Plasmid-mediated colistin resistance ( mcr-1 gene ) : three months later , the story unfolds

On 18 November 2015, Liu et al. reported the first description of plasmid-mediated colistin resistance (mcr-1 gene) in food animals, food and humans in China [1]. In this issue, Kluytmans-van den Bergh et al. report on their finding of the mcr-1 gene in Escherichia coli isolates from three (1.5%) of 196 samples of chicken meat collected at Dutch supermarkets, one in 2009 and two in 2014 [2]. This was done by whole genome sequencing of all E. coli isolates and then screening for the presence of the mcr-1 gene by comparing the assembled sequences with sequence data from two databases. The same study did not find any mcr-1-positive isolate among 2,275 extended-spectrum beta-lactamase-positive Escherichia coli (screening and clinical isolates) sampled in humans between 2009 and 2015. The exact origin of the sampled chicken meat was not known, with the two samples from 2014 being labelled ‘non-Dutch, European’. The fact that the genomes of the two isolates from 2014 differed by only three loci and were from the same lot of chicken meat strongly suggest cross-contamination from a common source.

On 18 November 2015, Liu et al. reported the first description of plasmid-mediated colistin resistance (mcr-1 gene) in food animals, food and humans in China [1].In this issue, Kluytmans-van den Bergh et al. report on their finding of the mcr-1 gene in Escherichia coli isolates from three (1.5%) of 196 samples of chicken meat collected at Dutch supermarkets, one in 2009 and two in 2014 [2].This was done by whole genome sequencing of all E. coli isolates and then screening for the presence of the mcr-1 gene by comparing the assembled sequences with sequence data from two databases.The same study did not find any mcr-1-positive isolate among 2,275 extended-spectrum beta-lactamase-positive Escherichia coli (screening and clinical isolates) sampled in humans between 2009 and 2015.The exact origin of the sampled chicken meat was not known, with the two samples from 2014 being labelled 'non-Dutch, European'.The fact that the genomes of the two isolates from 2014 differed by only three loci and were from the same lot of chicken meat strongly suggest cross-contamination from a common source.
This study adds to the already long list of articles on plasmid-mediated colistin resistance published in this and other journals  (Figure and Table).Within just three months of the first description, we learned that the mcr-1 gene (i) had spread to most continents (Figure), (ii) had been found in bacteria isolated from various food animals, from the environment including river water, from various types of meat and vegetables, and from infected patients and asymptomatic human carriers including international travellers, (iii) had been found in various bacterial species, mostly E. coli, and on several different plasmids, and (iv) was highly transferrable with in vitro transfer rates as high as 10 −1 .The fact that we have gained much additional information in such a short time highlights the strength of whole genome sequencing and publicly available sequence databases.
Another important piece of information is that the mcr-1 gene has been present, though not detected, for a long time.Shen et al. reported an mcr-1-positive isolate from chickens in China dating back to the 1980s [21].In Europe, the oldest isolate reported so far is an E. coli from a diarrhoeic veal calf in France in 2005 [10].The earliest reported isolate from humans is a Shigella sonnei from Vietnam in 2008.Trends are available in one study from China and show that the proportion of mcr-1-positive isolates in E. coli from chickens has been increasing sharply since 2009 [21].For most studies, it is impossible to calculate the prevalence of mcr-1-positive isolates because detection of the mcr-1 gene was only performed on colistin-resistant isolates.In France, systematic screening of all isolates from the routine European Union surveillance of antimicrobial resistance in zoonotic commensal bacteria showed that prevalence of the mcr-1 gene ranged from 0.5% in E. coli from pigs to 5.9% in E. coli from turkeys in the period 2013 and 2014 [16].
Plasmid-mediated colistin resistance lies at the interface between animal health and human health.Polymyxins, and in particular colistin, have been used, both in human and veterinary medicine, for more than 50 years, although their parenteral usage in humans has been limited because of concerns about nephrotoxicity and neurotoxicity.In veterinary medicine, colistin is widely used, especially for controlling diarrhoeal diseases in pig and poultry production [31].However, its use varies widely between countries; in Europe, from 0 mg (Finland, Iceland, Norway) to more than 20 mg (Italy, Spain) per kg animal biomass were used in 2013 [32].Data from other parts of the world are more scarce, however Liu et al. reported that the market value for colistin for veterinary usage increased from USD 8.7 billion (EUR 8.0 billion) in 1992 to a projected USD 43 billion (EUR 39.6 billion) in 2018, with China being the largest user of a projected 12,000 tonnes in 2015 [1].The Committee for Medicinal Products for Veterinary  Use (CVMP) of the European Medicines Agency (EMA) reviewed all veterinary medicinal products containing colistin oral use and recommended variations to the terms of their marketing authorisations, for example that the indication is restricted to enteric infections caused by non-invasive E. coli susceptible to colistin and that presence of the disease in the herd should be established before metaphylactic treatment [33].This opinion of the CVMP was converted into a Decision by the European Commission on 16 March 2015 [34], and a similar review is currently being performed for combination products containing colistin.In addition, in view of the recent developments with plasmid-mediated colistin resistance and at the request of the European Commission, the Antimicrobial Advice ad hoc Expert Group of the EMA is currently working on an update of its 2013 advice on the "use of colistin products in animals within the European Union: development of resistance and possible impact on human and animal health" [35].
In human medicine, colistin is increasingly used parenterally for the treatment of patients infected with highly resistant bacteria such as carbapenem-resistant Enterobacteriaceae and Acinetobacter spp.for which other treatment options are limited.In addition it is used topically by inhalation, especially in cystic fibrosis patients, as well as part of the regimen for selective decontamination of the digestive tract and of the oropharynx.As a result, consumption of polymyxins, mainly colistin, in European healthcare increased by 50% between 2010 and 2014, although with wide variation in the consumption rate depending on the country [36].In some European countries, this has resulted in increasing percentages of isolates and outbreaks of Enterobacteriaceae, mainly Klebsiella pneumoniae, that  are resistant to both carbapenems and colistin, the latter because of chromosomal point mutations [37,38].
In 2012, consumption of polymyxins, mainly colistin, was on average more than 600 times higher in food animals than in humans for those 19 Member States in the European Union and European Economic Area that reported complete data both for food animals and for humans and after controlling for biomass (analysis of data from the first joint report by the European Centre for Disease Prevention and Control (ECDC), the European Food Safety Agency (EFSA) and EMA on the integrated analysis of the consumption of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from humans and food-producing animals [39], data not shown).The fact that plasmid-mediated colistin resistance originated from animals combined with the much larger use of colistin in animals than in humans, has contributed to the perception that the problem needs to be tackled first in veterinary medicine.As documented by Kluytmans-van den Bergh et al., mcr-1-positive isolates have so far only been found sporadically in humans in Europe [2].This could be due to absence of selection in a non-favourable environment as indicated by the fact that all travellers that were tested positive for mcr-1 upon return were negative after one month [3].However, the presence of plasmid-mediated colistin resistance in foods and asymptomatic human carriers combined with increasing colistin use in European hospitals may be a game changer.In addition, mcr-1-positive isolates often carry multiple resistance genes, including genes encoding for an extended-spectrum beta-lactamase or a carbapenemase (Table ), and may thus be selected by usage of most antibiotics.Ultimately, if index cases are not detected early and proper control measures are not implemented, Europe may face hospital outbreaks of infections for which there will be little, and possibly no, antibiotic treatment options.
Hospitals must be aware of this new threat to patient safety and may want to consider a few practicable and proportionate preparedness options.Clinical microbiology laboratories should consider testing for colistin susceptibility more frequently, within their available resources, for example in situations involving multidrug-resistant Gram-negative bacteria, isolates from patients that receive or have received colistin, or isolates from patients transferred from or recently hospitalised in a foreign country.It should be noted that disk diffusion is not a reliable test for colistin susceptibility, which should rather be assessed by a method measuring the minimum inhibitory concentration [40].Enhanced infection control precautions, including patient isolation, should be considered already at the suspicion of colistin resistance and not await confirmation from a reference laboratory.Finally, measures aiming at strengthening infection prevention and control (hospital hygiene) as well as a more prudent use of antibiotics are essential to prevent and control antimicrobial resistance in general, and should be considered for plasmid-mediated colistin resistance.
There is no doubt that more information will surface in the coming months.In the meantime, increased awareness and preparedness may prevent spread of mcr-1-positive Enterobacteriaceae in hospitals and other healthcare settings in Europe and elsewhere.

Figure
FigureGeographic distribution of the mcr-1 gene (as of 1st March 2016)A.Food animals

Table a
Characteristics of mcr-1-positive isolates from food animals, the environment, food and humans, 1980s-2015 (as of 1st March 2016) NA: not available; E. coli: Escherichia coli; K. pneumoniae: Klebsiella pneumoniae.a Same as reporting country.

Table b
Characteristics of mcr-1-positive isolates from food animals, the environment, food and humans, 1980s-2015 (as of 1st March 2016)