial species [
34], the surveillance of
mcr -carrying
P. aeruginosa strains remains highly encouraged.
The relatively low prevalence of
mcr genes among colistin-resistant Gram-negative bacteria found in this study (16.7%) is higher to the values reported for
E.coli in Egypt (7.5%) [
35] and similar to those described in Ecuador (20.0%) [
36], but lower than the prevalence shown in a study in Nepal [
37] and Peru [
38], where the prevalence of strains that serve as carriers of
mcr genes was reported to be high among colistin-resistant
E. coli and
K. pneumoniae. The results of the current research confirm the fact that plasmid-mediated colistin resistance has spread at different rates among different geographical areas of the world. It is also important to note that in this study,
mcr genes were searched for in colistin-resistant strains and only multi-resistant isolates were included, but different studies have shown that colistin susceptible
Enterobacterales can carry
mcr genes [
39,
40]. Thus, the prevalence of
mcr -carrying strains at the Microbiology laboratory from Centro Médico ISSEMYM Toluca, Mexico, could be higher than the levels indicated by the results in this study. The low prevalence of
mcr genes found in this investigation among colistin-resistant strains suggest that in our population, resistance to this agent is mainly driven by chromosomal mutations or plasmid-mediated genes not included in the study. Further research is needed to understand additional mechanisms of colistin resistance to those identified in the current study.
Several reports have shown that Gram-negative bacilli can co-harbor
mcr genes and other plasmid-mediated antibiotic resistance traits such as those encoding for carbapenemase and ESBL production. Since colistin is mainly indicated as a last-resource antibiotic, the co-expression of colistin and carbapenemase resistance is worrisome among the medical community and, understandably, has been more thoroughly studied and more commonly demonstrated among
Enterobacterales [
30,
41] than its co-expression with extended-spectrum-β-lactamases. However, bacterial strains co-harboring
mcr and ESBL genes should also be carefully monitored. In addition to encode for intrinsic resistance against cephalosporines, penicillins and monobactams, plasmids carrying ESBL-encoding genes, can also harbor resistance genes against other commonly used antibiotics such as ciprofloxacin, trimethoprim/sulfametoxazol and aminoglycosides [
42]. When administered, any of these antibiotics could co-select for colistin-resistant strains and contribute to the spread of resistance against this antimicrobial. In the present study, the two
mcr-1-carrying
E.coli strains were additionally found to carry
blaCTX-M genes, a phenomenon that has also been demonstrated in human isolates in Qatar [
43], Peru [
44] and the Indian Ocean Commission [
45], showing that the co-expression of plasmid-mediated resistance against colistin and ESBL production has spread to different regions of the world.
In the current study, both
mcr-1- and CTXM-carrying
E. coli strains were able to horizontally transmit their colistin resistance and ESBL production to the recipient strain, suggesting that conjugation may play a role in the spread of colistin resistance. In this study, only a minority of colistin-resistant isolates were found to carry
mcr genes and in the two
mcr-1-carrying strains no carbapenem resistance was detected, leaving other therapeutic options available for the treatment of infections caused by these strains. However, as different studies have shown that the horizontal transmission of
mcr-1 genes occurs in food [
46] and animals [
47], this mechanism might be responsible around the globe for the steady increase in colistin resistance among Gram-negative bacilli.
The highly antibiotic resistant clone ST131, predominantly serogroup O25b, is considered the dominant extraintestinal pathogenic
E. coli around the world, including being a frequent cause of urinary tract infections [
12,
13]. In the current study, the two
E. coli strains carrying
mcr-1 and
blaCTX-M genes were isolated from patients with urinary tract infections and were resistant to most classes of antibiotics, demonstrating sensitivity only to the carbapenems and, in the case of one strain (2207), to cefepime. In addition, these two strains were found to belong to the ST131-O25b clone, an
E. coli clone that commonly exhibits resistance to quinolones, trimethoprim-sulfamethoxazole and aminoglycosides, and is recognized as the primary lineage responsible for the spread of
blaCTX-M genes [
48]. The results of this study support previous findings on strains isolated from humans [
15,
49], showing that the already highly resistant clone ST131 can acquire plasmid-mediated colistin resistance genes. As this clone can be transmitted from person to person and through the consumption of contaminated food [
13], and given that its prevalence in the feces of healthy humans is on the rise [
50], the spread of this multi-resistant clone could be facilitated by a lack of hygiene, which suggests that less privileged areas of the world might see an increase in the prevalence of this clone. If
mcr-1-carrying O25b-ST131
E. coli clones expand to different geographical areas, commonly used antibiotics, such as ciprofloxacin, cephalosporines and trimethoprim-sulfametoxazol, could co-select for colistin-resistant strains and contribute to the spread of resistance against this last-resort antibiotic.
Limitations of the current study include its small sample size, unicentric design and short timeframe. Lastly, not all known mcr variants were analyzed; thus, the actual plasmid-mediated colistin resistance prevalence at our institution might be higher than the levels suggested by these results.
In conclusion, this study, albeit small, demonstrated that the emergence and spread of mcr-carrying strains among humans is a reality in Mexico. In addition, the presence of this resistant trait among the highly resistant and easily transmissible O25b-ST131 E. coli clone further complicates the antibiotic resistance scenario in our country. Further surveillance studies are needed among other hospitals and ambulatory patients in Mexico to determine the magnitude of the problem of colistin resistance, especially that mediated by mcr genes, in order to establish policies aimed at optimizing antibiotic stewardship programs to reduce the dissemination of resistance against this last-resource antibiotic.