Nasal carriage of CTX-M-55-producing Escherichia coli ST8369 in a healthy cohort in the city of Yangzhou, China

This study aimed to investigate the prevalence and diversity of extended-spectrum β-lactamases (ESBL)-producing Escherichia coli isolates from healthy individuals in a community and to elucidate their dissemination mechanism. Cefotaxime-resistant E. coli were isolated from 95 samples of healthy persons from one community in Yangzhou, China, and were tested for minimal inhibitory concentrations of 14 antimicrobial agents. The isolates were subjected to whole genome sequencing by Illumina Hiseq or PacBio single-molecule real-time sequencing. A total of 30 cefotaxime-resistant E. coli isolates were obtained, carrying bla CTX-M (n=29) or bla DHA (n=1), of which the bla CTX-M-55 (n=19) was the most predominant genotype. One novel bla CTX-M variant bla CTX-M-252 was identified. Thirteen CTX-M-55-producing E. coli isolates belonged to ST8369 from nasal (n=12) or faecal (n=1) samples shared the identical cgMLST type, resistance profiles, resistance genes, plasmid replicons, and a 5,053-bp bla CTX-M-55 structure ΔIS26-ΔISEcp1-bla CTX-M-55-Δorf477-ΔTn2. The bla CTX-M-55 gene was located on IncHI2/ST3 plasmid in E. coli ST8369. The lengths of bla CTX-M/bla DHA-carrying contigs in the remaining 17 E. coli strains ranged from 1,663 to 382,836 bp, located on chromosome (n=4) or plasmids (n=5); the location of the other eight contigs could not be determined due to incomplete assembly. The bla CTX-M was associated with ISEcp1 as previously reported. Nasal colonization of CTX-M-55-producing ST8369 E. coli strains has occurred among healthy individuals in one community. There is a potential risk of antimicrobial resistance dissemination between humans within one community through close contact or environment via aerosols or dust. Therefore, surveillance of nasal carriage of bla CTX-M in communities is warranted to further monitor the spread of the antimicrobial resistance genes in China.


Introduction
Extended-spectrum cephalosporins are widely used in human clinics and veterinary medicine to treat infections caused by multidrug-resistant Gram-negative bacteria; thus, extended-spectrum b-lactamases (ESBL), particularly CTX-M enzymes, have been increasingly reported in human clinical settings and animals worldwide (Bevan et al., 2017). Globally, incidence of CTX-M ESBLs is increasing, bla CTX-M-15 and bla CTX-M-14 are the predominant genotypes detected in many parts of the world (Woerther et al., 2013;Bevan et al., 2017). The global dissemination of bla CTX-M is mainly due to the rapid horizontal transfer mediated by conjugative plasmids; the epidemic plasmids such as IncF, IncI, and IncHI2 facilitate the global spread of bla CTX-M in Enterobacteriaceae from humans, animals and the environment, particularly in Escherichia coli (Bevan et al., 2017;Rozwandowicz et al., 2018;Partridge et al., 2018). Mobile elements such as ISEcp1, IS26, and ISCR1 have also played an essential role in the blaCTX-M transmission (Bevan et al., 2017;Partridge et al., 2018). In addition, some successful E. coli clones, such as the E. coli clone ST131 lineage diffused worldwide, are also responsible for bla CTX-M-15 global dissemination, mostly in human clinics (Bevan et al., 2017).
To date, many studies have focused on CTX-M-producing Enterobacteriaceae from clinical patients. However, the high prevalence of CTX-M-producing E. coli colonizing the intestinal tract of healthy persons in communities is of particular concern, since they could be a major reservoir of bla CTX-M (Woerther et al., 2013;Bevan et al., 2017;Chen et al., 2021). In this study, we aimed to investigate the prevalence and characterization of extended-spectrum b-lactamases (ESBL)-producing E. coli isolates from healthy individuals in a community from Yangzhou, China, to elucidate their dissemination mechanism within this small-scale community.

Materials and methods
Sample collection and antimicrobial susceptibility testing From April 9 th to May 17 th 2021, 58 fecal samples and 37 nasal swabs of 72 healthy volunteers (3 male and 69 female) with no obvious disease symptoms at the age of 15-46 were collected from 37 apartments located in three buildings in one community in Yangzhou, China. Participants had been exposed to antimicrobial agents in the three months prior to sample collection were excluded from this study. This small-scale community with approximately 2000 individuals located in the urban area of Yangzhou, and included three main areas, and one building was randomly selected to sample in each area (Supplementary Figure S1). Individual written informed consent for samples was obtained from all volunteers. Samples were incubated in LB broth (OXOID, Basingstoke, UK) for 18~24 h and then cultured on the MacConkey agar (Haibo, Qingdao, China) with 2 mg/L cefotaxime. One E. coli isolate per plate was selected and identified by 16S rRNA gene sequencing using PCR and Sanger sequencing (Kim et al., 2010). The cefotaxime-resistant E. coli isolates were tested susceptibility to 14 antimicrobial agents including ampicillin, cefotaxime, meropenem, gentamicin, amikacin, streptomycin, tetracycline, chloramphenicol, florfenicol, nalidixic acid, ciprofloxacin, colistin, fosfomycin, and sulfamethoxazole/trimethoprim by using the agar dilution or broth microdilution method (limited to colistin). The results were interpreted according to Clinical Laboratory Standards Institute (CLSI) M100, 30 th edition (CLSI, 2020). E. coli ATCC 25922 was used as the quality control strain.

Whole genome sequencing and analysis
All cefotaxime-resistant E. coli isolates were sequenced by Illumina Hiseq. The library was constructed using NEB NEXT Ultra DNA Library Prep Kit for Illumina (New England Biolabs, USA) and 150 bp paired-end reads were obtained. For each E. coli isolate performed WGS, at least 100-fold coverage of raw reads was collected. The 150 bp pair-end raw reads were trimmed and filtered by the NGSQC toolkit 2.3.3, then were assembled into contigs using SPAdes 3.8.2 (Bankevich et al., 2012). One representative ST8369 E. coli isolate YZ21HCE18 was sequenced using PacBio single-molecule real-time sequencing. The phylogenetic groups of E. coli were confirmed according to previously described protocol by using assembled contigs (Clermont et al., 2000). The genomes were subjected to analysis of multilocus sequence typing (MLST), core genome multilocus sequencing typing (cgMLST), resistance genes, mutations and plasmids by using the Center for Genomic Epidemiology (CGE) pipelines (http://www.genomicepidemiology/org/). The phylogenetic tree of these isolates was constructed using Parsnp (https://harvest.readthedocs.io/en/latest/content/parsnp.html) and visualized by iTOL (Letunic and Bork, 2016). The bla CTX-M/ bla DHA -carrying contigs were retrieved from the draft genomes and analyzed by ISfinder (https://www-is.biotoul.fr/) and BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi). The bla CTX-M -bearing plasmid pYUYZ18-1 in strain YZ21HCE18 was compared with other ST8369 E. coli isolates using BRIG.

Conjugation assay
The transferability of cefotaxime resistance was determined using conjugation experiments as previously described (Chen et al., 2007) and streptomycin-resistant E. coli C600 as the recipient. Transconjugants were selected using 2 mg/L cefotaxime and 3,000 mg/L streptomycin, and were confirmed by detecting bla CTX-M or bla DHA-1 using PCR and sequencing (Chen et al., 2004;Liu et al., 2007).
Thirty ESBL-producing isolates were assigned to 14 known STs and three novel STs (ST12741, ST12743 and ST12744) ( Figure 1). The most prevalent STs among them were ST8369 (n=13). So far, only nine ST8369 E. coli isolates were retrieved from EnteroBase (https://enterobase.warwick.ac.uk/) originating from humans, wild animals, and the environment (Table S1), and none of them carried resistance genes or mutations associated with quinolone resistance. To reveal the genetic differences between 22 E. coli ST8369 isolates, we analyzed their cgMLST profiles (cgSTs) based on 2513 alleles (Zhou et al., 2020). Among the identified seven cgSTs, cgST 75038 (n=14) was the dominant type shared by 13 strains in our study and one isolate PS00212 (Figure 1). Thirteen E. coli ST8369 strains from nasal (n=12) or fecal (n=1) samples of different individuals in this study shared the same cgST, resistance profiles, resistance genes, and plasmid replicons ( Figure 1 and Table 1), indicating that there is a reservoir of this lineage in the community.
The genetic structures of bla CTX-M /bla DHA in 30 ESBL-producing E. coli isolates The lengths of bla CTX-M /bla DHA -carrying contigs ranged from 1,663 to 382,836 bp, located on chromosome (n=4) or plasmids (n=5). Twenty-one contigs were short (1,663 to 5,053 bp) due to incomplete assembly and the high number of insertion elements; they did not have replicon genes or plasmid backbone; thus, it is difficult to determine their location (Table 1; Figure 2).
The bla DHA -carrying contig (18,905 bp) of YZ21HCE2 was identical to the corresponding regions of multiple plasmids such as pM2901 (Shigella sonnei, CP061363), except for the insertion of one Tn3 family transposon within pspC flanked by 5-bp DRs in YZ21HCE2. The core structure sul1-qacED1-ampR-bla DHA-1 -pspDCBAF-qnrB4 is commonly observed in numerous plasmids from various species (e.g., Salmonella, Klebsiella pneumoniae, Citrobacter freundii), highlighting the co-transfer ability of bla DHA-1 and qnrB4.

Discussion
The bla CTX-M gene has been globally disseminated in different sources, with bla CTX-M-14 and bla CTX-M-15 being dominant (Bevan et al., 2017). In this study, bla CTX-M-55 is the most predominant genotype in healthy individuals, which agrees with the increasing prevalence of bla CTX-M-55 in both animals and patients in China (Bevan et al., 2017). As a variant of bla CTX-M-15 , bla CTX-M-55 was first reported in clinical E. coli and K. pneumoniae isolates in Thailand in 2007 (Kiratisin et al., 2007). Recently, bla CTX-M-55 has become the predominant CTX-M genotype in E. coli and Salmonella from food animals, food products, and patients in China (Rao et al., 2014;Zhang et al., 2014;Fu et al., 2020;Huang et al., 2020;Liu et al., 2022;Zeng et al., 2022). One novel bla CTX-M variant bla CTX-M-252 was identified in this study. CTX-M-252 and CTX-M-65 differ by a single amino acid and share a similar genetic structure, suggesting a common lineage. Further investigation of bla CTX-M-252 is needed.
In communities, faecal carriage rates of bla CTX-M are increasing, particularly in developing countries (Woerther et al., 2013;Bevan et al., 2017). However, investigation of bla CTX-M in nasal samples of healthy humans is rare. Previously, one (1/77, 1.30%) CTX-M-producing E. coli isolate was obtained from the nasal sample of a healthy human working at a pig abattoir in Cameroon (Founou et al., 2018). A high nasal carriage rate (32.43%) of bla CTX-M was observed in this study, and nasal colonization of ST8369 E. coli producing CTX-M-55 among healthy persons occurred in one community in Yangzhou, China. It suggests that nasal carriage of bla CTX-M is possibly common in humans. However, the small number of samples and communities is a limitation of this study. The acquisition of bla CTX-M-55 and other resistance genes by ST8369 is mediated by the horizontal transfer of IncHI2 plasmid, followed by clonal dissemination. E. coli ST8369 is rarely described worldwide and may represent an emerging clone in humans, animals, and the environment. Nasal colonization of bla CTX-M-55 -carrying ST8369 E. coli suggests a potential risk of antimicrobial resistance dissemination between humans by the spread of clonal lineages in the small-scale community through close contact or environment via aerosols or dust. Therefore, the clinical importance of nasal carriage of CTX-M-producing E. coli might be underestimated. Although horizontal transfer is the main reason for bla CTX-M dissemination, clonal spread of bla CTX-M -harbouring strains, such as E. coli ST8369 in this study, CTX-M-15-producing E. coli ST949 in water surfaces, E. coli ST2179 encoding CTX-M-65 in retail meat, and bla CTX-M-55 -carrying Salmonella Typhimurium ST34 in patients (Bevan et al., 2017;Falgenhauer et al., 2021;Leão et al., 2021;Zeng et al., 2022) is another important route for bla CTX-M transmission.
Horizontal transfer mediated by plasmids and mobile elements is responsible for the global spread of bla CTX-M (Bevan et al., 2017;Partridge et al., 2018). For example, IncI, IncFII, and IncHI2 plasmids facilitate the horizontal transmission of bla CTX-M in E. coli and Salmonella from various sources Guo and Zhao, 2021;Yang et al., 2014;Nadimpalli et al., 2019;Zhang et al., 2021;Zeng et al., 2021). In this study, various plasmids such as IncHI2, IncK1, IncX1, and IncI1 were associated with bla CTX-M . Although we were not able to determine the location of bla CTX-M in some E. coli isolates in this study due to incomplete assembly, sequence analysis indicates that ISEcp1 plays an important role in bla CTX-M dissemination among E. coli isolates and facilitates the horizontal transfer of bla C T X -M from plasmids to chromosomes in distinct integration sites. The chromosomal integration of bla CTX-M is increasingly reported in E. coli, K. pneumoniae, Salmonella, Proteus mirabilis and some other species of Enterobacteriaceae with the help of mobile elements (Huang et al., 2017;He et al., 2017;Zeng et al., 2022;Yoon et al., 2022). Chromosomal integration of bla CTX-M seems to be an adaptive evolution in response to antimicrobial pressure (Yoon et al., 2022).
In conclusion, we report nasal colonization of CTX-M-55producing E. coli ST8369 associated with IncHI2 plasmid in healthy individuals in one community from Yangzhou, China. Therefore, continued surveillance of nasal carriage of bla CTX-M in communities is warranted.

Accession Numbers
The sequences have been deposited in the GenBank under accession number: PRJNA819533.

Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material.

Ethics statement
The studies involving human participants were reviewed and approved by Yangzhou University. The participants provided their written informed consent to participate in this study.