Colistin: from the shadows to a One Health approach for addressing antimicrobial resistance

Antimicrobial resistance (AMR) poses a serious threat to human, animal and environmental health worldwide. Colistin has regained importance as a last-resort treatment against multi-drug-resistant Gram-negative bacteria. However, colistin resistance has been reported in various Enterobacteriaceae species isolated from several sources. The 2015 discovery of the plasmid-mediated mcr-1 (mobile colistin resistance) gene conferring resistance to colistin was a major concern within the scientific community worldwide. The global spread of this plasmid – as well as the subsequent identification of 10 MCR-family genes and their variants that catalyse the addition of phosphoethanolamine to the phosphate group of lipid A – underscores the urgent need to regulate the use of colistin, particularly in animal production. This review traces the history of colistin resistance and mcr-like gene identification, and examines the impact of policy changes regarding the use of colistin on the prevalence of mcr-1-positive Escherichia coli and colistin-resistant E. coli from a One Health perspective. The withdrawal of colistin as a livestock growth promoter in several countries reduced the prevalence of colistin-resistant bacteria and its resistance determinants (e.g. mcr-1 gene) in farm animals, humans and the environment. This reduction was certainly favoured by the significant fitness cost associated with acquisition and expression of the mcr-1 gene in enterobacterial species. The success of this One Health intervention could be used to accelerate regulation of other important antimicrobials, especially those associated with bacterial resistance mechanisms linked to high fitness cost. The development of global collaborations and the implementation of sustainable solutions like the One Health approach are essential to manage AMR.     

Molecular genetic characteristics of mcr-9-harbouring Salmonella enterica serotype Typhimurium isolated from raw milk

Among the 10 reported mcr genes, mcr-9 was first identified in Salmonella enterica serotype Typhimurium, which is a leading cause of foodborne illness worldwide. However, information about the prevalence and genetic features of mcr-9 is still lacking, especially among food samples. This study reports the presence of mcr-9 in raw milk samples from China; the prevalence rate was low (0.83%, 1/120). mcr-9 was located on a transferable plasmid, and was stable in wild-type S. enterica. However, it had a biological fitness cost when transferred to an Escherichia coli recipient. Whole-genome sequencing revealed that mcr-9 was located on the IncHI2A-type plasmid, and was surrounded by IS903B and IS26 in its flanking regions. The mcr-9-carrying S. enterica 19SE belonged to ST26 and had a multi-drug-resistant phenotype. It was confirmed that mcr-9 did not mediate colistin resistance in this study, indicating that its transfer may not facilitate the dissemination of colistin resistance.     

Inactivation of mgrB gene regulator and resistance to colistin is becoming endemic in carbapenem-resistant Klebsiella pneumoniae in Greece: A nationwide study from 2014 to 2017

IntroductionIn Greece, the spread of carbapenem-resistant Enterobacteriaceae in humans has led to the reintroduction of colistin as a therapeutic agent. Unfortunately, colistin resistance with different mechanisms has emerged. The present work aims to determine the prevalence of carbapenem and colistin resistance and the corresponding mechanisms in Klebsiella pneumoniae clinical isolates from Greece.MethodsFrom 2014 to 2017, 288 carbapenem-resistant K. pneumoniae clinical strains were gathered from a collection of 973 isolates from eight different hospitals in Greece. Antibiotic susceptibility testing was performed using three different methods. Screening of carbapenem and colistin resistance genes was conducted using polymerase chain reaction (PCR) amplification and sequencing.ResultsAmong the 288 (29.6 %) carbapenem-resistant isolates, 213 (73.9%) were colistin-resistant (minimum inhibitory concentration [MIC] >2 mg/L). The KPC type was the most common carbapenemase gene (116; 40.3%), followed by VIM (41; 14.2%), NDM (33; 11.5%) and OXA-48 (22; 7.6%). Moreover, 44 (15.3%) strains co-produced two types of carbapenemases. No mcr genes were detected for colistin resistance but mutations in chromosomal genes were found. These included inactivation of the mgrB gene for 148 (69.5%) strains, including insertion sequences for 94 (44.1%), nonsense mutations for 4 (1.9%) and missense mutations for 24 (11.3%). Moreover, PCR amplification of mgrB gene was negative for 26 (12.2%) strains. Finally, 65 (30.5%) colistin-resistant strains exhibited a wild-type mgrB, the mechanisms of which remain to be elucidated.ConclusionThis study shows that K. pneumoniae clinical strains in Greece are resistant to both carbapenems and colistin and this is endemic and is likely chromosomally encoded.     

Mefloquine enhances the activity of colistin against antibiotic-resistant Enterobacterales in vitro and in an in vivo animal study

Infections caused by carbapenem-resistant Enterobacterales are difficult to treat. Colistin is the last-resort drug for the treatment of these infections, however colistin resistance has emerged in animals and humans. This study investigated the in vitro efficacy of mefloquine in combination with colistin against 114 antibiotic-resistant Enterobacterales isolates including NDM-1, extended-spectrum β-lactamase (ESBL) and mcr-1 containing strains from a broad range of origins. The effect of the mefloquine and colistin combination was examined in vitro by chequerboard method and time–kill analysis and in vivo in a murine peritoneal infection model. The fractional inhibitory concentration index (FICI) of the combination indicated that synergy was detected for all NDM-1 and mcr-1 containing strains, 87.5% of ESBL producing Escherichia coli and 97.9% of ESBL producing Klebsiella pneumoniae strains. Time–kill curves demonstrated significant synergistic activity with low concentrations of colistin that were boosted by mefloquine. The combination showed enhanced activity against infection with NDM-1- or mcr-1 containing Enterobacteriaceae in mice at 4 h and 6 h after treatment. These findings suggest that the combination of mefloquine and colistin has the potential for rejuvenating the activity of colistin against multidrug-resistant Enterobacterales.     

The prevalence of colistin resistance in Escherichia coli and Klebsiella pneumoniae isolated from food animals in China: coexistence of mcr-1 and blaNDM with low fitness cost

Increasing colistin resistance is a global concern because colistin is used as a last resort for the treatment of carbapenem-resistant Enterobacteriaceae infections. The plasmid-mediated colistin resistance gene, mcr-1 was found in distinct bacterial species isolated from humans, animals, and the environment. In this study, farms in four different agricultural provinces in China were investigated to determine the occurrence of the antimicrobial resistance and related genes. A total of 373 Escherichia coli and 54 Klebsiella pneumoniae were isolated from 510 non-duplicated samples. Of the E. coli and K. pneumoniae isolates, 72.7% and 66.7%, respectively, were susceptible to colistin. Isolates resistant to colistin comprised 46.6% of the samples isolated from Shandong, and 17.8% and 16.4% of the samples from Jilin and Henan, respectively. Twenty-six carbapenem-resistant E. coli isolates were resistant to colistin, in which both mcr-1 and bla_NDM were present. Specifically, the co-existence was found in isolates from animals and sewage. Most of the resistance genes were located on plasmids and were 40–244 kilobases. Growth curves of transconjugants carrying mcr-1, bla_NDM-1, bla_NDM-4, bla_NDM-5, and bla_NDM-9 showed a low fitness cost compared with the recipient. In conclusion, mcr-1 was widespread in E. coli and K. pneumoniae isolated from farms in China. Co-existence of mcr-1 and bla_NDM-9 was identified in different sequence types of E. coli with low fitness cost from various origins, indicating an urgent need to take measures for decreasing dissemination.     

Isolation of an IncP-1 plasmid harbouring mcr-1 from a chicken isolate of Citrobacter braakii in China

The plasmid-mediated colistin resistance gene mcr-1 has been found worldwide, but the diversity of organisms harbouring this gene is unknown. In this study, 12 colistin-resistant Citrobacter spp. isolates were obtained from diseased or dead chickens in China, and PCR analysis indicated that five were positive for mcr-1. One Citrobacter braakii strain (SCC4) with a multidrug-resistant phenotype was chosen for further analysis. SCC4 was resistant or intermediate-resistant to ten of the tested antibiotics, and the colistin minimum inhibitory concentration (MIC) was >4?µg/mL. A conjugation assay demonstrated successful transfer of colistin resistance to Escherichia coli strain J53 at a frequency of 10^–7 cells per recipient cell. Whole-genome sequencing revealed that SCC4 contained 13 antibiotic resistance genes in its genome, and the mcr-1 gene resided on a 44-kb self-transmissible IncP-type plasmid of a recently discovered IncP-1 clade. In addition, the mcr-1 gene was part of an insertion element (ISApl1–mcr-1–orf–ISApl1) that was excised from the plasmid as a circular intermediate form. This is the first report of mcr-1-posiitve C. braakii of animal origin and these findings highlight the fact that the mcr-1 gene can be found in normal enteric flora as part of broad-host-range plasmids.     

Silent transmission of an IS1294b-deactivated mcr-1 gene with inducible colistin resistance

Global dissemination of the mobile colistin resistance mcr-1 is of particular concern as colistin is one of the last-resort antibiotics for the treatment of severe infections caused by carbapenem-resistant Gram-negative bacteria. In this study, an inactive form of mcr-1 in a fluoroquinolone-resistant and colistin-susceptible uropathogenic Escherichia coli isolate (ECO3347) was characterised. The mcr-1 gene was deactivated by insertion of a 1.7-kb IS1294b element flanked by two tetramers (GTTC) and located on a 62-kb pHNSHP45-like plasmid (p3347-mcr-1). Single-step and multistep selections were used to induce colistin resistance in vitro in ECO3347. ECO3347 acquired colistin resistance (MIC?=?16–32?mg/L) only after a serial passage selection with increasing concentrations of colistin (2–8?mg/L). Deactivated mcr-1 was re-activated by loss of IS1294b without any remnants in most colistin-resistant mutants. In addition, a novel amino acid variant (Leu105Pro) in the CheY homologous receiver domain of PmrA was detected in one colistin-resistant mutant. Plasmid p3347-mcr-1⁺ carrying the re-activated mcr-1 gene is transferrable to E. coli J53 recipient with a high conjugation rate (ca. 10^–1 cells per recipient cell). Transconjugants showed an identical growth status to J53, suggesting lack of a fitness cost after acquiring p3347-mcr-1⁺. These results highlight that the disrupted mcr-1 gene has the potential for wide silent dissemination with the help of pHNSHP45-like epidemic plasmids. Inducible colistin resistance may likely compromise the success of clinical treatment and infection control. Continuous monitoring of mcr-1 is imperative for understanding and tackling its dissemination in different forms.     

Longitudinal study on antimicrobial consumption and resistance in rabbit farming

Reliable indicators of antimicrobial consumption (AMC) measured with harmonised data and supported by indicators for antimicrobial resistance (AMR) at herd level are necessary to target antimicrobial misuse in food-producing animals. AMC data in 2010–2015 in 32 Italian industrial rabbit holdings weighted with semester production and standardised with animal daily doses (ADDs) were collected. Herd-level AMR against eight antimicrobials was assessed in Escherichia coli, Enterococcus faecalis and Enterococcus hirae collected in 2014–2015. Escherichia coli were assessed for mcr-1 and mcr-2 genes. To produce 1?kg of live rabbit, a mean of 71.8 ADDs was used. Overall AMC reduced over time (P?<?0.05) owing to lowering consumption of tetracyclines (P?<?0.05) and colistin (P?<?0.01), but consumption of quinolones (P?<?0.05), bacitracin (P?<?0.01) and sulfonamides (P?=?0.017) increased. All except one indicator E. coli were wild-type for cefotaxime, whereas 97% displayed reduced susceptibility to tetracyclines, 89% to trimethoprim, 63% to enrofloxacin, 24% to chloramphenicol and 21% to colistin. mcr-1 was detected in 50/320 E. coli isolates from 15/32 holdings; mcr-2 was not detected in 58 isolates with colistin MIC?≥?2?mg/L. All 305 enterococci were wild-type for ampicillin, ciprofloxacin and vancomycin and displayed reduced tetracycline susceptibility. The mean antimicrobial resistance index (ARI) was 0.5 for E. coli and 0.3 for enterococci. ARI was significantly correlated with AMC at herd level for enterococci (P?=?0.008) but not E. coli where high ARI levels were found in a few holdings with low AMC.     

Comprehensive proteomic and metabolomic profiling of mcr-1-mediated colistin resistance in Escherichia coli

Spread of the mcr-1 gene in human and veterinary medicine has jeopardised the use of polymyxins, last-resort antibiotics against life-threatening multidrug-resistant Gram-negative bacteria. As a lipid-modifying gene, whether mcr-1 causes proteomic and metabolomic changes in bacteria and affects the corresponding metabolic pathway is largely unknown. In this study, label-free quantitative proteomics and untargeted metabolomics were used to profile comprehensive proteome and metabolome characteristics of mcr-1-mediated colistin-resistant and -susceptible Escherichia coli in order to gain further insight into the colistin resistance mechanism. Large sets of differentially expressed proteins (DEPs) and metabolites were identified that contributed to mcr-1-mediated antimicrobial resistance, predominantly in different growth conditions with and without colistin. mcr-1 caused downregulated expression of most proteins in order to adapt to drug pressure. Pathway analysis showed that metabolic processes were significantly affected, mainly related to glycerophospholipid metabolism, thiamine metabolism and lipopolysaccharide (LPS) biosynthesis. The substrate phosphoethanolamine (PEA) for mcr-1 to mediate colistin resistance was accumulated in colistin-resistant E. coli. Notably, mcr-1 not only caused PEA modification of the bacterial cell membrane lipid A but also affected the biosynthesis and transport of lipoprotein in colistin resistance by disturbing the expression of efflux pump proteins involved in the cationic antimicrobial peptide (CAMP) resistance pathway. Overall, disturbed glycerophospholipid metabolism and LPS biosynthesis as well as accumulation of the substrate PEA was closely related with mcr-1-mediated colistin resistance. These findings could provide further valuable information to inhibit colistin resistance by blocking this metabolic process.     

Contributions of insertion sequences conferring colistin resistance in Klebsiella pneumoniae

BackgroundIncreasing colistin consumption is leading to expanding colistin resistance in Klebsiella pneumoniae worldwide, but particularly in Asia. Epidemiological studies indicate a link between specific insertion sequences (ISs) and colistin resistance; however, proof of a colistin-IS correlation is lacking.ObjectivesColistin-resistant mechanisms, and in vitro and in vivo efficacies of colistin against K. pneumoniae with ISs were investigated.MethodsColistin-resistant genes, including mcr-1 gene, were detected in 49 colistin- and carbapenem-resistant K. pneumoniae isolates. crrCAB genetic environments were analysed using whole-genome sequencing and polymerase chain reaction (PCR) mapping. Identified ISs were cloned into pRK415 vectors and investigated for potential contributions to colistin resistance. A Caenorhabditis elegans model was employed for in vivo analysis.ResultsmgrB gene alterations (32/49, 65.3%) were identified as the major colistin-resistant mechanism, followed by variations in crrB (57.1%), pmrB (32.7%), phoQ (20.9%), pmrA (16.3%) and phoP (8.2%) genes. Furthermore, 21 of the 49 tested isolates (42.9%) contained the IS elements, ISKpn26, ISEcp1, IS10R, IS903B or ISKpn14 in mgrB or in the surrounding region of crrCAB, indicating an association between these ISs and colistin resistance. The frequencies of colistin resistance significantly increased in colistin-susceptible K. pneumoniae laboratory strains, with plasmids carrying different ISs from clinical strains. In vivo analysis revealed that K. pneumoniae harboring ISKpn26 was associated with decreased lifespan during colistin treatment, leading to an increased risk for colistin treatment failure.ConclusionsThese findings indicate a correlation between diverse ISs and colistin resistance in K. pneumoniae and confirm a role for ISs in colistin treatment.     

Co-occurrence of mcr-1 in the chromosome and on an IncHI2 plasmid: persistence of colistin resistance in Escherichia coli

Two colistin-resistant Escherichia coli strains (FS13Z2S and FS3Z6C) possessing chromosomally encoded mcr-1 isolated from swine were characterised. Whole-genome sequencing revealed that in strain FS13Z2S mcr-1 occurred in triplicate in the chromosome with another copy encoded on a pHNSHP45-2-like IncHI2 plasmid, whereas in strain FS3Z6C only one copy mcr-1 was inserted in the chromosome. It seems likely that the triplication of chromosomal copies of mcr-1 in FS13Z2S is due to intramolecular transposition events via a composite transposon containing an mcr-1 cassette bracketed by two copies of insertion sequence ISApl1, and the pap2 gene at the insertion site was truncated by an IS1294-like element. In plasmid pFS13Z2S and the chromosome of strain FS3Z6C, only a single copy of ISApl1 was present upstream of the mcr-1 cassette. The two strains exhibited similar colistin minimum inhibitory concentrations (MICs) and featured phosphoethanolamine addition to lipid A, without regard to the copy number of mcr-1. The mcr-1-harbouring plasmid was unstable in wild-type strain FS13Z2S and was quickly lost after 7 days of passage on colistin-free Luria–Bertani broth containing 0.5% SDS, but the mcr-1 copies on the chromosome persisted. These results reveal that the single copy of mcr-1 could result in modification of lipopolysaccharide (LPS) and cause colistin resistance in E. coli. Acquisition of multiple copies of mcr-1, especially on the chromosome, would facilitate stable persistence of colistin resistance in the host strain.     

Enhanced bacterial killing with colistin/sulbactam combination against carbapenem-resistant Acinetobacter baumannii

Aims: Polymyxin-based combination therapy is often used to treat carbapenem-resistant Acinetobacter baumannii (A. baumannii) infections. Although sulbactam is intrinsically active against A. baumannii, few studies have investigated colistin/sulbactam combinations against carbapenem-resistant A. baumannii.Methods: Whole genome sequencing was undertaken on eight carbapenem-resistant (colistin-susceptible) isolates of A. baumannii from Chinese patients. Bacterial killing of colistin and sulbactam, alone and in combination, was examined with checkerboard (all isolates) and static and dynamic time-kill studies (three isolates). In the dynamic studies, antibiotics were administered in various clinically-relevant dosing regimens that mimicked patient pharmacokinetics.Results: The eight isolates consisted of ST195, ST191 and ST208 belonging to clonal complex 208, which is the most epidemic clonal type of A. baumannii globally. All isolates possessed Acinetobacter-derived cephalosporinase (ADC-61 or ADC-78) and seven of eight isolates contained the carbapenem-resistance gene bla_OXA-23. The colistin/sulbactam combination was synergistic against two of eight isolates in checkerboard studies. In time-kill studies, rapid bacterial killing of ca. 3–6 log₁₀ CFU/mL was observed with colistin monotherapy, followed by steady regrowth. Sulbactam monotherapy was generally ineffective. Substantially enhanced bacterial killing was observed with colistin/sulbactam combinations in both static and dynamic models, especially with the higher sulbactam concentration (2 g) and/or longer sulbactam infusion time (2 hours) in the dynamic model.Conclusions: This study was the first to use a pharmacokinetics/pharmacodynamics model to investigate synergistic activity of colistin/sulbactam combinations against A. baumannii. It showed that clinically-relevant dosing regimens of colistin combined with sulbactam may substantially improve bacterial killing of multidrug-resistant and carbapenem-resistant A. baumannii.     

Impact of the colistin resistance gene mcr-1 on bacterial fitness

A Klebsiella pneumoniae isolate harbouring a 217 kb IncHI2-type plasmid (pKP2442) encoding the colistin resistance gene mcr-1 was isolated from a leukaemia patient. pKP2442 was mobilised by intragenus and intergenus transconjugation from the clinical isolate to Escherichia coli J53 (transconjugation frequency 6.86?×?10^?8?±?5.57?×?10^?8) and K. pneumoniae PRZ (transconjugation frequency 4.04?×?10^?8?±?3.03?×?10^?8), respectively. Since acquisition of resistance determinants often results in a loss of fitness, the impact of mcr-1 on the fitness of E. coli and K. pneumoniae was investigated. Escherichia coli J53 and K. pneumoniae PRZ transformants harbouring the TOPO expression vector encoding mcr-1 displayed significantly decreased growth rates compared with isogenic parental strains and controls. In contrast, competitive growth experiments revealed equal growth rates between E. coli J53 pKP2442 transconjugants (Tc_pKP2442) and the parental strain, whereas K. pneumoniae PRZ Tc_pKP2442 showed significantly reduced growth rates compared with their parental strain (selection rate constant ?1.62?±?0.49), indicating a decrease in fitness. Infection of A549 human lung epithelial cells with Tc_pKP2442 or mcr-1 transformants and controls revealed equal lactate dehydrogenase activities, indicating no significant impact of mcr-1 on cytotoxicity. Likewise, survival of Galleria mellonella larvae infected with mcr-1-expressing strains and isogenic controls was similar. These data indicate that expression of mcr-1 is able to cause a fitness cost when encoded on expression vectors and that acquisition of natural plasmid-borne mcr-1 does not impair fitness in E. coli J53 but negatively influences growth rates in K. pneumoniae PRZ.     

Within-host heterogeneity and flexibility of mcr-1 transmission in chicken gut

ObjectivesTo characterize the colistin-resistant bacterial population in the gut and assess diversity of mcr-1 transmission within a single individual.MethodsLarge numbers of isolates (>100 colonies/chicken cecum sample) were collected from nine randomly selected mcr-1-positive chickens in China and used for comprehensive microbiological, molecular and comparative genomics analyses.ResultsOf 1273 colonies, 968 were mcr-1 positive (962 Escherichia coli, two Escherichia fergusonii, two Klebsiella pneumoniae and two Klebsiella quasipneumoniae). One to six colistin-resistant species and three to 10 E. coli pulsed-field gel electrophoresis (PFGE) clusters could be identified from each sample. Whole-genome sequencing (WGS) analysis of the representative E. coli strains revealed three to nine sequence types observed in a single chicken host. The mcr-1 genes are located in either chromosomes or plasmids of different types, including IncI2 (n=30), IncHI2 (n=14), IncX4 (n=4), p0111(n=2) and IncHI1(n=1). Strikingly, in single cecum samples, one to five Inc type plasmids harbouring mcr-1 could be identified. Great diversity was also observed for the same IncI2 plasmid within a single chicken host. In addition, up to eight genetic contexts of the mcr-1 gene occurred within a single chicken.ConclusionsThere is extensive heterogeneity and flexibility of mcr-1 transmission in chicken gut due to bacterial species differences, distant clonal relatedness of isolates, many types and variations of mcr-positive plasmids, and the flexible genetic context of the mcr-1 gene. These compelling findings indicate that the gut is a ‘melting pot’ for active horizontal transfer of the mcr-1 gene.     

Expansion of KPC-producing Klebsiella pneumoniae with various mgrB mutations giving rise to colistin resistance: the role of ISL3 on plasmids

mcr-1 has been reported as the first plasmid-encoded gene conferring colistin resistance. In KPC-producing Klebsiella pneumoniae (KPC-KP), however, colistin resistance is rapidly emerging through other mechanisms. Resistance is frequently due to disruption of the mgrB gene by insertion sequences, e.g. ISL3. The aim of this study was to investigate the expansion of mgrB-mutated KPC-KP isolates. In addition, the localisation and targets of ISL3 sequences within the core and accessory genome of common KPC-KP lineages were identified. A total of 29 clinical K. pneumoniae isolates collected from Italian patients were randomly selected. Whole genome sequences were analysed for resistance genes, plasmids and insertion sequences. In addition, 27 colistin-resistant KPC-KP isolates from a previous study from Crete (Greece) were assessed. Clonal expansion of KPC-KP isolates with various mutations in mgrB among all lineages was observed. In two Italian MLST ST512 isolates and eight Greek ST258 isolates, an identical copy of ISL3 was inserted in mgrB nucleotide position 133. ISL3, a transposable restriction–modification system of 8154 nucleotides, was located on pKpQIL-like plasmids and may transpose into the chromosome. In four isolates, chromosomal integration of ISL3 in diverse inner membrane proteins other than mgrB was identified. Colistin resistance is most often explained by clonal expansion of isolates with mutated mgrB. pKpQIL-like plasmids, which are omnipresent in KPC-KP, carry insertion sequences such as ISL3 that have mgrB as a target hotspot for transposition. Transposition of insertion sequences from plasmids and subsequent clonal expansion may contribute to the emerging colistin resistance in KPC-KP.     

KPC-2-producing Klebsiella pneumoniae ST147 in a neonatal unit: Clonal isolates with differences in colistin susceptibility attributed to AcrAB-TolC pump

This study characterizes four KPC-2-producing Klebsiella pneumoniae isolates from neonates belonging to a single sequence type 147 (ST147) in relation to carbapenem resistance and explores probable mechanisms of differential colistin resistance among the clonal cluster.Whole genome sequencing (WGS) revealed that the isolates were nearly 100% identical and harbored resistance genes (bla_KPC-2,_OXA-9,_CTX-M-15,_SHV-11,_OXA-1,_TEM-1B, oqxA, oqxB, qnrB1, fosA, arr-2, sul1, aacA4, aac(6′)Ib-cr, aac(6′)Ib), and several virulence genes. bla_KPC-2 was the only carbapenem-resistant gene found, bracketed between ISKpn7 and ISKpn6 of Tn4401b on a non-conjugative IncFII plasmid. Remarkably, one of the clonal isolates was resistant to colistin, the mechanistic basis of which was not apparent from comparative genomics. The transmissible colistin resistance gene, mcr, was absent. Efflux pump inhibitor, carbonyl cyanide 3-chlorophenylhydrazone (CCCP) rendered a 32-fold decrease in the minimum inhibitory concentration (MIC) of colistin in the resistant isolate only. acrB, tolC, ramA, and soxS genes of the AcrAB-TolC pump system overexpressed exclusively in the colistin-resistant isolate, although the corresponding homologs of the AcrAB-TolC pump, regulators and promoters were mutually identical. No change was observed in the expression of other efflux genes (kpnE/F and kpnG/H) or two-component system (TCS) genes (phoP/phoQ, pmrA/pmrB).Colistin resistance in one of the clonal KPC-2-producing isolates is postulated to be due to overexpression of the AcrAB-TolC pump. This study is probably the first to report clinical clonal K. pneumoniae isolates with differences in colistin susceptibility. The presence of carbapenem-resistant isolates with differential behavior in the expression of a genomically identical pump system indicates the nuances of the resistance mechanisms and the difficulty of treatment thereof.     

Activity of ceftolozane-tazobactam and comparators when tested against Gram-negative isolates collected from paediatric patients in the USA and Europe between 2012 and 2016 as part of a global surveillance programme

Ceftolozane-tazobactam is a combination of an antipseudomonal cephalosporin and a β-lactamase inhibitor. Ceftolozane-tazobactam was approved by the US Food and Drug Administration in 2014 and by the European Medicines Agency in 2015 for use in adults to treat complicated urinary tract infections, acute pyelonephritis, and complicated intra-abdominal infections with metronidazole. Studies for paediatric indications are planned. The Programme to Assess Ceftolozane-Tazobactam Susceptibility monitors the resistance of ceftolozane-tazobactam to Gram-negative isolates worldwide. In total, 6240 Gram-negative isolates were collected between 2012 and 2016 from paediatric patients (<18 years old) in 31 US hospitals (4207 isolates) and 48 European hospitals (2033 isolates), and tested for susceptibility (S) to ceftolozane-tazobactam by broth microdilution. Other antibiotics tested included amikacin, colistin and meropenem. The most common infection type in hospitalized paediatric patients was pneumonia (n=2018), followed by urinary tract infection (n=1569) and bloodstream infection (n=1236). In total, 4316 Enterobacteriaceae and 1765 non-enterics were isolated. The most common species were Escherichia coli (n=1919), Pseudomonas aeruginosa (n=1236) and Klebsiella pneumoniae (n=709). In all regions, the three most active antimicrobials against paediatric Enterobacteriaceae isolates were amikacin (99.0%S), meropenem (98.9%S) and ceftolozane-tazobactam (94.6%S). For all P. aeruginosa, colistin (98.9%S) and ceftolozane-tazobactam (97.4%S) were the most active. In conclusion, for all Enterobacteriaceae, ceftolozane-tazobactam was the most potent cephalosporin tested, with only meropenem and colistin having higher susceptibility rates. For P. aeruginosa, ceftolozane-tazobactam was the most potent β-lactam and had a similar susceptibility rate to colistin.     

Genomic analysis of in vivo acquired resistance to colistin and rifampicin in Acinetobacter baumannii

Acinetobacter baumannii is an opportunistic pathogen in healthcare facilities responsible for nosocomial infections mostly in immunocompromised patients. Colistin resistance is increasingly reported worldwide in A. baumannii. Here we describe the in vivo selection of colistin and rifampicin resistance in carbapenem-resistant A. baumannii. Antimicrobial susceptibility testing, plasmid analysis and whole-genome sequencing (WGS) were performed to fully characterise the resistome of two clinical isolates (AbS1 and AbS2) selected during treatment. Clinical isolate AbS1 remained susceptible to colistin, rifampicin and tigecycline, whilst AbS2 was susceptible only to tigecycline. PCR analysis revealed the presence of a bla_OXA-23-like carbapenemase gene. Kieser extraction revealed an ca. 74?kb plasmid harbouring bla_OXA-23. WGS revealed genomes of 3.8 Mbp in size with a G?+?C content of 38.9%, and both belonged to ST281 according to the Oxford MLST scheme and ST641 according to the Institut Pasteur scheme. The resistome was also composed of naturally occurring β-lactamases, i.e. ADC-25 cephalosporinase and OXA-82 oxacillinase, aminoglycoside resistance genes [aac(3)-Ia, aadA1 and aph(3')-VIa (aphA6)], and mutations in DNA gyrases explaining fluoroquinolone resistance. Single nucleotide polymorphism analysis revealed that both isolates were identical except for a 30-nucleotide duplication within the pmrB gene and a point mutation in the rpoB gene resulting in colistin and rifampicin resistance, respectively. This study highlights the genomic plasticity of A. baumannii under antibiotic pressure. The 10-amino acid duplication in PmrB affects colistin susceptibility by regulating lipopolysaccharide modification through the PmrAB two-component system. These findings provide further information on the molecular mechanisms leading to colistin resistance in A. baumannii.     

Risk factors and outcomes associated with the isolation of polymyxin B and carbapenem-resistant Enterobacteriaceae spp.: A case–control study

Increasing resistance to polymyxin, a last-line antibiotic, is a growing public health concern worldwide. The primary objective of this study was to identify predictors for the isolation of polymyxin-resistant (PR) carbapenem-resistant Enterobacteriaceae (CRE) among hospitalized patients. The secondary objective was to describe the clinical outcomes of patients with PR-CRE infections. A retrospective case–control study including patients admitted to Singapore General Hospital between June 2012 and June 2016 was conducted. Cases were defined as patients who had clinical cultures from which a PR-CRE was isolated. Controls were randomly selected from patients with polymyxin-susceptible (PS) CRE admitted during the same period, and frequency-matched to site of isolation. We included 37 PR cases and 111 PS controls. Polymyxin resistance was detected predominantly in Enterobacter spp. (54.1%) and Klebsiella pneumoniae (43.2%). Multilocus sequence typing showed little clonal relatedness among the isolates. mcr-1 was detected in two PR-CRE isolates. Multivariable analyses showed that PR-CRE isolation was associated with prior polymyxins (adjusted odds ratio (OR), 21.31; 95% confidence interval (CI), 3.04–150.96) and carbapenem exposures (OR 3.74; CI 1.13–12.44), when adjusted for time at risk and bacteria species. In PR-CRE patients with infections, the 30-day all-cause in-hospital mortality was 50.0% as compared to 38.1% in patients with PS-CRE (P?=?0.346). Prior polymyxin and carbapenem exposures were independent risk factors for isolation of PR-CRE. Outcomes of PR-CRE and PS-CRE infections were similar in this study.