Potential sources and characteristic occurrence of mobile colistin resistance (mcr) gene-harbouring bacteria recovered from the poultry sector: a literature synthesis specific to high-income countries

Asia

The high population density, increasing economies of many nations, high disease burden, and increasing livestock intensification are factors that potentially facilitate the development of AMR in Asia and its dissemination from this region to other parts of the globe (Coyne et al., 2019). Eleven studies investigated PMCR in 4596 isolates from the poultry sector in four HICs (out of 22 HICs/territories) in Asia (Table 1). Ninety-nine isolates (97 E. coli and 2 Salmonella enterica subspecies enterica serovar Typhimurium (S. Typhimurium) and one Aeromonas sobria) were reported to harbour the mcr-1 and mcr-3 gene, respectively.

-Eastern Asia

The poultry sector in Taiwan has been reported as a reservoir for mcr-1-habouring Enterobacteriaceae (Kuo et al., 2016; Chiou et al., 2017; Liu et al., 2018). Thirty-seven isolates (35 E. coli and 2 S. Typhimurium) carrying mcr-1 on ~60 kb IncI2, ~43 kb IncX4, and 60–300 kb IncHI2 plasmids and chromosome (in eight E. coli strains) were detected among 122 E. coli (30.3%) recovered 2012–2016 from chickens and samples of chicken meats (Kuo et al., 2016; Chiou et al., 2017; Liu et al., 2018), indicating that diverse promiscuous plasmids have widely spread mcr-1 among Enterobacteriaceae circulating in Taiwanese poultry meat sector since at least 8 years ago. It further indicates that mcr-1 was transferred vertically to progenies, thus persisting among E. coli clones in Asia. ISApl1 was upstream of mcr-1 in the plasmids of some mcr-1-positive isolates, which were heterogeneous belonging to three STs (Kuo et al., 2016) (Table 1) dominated by high-risk (HiR) zoonotic pandemic extraintestinal pathogenic E. coli (ExPEC) clone ST38 (Manges et al., 2019). Thus, in Taiwan, diverse genetic elements (plasmids and transposons) drive the acquisition/transfer of mcr-1 among commensal and virulent E. coli clones. Some of the mcr-1-positive E. coli isolates (recovered 2013) were producers of extended-spectrum β-lactamases (ESBL) (Kuo et al., 2016) while the mcr-1-positive salmonellae were multidrug-resistant Ampicillinase C (AmpC) producers (Chiou et al., 2017), suggesting that potentially pandrug-resistant Enterobacteriaceae coproducing MCR-1 and ESBL/AmpC has been present Taiwan since at least 7 years ago, thus posing a huge danger to animal and public health. Fortunately, there is now a ban on non-therapeutic COL use in the Taiwanese livestock sector (Liu et al., 2018).

Japanese poultry sector has also been noted as a reservoir for MGHB (Kawanishi et al., 2017; Nishino et al., 2017; Ohsaki et al., 2017). A total of thirty-seven E. coli carrying mcr-1 (mcr-1.21 in one strain) on ~60 kb IncI and 30 kb IncX4 plasmids and one chromosomal mcr-3.25-positive Aeromonas sobria) were detected among 2147 Enterobacterales (1.8%) isolated 2000–2018 from healthy broilers, retail chicken meats, and chicken meats sourced locally and imported from Brazil into Japan (Kawanishi et al., 2017; Nishino et al., 2017; Ohsaki et al., 2017; Odoi et al., 2021). This points to the fact that both local and external poultry meat supply chains are very potential routes through which PMCR might have been disseminated in Japan, albeit at a low prevalence. It also suggested that in Japan, COLROS possibly cross-contaminate poultry meat from the chicken meat handlers and/or fomites during the processing (Figueiredo et al., 2016). The mcr-1-positive isolates were extensively diversified belonging to nine STs (Nishino et al., 2017; Ohsaki et al., 2017) (Table 1), including HiR pandemic ExPEC clone ST117 (Manges et al., 2019), indicating that diverse commensal and virulent E. coli clones are spreading mcr-1 in Japan. Lamentably, the mcr-1-IncI conjugated with a recipient organism (Kawanishi et al., 2017), indicative that the isolates could quickly transfer mcr-1 to other organisms. Although most of the mcr-1-harbouring strains from the imported meats were ESBL-producers exhibiting multidrug resistance (Nishino et al., 2017; Ohsaki et al., 2017), some of them were susceptible to all the 13 antimicrobial agents tested (Nishino et al., 2017). This suggests that organisms coproducing MCR-1 and ESBL were possibly imported into Japan and that mcr gene does not necessarily confer multi-, extensive or pandrug resistance. This highlights the need to conduct antimicrobial susceptibility testing (AST) prior to the prescription/administration of antibiotics. Howbeit, Japan imports food from many countries, which predisposes it to the imported AMR. But encouragingly, risk management for COL in livestock animals, including enhanced monitoring of the antimicrobial-resistant bacteria, restriction of COL to a second choice drug status, and revocation of its designation as a feed additive, is being promoted in Japan (Ling et al., 2020).

Evidence has shown that the South Korean poultry sector constitutes a reservoir for COLROS (Lim et al., 2016; Yoon et al., 2018; Kim et al., 2019; Oh et al., 2020). Twenty-five strains carrying mcr-1 on IncI2 plasmid were detected among 122 E. coli (20.5%) isolated 2000–2018 from healthy/diseased chickens, and samples of chicken meats collected from processing facilities (Lim et al., 2016; Yoon et al., 2018; Kim et al., 2019; Oh et al., 2020), indicating that mcr-1 has been widely spread by IncI in South Korean poultry industry since at least in 2013. ISApl1 flanked mcr-1 upstream in one of the strains (Yoon et al., 2018), meaning that diverse genetic elements (plasmids and transposons) facilitate the acquisition/spread of COL resistance in South Korea. The mcr-1-positive isolates extensively diversified belonging to 12 STs (Lim et al., 2016; Oh et al., 2020), including HiR pandemic ExPEC clones ST410, ST10, and ST88 (Manges, 2016), and there were 20 extra resistance genes (including ESBL, fosfomycin and plasmid-mediated fluoroquinolones resistance (PMQR) genes) belonging to eight antimicrobial families in them (Table 1). This suggests that the transmission of mcr-1 among E. coli strains in South Korea is non-clonal. These strains are spreading resistance against last resort antimicrobial agents in the country. Unfortunately, these organisms could rapidly transfer multi to pandrug resistance to other organisms, having transferred mcr-1 to a recipient organism at a very high frequency of 10⁻² to 10⁻⁶ (Kim et al., 2019). However, there was no other resistance gene in one of the mcr-1-positive isolates (Yoon et al., 2018), implying that mcr-1 is acquired without selective pressure being exerted by non-COL antimicrobial agents.

Since COL has been used for a long time in the South Korean livestock sector with annual consumption of 6–16.3 tons in 2005–2015 (Kim et al., 2019), the selective pressure for the acquisition of mcr genes very likely originated from the livestock sector from where PMCR disseminated into the human-environmental ecosystem (Yoon et al., 2018), possibly by contact with/consumption of livestock/related products in the country. Nevertheless, mutations in chromosomally-encoded pmrB, phoP, phoQ, mgrB, and pmrD have also been detected in isolates from food animals in South Korea (Kim et al., 2019), further supporting that non-mcr mechanism also mediate COL resistance in Eastern Asia.

-Southeastern Asia

The presence of mcr-1.8 in E. coli isolates of poultry origin has been documented in Brunei (GenBank Accession Number: NG_054697). This means that mcr-1 is circulating in the only HIC in South East Asia (SEA) (World Bank, 2018), a region made up of countries that are heavy producers/exporters of poultry and aquaculture products (Coyne et al., 2019).

-Northern Europe

The poultry meat supply chain in northern Europe has been noted as a reservoir of the mcr-1 gene (Hasman et al., 2015; Doumith et al., 2016; Sia et al., 2020). A total of five AmpC-/ESBL-producing strains carrying mcr-1 on IncI2 plasmid (in four strains) were detected among 480 E. coli (1%) isolated from samples of chicken meats sourced locally and imported into Denmark during 2012–2014 (Hasman et al., 2015). There were 17 other resistance genes in eight different antimicrobial families (including ESBL and AmpC genes and chromosomal resistance genes) in the extensively diverse strains belonging to five STs, including HiR pandemic ExPEC clone ST131 (Manges et al., 2019). This indicates that there has been a low circulation of commensal and virulent E. coli strains coproducing MCR-1 and ESBL/AmpC in the Nordic region for at least 8 years ago. The ST131 E. coli is known to have a broad-host-range and is mostly associated with difficult-to-treat urinary tract and bloodstream infections in humans related to their capacity to carry multi-resistance virulence genes without fitness cost (Manges et al., 2019). Thus, its presence in the poultry industry poses a significant danger to public health, especially to handlers and consumers of these meats. Unhappily, MGHB has already disseminated into the human-environmental ecosystem in Denmark, though later than in the poultry sector (Zurfluh et al., 2016).

In the United Kingdom, the mcr-1 carried on IncHI2 plasmid was detected in two Salmonella Paratyphi B var Java PT Colindale isolated from poultry meat imported (in 2014) into England/Wales from Europe (Doumith et al., 2016), suggesting that mcr-1-carrying organisms might have been imported into the UK poultry meat supply chain since at least in about 2014. Recently, mcr-1 was also detected in a Salmonella enterica serovar Java isolated from poultry meat equally sampled in 2014 (Sia et al., 2020), further supporting that mcr-1 has been present in the UK’s poultry meat supply chain since at least 6 years ago. Because COL is used more in patients in the UK than in other European countries (Catry et al., 2015; Doumith et al., 2016) and mcr-1 has been detected earlier in humans in the UK (Public Health England (PHE), 2015), the mcr-1 in poultry could be of anthropogenic origin. Other COL resistance determinants such as mcr-2-and mcr-3 were detected in salmonellae recovered from humans and the environment. It also suggests that the protocol used for treating anthropogenic/agricultural sewages/wastes cannot destroy mcr genes, thereby allowing the genes to escape into the environment. This highlights the need to improve sewage treatment protocols to ensure the complete elimination of ARGs in waste before releasing them into the environment (Anyanwu, Jaja & Nwobi, 2020). There is also a high possibility of farm-to-plate transmission since the COL resistance determinants were also detected in the study’s food samples. This further highlights the importance of cooking food properly and the maintenance of strict hygiene by food vendors. Diverse MGEs, including plasmids, insertions sequences, and transposons, was also found in the strains (Sia et al., 2020), suggesting that these are the drivers of COL resistance in the UK.

-Southern Europe

The presence of MGHB in the poultry sector in Southern Europe has also been documented (Carnevali et al., 2016; Figueiredo et al., 2016; Quesada et al., 2016; Zogg et al., 2016; Manageiro et al., 2017; Alba et al., 2018; Clemente et al., 2019; Apostolakos & Piccirillo, 2018; Koutsianos et al., 2021). In Portugal, 57 (23.6%) mcr-1-harbouring strains (56 E. coli and 1 S. Typhimurium) were detected among 242 Enterobacteriaceae recovered in 2011/2014 from caeca of poultry birds and meats (Figueiredo et al., 2016; Manageiro et al., 2017; Clemente et al., 2019), suggesting a high prevalence of mcr-1 among Enterobacteriaceae in the Portuguese poultry meat sector since the year 2011. It could not be categorically stated that MGHB colonized poultry birds in Europe at that period because the exact geographical origin of the meats from which some of the mcr-1-positive E. coli were isolated in 2011 could not be ascertained. Also, there was a high possibility of cross-contamination during meat processing (Figueiredo et al., 2016). Nevertheless, the Salmonella transferred mcr-1 to a recipient organism at a frequency of ~10⁻⁴, meaning that the strain could rapidly transfer mcr-1 to other microorganisms, thus posing a risk to public health. There were also β-lactam resistance genes (including ESBL gene) in the mcr-1-positive E. coli isolates (Manageiro et al., 2017; Clemente et al., 2019) (Table 2), indicating that E. coli coproducing mcr-1 and ESBL has been present in Portugal since at least around 2014. The β-lactam could co-select for COL resistance. Slaughtered birds are a potential source for contamination of slaughterhouse environment/personnel, and the manure from them used as farm fertilizers could introduce MGHB into the environment. Unfortunately, mcr-1-positive E. coli has also been isolated from humans and botanical ecosystems in Portugal (Anyanwu, Jaja & Nwobi, 2020). In Greece, one MDR mcr-1-positive strain was detected among 150 E. coli (0.7%) isolated from commercial layers with colibacillosis (Koutsianos et al., 2021). This specific finding would suggest that mcr-1 would be circulating probably at low rate in the Greece poultry sector.

In Spain, three faecal E. coli (isolated 2014) carrying mcr-1 on 50–70 kb IncI plasmid were isolated from 1,700 turkeys (Quesada et al., 2016), indicating that IncI has spread mcr-1among E. coli strains colonizing poultry birds in Spain since at least 6 years ago. Conjugation was positive at a frequency of 2.6 × 10⁻¹ to 5.1 × 10⁻², meaning that the organisms could rapidly transfer the mcr-1 to other organisms. COL selective pressure in Spain is likely due to use in food animals as reported consumption of COL was > 100 tons with more than 20 mg/kg of animal biomass of COL used in 2013 (Webb et al., 2017). Contact with livestock, use of insufficiently-treated animal manure in farmlands and/or discharge of these manures into water bodies might have facilitated the spread of PMCR into the human-environmental ecosystems in Spain (Anyanwu, Jaja & Nwobi, 2020). PMCR has been reported in Italy’s poultry sector, the second member of the European Union with the most extensive use of polymyxins in veterinary medicine (Carnevali et al., 2016). Atotal of one hundred and thirty-one strains (115 E. coli and 16 salmonellae) possessing mcr-1.1, mcr-1.2, and a new mcr-1 variant, named mcr-1.13 were detected among 652 MDR Enterobacteriaceae (20.1%) isolated from poultry birds, meats, and eggs (Carnevali et al., 2016; Zogg et al., 2016; Ghodousi, Bonura & Mammina, 2017; Alba et al., 2018; Apostolakos & Piccirillo, 2018; Carfora et al., 2018). This discovery suggests that mcr-1 variants are widely spread among Enterobacteriaceae in the Italian poultry sector. In the E. coli isolates, mcr-1 was on various plasmids, mostly IncX4 (Table 2), IS66 and IS110 were upstream and downstream in one strain, and class 1 and 2 integrons were also in some of them (Alba et al., 2018). These strains were extensively diverse, belonging to 23 STs (dominated by ST10) (Zogg et al., 2016; Alba et al., 2018) (Table 2), including zoonotic HiR virulent pandemic ExPEC clones ST10, ST131, ST38, ST69, ST410, and ST354 (Manges et al., 2019) and haboured 31 extra resistance genes (including ESBL and AmpC genes) belonging to eight antimicrobial families (Zogg et al., 2016; Ghodousi, Bonura & Mammina, 2017; Alba et al., 2018; Apostolakos & Piccirillo, 2018). Some of the mcr-1-harbouring salmonellae were recovered in 2012 (Carnevali et al., 2016), and they contained 11 other resistance genes (including ESBL gene) in six antimicrobial classes and many virulence/fitness-enhancing genes in some others (Carfora et al., 2018; Alba et al., 2018) (Table 2). These findings indicate that virulent Enterobacteriaceae coproducing MCR-1 and ESBL has been present in Italy since at least 8 years ago. Those diverse genetic elements (plasmids, transposons, and integrons) facilitate the acquisition/spread of multi to extensive resistance among Enterobacteriaceae in the Italian poultry industry. Unfortunately, the mcr-1 could be transferred to other organisms from the recipient organism (Apostolakos & Piccirillo, 2018). Regrettably, the dissemination of PMCR into the Italian human-environmental ecosystems has already been documented (Anyanwu, Jaja & Nwobi, 2020). This further highlights the need to reduce all antimicrobial agents at the primary level of livestock production. The reduction will mitigate the effects of complex mechanisms of co-selection and multidrug resistance in “Consumer Protection” and “One Health” perspective (Alba et al., 2018).

Remarkable, some of the mcr-1-positive strains exhibited wild-type COL MIC (<2 µg/mL) (Ghodousi, Bonura & Mammina, 2017; Apostolakos & Piccirillo, 2018), suggesting that the magnitude of PMCR in an ecological niche could be underestimated by testing only isolates within the recommended COL ecological value (ECV) (≥2 µg/mL). This means that the accurate prevalence of mcr gene can solely be determined by screening all isolates from an ecological niche for the mcr gene irrespective of the organisms’ COL MIC (Apostolakos & Piccirillo, 2018). Nevertheless, direct sample testing before isolation in COL resistance surveillance remains the best approach for determining the magnitude of mcr gene in a potential reservoir.

-Eastern Europe

The poultry industry in Eastern Europe has been reported to be a reservoir for MGHB (Karpíšková et al., 2017; Zając et al., 2019; Gelbíčová et al., 2019; Gelbicova et al., 2020, Ćwiek et al., 2021). In Poland, 97 isolates carrying mcr-1.1 on diverse plasmids (IncX4, IncFI, IncHI2, IncQ1, TrfA, IncB/O/K/Z, and many others) were detected among 128 faecal COL-resistant E. coli (62.5%) recovered from poultry birds during 2011–2020 (Zając et al., 2019; Ćwiek et al., 2021), suggesting that diverse promiscuous plasmids evolved mcr-1 in Poland since at least in 2011. There was an increase in mcr-1-carrying strains from 1.1% in 2011 to 11.6% in 2016, suggesting possible increasing use of COL in breeder farms in countries where day-old chicks are imported into Poland (Zając et al., 2019). A total of one of the mcr-1-positive isolates had a mutation in the chromosomal pmrB gene, further indicating that chromosomal and plasmid-mediated mechanisms simultaneously mediate COL resistance in E. coli strains of avian origin. There were 29 additional resistance genes (including ESBL, AmpC, and PMQR genes) in the mcr-1-harbouring isolates, which were extensively diverse belonging to 50 STs (Zając et al., 2019; Ćwiek et al., 2021) (Table 2), including HiR pandemic ExPEC clones ST167, ST38, ST117, ST58, ST69, ST88, ST10 and ST410 (Manges et al., 2019). This diversity proves that various plasmids resulted in a diverse range of clones carrying mcr-1 and genes coding against last-resort antimicrobials in Eastern Europe. Some of the mcr-1-harbouring isolates also exhibited resistance against tigecycline, and some haboured virulence-associated genes, including astA gene encoding heat-stable enterotoxin-1 associated with diarrhoea in humans (Anyanwu, Jaja & Nwobi, 2020). mcr-1-carrying E. coli has been associated with diarrhoea in an individual who possibly contacted livestock in Poland (Izdebski et al., 2016). Since tigecycline is a last-line agent for treating COL-resistant infections, infection by a tigecycline-resistant organism could result in death. Therefore, increased surveillance of PMCR and tigecycline resistance in livestock/animal products is urgently warranted to mitigate an impending global health catastrophe.

In Romania, 17 E. coli strains carrying mcr-1 on IncF, IncX, and IncI plasmids were detected among 92 AmpC-producing Enterobacterales (18.5%) isolated in 2011/2012 from caeca of 92 slaughtered chickens (Maciuca et al., 2019). This illustrates that diverse plasmids have widely spread mcr-1 in MDR organisms colonizing poultry birds in Romania since 2011. The mcr-1 was transferred to a recipient organism, and a composite transposon (Tn6330) and ISApl1 were upstream and downstream of mcr-1 in the isolates. This means that the mcr-1 was acquired horizontally and transferrable to other organisms. Eight additional resistance genes belonging to five antimicrobial families were present in theisolates which were heterogenous belonging to phylogroup B/A/D and four STs (dominated by ST744 and ST57) (Table 2), including pandemic HiR pandemic ExPEC clones ST10 and ST57 (Manges et al., 2019), indicating that Romanian poultry sector constitutes a potential reservoir for highly-virulent E. coli clones thus posing a threat to public health. Fortunately, none of the poultry abattoir workers sampled in Romania harboured organisms carrying mcr-1 or mcr-2 (Maciuca et al., 2019). This is possibly due to hygienic slaughter techniques employed in the sampled slaughterhouses.

In the Czech Republic, 113 strains (110 E. coli and 3 K. pneumoniae) carrying mcr-1 were isolated from poultry meats sourced locally and imported into the country from Poland, Germany, and Brazil (Karpíšková et al., 2017; Gelbíčová et al., 2019; Gelbicova et al., 2020; Kubelova et al., 2021; Zelendova et al., 2021), suggesting that PMCR in Czechia poultry meat supply chain originate locally. Hence meat trade is a route for global dissemination of PMCR. The mcr-1 was on 34–45 kb IncX4, 55–60 kb IncI2, and 250–260 kb IncHI2 plasmids (with IncX4 predominating) in some of the isolates (Gelbíčová et al., 2019; Zelendova et al., 2021), indicating that IncI and IncH plasmids are the commonest drivers of PMCR resistance in the poultry environment in the countries from where the meats originated. There were nine other resistance genes (including ESBL and PMQR) in the strains (Gelbíčová et al., 2019; Zelendova et al., 2021) (Table 3), further supporting that poultry meat remains a potential source for spreading multi-to extensively drug-resistant organisms that would pose severe risk to public health. Conjugation was positive, and the microorganisms were extensively diverse, with the E. coli isolates belonging to 30 STs, including HiR pandemic ExPEC clones ST10, ST58, ST354, ST69, and 410 (Gelbíčová et al., 2019; Manges et al., 2019; Kubelova et al., 2021; Zelendova et al., 2021). In comparison, the K. pneumoniae isolates belonged to four STs (Gelbíčová et al., 2019; Zelendova et al., 2021) (Table 3). This suggests the exchange of mcr-1 between diverse clones of enterobacteria, thus increasing public health risks. It also implies that several enterobacterial lineages are carrying mcr-1 in European Economic Area (EEA). Hence the need for post-slaughter/pre-exportation assessment of meats for the presence of MGHB to minimize the risk posed by contaminated meats to public health. Since meats are not cooked for a long duration by many Europeans as done by people from some other parts of the world, MGHB in meats consumed in Europe can easily colonize the consumers, potentially jeopardizing antimicrobial therapy.

-Western Europe

The poultry production chain in Western Europe has been reported to be a reservoir for MGHB (Allerberger et al., 2016; Ewers et al., 2016; Falgenhauer et al., 2016; Irrgang et al., 2016; Kluytmans–van den Bergh et al., 2016; Veldman et al., 2016; Zogg et al., 2016; Perrin-Guyomard et al., 2016; Borowiak et al., 2017; Donà et al., 2017; Schrauwen et al., 2017; El Garch et al., 2018; Eichhorn et al., 2018; Garcia-Graells et al., 2018; Webb et al., 2017; Hornsey et al., 2019; Borowiak et al., 2020). In the Netherlands, 63 strains (48 E. coli, 13 salmonellae, and two K. pneumoniae) carrying mcr-1 were detected among 278 Enterobacteriaceae (22.7%) isolated 2002–2014 from chickens and samples of poultry meat sourced domestically and imported from other European countries (Veldman et al., 2016; Kluytmans–van den Bergh et al., 2016; Schrauwen et al., 2017; El Garch et al., 2018). The mcr-1 was carried on diverse plasmids, especially 225–290 kb IncHI2 and 20–60 kb IncX4 in some isolates (Table 2), while it was flanked upstream by ISApl1 in others (Veldman et al., 2016). These findings suggest a wide circulation of mcr-1 among Enterobacteriaceae in the Dutch poultry meat supply chain. Diverse genetic elements (plasmids and transposons) have been driving the spread of mcr-1 in the Netherlands since at least more than a decade ago. It further proves that the trade of poultry meat is a route for the dissemination of MGHB across the Eurozone. Eighteen additional resistance genes belonging to seven antimicrobial families were present in some of the mcr-1-positive E. coli isolates (Kluytmans–van den Bergh et al., 2016; Veldman et al., 2016), and the genomes of most of these strains conjugated with that of a recipient organism (Veldman et al., 2016). Moreover, the mcr-1-positive E. coli isolates were extensively diverse belonging to ten STs (Kluytmans–van den Bergh et al., 2016; Veldman et al., 2016), including zoonotic HiR pandemic ExPEC clones ST10 and ST38 (Manges et al., 2019). Thus, diverse commensal and pathogenic E. coli clones in the Dutch poultry industry could easily transfer multi-resistance to other organisms, thereby posing a massive threat to public health. Unfortunately, mcr-1 has disseminated into the human ecosystem in the Netherlands, possibly from the livestock sector or due to the frequent use of COL for selective gut decontamination in intensive care unit and stem cell transplantation patients (Nijhuis et al., 2016; Terveer et al., 2017).

A total of thirty-four of the mcr-1-harbouring isolates (32 E. coli and 2 K. pnuemoniae) were recovered from 53 mcr-1-positive chicken meat samples (Schrauwen et al., 2017), further indicating that by only isolation, the magnitude of COL resistance in food samples could be underestimated. Therefore, direct sample testing followed by isolation is a better approach for surveillance of PMCR in animal origin food. Furthermore, the culture approach showed that the mcr-1-carrying strains were susceptible to cephalosporins, carbapenems, and aminoglycosides. This means that the magnitude of PMCR in an ecological niche could also be underestimated if the presence of mcr gene is assessed in isolates recovered by selection approach using non-COL antibiotics (Schrauwen et al., 2017). It is worthy to note that in NDARO database, sequence of mcr-4.6 and other resistance genes were detected in E. coli isolated 2019 from Dutch poultry, meaning that mcr-4 is present in the Netherlands.

In Belgium, an ST3663 MDR strain (isolated 2015) carrying mcr-1 on IncX4 plasmid and four other resistance genes in four antimicrobial families (Table 2) on other various plasmids was detected among 68 COL-resistant salmonellae (1.5%) recovered from chickens/poultry meats (Garcia-Graells et al., 2018). This means there has been a low circulation of mcr-1-carrying Salmonella in the Belgian poultry meat production chain since at least 5 years ago, posing a risk to public health.

In Switzerland, seven strains carrying mcr-1 were detected among 537 E. coli (1.3%) isolated from chicken meats imported into Switzerland from Germany during 2012–2016 (Donà et al., 2017; Zogg et al., 2016). The low circulation of mcr-1 in the Swiss poultry meat sector is associated with importing poultry meat from Germany. The mcr-1-positive isolates were extensively diverse, belonging to five STs (Donà et al., 2017), including zoonotic HiR pandemic ExPEC clones ST58 and ST38, which dominated (Manges et al., 2019) and16 other resistance genes (including AmpC and ESBL genes) in seven different antimicrobial families were also in the isolates (Donà et al., 2017). This finding suggests that organisms coproducing MCR-1 and ESBL might have been imported into Switzerland at least 8 years ago and further supporting that meat is a potential vehicle for disseminating diverse E. coli clones capable of causing hard-to-treat diseases. ISApl1 was upstream of mcr-1 in 65 kb IncI2 plasmid (in four isolates), and a new 100 kb IncK2 plasmid (in two isolates) (Donà et al., 2017) and chromosome (Donà et al., 2017; Zogg et al., 2016) of the organisms and it was transferred to a recipient organism at a very high frequency of 2.6 × 10⁻⁴ to 6.3 × 10⁻⁶. This means that ISApl1 translocatesmcr-1 into the chromosome of E. coli enabling its maintenance/persistence in the poultry meat supply chain by vertical transmission of the gene to progenies among diverse clones and that these strains could rapidly transfer/acquire mcr-1 to/from other organisms and thus posing a worrisome threat to public health. It further shows that poultry meat is a vehicle for dispersing novel mcr-associated genetic elements. Sadly, mcr-1 has disseminated into Switzerland’s human-environmental ecosystem though at a low level (Zogg et al., 2016). The low incidence could be due to either the non-use of COL in treating community-acquired infection in Switzerland (Liassine et al., 2016; Büchler et al., 2018) or the limitation of the methods used to screen the isolates (Liassine et al., 2016). However, since COL is used in treating sick animals in Switzerland and given the continued importation of poultry meat (into Switzerland) and potential travel to countries with high endemicity, the human carriage of MGHB in the country could rise if unchecked (Büchler et al., 2018). This further highlights the need for post-slaughter assessment of meats for mcr even before exportation/importation, especially in Europe, where there is a tendency to cook meat for a short duration. Nevertheless, it is worthy to note that mcr-1 was not detected in 40 chicken meat samples collected from Switzerland in 2016 (Zogg et al., 2016).

In Austria, three (1.8%) strains of ST10, ST43, and ST616 carrying mcr-1 on diverse plasmids (IncFIC/FII, IncFIB, and IncX4) were isolated from caeca of 164 poultry birds and chicken meats of domestic and Italian origin collected in 2016 from slaughterhouses and retail outlets (Allerberger et al., 2016). Hence, various promiscuous plasmids spread mcr-1 but at a low prevalence in Austria since at least 4 years ago, and that PMCR might also have been imported into Austria through meat trade. It equally suggested that slaughterhouse and retail points are critical points for contamination of meat by COLROS, highlighting the need for regular hazard analysis of critical meat-contamination points to prevent transmission of foodborne superbugs to potential meat consumers. Dismally, however, mcr-1-habouring E. coli have also been isolated from a human patient without a history of travel in Austria (Hartl et al., 2017), suggesting possible acquisition from livestock and community transmission. In 2015, Austria’s livestock sector consumed 1,548 kg of COL, which is 300 times more than the five kg amount used in human medicine (Kirchner et al., 2017). Therefore, the selective pressure for mobilization of mcr in Austria might have originated from the livestock setting. However, it is worthy to note that mcr-1 was not detected in E. coli isolates from chicken meats collected from Austria in 2016 (Zogg et al., 2016).

In France, 26 strains (23 E. coli and 3 salmonellae) carrying mcr-1 and mcr-1-like genes were detected among 275 Enterobacteriaceae (9.4%) isolated from feces of poultry birds, chicken meat, ready-to-cook guinea fowl pie, chicken-farm boot (Perrin-Guyomard et al., 2016; Webb et al., 2017; El Garch et al., 2018), suggesting low circulation of mcr-1 among Enterobacteriaceae in the French poultry sector, that farmer’s wears are potential vehicles for transporting COLROS from livestock farms to other places. The mcr-1 was on IncI, IncX4, and IncP plasmids in the Salmonella isolates (Perrin-Guyomard et al., 2016; Webb et al., 2017; El Garch et al., 2018), meaning that diverse plasmids evolved mcr-1 in France. This highlights the need for adequate biosecurity measures in livestock farms, especially disinfectant foot dips at farm entrances/exits to curtail the spread of MGHB from farm-to-farm and to public places. Lamentably, PMCR has already been disseminated into the French human-environmental ecosystem, possibly through livestock manure and/or livestock-to-human-to-human transmission (Caspar et al., 2017). A gene encoding AmpC was present in some of the mcr-1-harbouring E. coli isolates (Perrin-Guyomard et al., 2016), suggesting that E. coli coproducing MCR-1 and AmpC is present in the French poultry sector. Notably, in the NDARO database, a sequence of mcr-5.1 was detected in S. enterica isolated from boot used in a French chicken farm, suggesting that mcr-5 has been present in France.

The German poultry sector has been reported as a reservoir for MGHB (Ewers et al., 2016; Falgenhauer et al., 2016; Irrgang et al., 2016; Zogg et al., 2016; El Garch et al., 2018; Eichhorn et al., 2018; Hornsey et al., 2019; Borowiak et al., 2020). Reported consumption of COL in food animals in Germany was > 100 tonnes (Webb et al., 2017). Compared to 30 European countries, Germany had the sixth-highest polymyxin sale for food animals in 2016 (Borowiak et al., 2020). Thus, COL selective pressure has been exerted in the German livestock sector. A total of six hundred and eighty strains carrying various mcr genes (mcr-1–585 E. coli and 53 salmonellae, mcr-2–10 salmonellae, mcr-3–10 salmonellae and 1 Aeromonas media, mcr-5–19 salmonellae, mcr-4 and mcr-9–1 Salmonella each) were detected among 1,190 isolates (57.1%) recovered 2008–2018 from poultry birds (live and dead), meats and environment, including meats exported from Germany to Switzerland (Ewers et al., 2016; Falgenhauer et al., 2016; Zogg et al., 2016; Borowiak et al., 2017; Eichhorn et al., 2018; Eichhorn et al., 2018; Hornsey et al., 2019; Borowiak et al., 2020). This further shows a high prevalence of mcr-harbouring organisms in German poultry since at least 12 years ago and that these organisms are causing difficult-to-treat diseases in the birds. The mcr-1 was on > 200 kb IncHI2 and IncX4 plasmids in some of the isolates (Ewers et al., 2016; Hornsey et al., 2019), while in others, it was also in the chromosome (flanked by ISApl1) together with the ESBL gene (bla_CTX-M-15) flanked by ISEcp1 (Falgenhauer et al., 2016). The mcr-3 was a novel gene, named mcr-3.7 on 80 kb plasmid while the mcr-5 (a new gene) associated with transposon Tn3 as well as mcr-1 and mcr-3 were on ColE plasmid in more than half of the mcr-5-positive salmonellae (Borowiak et al., 2017; Eichhorn et al., 2018). This suggests that various genetic elements (transposons and common and uncommon plasmids) have circulated diverse mcr genes in Germany. These genes could persist in the farm through vertical transmission among clones. It also revealed that the German poultry sector constitutes a reservoir for the emergence of novel COL determinants. Two S. Paratyphi B var. Java possessed both mcr-1 and mcr-9 (Borowiak et al., 2020), suggesting that selective pressure by different antimicrobials could facilitate carriage of diverse mcr genes on a plasmid and that the German poultry industry constitutes a source for organisms capable of causing untreatable diseases. There was a chromate and ESBL gene in some of the mcr-5-harbouring salmonellae even in tigecycline-resistant isolates (Borowiak et al., 2017; Borowiak et al., 2020), 13 other resistance genes in five different antimicrobial families were in the mcr-3-positive Aeromonas (Eichhorn et al., 2018) whereas 18 extra resistance genes (including ESBL genes) were present in the mcr-1-positive isolates which were extensively diverse belonging to seven STs (dominated by ST156 and ST1431) (Ewers et al., 2016; Falgenhauer et al., 2016; Zogg et al., 2016; Horseny et al., 2019) (Table 2), including HiR pandemic ExPEC clones ST69, ST131, ST410, ST10 and ST58 (Manges et al., 2019). Thus indicating that multi to pandrug-resistance transmission among Enterobacteriaceae in Western Europe is clonally unrestricted. It also demonstrates that Aeromonas could transfer cocktails of multi-resistance determinants to other organisms. The heavy metal-resistant Salmonella coproducing MCR-5 and ESBL have been present in the German poultry environment, posing a significant public health risk. The chromate gene could code resistance against disinfectants, thereby causing a breach of biosecurity measures in livestock farms. Although the mcr-1 was located on a plasmid in the E. coli strains, conjugation was negative (Hornsey et al., 2019), further supporting that plasmidal location of mcr gene does not necessarily imply transferability. However, mcr-1-carrying E. coli has already disseminated into the human-environmental ecosystems in Germany (Guenther et al., 2017), possibly through contact with and/or consumption of contaminated livestock/livestock-related products. Nonetheless, the genome of mcr-3-and mcr-5-habouring strains conjugated with that of recipient organisms (Borowiak et al., 2017), implying that mcr-3 and mcr-5 could be transferred to other organisms.

Since mcr-positive isolates were from meat samples, it suggests that meat handlers and the meat-processing environment in Germany are potential sources for cross-contamination of meats with multi to pandrug resistant virulent E. coli clones, posing a grave danger to handlers of raw and/or undercooked meats. This calls for urgent attention since Germany is a prominent exporter of poultry meat in the EEA. It supports the need for post-processing assessment of meat for COLROS/mcr genes and adequate meat cooking. Moreover, since Aeromonas is a known reservoir for mcr-3 and also a common inhabitant of aquatic system (from where mcr-3, mcr-4, and mcr-7 originated), avian and human gut (Shen et al., 2020a), the fish meal often used as a source of protein in livestock feed constitute a potential source of mcr-3-harbouring Aeromonas.

Furthermore, 85% of the mcr-positive salmonellae were susceptible to COL (MIC ≤ 2 mg/L); hence they were not considered earlier for PCR screening (Borowiak et al., 2020). This indicates that by using methods targeting a wide range of mcr genes, there is a high likelihood of estimating the actual prevalence of mcr-carrying isolates from an ecological niche. It also indicates that studies in which all the currently known mcr genes (mcr-1 to mcr-10) were not screened possibly underestimated the magnitude of PMCR. This notwithstanding, the salmonellae might also be harbouring yet unknown mcr genes, which further causes the spread of PMCR. Hence, there is a high likelihood that novel mcr genes will continue to emerge. Therefore, revision of the current ECV breakpoint for COL resistance, affordable rapid test kits capable of detecting all the currently known mcr genes and possibly the ones yet to emerge, and increased application of high throughput methods (such as whole-genome sequencing (WGS)) in surveillance of PMCR is much needed.