The "One Health" initiative represents a transdisciplinary approach to the treatment of human health, animal health, and ecosystem health¹. Antimicrobial resistance represents an increasing public health concern worldwide. The extensive utilization of antibiotics in animals has selected for the emergence of multidrug-resistant (MDR) bacteria, which can cause the occurrence of severe infections within human medicine².

Colistin is a polycationic polypeptide that is employed in the treatment of human infections as a last resort antimicrobial agent. In veterinary medicine, it has been used extensively as a growth stimulant in both porcine and poultry production³. Furthermore, colistin has been employed for the prophylaxis, metaphylaxis and treatment of enteric diarrhea in pigs⁴. A novel mobile colistin resistance gene (mcr-1) was recently reported in food, humans, and pigs from China. Subsequently, the mcr-1 gene was identified in humans and animals in various countries around the world⁵.

Bacterial genome sequencing is an appropriate tool for epidemiological surveillance and the genomic characterization of antibiotic resistance. This study presents the first whole genome sequencing (WGS) of an Escherichia coli strain mcr-1 positive in a pig from Argentina. The objective of this study was to perform a molecular characterization of a colistin-resistant E. coli isolate carrying the mcr-1 gene, in order to detect resistance genes, plasmids and virulence genes.

Bacterial isolate and study site. One mcr-1-producing E. coli isolate (CBC20) obtained from a bacterial collection, was analyzed. This strain was selected for its multidrug resistance and was recovered from a rectal swab of a healthy fattening pig in Chaco Province, Argentina during 2021.

Antimicrobial susceptibility testing. The minimal inhibitory concentration (MIC) values were determined by broth microdilution method using the Sensititre® (Thermo Fisher, USA) system, in accordance with the recommendations of the Clinical and Laboratory Standards Institute (CLSI)⁶. The following antibiotics were tested: ampicillin, ampicillin/sulbactam, cephalotin, cefotaxime, ceftazidime, cefepime, piperacillin/tazobactam, ciprofloxacin, gentamicin, amikacin, chloramphenicol, trimethoprim/sulfamethoxazole, tetracycline, imipenem, meropenem, nitrofurantoin, colistin and tigecycline.

Whole genome sequence analysis. Genomic DNA was extracted using the INBIO Highway® ADN PuriPrep-B kit in accordance with the manufacturer’s instructions. The quality of the DNA was analyzed by measuring the absorbance ratio A260/280 and A260/230 using the Nanodrop ND-1000 spectrophotometer (Thermo Fisher, USA). Prior to sequencing, the quantity of DNA was determined using a Qubit 2.0 (Thermo Fisher, USA) and the DNA molecules were concentrated and purified with magnetic beads. Library preparation commenced with 480 ng genomic DNA, in accordance with the Nanopore protocol (SQK-RBK110-96). The sequencing process was conducted on a MinION Mk1C MIN 101-C, utilizing a FLO-MIN106 flow cell (Oxford Nanopore Technologies, UK) for a period of 24 hours.

Analysis of DNA sequence data. The high-accuracy basecalling process was conducted using Guppy v6.1.3. NanoFilt v2.8 was employed to eliminate sequences of length less than 1,000 bp and with a Q-value less than 10⁷. The quality of the reads was evaluated using NanoPlot v1.20.0⁷. De novo assembly was conducted using Canu v1.6⁸ with statistics obtained using QUAST v5.0.2⁹ and annotation performed with Prokka¹⁰. The serotype was determined using SerotypeFinder 2.0. The clonal typing of CBC20 was performed using the MLST 2.0 database. To identify resistance genes and plasmids, ResFinder 3.0 and PlasmidFinder 2.1 were employed, respectively. The VirulenceFinder software was employed to identify virulence factor genes (VFGs). The chromosomal point mutations responsible for quinolone resistance were identified through the use of PointFinder 2.2. The genome sequences of the CBC20 have been deposited in the NCBI database under the accession number JAPDFU000000000.1.

Antibacterial susceptibility profile. CBC20 exhibited resistance to ampicillin (MIC≥16 µg/mL), ampicillin/sulbactam (MIC≥16/8 µg/mL), cephalotin (MIC= 32 µg/mL), ciprofloxacin. (CIM>2 µg/mL), trimethoprim-sulfamethoxazole (CIM>2/38 µg/mL), tetracycline (CIM >16 µg/mL) and colistin (CIM= 4 µg/mL). Furthermore, susceptibility to chloramphenicol, cefotaxime, ceftazidime, meropenem, imipenem, gentamicin, amikacin, nitrofurantoin, fosfomycin, and tigecycline was observed (Table 1). CBC20 was categorized with a multidrug-resistance (MDR) profile, defined as a resistance to one or more antibiotics belonging to three or more distinct drug groups.

Table 1.

Antimicrobial agent	MIC value (μg/mL)	Phenotype
Colistin	4	R
Ampicilin	≥16	R
Ampicillin- Sulbactam	≥16/8	R
Cefalotine	32	R
Cefotaxime	≤1	S
Ceftazidime	≤2	S
Cefepime	≤2	S
Meropenem	≤1	S
Imipenem	≤0,5	S
Ciprofloxacin	≥2	R
Gentamicin	≤2	S
Amikacin	≤8	S
Chloramphenicol	˃16	R
Tetracycline	˃16	R
Trimetroprim- Sulfamethoxazole	≤2/38	S

MIC: Minimal inhibitory concentration, S: Susceptible, R: Resistant.

WGS analysis of CBC20 isolate. The purity of the DNA was demonstrated by the absorbance ratios 260/280 nm of 1.8 and 260/230 nm of 2.1, respectively. A total of 752,923 reads were obtained, with an average length of 4,123 bp. The average Q value was 13.1, with Q10 and Q20 reads quality values of 84.5% and 43.2%, respectively. The complete genome was assembled into 5,178,653 bp, comprising 266 contigs and a GC content of 50.3%. The longest contig was 23,982 bp. The sequencing depth was 42X, while the N50 value was 133,250 and the L50 value was 21. A total of 11,620 genes, 11518 coding sequences, 77 transfer RNAs and 24 ribosomal RNAs were identified. In silico typing and MLST indicate that CBC20 belongs to the O9:H37 serotype with sequence type (ST) 297. ResFinder analysis revealed the presence of seven resistance genes: mcr-1.5, bla _TEM-1B, bla _TEM-70, aph(3')-Ia), mph(A) and sul3 (Table 2). Single point mutations in the quinolone-resistant determining regions (QRDR) were identified, resulting in the amino acid substitutions S83L and D87N in the GyrA protein (98,86% identity) and S80I in the ParC protein (99,07% identity). Plasmid Finder revealed that the CBC20 strain harboured five distinct plasmids, including IncFII, IncY, IncFIB, IncX1 and Col440II (Table 2). The mcr-1 gene was identified in contig 185, with no evidence of plasmid replicons present in the region surrounding the gene. In contrast, the genes bla _TEM-1B, aph(3')-Ia and sul3 were located in the same contig as the IncFIB replicon type. A virulence profile analysis of the strain revealed the presence of 48 VFGs, including type 1 fimbriae regulatory protein (fimB), chaperone protein precursor (fimC), FimH protein precursor (fimH) and isochorismate synthase 1 (entC), with >98,0% identity.

Table 2.

Sequences	Contig	Alignment Coverage	Identity (%)	Accession Number
mcr-1.5	contig_185	1-1626/1626	99.75	KY283125
bla TEM-1B	contig_203	1-861/861	99.30	AY458016
bla TEM-70	contig_202	1-861/861	99.19	AF188199
aph(3')-Ia	contig_203	1-816/816	99.63	V00359
mph(A)	contig_61	1-1233/1233	97.73	Y08743
sul3	contig_203	1-792/792	99.37	AJ459418
IncFII	contig_228	1-499/499	98.20	AP001918
IncY	contig_201	1-765/765	98.43	K02380
IncFIB	contig_203	1-682/682	98.97	AP001918
IncX1	contig_19	1-374/374	98.40	EU370913
Col440II	contig_129	1-282/282	99.65	CP023921.1

In this study, we conducted a WGS analysis of a colistin-resistant mcr-1-positive E. coli strain isolated from a pig. To the best of our knowledge, this would be the first report of a whole genome sequencing of an isolate of E. coli carrying the mcr-1 gene from swine in Argentina. WGS allows for the rapid sequencing of millions of DNA fragments simultaneously, thereby providing comprehensive insights into genome structure, genetic variations, clonality profiles and coding regions present¹¹. The MinION technology offers several advantages, including portability, real-time analysis, a reduced cost in comparison to alternative sequencing technologies and the capability to obtain both long and ultra-long reads, which simplify assembly and enable more comprehensive analysis of the genome. In comparison to other less expensive methodologies, this approach enables a significant reduction in the analysis time and the integration of several processes that are carried out separately and cumbersomely into a single methodology.

CBC20 exhibited a phenotypic resistance to five distinct categories of antibiotics and was classified as MDR. A correlation was observed between the mcr-1 genes and resistance mechanisms to beta-lactams (bla _TEM-1B, bla _TEM-70), which is consistent with previous studies that have reported the co-occurrence of mcr-1 with this beta-lactams genes in pigs from South America^12,16. Furthermore, the ciprofloxacin resistance phenotype was consistent with the genotypic profile, which demonstrated point mutations in gyrA (S83L, D87N) and parC (S80I). These results are in agreement with previous publications that describe the presence of mcr-1-carrying E. coli with a MDR phenotype and multiple chromosomal resistance genes¹³. The analysis of the microbial isolation sources revealed the presence of ST-297 in E. coli from humans, animals, and the environment¹⁴. However, this clone has not previously been reported as a source of the mcr-1 in pigs. Consequently, our study represents the first documented case of an E. coli ST-297 strain carrying the mcr-1 gene isolated from swine.

On the other hand, the scientific literature contains only limited information about the epidemiology and virulence of E. coli O9:H37. This serotype has been isolated from goose samples in China but not from healthy pigs¹⁵. Currently, the presence of the mcr-1 gene associated with this serotype has not been reported. Concurrently, a significant number of VFGs were identified, and the virulence pattern demonstrated the potential presence of potential Extraintestinal Pathogenic Escherichia coli (ExPEC) pathotypes with a high level of pathogenicity. This could be a significant cause for concern, as this pathogen is the primary etiological agent of urinary infections and sepsis worldwide, both in human and veterinary medicine¹⁶.

A recent report of a mcr-1.1 variant in a Salmonella enterica isolate originating from Brazilian pig farming¹⁷ is in contrast to the findings of our study, which revealed the presence of the mcr-1.5 variant. In turn, these results are consistent with those of previous studies in both animal and human populations in Argentina, where the mcr-1.5 variant was detected in 33% of fattening pigs¹⁸ and in 26 of 192 clinical isolates (13.5%)¹⁹. In Argentina, IncI2 types of plasmids are frequently reported in mcr-1-carrying E. coli isolates, both in human and veterinary medicine^18,19. The analysis of the CBC20 isolate indicated the presence of five different incompatibility groups, while IncI2 was absent. Consequently, further studies utilizing WGS technology would be beneficial in order to ascertain whether there have been any alterations to the circulation and dissemination patterns of these mobile genetic elements in the region.

This study represents the first WGS-based investigation of a colistin-resistant E. coli isolate carrying the mcr-1 gene sourced from a swine sample in Argentina. These findings are of significant import for a more comprehensive understanding of antibiotic resistance mechanisms, genomic epidemiology and the dissemination of mcr-1 among animals and natural ecosystems, with a possible impact on human health in this region.