Antibiotic resistant Escherichia coli from diarrheic piglets from pig farms in Thailand that harbor colistin-resistant mcr genes

Antibiotic-resistant Escherichia coli is one of the most serious problems in pig production. This study aimed to determine the antibiotic susceptibility and genotypes profiles of diarrhoeagenic E. coli that causes diarrhea in piglets. Thirty-seven pathogenic E. coli strains were used in this study. These were isolated from rectal swabs of diarrheic piglets from farms in Thailand from 2018 to 2019. Escherichia coli isolates were highly resistant to amoxicillin (100%), followed by oxytetracycline (91.9%), enrofloxacin (89.2%), trimethoprim/sulfamethoxazole (86.5%), amoxicillin: clavulanic acid (81.1%), colistin and gentamicin (75.7%), ceftriaxone and ceftiofur (64.9%), ceftazidime (35.1%) and 97.3% showed multidrug-resistance (MDR). There were 8 (21.6%) mcr-1 carriers, 10 (27.0%) mcr-3 carriers and 10 (27.0%) co-occurrent mcr-1 and mcr-3 isolates. The phenotype-genotype correlation of colistin resistance was statistically significant (performed using Cohen’s kappa coefficient (κ = 0.853; p < 0.001)). In addition, PCR results determined that 28 of 37 (75.7%) isolates carried the int1 gene, and 85.7% int1-positive isolates also carried the mcr gene. Genetic profiling of E. coli isolates performed by ERIC-PCR showed diverse genetics, differentiated into thirteen groups with 65% similarity. Knowledge of the molecular origins of multidrug-resistant E. coli should be helpful for when attempting to utilize antibiotics in the pig industry. In terms of public health awareness, the possibility of transmitting antibiotic-resistant E. coli from diarrheic piglets to other bacteria in pigs and humans should be of concern.


Materials and methods
Escherichia coli collection and virulence genes detection. Thirty-seven pathogenic E. coli isolates were previously isolated and identified in routine microbiology service at the Laboratory of Bacteria, Veterinary Diagnostic Center, Faculty of Veterinary Science, Mahidol University. The isolates were obtained from rectal swabs of diarrheic piglets from farms in Thailand during Edema disease outbreak from 2018 to 2019. All isolates were stored at − 80 °C at the Veterinary Diagnostic Center, Faculty of Veterinary Science, Mahidol University, Thailand. Nineteen specific virulence genes of pathogenic E. coli, including 7 toxins genes (lt, sth, stp, stx1A, stx2A, stx2e and astA), 7 adhesin genes (bfpA, eaeA, ipaH, aggR, pCDV432, paa and aidA) and 5 fimbriae genes (F4, F5, F6, F18 and F41) were detected using multiplex PCR. Two groups of primer sets were used to detect and discriminate virulence genes in E. coli isolates as listed in Supplementary Table S1. Multiplex PCR was carried out using a BiometraTOne96G thermocycler (AnalytikJena, Germany). For the group A primer (1A and 2A), amplification was performed with an initial denaturation step at 95 °C for 5 min, followed by 35 cycles of denaturation at 95 °C for 1 min, annealing at 52 °C for 1 min, extension at 72 °C for 1 min, and a final extension at 72 °C for 5 min. In the case of amplification with the group B primer (1B, 2B and 3B), PCR was conducted with an initial denaturation step at 95 °C for 5 min, followed by 30 cycles of denaturation at 95 °C for 30 s, annealing at 61 °C, 57 °C, and 56 °C for fimbriae, toxins, and alternate adhesins, respectively, for 45 s, extension at 72 °C for 1 min, and a final extension step at 72 °C for 7 min. The PCR products were separated using 1.5% agarose gel electrophoresis, stained with 1× GelRed (Sigma Aldrich, USA), and visualized under a UV transilluminator UVP GelStudio (AnalytikJena, USA). www.nature.com/scientificreports/ Antimicrobial susceptibility testing. All isolates were tested for antimicrobial susceptibility by broth microdilution according to the guidelines of the Clinical and Laboratory Standards Institute (CLSI) (VET01S) 17 . Broth microdilution in 96-well microdilution plates was used to determine minimal inhibitory concentrations (MICs). The antibiotic stock solution (256 µg/mL) was diluted by serial two-fold dilutions in Mueller Hinton broth (MHB) and a quality control was composed of media without antibiotic. The following antibiotics were tested: amoxicillin (AMX), amoxicillin: clavulanic acid (AMC), ceftiofur (CEF), ceftazidime (CAZ), ceftriaxone (CRO), colistin (CT), enrofloxacin (ENR), gentamicin (CN), oxytetracycline (OTC), trimethoprim/sulfamethoxazole (SXT). The inoculum was prepared by taking colonies from nutrient agar (NA) by a sterile swab and preparing a McFarland standard. The inoculum was dispensed into the microdilution plate with the serialy diluted antibiotic and incubated at 37 °C for 16-20 h. Escherichia coli ATCC 25922 was used in each assay as a quality control. Results were recorded as the lowest concentration of an antimicrobial that inhibited visible growth of a microorganism.
Enterobacterial repetitive intergenic consensus polymerase chain reaction (ERIC-PCR) for E. coli isolates. ERIC-PCR was performed on a total of 37 E. coli isolates with primers ERIC-1 (5′-ATG TAA GCT CCT GGG GAT TCA C-3′) and ERIC-2 (5′-AAG TAA GTG ACT GGG GTG AGC G-3′) that were described in previous studies 20 . Genomic DNA was extracted from 1 mL of overnight culture using a G-spin™ Genomic DNA Extraction Kit (iNtRON, Korea) and following the manufacturer's instructions. ERIC-PCR was performed in a total volume of 20 μL containing 0.4 μM concentrations of each forward and reverse primer, 20 ng of DNA template and 1× Green PCR master mix kit (Biotechrabbit, Germany). The amplification steps were completed using a BiometraTOne96G thermocycler (AnalytikJena, Germany) with the following thermal cycles: the initial denaturation at 94 °C for 5 min, followed by 35 cycles (denaturation at 94 °C for 1 min, annealing at 52 °C for 1 min, and extension at 72 °C for 5 min), and a final extension step at 72 °C for 10 min. PCR products were separated using 2.0% agarose gel electrophoresis, stained with 1X GelRed (Sigma Aldrich, USA), and visualized under a UV transilluminator UVP GelStudio (AnalytikJena, USA). ERIC-PCR results were analyzed by online data analysis services (insilico.ehu.es). ERIC profiles were compared using the Dice coefficient method, and a dendrogram was made via the unweighted pair-group method using arithmetic averages (UPGMA).

Statistical analysis. Genotype-phenotype correlations of harboring mcr genes and colistin resistance in E.
coli isolates were performed using Cohen's kappa coefficient in SPSS version 23 (IBM Corp. in Armonk, NY). p value of < 0.05 were considered statistically significant.
Ethics approval and consent to participate. The study was carried out in compliance with the ARRIVE guidelines. This research project was approved by the Faculty of Veterinary Science-Animal Care and Use Committee (FVS-ACUC-Protocol No. MUVS-2019-06-31 and MUVS-2021-10-40). All methods were performed in accordance with the relevant guidelines and regulations.

Results
Detection of virulence genes in E. coli. The (Fig. 1). The results showed that 36 of 37 isolates (97.3%) were resistant to at least four different antimicrobial classes, which indicates multidrug-resistance (MDR). About 75.68% showed resistance to β-lactams, fluoroquinolone and aminoglycosides/polymyxin E, and 45.95% of isolates were resistant to all seven antimicrobial classes with different patterns ( Table 2).  www.nature.com/scientificreports/ The majority of E. coli isolates (29 of 37) were from Western Thailand, which has the highest pig farm density in Thailand. Minority isolates were from Central and Eastern Thailand (i.e., 2 and 6 isolates, respectively). The result of antimicrobial susceptibility showed resistance to amoxicillin, amoxicillin: clavulanic acid, colistin, enrofloxacin, oxytetracycline, trimethoprim/sulfamethoxazole, which is common across all three regions from 66.7 to 100% resistance (Fig. 2). Isolated E. coli (2/2) from Central Thailand were resistant to ceftiofur, ceftriaxone and susceptible to gentamicin, ceftazidime. All of the samples from Eastern Thailand were resistant to gentamicin with 83.3% and 100% of samples (6/6) susceptible towards third-generation cephalosporins including ceftiofur, ceftazidime, and ceftriaxone. Escherichia coli isolates from Western Thailand were resistant towards gentamicin (79.3%), ceftiofur (75.9%), ceftriaxone (75.9%) and ceftazidime (44.8%).
Genetic profiling of E. coli isolates. ERIC sequences of 37 isolates were amplified using PCR with ERIC-1 and ERIC-2 primers. Bands for each sample were recorded according to their molecular weights based on a molecular marker (100 bp DNA Ladder). All 37 pathogenic E. coli isolates had bands and were genotyped. The dendrogram from ERIC-PCR banding pattern was analyzed and the isolates were differentiated into thirteen groups with 65% similarity (Fig. 3). ERIC-PCR profile of some isolates showed a difference from others resulting in separate groups with only one isolate in each group (II, V, IX, X and XII). However, groups IV, VII and VIII had a greater number of isolates with 5, 8 and 7 isolates, respectively.

Discussion
Diarrheal disease caused by E. coli is one of the most common diseases in neonatal and weaned piglets. Although there are many approaches to prevent pathogenic E. coli infection in piglets, antibiotics are still commonly used to treat enteric colibacillosis in swine 16 . However, using antibiotics to control infection led to increased selection pressure and resulted in the selection of antibiotic-resistant bacteria. Governments sector worldwide, including Thailand, issued regulations on the use of antibiotics in livestock; however, the risk of antimicrobial resistance gene transmission is still a great global concern. Escherichia coli that carry the antimicrobial resistance gene www.nature.com/scientificreports/ may transfer these to other pathogens and human pathogens, particularly the mcr gene which mediates colistin resistance in animals and humans 9 . Genotype and antimicrobial susceptibility-profiles of E. coli would be helpful for the clinical use of antibiotics. In this study, we investigated whether or not E. coli found in rectal swabs of diarrheic piglets from farms in Thailand were resistant to antimicrobes and if they harbored colistin-resistance mcr genes. Our data showed that most of these strains contain one or more virulence factor genes ( Table 1). The result of antimicrobial susceptibility testing showed that 37 E. coli isolates were resistant to at least three antibiotics with 21 different patterns. The resistance rate to ceftazidime were the lowest with 35.1% and the resistance rate to amoxicillin were the highest with 100%. 97.3% of the isolates were multidrug-resistant (MDR), which means resistant to at least one agent in three or more antimicrobial categories. Among them, the resistance to  www.nature.com/scientificreports/  www.nature.com/scientificreports/ β-lactams and fluoroquinolone antimicrobial were the most frequent. Consequently, a diarrheal disease caused by multidrug-resistant E. coli (MDR-E. coli) can be difficult to treat in pig farms. These results contribute to the overall picture of antimicrobial resistance in E. coli in the pig industry. Studies from many countries showed that MDR-E. coli isolates from pigs were 86.2% in US (between November 2013 and December 2014) 16  In this study, colistin-resistance genes were detected in 28 isolates (75.7%). In these 28 mcr-positive E. coli, the mcr-1 and mcr-3 genes were detected whereas mcr -2, mcr-4, mcr-5, mcr-6, mcr-7, mcr-8, mcr-9 and mcr-10 genes were not detected. Eight (21.6%) were mcr-1 carriers, 10 (27.0%) were mcr-3 carriers and 10 (27.0%) demonstrated co-occurrence of mcr-1 and mcr-3 isolates. The mcr-1 is the first gene reported and is globally distributed whereas others are less distributed 8 . In recent studies, the mcr-3 was also found along with mcr-1 0.43% and 3% respectively in E. coli isolates from pig farms in Thailand and Vietnam 9,26 . This is the first result that showed the mcr-3 gene found with a high rate (54%), which should be an alert for the rapid emergence of mcr-3-mediated colistin resistance in pig farm in Thailand.
On the other hand, one of the 28 colistin-resistant isolates did not carry the mcr gene (tested mcr genes) in this study. On the contrary, one mcr-positive (mcr-1 and mcr-3) isolate was susceptible to colistin, which is in agreement with an earlier report by García et al. 8 in which six of 143 colistin-resistant isolates did not contain mcr genes and three mcr-positive isolates were susceptible to colistin. This inactive form of mcr might be explained by the occurrence of an insertion of a 1.7 Kb IS1294b element into mcr-1 8 . It does not escape our attention that there was one mcr-negative colistin-resistant isolate. There might possibly be an alternative colistin-resistant mechanism that does not involve the mcr gene. Such mechanism would be of interest for further study. Nonetheless, the correlation between genotypes and phenotypes of colistin resistance was considered strongly statistically significant (κ = 0.853; p < 0.001). The high proportion of mcr-positive isolates (75.7%) is a risk for public health. In addition, 75.7% of isolates carried the int1 gene and 24/28 (85.7%) int1-positive isolates carried the mcr gene. It has been reported that the presence of the int1 gene (a mobile genetic element) in pathogens, may increase the ability to transfer resistance genes to other bacteria in the environment 27 . ERIC-PCR profiles of isolated E. coli showed that isolates have diverse genetic structures. A total of 37 isolates were differentiated into thirteen groups with 65% similarity. Beside various gene sources, the rapid and easy genetic modification to adapt in E. coli may also cause genetic diversity. Frequent use of antibiotics in pig farms creates selective pressure leading to genetic modification in E. coli. In addition, each pig farm may have its own treatment approach for the selection of antibiotics in order to control piglet diarrhea. This can also result in the genetic diversity of isolates among pig farms 28 . In addition, Guenther et al. showed that antibiotic-resistant strains can transmit resistance genes to other pathogenic bacteria, especially to human pathogens 29 . It is worth noting that colistin is an antibiotic that is widely used in both humans and animals. Thus, the possibility of spreading the mcr gene from animals to humans is a serious public health concern. Bacteria carrying mcr gene have been found in humans and animals in many countries. Recently, it has been demonstrated that E. coli strains carrying mcr-1 and mcr-3 genes were found not only in pig feces, but also from contaminated pig carcasses and pork 30,31 . This indicates a high risk of spreading bacteria-harboring mcr genes to humans and other environments.

Conclusion
In the present study, multidrug-resistance (MDR) E. coli was found in a high proportion of fecal sample (97.3%) isolated from diarrheic piglets from pig farms in Thailand. From 10 mcr (1-10) genes tested, a large number of isolates harbored colistin-resistance genes mcr-1 and mcr-3. The correlation between the colistin resistance phenotype and genotype among isolates was significant (p < 0.001). Taken together, the results of this study provide informative scientific evidence regarding bacterial resistance to antibiotics in pig farms, and it should also raise public health awareness regarding transmitting resistance gene from animals to humans.

Data availability
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request. www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.