Traces of pandemic fluoroquinolone-resistant Escherichia coli clone ST131 transmitted from human society to aquatic environments and wildlife in Japan

Transmission of antimicrobial-resistant bacteria among humans, animals, and the environment is a growing concern worldwide. The distribution of an international high-risk fluoroquinolone-resistant Escherichia coli clone, ST131, has been documented in clinical settings. However, the transmission of ST131 from humans to surrounding environments remains poorly elucidated. To comprehend the current situation and identify the source of ST131 in nature, we analyzed the genetic features of ST131 isolates from the aquatic environment (lake/river water) and wildlife (fox, raccoon, raccoon dog, and deer) and compared them with the features of isolates from humans in Japan using accessory and core genome single nucleotide polymorphism (SNP) analyses. We identified ST131 isolates belonging to the same phylotype and genome clusters (four of eight clusters were concomitant) with low SNP distance between the human isolates and those from the aquatic environment and wildlife. These findings warn of ST131 transmission between humans and the surrounding environment in Japan.


Antimicrobial resistance Escherichia coli Fluoroquinolone resistance ST131
A B S T R A C T Transmission of antimicrobial-resistant bacteria among humans, animals, and the environment is a growing concern worldwide.The distribution of an international high-risk fluoroquinolone-resistant Escherichia coli clone, ST131, has been documented in clinical settings.However, the transmission of ST131 from humans to surrounding environments remains poorly elucidated.To comprehend the current situation and identify the source of ST131 in nature, we analyzed the genetic features of ST131 isolates from the aquatic environment (lake/river water) and wildlife (fox, raccoon, raccoon dog, and deer) and compared them with the features of isolates from humans in Japan using accessory and core genome single nucleotide polymorphism (SNP) analyses.We identified ST131 isolates belonging to the same phylotype and genome clusters (four of eight clusters were concomitant) with low SNP distance between the human isolates and those from the aquatic environment and wildlife.These findings warn of ST131 transmission between humans and the surrounding environment in Japan.

Text
Extraintestinal pathogenic Escherichia coli (ExPEC) is the most common pathogenic gram-negative bacteria that causes urinary tract and bloodstream infections in clinical settings.The international high-risk fluoroquinolone-resistant ExPEC clone, sequence type (ST) 131, has caused clinical concerns in the past two decades owing to its pathogenicity, pandemic spread, and frequent co-resistance to third-generation cephalosporins by producing CTX-M-type extended-spectrum β-lactamases (ESBL) [1].Certain ST131 strains reportedly exhibit multidrug resistance including carbapenem resistance and/or resistance to last-line drugs, including colistin and tigecycline [1][2][3].In recent years, ST131 prevalence has also increased in companion animals (dogs and cats) [4,5]; however, it remains lower than that in humans [6].ST131 was isolated among ESBL-producing E. coli from wildlife in Japan [7][8][9].The spread of ST131 among wildlife and subsequently to the environment is a cause of concern [10] because such dissemination makes the eradication of ST131 more challenging by establishing its circulation throughout human society and the surrounding environment.In the current study, we analyzed ST131 isolates from the natural environment (aquatic environment and wildlife) and humans in Japan to estimate the current situation and future possibilities of ST131 dissemination across humans and the surrounding environment.
We identified 16 ST131 isolates from feces of wildlife (5 isolates from Abbreviations: ESBL, extended-spectrum β-lactamases; SNP, single nucleotide polymorphisms; MLST, multilocus sequence typing. raccoon dogs, deer, fox, and raccoon [7][8][9]) and from aquatic environments (11 isolates from river and lake water) between 2016 and 2021 using the isolation method described previously [11] (Table 1).The regions from where the isolates were derived are in neighboring prefectures in Japan, except the Kanagawa prefecture.The ST131 and the clades were identified using multiplex PCR, as described previously [12].The antimicrobial susceptibility was tested using the broth microdilution method, and a minimum inhibitory concentration > 1 mg/L for ciprofloxacin was defined to indicate fluoroquinolone resistance according to the Clinical and Laboratory Standards Institute [13].
To compare the genetic background of environmental isolates, we also analyzed 57 fluoroquinolone-resistant E. coli isolates derived from urine samples of patients in Gifu and Shiga prefectures from 2016 to 2021.These isolates were obtained during clinical diagnoses using 5% Sheep Blood Agar (Becton Dickinson and Company, Franklin Lakes, NJ).Identification of E. coli was performed using MALDI Biotyer (Becton Dickinson and Company).Whole-genome sequencing of DNA library from fluoroquinolone-resistant E. coli isolates prepared using using Illumina DNA prep kit was performed using MiSeq or iSeq (Illumina DNA prep, Illumina, San Diego, CA).Quality check was performed using FastQC, and trimming and de novo assembly were performed using a CLC Genomics Workbench (Qiagen, Hilden, Germany).ST131 and serotypes were identified using multilocus sequence typing (MLST 2.0) and SeroTypeFinder 2.0, respectively, at the Center for Genomic Epidemiology (https://www.genomicepidemiology.org).The assembled genomes were annotated using Prokka 1.14.6 (https://github.com/tseemann/prokka) [14], followed by pan-and accessory genome analysis using Roary (https://github.com/sanger-pathogens/Roary)[15], together with 65 complete E. coli ST131 genomes with complete epidemiological information (origin, year of isolation, and country) available in the NCBI database.The thresholds for the phylotypes and clusters were defined at tree-scale values of 0.01 and 0.005, respectively.Phylogenetic analysis based on the core genome single nucleotide polymorphisms (SNP) was performed using the kSNP4 program [16], and Kmer size 17 and E. coli EC958 (Accession number was HG941718.1)were used as the reference.Pairwise SNP distance matrix was generated using snp-dists 0.8.2 (https://github.com/tseemann/snp-dists)and illustrated as a matrix heatmap using SRplot (https://www.bioinformatics.com.cn/en).
In the current study, except for six ST131 clade C2 isolates, the others were ST131 clade C1, which consists of subclades C1-nM27 and C1-M27, carrying CTX-M-type ESBL genes (bla CTX-M-14 and bla CTX-M-27 , respectively) [17].Clade C1 is a distinct ST131 cluster that has spread throughout Japan since the 2000s [17].Previous studies have reported the isolation of clade C1 from companion animals in Japan [4], as observed in the current study (Fig. S1).The isolation of clade C1 from environment has also been reported in other countries (for example, from water birds, rivers, and lakes in Central Europe and Switzerland) [17].However, the appearance was low, and genetic relationships with human ST131 remain unclear from the study.In addition, the isolation of ST131 clade B that carries the fimH22 allele from poultry and humans has been reported, suggesting the zoonotic potential [10,18], which was not isolated in this study.
Using Roary, we observed a total of 1590 core and 14,533 accessory genes.Using the accessory genome analysis, ST131 isolates were divided into two phylotypes (I and II) and eight clusters (from a to h) (Fig. S1).Phylotype I only comprised Japanese isolates (n = 51).The ST131 isolates derived from river/lake water and wildlife included 11 and 5 isolates belonging to phylotypes I and II, respectively (Table 1).Among them, 11 isolates of phylotype I were serotype O25:H4 and belonged to four clusters, c (n = 4 isolates), e (n = 5), and f and h (n = 1, respectively) (Table 1 and Fig. S1), as determined using accessory genome analysis.This phylotype from river/lake water and wildlife-derived isolates in Japan consisted of three ST131 clades, C1-M27 (clusters c and f), C1-nM27 (cluster e), and C2 (cluster h) (Table 1).The five isolates of phylotype II were identified as another ST131, serotype O16:H5 (Table 1).Therefore, these observations indicate that ST131 spreads in the aquatic environment and wildlife in Japan at a certain frequency.
Collectively, we revealed the concurrence of ST131 isolates from humans, water environments, and wildlife in several accessory-genomebased clusters (four of eight clusters, c, e, f, and h, in phylotype I).This observation suggests the spread and transmission of ST131 between human society and the surrounding nature in Japan, although a direct epidemiological relationship for each sample could not be established.
Core genome SNP of ST131 phylotype I showed clustering of some isolates from humans, river/lake, and wildlife (Fig. S2A).In particular, some ST131 strains exhibited low SNP distance (SNP number ranging from 18 to 62) between the isolates from humans and river and lake water/companion animals (Fig. S2B).The low SNP distance between ST131 isolates from humans and wild birds/environment (number of SNP number ranging from 16 to 75) was reported in Australia around 2019 [10].These observations consistently indicated similar genetic backgrounds between some ST131 (mainly clade C1) derived from humans, wildlife, and river/lake water samples and transmitted among each other in some countries.Accordingly, we suggest that ST131 disseminates from humans to the aquatic environment/wildlife based on the following observations in Japan: i) identical ST131 phylogeny and clade C1 between humans, wildlife, and river/lake water, and previous observations (the current study); ii) higher prevalence of fluoroquinolone-resistant E. coli in humans (approximately 40.4% (154,100/381,447) isolates in patients [19] but 9.6% (35/336 isolates) in wildlife [8]) and clade C1 dominance in humans (>50% of total fluoroquinolone-resistant E. coli; Fig. S1, [6]); iii) the occurrence of ST131 clade C1 in human isolates around 2004 [17] but no reports from the nature prior to the report from humans in Japan; and iv) a recent report documenting the isolation of ST131 in wastewater in other regions of Japan [20].
In the current study, the lake water-isolated ST131 was obtained from Lake Biwa, the largest lake in Japan.Lake Biwa is the source of approximately 450 rivers, including 117 class A rivers, and thus, the flow is the primary source of tap water for 14.5 million people, covering approximately 11.5% of Japan's population, and agricultural and industrial water in Japan.Therefore, these results imply that ST131 overflowed from human society to the aquatic environment/wildlife via the effluent.This dissemination may also apply to another ST131 clade, C2, isolated from humans and deer and belonging to an identical cluster (cluster h in Fig. S1).The limitation of this study is that the observations made were circumstantial and more direct evidence is required to confirm the transmission.
In conclusion, the fluoroquinolone-resistant E. coli ST131 isolated from aquatic environments and wildlife was genetically associated with ST131 clone spreading in clinical settings in Japan.This finding is a warning regarding the overflow and pollution of pandemic ST131 from human society to the surrounding environments, warranting continuous surveillance based on the One Health approach.MLST, multilocus sequence typing: CIP, ciprofloxacin: MIC, minimum inhibitory concetrration.
T. Sato et al.

Table 1
Origin and characterization of environmental ST131 isolates used in the current study.

Table 2
Phylotype of ST131 isolates from human patients from Gifu and Shiga prefectures.