Lack of experimental evidence to support mcr-1-positive Escherichia coli strain selection during oral administration of colistin at recommended and higher dose given by gavage in weaned piglets

Occurrence of mcr-1 and mcr-2 colistin resistance genes in porcine Escherichia coli isolates (2010-2020) and genomic characterization of mcr-2-positive E. coli

Introduction

The global emergence of plasmid-mediated colistin resistance is threatening the efficacy of colistin as one of the last treatment options against multi-drug resistant Gram-negative bacteria. To date, ten mcr-genes (mcr-1 to mcr-10) were reported. While mcr-1 has disseminated globally, the occurrence of mcr-2 was reported scarcely.

Methods and results

We determined the occurrence of mcr-1 and mcr-2 genes among Escherichia coli isolates from swine and performed detailed genomic characterization of mcr-2-positive strains. In the years 2010-2017, 7,614 porcine E. coli isolates were obtained from fecal swine samples in Europe and isolates carrying at least one of the virulence associated genes predicting Shiga toxin producing E. coli (STEC), enterotoxigenic E. coli (ETEC) or enteropathogenic E. coli (EPEC) were stored. 793 (10.4%) of these isolates carried the mcr-1 gene. Of 1,477 additional E. coli isolates obtained from sheep blood agar containing 4 mg/L colistin between 2018 and 2020, 36 (2.4%) isolates were mcr-1-positive. In contrast to mcr-1, the mcr-2 gene occurred at a very low frequency (0.13%) among the overall 9,091 isolates. Most mcr-2-positive isolates originated from Belgium (n = 9), one from Spain and two from Germany. They were obtained from six different farms and revealed multilocus sequence types ST10, ST29, ST93, ST100, ST3057 and ST5786. While the originally described mcr-2.1 was predominant, we also detected a new mcr-2 variant in two isolates from Belgium, which was termed mcr-2.8. MCR-2 isolates were mostly classified as ETEC or ETEC-like, while one isolate from Spain represented an atypical enteropathogenic E. coli (aEPEC; eae+). The ST29-aEPEC isolate carried mcr-2 on the chromosome. Another eight isolates carried their mcr-2 gene on IncX4 plasmids that resembled the pKP37-BE MCR-2 plasmid originally described in Belgium in 2015. Three ST100 E. coli isolates from a single farm in Belgium carried the mcr-2.1 gene on a 47-kb self-transmissible IncP type plasmid of a new IncP-1 clade.

Discussion

This is the first report of mcr-2 genes in E. coli isolates from Germany. The detection of a new mcr-2 allele and a novel plasmid backbone suggests the presence of so far undetected mcr-2 variants and mobilizable vehicles.

Current Insights Regarding the Role of Farm Animals in the Spread of Antimicrobial Resistance from a One Health Perspective

Antimicrobial resistance (AMR) represents a global threat to both human and animal health and has received increasing attention over the years from different stakeholders. Certain AMR bacteria circulate between humans, animals, and the environment, while AMR genes can be found in all ecosystems. The aim of the present review was to provide an overview of antimicrobial use in food-producing animals and to document the current status of the role of farm animals in the spread of AMR to humans. The available body of scientific evidence supported the notion that restricted use of antimicrobials in farm animals was effective in reducing AMR in livestock and, in some cases, in humans. However, most recent studies have reported that livestock have little contribution to the acquisition of AMR bacteria and/or AMR genes by humans. Overall, strategies applied on farms that target the reduction of all antimicrobials are recommended, as these are apparently associated with notable reduction in AMR (avoiding co-resistance between antimicrobials). The interconnection between human and animal health as well as the environment requires the acceleration of the implementation of the ‘One Health’ approach to effectively fight AMR while preserving the effectiveness of antimicrobials.

Impact of colistin and colistin-loaded on alginate nanoparticles on pigs infected with a colistin-resistant enterotoxigenic Escherichia coli strain

Colistin is frequently used for the control of post-weaning diarrhoea in pigs. Colistin resistance caused by plasmidic genes is a public health issue. We evaluated, in experimental animal facilities, whether free colistin or colistin-loaded on alginate nanoparticles (colistin/Alg NPs) could select a colistin-resistant Enterotoxigenic Escherichia coli. The Alg NPs were produced by a simple top-down approach through ball milling of sodium alginate polymer precursor, and colistin loading was achieved through physical adsorption. Colistin loading on Alg NPs was confirmed using various tools such Fourier transform infrared spectroscopy and dynamic light scattering measurements. Thirty-four piglets were orally inoculated or not with a mcr-1-positive, rifampicin-resistant enterotoxigenic E. coli strain, and the inoculated pigs were either treated or not during five days with commercial colistin (100,000 IU/kg) or colistin/Alg NPs (40,415 IU/kg). Clinical signs were recorded. Fecal and post-mortem samples were analyzed by culture. The result clearly indicated that colistin/Alg NPs had a slightly better therapeutic effect. Both treatments led to a transitory decrease of the total E. coli fecal population with a majority of colistin-resistant E. coli isolates during treatment, but the dominant E. coli population was found susceptible at the end of the trial. Further studies are needed to evaluate, in diverse experimental or field conditions, the therapeutic efficacy of colistin/Alg NPs for post-weaning diarrhoea.

Pharmacokinetics of Colistin in the Gastrointestinal Tract of Poultry Following Dosing via Drinking Water and Its Bactericidal Impact on Enteric Escherichia coli

Colistin, a last-line antibiotic of major importance in veterinary medicine and of critical importance in human medicine, is authorized to treat gastrointestinal (enteric) infections caused by non-invasive Escherichia coli in multiple veterinary species including poultry. Its use in veterinary medicine has been implicated in the widespread prevalence of mobilized colistin resistance. The objectives of this study were to determine the intestinal content reached in broiler chickens during 72-h treatment with colistin, to evaluate the associated impact on intestinal E. coli density, and to select less susceptible E. coli populations. In this study, 94 broiler chickens were administered a dose of 75,000 IU/kg/day via drinking water. Intestinal samples were collected pre-, during-, and post-dosing. Luminal intestinal content was assessed for colistin content by ultra-high-performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS), and E. coli were isolated and enumerated on UriSelect agar™. Minimum inhibitory concentration (MIC, for eight isolates per intestine per animal) was determined, and when higher than the epidemiological cutoff (ECOFF 2 mg/l), isolates were screened for mobilized colistin resistance (mcr)-1 to 5. Colistin content increased during treatment to a maximum of 5.09 mg/kg. During this time, the total population of E. coli showed an almost 1,000-fold reduction. An apparent increase in the relative abundance of E. coli with an MIC ≥ ECOFF, either mcr-negative (6.25–10.94%) or mcr-1-positive (4.16–31.25%) was observed, although this susceptibility shift was not maintained post-treatment. Indeed, following cessation of dosing, colistin was eliminated from the intestine, and content was below the limit of quantification (LOQ, 1.1 mg/kg) within 4 h, and the median MIC of E. coli isolates returned below baseline thereafter. Few isolates with a lower susceptibility (mcr-1-positive or negative) were however observed at the end of the study period, indicating maintained sub-populations in the chicken gut. The results of this study show a limited impact on long-term maintenance of less susceptible E. coli populations as a direct result of colistin treatment in individual birds.

Impact of colistin administered before or after inoculation on the transmission of a mcr-1 colistin-resistant Escherichia coli strain between pigs

Colistin resistance associated with plasmidic resistance genes is a serious public health issue. We aimed at studying the transmission of an mcr-1 colistin- and rifampicin-resistant Escherichia coli strain between inoculated pigs and sentinels in different controlled conditions. Three groups of four pigs were bred in separated animal rooms and inoculated on Day 0 (D0). In each inoculated group, six contact pigs were introduced on D2. The first inoculated-and-contact group was left untreated. The ten pigs in the second inoculated-and-contact group received colistin (100 000 IU/kg) before inoculation or contact (D-7 to D-5), simulating prophylactic administration. Pigs in the third inoculated-and-contact group were treated just after inoculation or before transfer (D0 to D2), simulating metaphylactic administration. Faecal samples were regularly collected and segments of intestinal tracts were obtained at necropsy, on D20-D22. Samples were cultured on rifampicin-supplemented media, and PCR was used to detect the mcr-1 gene. The kinetics of infection, based on culture results, were analysed using an SIR model. The inoculated strain was detected in all inoculated and contact pigs. The SIR model showed that one infected pig could transmit the resistant bacteria to one susceptible individual in less than 3 h on average. Prophylactic administration significantly enhanced the transmission rate and resulted in more samples containing the mcr-1 resistance gene at necropsy. No effect of metaphylactic administration could be detected on the transmission rate, nor on the carriage of the resistant strain. Our study confirms that colistin should not be used in a prophylactic manner.