A total of 51 isolates of E. albertii were recovered from various samples collected in Zigong city, Sichuan province between 2013 and 2015, including diarrheal patient feces (3), healthy carrier feces (3), duck intestine (19), chicken intestine (18), chicken meat (3), duck meat (2), raw mutton (1), raw pork (1), and egret excrement (1). All isolates were confirmed to be E. albertii based on combination of diagnostic multiplex PCR, 16S rDNA sequencing, and multi-locus sequence typing (MLST) analysis as described previously (Wang et al., 2016) and stored at -80°C in Luria-Bertani (LB) medium (Oxoid, United Kingdom) with 30% (vol/vol) glycerol.

Susceptibility to antimicrobials was determined by Kirby–Bauer disk diffusion method on Mueller Hinton agar (MHA). The antibiotics used in the study included imipenem (IMP), meropenem (MEM), piperacillin (PRL), ampicillin/sulbactam (SAM), amoxicillin/clavulanic acid (AMC), cefepime (FEP), cefuroxime (CXM), cephalothin (KF), ceftriaxone (CRO), aztreonam (ATM), kanamycin (K), streptomycin (S), gentamicin (CN), nalidixic acid (NA), levofloxacin (LEV), norfloxacin (NOR), ciprofloxacin (CIP), trimethoprim/sulfamethoxazole (SXT), tetracycline (TE), furadantin (F), and chloramphenicol (C) (Oxoid, United Kingdom). The inoculated plates were incubated for 24 h aerobically at 37°C. The diameters of the zones of inhibition was interpreted according to the Clinical Laboratory Standards Institute (CLSI) guidelines (CLSI, 2016).

The minimum inhibitory concentration (MIC) of colistin was determined by broth microdilution method recommended by the joint CLSI-EUCAST Polymyxin Breakpoints Working Group¹.

The double disk test was performed to confirm the ESBL phenotype. It was carried out on MHA with two pairs of disks (ceftazidime + ceftazidime/clavulanic acid and cefotaxime + cefotaxime/clavulanic acid) (BD Diagnostics, United States). The results were interpreted as recommended by the CLSI (CLSI, 2016). E. coli ATCC 25922 and Klebsiella pneumoniae ATCC 700603 were used as the quality control strains.

Escherichia albertii isolates were inoculated on LB agar and incubated overnight at 37°C. A colony was suspended in 50 μl of sterilized distilled water and boiled for 10 min. The cell suspension was centrifuged at 10,000 × g for 5 min, and the supernatant was used as template DNA. PCR was performed to screen the main β-lactamase gene groups, i.e., bla_TEM, bla_SHV, bla_CTX-M, bla_KPC, and bla_NDM genes. All primers and PCR conditions used in this study were presented in Supplementary Table S1. Each reaction tube contained 10 μl of master Mix (Qiagen, Germany), 0.5 μM of forward and reverse primers, and 1 μl of template DNA, and was made up to a total volume of 20 μl with sterile distilled water.

All 51 isolates were subjected to PCR for the presence of mcr-1 gene using primers described previously (Supplementary Table S1). The expected PCR products were sequenced using the ABI 3730 Automated DNA Analyzer (Applied Biosystems, United States).

The mcr-1 sequences obtained in this study have been deposited in GenBank under accession numbers: KX765477–KX765480.

Plasmid DNA profiles of all 51 E. albertii isolates were analyzed using the S1-nuclease pulsed-field gel electrophoresis (PFGE) method. Briefly, the bacterial cells embedded in agarose were lysed by SDS/proteinase K, and then were digested with 8 U S1-nuclease at 37°C for 30 min. Finally, each sample was resolved by PFGE in a Chef-Mapper (Bio-Rad, United States) at 14°C, with a switch time 2.16 to 54.17 s at 6 V/cm for 18 h. Each DNA band visualized was considered as a unit length of linear plasmid. The approximate size of each plasmid was determined by comparing profiles with XbaI-digested DNA from Salmonella serotype Braenderup strain H9812 (Bai et al., 2017).

Four of the mcr-1-positive isolates were selected for conjugation experiments. Filter conjugation was carried out using E. coli J53 (sodium azide-resistant) as the recipient. The donor and recipient were grown on LB medium to an optical density at 600 nm of 0.5, mixed equally, and then incubated on sterilized filter paper for 4 h. The filter was then resuspended in LB medium, and dilutions were plated on M-H agar containing sodium azide (150 μg/mL and colistin (4 μg/mL) to select for transconjugants. Mobilization efficiency was calculated as the number of transconjugant colonies divided by the number of donor colonies (Wang et al., 2011).

Genomic DNA was isolated from an overnight culture using the Wizard Genomic DNA purification kit (Promega, United States) according to the manufacturer’s instructions. Total DNA obtained was subjected to quality control by agarose gel electrophoresis and quantified by Qubit (Life Technologies, United States). The complete genome was sequenced by single molecule real-time (SMRT) technology using the Pacific Biosciences (PacBio) sequencing platform performed at the Beijing Novogene Bioinformatics Technology, Co., Ltd. (McCarthy, 2010). The filtered reads were assembled to generate one contig without gaps using SMRT Analysis 2.3.0 (Berlin et al., 2015). The protein-coding sequences (CDSs) were predicted using GeneMarkS (Besemer et al., 2001). ARDB (Antibiotic Resistance Genes Database)² was used to search for antimicrobial resistance genes (Liu and Pop, 2009).

The complete genome sequences of SP140150 isolate are available at GenBank under the accession numbers: CP025676–CP025679.

Samples were collected and detected as part of the infectious disease surveillance program led by National Institute for Communicable Disease Control and Prevention, China CDC and implemented by Zigong Center for Disease Control and Prevention. The study was approved by the ethics committee of National Institute for Communicable Disease Control and Prevention, China CDC, according to the medical research regulations of National Health and Family Planning Commission of the People’s Republic of China.

Antimicrobial resistance in E. albertii isolates was determined against 21 antibiotics. The highest prevalence was tetracycline resistance with a rate of 62.7% (32/51), followed by resistance to nalidixic acid and streptomycin with a rate of 56.9% (29/51) and 51.0% (26/51), respectively. Resistance rate to cefuroxime, piperacillin, and chloramphenicol was 45.1, 43.1, and 41.2%, respectively. Lower resistance was observed for ampicillin/sulbactam, cefepime, cephalothin, ceftriaxone, aztreonam, kanamycin, gentamicin, norfloxacin, ciprofloxacin, trimethoprim/sulfamethoxazole, and furadantin with a rate ranging from 17.6 to 39.2%. All isolates were sensitive or intermediate susceptible to imipenem, meropenem, amoxicillin–clavulanic acid, and levofloxacin (Figure 1).

Ten isolates (19.6%) were resistant to one antimicrobial agent, while the majority exhibited resistance to two or more antimicrobials tested. All isolates from diarrheal patients and healthy carriers were susceptible to imipenem, meropenem, piperacillin, cefepime, aztreonam, and levofloxacin (Supplementary Table S2).

Among 51 E. albertii isolates, 15 isolates from four sources, i.e., six isolates from duck intestine, six from chicken intestine, two from chicken meat and one from raw mutton samples, were ESBL-producing as confirmed by the double disk test. All isolates recovered from diarrheal patients and healthy carriers were non-ESBL-producing (Supplementary Table S2).

All ESBL-producing E. albertii isolates were resistant to piperacillin, cefuroxime, cephalothin, ceftriaxone, aztreonam, tetracycline and chloramphenicol, whereas resistance to streptomycin and nalidixic acid were observed in 14 of the 15 ESBL-producing isolates. ESBL-producing E. albertii isolates have higher resistance rate than non-ESBL-producing isolates, as expected (Table 1).

The main β-lactamase gene groups (bla_TEM, bla_SHV, bla_CTX-M, bla_KPC, and bla_NDM) were screened by PCR. Eight E. albertii isolates did not contained any β-lactamase from gene groups tested and none was positive for bla_KPC and bla_NDM. The bla_TEM, bla_SHV, and bla_CTX-M were detected in 17 (33.3%), 20 (39.2%), and 22 (43.1%) isolates, respectively. All 15 ESBL-producing isolates contained 1–3 β-lactamase genes (Supplementary Table S2).

Four out of 51 isolates (one from raw mutton, one from raw chicken meat, and two from chicken intestine) were positive for MCR-1 encoding gene by PCR. Sequencing analysis showed that the four nucleotide sequences are identical to the first reported mcr-1 sequence in plasmid pHNSHP45 (GenBank accession number KP347127).

The MICs of colistin of all 51 E. albertii isolates were determined by broth microdilution method. The four mcr-1-positive isolates were colistin-resistant with MIC of 8 mg/L. The MICs of colistin of all mcr-1-negative isolates were less than 4 mg/L.

The colistin resistance genes of all four mcr-1-positive isolates were successfully transferred by conjugation into sodium azide resistant E. coli J53. The transfer frequencies of isolates SP140128, SP140089, SP140149, and SP140150 were similar (4.2 × 10^-3, 8.8 × 10^-4, 3.8 × 10^-4, and 1.3 × 10^-3, respectively).

Except one isolate T150248 from healthy carrier, plasmids ranging in size from 36 to 283 kbp were identified by S1-nuclease-based PFGE in 50 E. albertii isolates. Among which, 23 isolates harbored one plasmid; 13 harbored two plasmids; 10 harbored three plasmids; three harbored four plasmids; and one isolate harbored five plasmids (Supplementary Table S2). Two different size plasmids (56 and 113 kbp) were identified in all four colistin-resistant isolates, and an additional 45 kbp plasmid was present in three out of four colistin-resistant isolates (Table 2).

All four isolates harboring mcr-1 were positive for bla_CTX-M group gene (bla_CTX-M-55 type) and were ESBL-producing. The four colistin-resistant isolates showed indistinguishable PFGE patterns and they were all typed as sequence type (ST) 4479 (Supplementary Figure S1). They exhibited multi-drug resistance. All four isolates were resistant to 12 antibiotics tested in this study, i.e., piperacillin, cefepime, cefuroxime, cephalothin, ceftriaxone, aztreonam, kanamycin, streptomycin, nalidixic acid, norfloxacin, tetracycline, and chloramphenicol (Table 2).

The completed genome sequence of SP140150 consists of a circular chromosome of 4,881,553 bp with an average GC content of 49.8% and three circular plasmids. Antimicrobial resistance genes were searched against these three plasmids. bla_CTX-M-55 was identified in pEA-1, a plasmid of 129,356 bp in size with an average GC content of 51.9%. In addition, pEA-1 also carried chloramphenicol resistance gene cml, tetracycline resistance determinant tetA, streptomycin resistance genes aph6id and aph33ib, sulfonamide-resistant dihydropteroate synthase gene sul2, and multidrug efflux RND transporter OqxA and OqxB. None of any antimicrobial resistance genes was identified in pEA-2, the second plasmid of 57,110 bp in size with an average GC content of 46.1%. A 1626 bp CDS encoding MCR-1 was found located downstream of an insertion sequence ISApl1 in the third plasmid pEA-3. pEA-3 is 68,747 bp in size with an average GC content of 42.5%. pEA-3 possesses an IncFII-type backbone and contains 90 predicted CDSs encoding plasmid replication, maintenance and stability functions, and a type IV protein secretion system (Figure 2).