Genomic analysis of Shiga toxin-producing Escherichia coli from patients and asymptomatic food handlers in Japan.

Shiga toxin-producing Escherichia coli (STEC) can cause severe gastrointestinal disease and colonization among food handlers. In Japan, STEC infection is a notifiable disease, and food handlers are required to undergo routine stool examination for STEC. However, the molecular epidemiology of STEC is not entirely known. We investigated the genomic characteristics of STEC from patients and asymptomatic food handlers in Miyagi Prefecture, Japan. Whole-genome sequencing (WGS) was performed on 65 STEC isolates obtained from 38 patients and 27 food handlers by public health surveillance in Miyagi Prefecture between April 2016 and March 2017. Isolates of O157:H7 ST11 and O26:H11 ST21 were predominant (n = 19, 29%, respectively). Non-O157 isolates accounted for 69% (n = 45) of all isolates. Among 48 isolates with serotypes found in the patients (serotype O157:H7 and 5 non-O157 serotypes, O26:H11, O103:H2, O103:H8, O121:H19 and O145:H28), adhesion genes eae, tir, and espB, and type III secretion system genes espA, espJ, nleA, nleB, and nleC were detected in 41 to 47 isolates (85-98%), whereas isolates with other serotypes found only in food handlers were negative for all of these genes. Non-O157 isolates were especially prevalent among patients younger than 5 years old. Shiga-toxin gene stx1a, adhesion gene efa1, secretion system genes espF and cif, and fimbrial gene lpfA were significantly more frequent among non-O157 isolates from patients than among O157 isolates from patients. The most prevalent resistance genes among our STEC isolates were aminoglycoside resistance genes, followed by sulfamethoxazole/trimethoprim resistance genes. WGS revealed that 20 isolates were divided into 9 indistinguishable core genomes (<5 SNPs), demonstrating clonal expansion of these STEC strains in our region, including an O26:H11 strain with stx1a+stx2a. Non-O157 STEC with multiple virulence genes were prevalent among both patients and food handlers in our region of Japan, highlighting the importance of monitoring the genomic characteristics of STEC.

Introduction

Shiga toxin-producing Escherichia coli (STEC) cause various gastrointestinal diseases in humans, including life-threatening hemolytic uremic syndrome (HUS) [1]. Although O157:H7 is the predominant pathogenic serotype, severe infections caused by non-O157 serogroups are increasingly reported worldwide [2]. STEC transmission occurs through intake of contaminated food or via person-to-person spread, with large-scale outbreaks having been reported [1]. In Japan, food safety control measures and a STEC surveillance system were instituted after a massive STEC epidemic occurred in Sakai city in 1996, and STEC infection became a notifiable disease [3]. To prevent the spread of infection via food, the Japanese Ministry of Health, Labor and Welfare requires food handlers to undergo routine stool examination for various infectious pathogens, including STEC, and asymptomatic STEC carriers are legally restricted from working as food handlers [4]. Despite these efforts, approximately 4,000 cases of STEC infection are still reported annually in Japan [3].

Shiga toxin (Stx) is the most important STEC virulence factor. Stx has 2 subtypes with variants, which are Stx1 (stx1a, stx1c, and stx1d) and Stx2 (stx2a, stx2b, stx2c, stx2d, stx2e, stx2f, and stx2g) [5]. In addition, highly pathogenic STEC possess other virulence factors that include adhesins, other toxins, and protein secretion systems [1]. Detection of genes encoding these virulence factors in STEC strains could provide useful information about risk factors that may contribute to human disease. In recent years, there has been a worldwide increase of reports about antimicrobial resistance (AMR) among STEC strains [6]. In STEC carriers taking antibiotics, resistant STEC strains may have a selection advantage over other intestinal bacteria. Because of the public health implications of STEC infection, a comprehensive investigation of virulence and AMR factors is required to assess the potential pathogenicity and antibiotic resistance of STEC isolates from patients and asymptomatic food handlers. Some European authors have investigated the molecular characteristics of STEC isolates [7, 8], but no molecular epidemiological studies have been done to assess the relationship between STEC isolates from patients and food handlers. In the present study, we investigated molecular epidemiology of STEC infection in Miyagi Prefecture, Japan, and performed whole-genome sequencing (WGS) to characterize the genomic features of STEC isolates from patients and asymptomatic food handlers including virulence factors and AMR genes.

Material and methods

Bacterial strains and clinical data

From April 2016 to March 2017, we collected all 65 epidemiologically unlinked STEC isolates detected through public health surveillance for infectious diseases in Miyagi Prefecture, which is located in central northeastern Japan and has a population of about 2.3 million. Thirty-eight isolates were obtained from fecal samples of hospital patients and 27 isolates were detected by routine stool examination of asymptomatic food handlers. Isolation of STEC from stool samples was done with sorbitol-MacConkey agar containing cefixime and tellurite in addition to conventional E. coli isolation agar (e.g., triple sugar iron agar and lysine-indole-motility medium). A latex agglutination test (VTEC-RPLA, Denka Seiken, Japan) and PCR with the EVT-1&2 and EVS-1&2 primers (TaKaRa Biomedicals, Tokyo, Japan) were used to detect Stx and Stx genes, respectively.

Patient data (e.g., age, sex, and clinical manifestation) were also collected through STEC public health surveillance. Patients were divided into four age groups: infants and small children (0–4 y), older children and adolescents (5–19 y), adults (20–64 y), and older people (>65 y) [9]. The age-specific incidence of STEC infections per 100,000 population by age group was calculated using Miyagi Prefecture population data obtained from the National Institute of Population and Social Security Research website (http://www.ipss.go.jp/). This study was approved by the institutional review board of Tohoku University Graduate School of Medicine (IRB no. 2018-1-368).

Whole-genome sequencing

Bacterial DNA from the 65 isolates was extracted as described previously [10], and a DNA library was prepared from each sample with a NEBNext Ultra DNA Library Prep Kit for Illumina (New England Biolabs, Ipswich, MA, USA) according to the manufacturer’s instructions. Then WGS was performed using a MiSeq (Illumina, San Diego, CA, USA) to generate paired-end 300-bp reads, resulting in an average of 5,556,073 read pairs per isolate. All samples showed a minimum average 30-fold coverage. The passing filter ranged from 90.88 to 96.74% (mean, 93.19%), and the average Q30 ranged from 78.30 to 87.21% (mean, 83.50%). All of the sequence data reported here have been deposited in DDBJ/EMBL/Genbank Sequence Read Archive (SRA) under accession numbers DRX149942 to DRX150006.

Genetic analysis

Sequence reads were trimmed of adaptors and filtered to remove reads shorter than 36 bp using Trimmomatic [11], followed by assembly using Platanus assembler v 1.2.4. [12]. Specific genes and alleles were identified with the bioinformatic pipeline of the Center for Genomic Epidemiology (http://www.genomicepidemiology.org), using the default setting of a 90% ID threshold and 60% minimum gene length overlap, except where otherwise stated. Specifically, SerotypeFinder 1.1 [13] was used to identify serogenotypes, MLST Finder 2.0 [14] was employed for multilocus sequence typing (MLST), VirulenceFinder 1.5 [15] was used for virulence genes, and ResFinder 3.0 [16] was employed for acquired AMR genes. We also searched for the major adhesin gene saa, which cannot be detected by VirulenceFinder, using BLAST (http://blast.ncbi.nlm.nih.gov).

A non-recombinogenic core genome single-nucleotide polymorphism (SNP)-based phylogeny was generated with Parsnp v 1.2 (https://harvest.readthedocs.io/en/latest/content/parsnp/quickstart.html#advanced-usage) [17] using the 65 STEC isolates and the following 6 reference genomes: STEC O157:H7 Sakai (GenBank accession numbers: BA000007), STEC O157:H7 strain EDL933 (AE005174), STEC O157:H7 strain EC4115 (CP001164), STEC O157:H7 strain TW14359 (CP001368), STEC O26:H11 strain 11368 (AP010953), and STEC O103:H2 strain 12009 (AP010958). Differences in the number of SNPs between STEC strains were calculated by Parsnp. Clonal STEC strains were defined as isolates with less than 5 SNPs [18]. A phylogenetic tree with 1000 boostrap replicates was constructed by using the randomized accelerated maximum likelihood (RAxML) program v 8.2.12 (https://cme.h-its.org/exelixis/web/software/raxml/index.html) [19], and it was visualized with FigTree (http://tree.bio.ed.ac.uk/software/figtree/).

Statistical analysis

Fisher’s exact test and the two-sample t-test were used for analysis of categorical variables and continuous variables, respectively. In all analyses, P<0.05 was considered statistically significant.

Results

Serogenotyping and MLST

Among the 65 STEC isolates, serogenotyping and MLST revealed 18 different serogroups and 20 different sequence types (Fig 1). Seven sequence types had already been reported as STECs causing human disease according to Enterobase (http://enterobase.warwick.ac.uk), while the other 13 sequence types, including O103:H8 ST2836, are new as STEC strains. An O-group was not detected in the one OUT (O-serotype untypable) isolate by SerotypeFinder. In this isolate, no O-processing genes (wzx, wzy, wzm, wzt) were detected by BLAST. It is possible that this isolate could be assigned to serogroups O14 or O57 since O-processing genes for these serogroups have not been found in their genomes [20], or it could represent a new serogroup.

Fig 1

Frequency of O/H serotypes and sequence types (STs) among Shiga-toxin producing Escherichia coli (STEC) isolates.

Red and blue bars represent the number of isolates from patients and asymptomatic food handlers, respectively. *New STEC STs. aMajor serotypes: serotypes found in patients in this study. bMinor serotypes: serotypes only found in food handlers in this study. cOUT: O-serotype untypable.

Non-O157 isolates accounted for 69% (n = 45) of all isolates. O157 isolates were significantly more frequent among the patients than the food handlers (19/38 isolates from patients versus 1/27 isolates from food handlers, P<0.001), whereas non-O157 isolates were significantly more frequent among the food handlers. Among isolates from the patients, the predominant serotype was O157:H7 (n = 19, 50%) followed by O26:H11 (n = 14, 37%), while O26:H11 was predominant (n = 6, 22%) among isolates from the food handlers, followed by O103:H2, O181:H49, O55:H12, O76:H19, O91:H14, and O8:H19 (n = 2, 7.4%, respectively). Isolates from patients belonged to the following 6 serotypes (O157:H7 and 5 non-O157 serotypes, which were O26:H11, O103:H2, O103:H8, O121:H19, and O145:H28) (hereinafter called “major serotypes”), while the remaining 12 serotypes were only found in the food handlers (hereinafter called “minor -serotypes”) (Fig 1). Isolates with the same serotype generally belonged to the same sequence type, except that O157:H7 isolates belonged to ST11 and ST2966, O26:H11 isolates belonged to ST21 and ST1705, and O8:H19 isolates belonged to ST88 and ST2385. There was only one SNP difference between ST11 and ST2966 in purA, as well as between ST21 and ST1705 in gyrA, suggesting a close relation between these sequence types.

Clinical characteristics of patients and food handlers

The majority of the patients (28/38, 74%) had bloody diarrhea and 1 patient (2.6%) developed HUS, while the symptoms of the remaining 10 patients (36%) were unknown. The patients were aged between 11 months and 85 years (median: 17.5 years), whereas the food handlers were aged from 21 to 71 years (median: 54 years). Sixty-one percent (15/23) of the patients and 74% (20/27) of the food handlers were female.

The annual age-specific incidence of O157 and non-O157 infections in Miyagi Prefecture during the study period is summarized in Table 1. In this region, the overall incidence of STEC infection was 1.6, with infants and small children having the highest incidence (13.5, P<0.001 vs. each other age group). Importantly, the incidence of non-O157 infections was significantly higher in infants and small children (11.3) than in the other age groups (P<0.001 vs. each other age group), while there was no significant difference in the incidence of O157 infections among the age groups.

Table 1

Annual age-specific incidence of Shiga toxin-producing Escherichia coli (STEC) infection in Miyagi Prefecture during the study period.

Age group	Population (n)	No. of cases			Incidence per 100,000 population
		All STEC infections	O157	non-O157	All STEC infections	O157	non-O157
Infants and small children	88,787	12	2	10	13.5	2.3	11.3
Older children and adolescents	311,185	8	3	5	2.6	1.0	1.6
Adults	1,296,353	11	8	3	0.9	0.6	0.2
Older people	588,240	7	6	1	1.2	1.0	0.2
Total	2,333,899	38	19	19	1.6	0.8	0.8

Virulence genes

Among a total of 76 virulence genes registered in VirulenceFinder, 44 genes (58%) were detected among the isolates and there was a median of 17 virulence genes per isolate (range: 1–26). Isolates from the patients harbored significantly more virulence genes than isolates from the food handlers (a median of 18 and 10 virulence genes per isolate, respectively, P<0.001), and O157 isolates had significantly more virulence genes than non-O157 isolates (a median of 19 and 16 virulence genes per isolate, respectively, P = 0.013). Eight different Stx subtypes (combinations) were detected among the isolates, with stx1a-only being most frequent (n = 27, 42%), followed by stx1a+stx2a (n = 12, 18%). The distribution of virulence genes among isolates from the patients or food handlers and among O157 or non-O157 isolates is shown in S1 Fig and S1 Table.

Isolates with major serotypes had significantly more virulence genes than isolates with minor serotypes (a median of 18 and 7 virulence genes per isolate, respectively, P<0.001). The Stx subtype stx1a was significantly more frequent among isolates with major serotypes, while stx2d and stx2e were detected significantly more often in isolates with minor serotypes. Among the 48 isolates with major serotypes, adhesion genes eae, tir, and espB were detected in 47 (98%), 46 (96%), and 45 (94%) isolates respectively, and secretion system genes espA, espJ, nleA, nleB, and nleC were detected in 46 (96%), 46 (96%), 41 (85%), 45 (94%), and 41 (85%) isolates respectively, whereas all isolates with minor serotypes were negative for all of these genes (all P<0.001) (Table 2).

Table 2

Distribution of putative virulence genes among Shiga toxin-producing Escherichia coli (STEC) isolates with major/minor serotypes and O157/non-O157 isolates from patients.

		No. of isolates (%)
		Majorserotype (n = 48)	Minor serotype (n = 17)	P	O157 from patients (n = 19)	Non-O157 from patients (n = 19)	P
Shiga-toxin pattern	stx1a	24 (50)	3 (18)	0.024	0	15 (79)	<0.001
	stx1c	0	2 (12)		0	0
	stx2a	3 (6)	2 (12)		1 (5)	2 (11)
	stx2c	8 (17)	1 (6)		8 (42)	0	0.003
	stx2d	0	3 (18)	0.016	0	0
	stx2e	0	4 (24)	0.015	0	0
	stx1a+stx2a	11 (23)	1 (6)		9 (47)	2 (11)	0.029
	stx1a+stx2c	1 (2)	0		0	0
Adhesins	eae	47 (98)	0	<0.001	19 (100)	19 (100)
	tir	46 (96)	0	<0.001	18 (95)	19 (100)
	espB	45 (94)	0	<0.001	19 (100)	18 (95)
	iha	32 (67)	8 (47)		18 (95)	5 (26)	<0.001
	efa1	21 (44)	0	<0.001	0	14 (74)	<0.001
Toxins	ehxA	44 (92)	8 (47)	<0.001	18 (95)	17 (89)
	toxB	37 (77)	0	<0.001	18 (95)	12 (63)	0.042
	astA	41 (85)	2 (12)	<0.001	19 (100)	15 (79)
	subA	0	8 (47)	<0.001	0	0
	cdtB	0	3 (18)	0.016	0	0
	sta1	0	1 (6)		0	0
	senB	0	2 (12)		0	0
Secretion system	espA	46 (96)	0	<0.001	18 (95)	19 (100)
	espF	28 (58)	0	<0.001	5 (26)	14 (74)	0.009
	espI	2 (4)	0		0	2 (11)
	espJ	46 (96)	0	<0.001	18 (95)	18 (95)
	nleA	41 (85)	0	<0.001	19 (100)	13 (68)	0.020
	nleB	45 (96)	0	<0.001	19 (100)	19 (100)
	nleC	41 (85)	0	<0.001	18 (95)	14 (74)
	etpD	19 (40)	0	0.001	17 (89)	0	<0.001
	cif	23 (48)	0	<0.001	0	16 (84)	<0.001
	tccP	23 (48)	0	<0.001	2 (11)	8 (42)
SPATEsa	espP	43 (90)	7 (41)	<0.001	19 (100)	17 (89)
	pic	1 (2)	2 (12)		1 (5)	0
	sepA	0	1 (6)		0	0
Colicins	cma	1 (2)	3 (18)		1 (5)	0
	cba	16 (33)	4 (24)		3 (16)	5 (26)
	celb	3 (6)	4 (24)		0	3 (16)
Microcins	mcmA	4 (8)	2 (12)		0	4 (21)
	mchB	4 (8)	2 (12)		0	4 (21)
	mchC	4 (8)	2 (12)		0	4 (21)
	mchF	4 (8)	2 (12)		0	4 (21)
Others	lpfA	25 (52)	14 (82)	0.043	1 (5)	17 (89)	<0.001
	katP	41 (85)	2 (12)	<0.001	19 (100)	15 (79)
	ireA	0	5 (29)	0.000	0	0
	gad	20 (42)	2 (12)	0.036	11 (58)	1 (5)	0.001
	iss	46 (96)	14 (82)		18 (95)	18 (95)
	CapU	0	1 (6)		0	0

aSPATE: Serine protease autotransporters of Enterobacteriaceae. P values are only shown if P<0.05.

Among the isolates from patients, O157 and non-O157 isolates had a comparable number of virulence genes (a median of 17 virulence genes per isolate, respectively, P = 0.32). Stx gene stx1a, adhesion gene efa1, secretion system genes espF and cif, and fimbrial gene lpfA were significantly more frequent among non-O157 isolates from patients than among O157 isolates from patients (Table 2). In addition, stx1a, lpfA, and secretion system gene tccP were significantly more frequent in isolates from patients that were infants and small children than in isolates from patients of other age groups (S2 Table). The isolate from the 1 patient with HUS had stx2c-only, as well as adhesion genes eae, tir, and espB, and secretion system genes espA, espJ, nleA, nleB and nleC.

The additional search for saa using BLAST showed that only 5 out of 65 isolates (7%) possessed this gene. All of these isolates were non-O157 and from food handlers.

AMR genes

WGS analysis identified 20 acquired AMR genes in 18 STEC isolates (28% of all 65 STEC isolates) (S1 Fig). The β-lactamase gene (blaTEM-1B) was detected in 7 isolates (11%). There were 14 isolates (22%), 16 isolates (25%), 11 isolates (17%), 3 isolates (5%), and 2 isolates (3%) with at least one of the sulfamethoxazole/trimethoprim, aminoglycoside, tetracycline, macrolide, and phenicol resistance genes, respectively. The distribution of AMR genes was similar among isolates from the patients or food handlers and among O157 or non-O157 isolates. Aminoglycoside resistance genes were less frequent among isolates with major serotypes than isolates with minor serotypes (22/48 isolates with major serotypes versus 16/17 isolates with minor serotypes, P<0.001).

Phylogenetic analysis

Phylogenetic analysis was performed using 132,711 SNPs identified within the core genome of 71 STEC isolates (including the 6 reference strains) (Fig 2). The STEC isolates were divided into two clades, O157 and non-O157, except that the O145:H28 isolate clustered with the O157 isolates. Isolates with the same O serotype formed a cluster together, except for O103 and O8. In addition, the O103:H8 ST2836 isolates clustered with the O26:H11 isolates and were separated from the O103:H2 ST17 isolate. Within the O157:H7 cluster, isolates positive for stx1a+stx2a formed a subcluster. Isolates positive for microcin genes mcmA, mchB, mchC, and mchF were assigned to a subcluster within the O26:H11 cluster.

Fig 2

Maximum likelihood tree based on the core genome shared by Shiga toxin-producing Escherichia coli (STEC) isolates, including reference strains.

Scale bar represents 8.0 nucleotide substitutions per site. A total of 132,711 SNPs were identified in the core genome. The average number of SNPs showing differences between each pair of STEC isolates was 33,603. Yellow shading = isolates with major serotypes; light green shading = isolates with minor serotypes. Closely related isolates (SNP differences <5) are highlighted with red lines behind them. aST: sequence type. bStx: Shiga toxin. cOUT: O-serotype untypable.

In pairwise comparisons, the median number of SNP differences between different core genomes was 423 within O157 (range, 0 to 584), compared to 22,054 (range, 0 to 54,528) for non-O157 genomes overall. Twenty isolates (6, 10, 2, and 2 isolates with O157:H7 ST11, O26:H11 ST21, O103:H8 ST2836, and O76:H19 ST675, respectively) were divided into 9 indistinguishable core genomes (<5 SNPs) (Fig 2). Among these 9 core genomes, 5 were from patients, 3 were from food handlers, and one was isolated from both a patient and a food handler. Two of the 4 O26:H11 strains, including one strain positive for stx1a+stx2a, were from patients under 5 years of age.

Discussion

To the best of our knowledge, this is the first genomic epidemiological study to investigate STEC isolates from patients and asymptomatic food handlers. Isolates with 5 non-O157 serotypes (O26:H11, O103:H2, O103:H8, O121:H19, and O145:H28), which had as many virulence genes as O157 isolates were prevalent among both patients and food handlers, whereas isolates with the remaining 12 serotypes were only found in food handlers and were negative for major virulence genes, eae, tir, espB, espA, espJ, nleA, nleB, and nleC. Non-O157 isolates were especially prevalent among children under 5 years of age. WGS analysis revealed clonal expansion of highly virulent STEC strains (e.g., O26:H11 strain with stx1a+stx2a) in our region (Miyagi Prefecture, Japan).

Although the overall isolation rate of non-O157 STEC strains was reported to be 30–40% in Japan [3], more than half of the STEC isolates were non-O157 strains in our region. Similar to the increment of non-O157 isolates revealed by this study, non-O157 infections have been increasingly reported worldwide [2, 21]. Notably, non-O157 infection was predominant among children under 5 years old in this study. Among the isolates from patients, adhesin gene efa1, secretion system genes espF and cif, and the gene lpfA encoding fimbriae, were significantly more frequent among non-O157 isolates than O157 isolates. lpfA was reported to be involved in prolonged shedding of STEC in young children [22]. In Japan, direct person-to-person contact is the suspected route of transmission for the majority of STEC infection outbreaks among children [23]. Prolonged shedding of STEC can facilitate its spread. Our findings suggested that these genes may be associated with a high frequency of non-O157 STEC infection among children.

The isolates with major serotypes (O157:H7, O26:H11, O103:H2, O103:H8, O121:H19, and O145:H28) harbored significantly more virulence genes than the isolates with minor serotypes. Apart from O103:H8, these major serotypes have been linked to epidemics and serious infections and have been frequently detected among clinical isolates worldwide [24]. Most of the isolates with major serotypes, including the isolate from a patient with HUS, possessed adhesion genes eae, tir, and espB, and secretion system genes espA, espJ, nleA, nleB and nleC, whereas the isolates with minor serotypes were negative for all of these genes. Studies have shown that the eae gene encoding intimin, an outer membrane protein involved in close attachment, is closely linked to the pathogenesis of STEC infection, along with other genes clustering on the bacterial chromosome (such as tir, espA, espB, and espJ) that form a pathogenicity island called the locus of enterocyte effacement [8, 25]. Other studies have shown that effectors outside this locus encoded by nleA, nleB, and nleC are required to form attaching and effacing lesions in the intestinal epithelium, which allow STEC to colonize the human gut [26]. Accordingly, these virulence factors may play a key role in the pathogenesis of STEC infections. The current STEC surveillance system for food handlers in Japan is only based on serotyping and detection of Stx [3]. However, we think that STEC surveillance should focus on the above-mentioned virulence genes, such as eae, tir, espB, espA, espJ, nleA, nleB and nleC.

O103:H8 ST2836 STEC with multiple virulence genes was newly detected in this study. O103:H8 ST2836 isolates formed a separate cluster from the known isolates of O103:H2 ST17 on the phylogenetic tree, suggesting that these two clusters of serogroup O103 had different origins. As previously reported, STEC are E. coli strains of different lineages that have acquired virulence genes independently at different time points [27], and STEC strains from the same O serogroups are polyphyletic since horizontal transfer of the O-antigen gene can occur among different E. coli strains [28]. These points raise the possibility that new serotypes of highly virulent STEC may emerge.

There have only been a limited number of epidemiological studies on AMR in STEC isolates [6]. The resistance genes with the highest prevalence among our STEC isolates were aminoglycoside resistance genes (e.g., aadA, aph(3’)-I, and str), followed by sulfamethoxazole/trimethoprim resistance genes (e.g., sul and dfrA). In Japan, sulfamethoxazole/trimethoprim and aminoglycosides are antibacterial agents commonly used in domestic animals [29], which are the main reservoir of STEC, and the prevalence of aminoglycoside resistance among STEC isolates from cows in Japan has increased during the past decade [30]. Antimicrobial therapy is generally not recommended for STEC infection due to the possible risk of HUS, but it may be beneficial for patients with persistent diarrhea or food handlers with long-term STEC carriage [31]. Transfer of mobile genetic elements was reported to facilitate the spread of AMR genes to other bacteria [6]. Accordingly, it is important to monitor AMR in STEC isolates and prevent misuse/overuse of antibiotics based on the One Health approach [32].

WGS-based phylogenic analysis revealed a variety of SNP variants among isolates from the same serogenotype or same ST clade, suggesting dissemination of diverse STEC strains throughout our region. The O157:H7 isolates with stx1a+stx2a formed a subcluster within the O157:H7 cluster, and the O26:H11 isolates positive for microcin genes formed a subcluster in the O26:H11 cluster. STEC isolates possessing stx1a+stx2a have been linked to outbreaks associated with a high frequency of HUS [33]. The presence of microcin genes indicates environmental plasticity of the isolates since microcin is a bactericidal antibiotic [34]. These results highlight the fact that diverse strains with differing levels of virulence can exist within the same STEC serogroup.

Our phylogenetic analysis also detected 9 clonal expansions of STEC strains suggesting circulation of these strains among patients and food handlers in our region. One of the strains was serogenotype O26:H11 ST21 strain harboring stx1a+stx2a, which differed from a newly emerging virulent O26:H11/H- ST29 STEC clade reported in Japan by Ishijima et al [35]. In general, the majority of STEC O26:H11 isolates are only positive for stx1a [36, 37], highly virulent stx2a-containing O26:H11 strains have been increasingly reported worldwide in recent years [36]. The spread of O26:H11 strains with stx2a could pose a threat in our region. WGS has been employed to investigate the molecular epidemiology of STEC [7], since it is a powerful tool for performing high-resolution molecular typing, population structure analysis, and detailed molecular characterization of microbes [38]. Further genome-based epidemiological studies are needed to provide a better understanding of STEC isolates for assistance in developing prevention and control strategies.

This study had several limitations. First, there were only a few of STEC strains from the same lineage or serotype, although we assessed all of the STEC isolates detected through public health surveillance in our region during the study period. Second, we could only obtain restricted epidemiological and clinical information. Third, while this in silico study was focused on putative virulence genes, the pathogenicity of STEC isolates needs to be clarified by in vitro and in vivo experimental studies.

In conclusion, we found that genetically diverse non-O157 isolates (O26:H11, O103:H2, O103:H8, O121:H19, and O145:H28) with as many important virulence genes as O157 isolates (including eae, tir, espB, espA, espJ, nleA, nleB and nleC) plus AMR genes (such as aminoglycoside and sulfamethoxazole/trimethoprim resistance genes) were prevalent among both patients and asymptomatic food handlers in Miyagi Prefecture, Japan. Our WGS analysis demonstrated the importance of monitoring the genomic characteristics of STEC isolates from asymptomatic food handlers in addition to symptomatic patients.

Characteristics and virulence/antimicrobial resistance (AMR) gene profiles of Shiga toxin-producing Escherichia coli (STEC) isolates.

Yellow shading = isolates with major serotypes; light green shading = isolates with minor serotypes. The presence (black) or absence (white) of virulence genes and AMR genes is shown.

aST: sequence type. bSPATE: Serine protease autotransporters of Enterobacteriaceae. cOUT: O-serotype untypable.

(TIF)

Click here for additional data file.

Distribution of putative virulence genes among Shiga toxin-producing Escherichia coli (STEC) isolates from patients/food handlers and O157/non-O157 isolates.

aSPATE: Serine protease autotransporters of Enterobacteriaceae. P values are shown only if P<0.05.

(XLSX)

Click here for additional data file.

Distribution of putative virulence genes among Shiga toxin-producing Escherichia coli (STEC) isolates from patients of infants and small children/patients of different age groups.

aSPATE: Serine protease autotransporters of Enterobacteriaceae. P values are shown only if P<0.05.

(XLSX)

Click here for additional data file.

5 Sep 2019

PONE-D-19-20606

Genomic analysis of Shiga toxin-producing Escherichia coli from patients and asymptomatic food handlers in Japan

PLOS ONE

Dear Dr. Baba,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please address all of the comments of the two reviewers, including adding bootstrap analysis, criteria for accepting a WGS run, and details about genome assembly (reviewer #2), and please carefully address all of the comments of reviewer #1.  The term non-patient isolates is not entirely clear.  The authors should reconsider using this term or explain it more clearly.  As reviewer #1 pointed out, some of these serotypes have been isolated from patients.

Furthermore, there are a number of errors in English usage and some clarifications are needed.  For example, Line 69:  name the location in Japan.  Lines 78-80:  Where were the patient fecal samples obtained (e.g., hospitals?) and how were the asymptomatic individuals selected?  Please clarify in the text.  Line 91:  Bacterial DNA from the 65 isolates was extracted…..   Line 179:  Again, here the phrase “isolates with patient serotypes” is awkward/unclear.  Also, on line 180:  What is meant by “detected in 41-47 isolates”?  Lines 183-184:  Consider changing this title to read more clearly.  Line 242:  Give examples/information in the text on which highly virulent STEC showed clonal expansion.

And finally, please make the following these changes in the text:

Line 21:  ….However, the molecular…..

Line 23:  ….and asymptomatic food handlers…..

Line 30:  Delete “which”

Line 39:  …genes were…..

Line 49:  ….after a massive…..

Line 58:  ….In addition, highly pathogenic STEC contain…..

Line 64:  ….a comprehensive….

Line 94:  MiSeq…….300-bp reads,…..

Line 146:  …..suggesting a close relationship between……

Line 151:  …..of the remaining….

Line 152:  …..whereas the ages of the food handlers…

Line 188:  ….had a comparable….

Lines 192-193:  Should this read. …..often found in isolates from patients that were infants and small children….”

Line 194:  …..from an HUS patient….

Line 205:  ….with patient serotypes…..

Line 233:  …patients younger than 5 years of age.

Line 247:  …..isolates from patients,….

Line 249:  …..present among non-O157 isolates than from O157.

Line 270:  The majority (17…..

Line 271:  ….low virulence non-patient serotypes.

Line 272:  ……focused on patient-related serotypes….

Line 276:  ……two clusters of serogroup O103 had different…..

Line 279:  …..O-antigen gene…..

Line 292:  ….antibiotics based on the One……

Line 297:  …..cluster, and the O26:H11 isolates…….

We would appreciate receiving your revised manuscript by Oct 20 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Pina Fratamico, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

1. Thank you for including your competing interests statement; "The authors have declared that no competing interests exist."

We note that one or more of the authors are employed by a commercial company: Miyarisan Pharmaceutical Co., Ltd.

Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form.

Please also include the following statement within your amended Funding Statement.

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement.

2. Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc.

Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests) . If this adherence statement is not accurate and  there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include both an updated Funding Statement and Competing Interests Statement in your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Major comments

In this study, the authors revealed the characteristics and phylogeny of STEC in genomic level. As scarce information is available for Japanese STEC isolates, the information is useful. However, some of the discussion is superficial and the authors should consider the results in deeper depth. For instance, when the results were compared to the previous surveillance results of Japan or other countries, what is the implication?

In addition, some sentences are difficult to intuitively understand. English editing is recommended if it was not done yet.

Specific comments

Line 23. our region -> Miyagi prefecture, Japan

Line 26. What “n=19, 29%” meant was unclear. There are 20 isolates of O157 and O26 according to Fig. S1.

Line 29. Add “type III” before “secretion system”.

Line 76. remove “During a one-year period”

Line 76. How were strains isolated? Please add the details of isolation.

Are there any epidemiological links between the isolates used in this study? Discuss it in the Discussion section.

Were all the strains isolated in Miyagi prefecture during the study period included in this study?

In this study, serotypes of the isolates were mentioned. However, it is not mentioned how they were determined. For instance, what did OUT or H- mean. No identical O genotype even in in silico analysis? Did H- mean “no motility”? If so, did you search fliC genotype?

Line 94. Add reagent kit name.

Line 97. It seems that the data is not published yet. Please make the data public when the manuscript is submitted.

Line 106. It is useful to search saa sequence, which is the major adhesin in LEE negative STEC. In Virulence Finder, saa cannot be detected, because almost half of the sequence is repeat sequences.

Line 108. What strain was used as reference?

Line 115. Please add the citation of RAxML, the model used, and the number of bootstrap replicates.

Did you remove recombinogenic region?

Line 127. What does “new” mean in this sentence? For instance, several strains of ST2836 of O103:H8 can be found in EnteroBase.

Line 142. According to the surveillance information of Japan, the “non-patient serotypes” in this study have been reported from the patients.

Line 146. How many SNPs are there in the different allele?

Line 158-160. Are they multiple comparison? If so, how was the P values were adjusted?

Line 199. Please remove “and all 18 isolates had at least one AMR gene”, because it is redundant.

Line 214. Is O26:H11 correct, rather than O157:H7?

Line 214. The O26 cluster in unclear in Fig. 1.

L230-233. The meaning of these sentences is unclear. The authors should clarify the implication of the information.

Line 240. What are major virulence genes? LEE-related genes?

Line 241. The etiologic agent of most of the young children patients was O26. Therefore, the serotype should be focused on.

Line 254-256. These are paradoxical statements. The fact “women are more likely to work in food related occupations” do not explain that more than half of the cases of food handlers are women.

Line 272. This suggestion is unclear. Please specify the recommendation (e.g. isolation method).

In Japan (and other countries), “non-patient serotypes” in this study are responsible for severe diseases, although the incidence rate is not high. How do the authors evaluate them?

Line 290. “under certain conditions” should be clarified.

Line 294-298. Subclusters of O157 is also unclear in Fig. 1. What about clade of O157?

Line 297-301. Virulence of O26:H11 was reported previously (Ishijima et. al. Sci Rep 2017). Please compare them and discuss the relevance.

Line 321. It is vague conclusion.

Fig 1:

The resolution is too low. Please replace it.

Number of isolates would be better rather than number of cases

Please mention the definition of “patient serotype” and “non-patient serotype” in the legend.

Table S1. It seems that this table is identical to Table 1.

Table S1 and S2.

Table title should be included.

Fig. S1

Stx2e positive O157 would be rare. Did you confirm the result by PCR or other methods?

Reviewer #2: The manuscript, Genomic analysis of Shiga toxin-producing E. coli from patients and asymptomatic food handlers in Japan, written by Baba et al is a good manuscript that will enrich our understanding of molecular epidemiology of STEC and correlation of STEC, virulence factors, and HUS.

I have the following comments:

1- Page 18 Line 247: please change the word form to from (the isolates form patients).

2- Page 18 Lines 248-249: reword the following sentence because it is not clear what you mean by that: were significantly more often present among frequent among non-O157 isolates than O157.

3- Add bootstrap analysis to the phylogenetic tree.

4- In Materials and methods, WGS section: can you add the criteria for accepting a WGS run.

5- In genetic analysis section: Can you please elaborate on genome assembly. How did you trim the reads, does Platanus assembler do de novo assembly, what kind of software is used for this assembly (spades, shovil, or velvet), did you filter out small contigs? what sizes did you filter out if any?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

18 Oct 2019

Dr. Joerg Heber

Editor-in-Chief of PLOS ONE

Re: Manuscript ID: PONE-D-19-20606

Dear Dr. Heber,

We thank you very much for evaluating our manuscript entitled “Genomic analysis of Shiga toxin-producing Escherichia coli from patients and asymptomatic food handlers in Japan” (PONE-D-19-20606) and inviting a revision. We are pleased that the reviewers found our study interesting and are grateful for their thoughtful comments.

We have addressed each of the editor and the reviewers’ concerns in a point-by-point response, with each comment and response numbered sequentially as C and R. Each comment is stated in bold font, and our response is in standard font.

We again thank the reviewers for their careful review of our manuscript and hope that the manuscript is now acceptable for publication in PLOS ONE.

Hiroaki Baba, M.D., PhD

Department of Infectious Diseases, Tohoku University Graduate School of Medicine

1-1 Seiryo-machi, Aoba-ku, Sendai, Miyagi Prefecture 980-8574, Japan

TEL: 81-22-717-7373

E-mail: hbaba48@med.tohoku.ac.jp

Our responses to the comments from the Editor:

C1-1 Line 69: name the location in Japan

R1-1 We thank the editor for this comment. We have added “Miyagi Prefecture” on p.5, line 72.

C1-2 Lines 78-80: Where were the patient fecal samples obtained (e.g., hospitals?) and how were the asymptomatic individuals selected? Please clarify in the text.

R1-2 We agree that these points require clarification. We obtained fecal samples of hospital patients, so we have added “Thirty-eight isolates were obtained from fecal samples of hospital patients” on p.6, lines 82-83. Regarding the asymptomatic individuals, we collected all the STEC isolates detected through public health surveillance. We have clarified this on p.6, lines 79-81.

C1-3 Line 91: Bacterial DNA from the 65 isolates was extracted….

R1-3 This change has been made.

C1-4 Line 179: Again, here the phrase “isolates with patient serotypes”is awkward/unclear.

R1-4 We agree with your assessment. Since the terms “patient serotype” and “non-patient serotype” were confusing, we have replaced these terms throughout the paper with “major serotype” and “minor serotype”, respectively. The terms “major serotypes” and “minor serotypes” were defined as “serotypes found in patients in this study” and “serotypes only found in food handlers in this study” (Please see p.11, lines 161-165).

C1-5 Line 180: What is meant by “detected in 41-47 isolates”?

R1-5 We thank the editor for this comment. We have changed this sentence to “Among the 48 isolates with pathogenic serotypes, adhesion genes eae, tir, and espB, and secretion system genes espA, espJ, nleA, nleB, and nleC were detected in 41/48 (85%) isolates to 47/48 (98%) isolates” (p.14, lines 206-210).

C1-6 Lines 183-184: Consider changing this title to read more clearly.

R1-6 Please see R1-4 above.

C1-7 Line 242: Give examples/information in the text on which highly virulent STEC showed clonal expansion.

R1-7 We have added an example for highly virulent STEC as follows: “WGS analysis revealed clonal expansion of highly virulent STEC strains (e.g., O26:H11 strain with stx1a+stx2a) in our region” on p.20, lines 278-280.

C1-8 Please make the following these changes in the text:

1. Line 21: ...However, the molecular…

2. Line 23: ...and asymptomatic food handlers…

3. Line 30: Delete “which”

4. Line 39: …genes were…

5. Line 49: …after a massive…

6. Line 58: …In addition, highly pathogenic STEC contain…

7. Line 64: …a comprehensive…

8. Line 94: MiSeq…300-bp reads, …

9. Line 146: …suggesting a close relationship between…

10. Line 151: …of the remaining…

11. Line 152: …whereas the ages of the food handlers…

12. Line 188: …had a comparable…

13. Lines 192-193: Should this read…often found in isolates from patients that were infants and small children…”

14. Line 194: …from an HUS patient…

15. Line 205: …with patient serotypes…

16. Line 233: …patients younger than 5 years of age.

17. Line 247: …isolates from patients,….

18. Line 249: …present among non-O157 isolates than from O157.

19. Line 270: The majority (17…

20. Line 271: …low virulence non-patient serotypes.

21. Line 272: …focused on patient-related serotypes…

22. Line 276: …two clusters of serogroup O103 had different…

23. Line 279: …O-antigen gene…

24. Line 292: …antibiotics based on the One…

25. Line 297: …cluster, and the O26:H11 isolates…

R1-8 We thank the editor for pointing out grammatical and typographical errors and have corrected these throughout the manuscript.

Our responses to the comments from Reviewer 1:

In this study, the authors revealed the characteristics and phylogeny of STEC in genomic level. As scarce information is available for Japanese STEC isolates, the information is useful. However, some of the discussion is superficial and the authors should consider the results in deeper depth. For instance, when the results were compared to the previous surveillance results of Japan or other countries, what is the implication?

In addition, some sentences are difficult to intuitively understand. English editing is recommended if it was not done yet.

We appreciate the reviewer’s comment on this point. To the best of our knowledge, this is the first genomic epidemiological study to investigate STEC isolates from asymptomatic food handlers. In Japan, STEC surveillance was instituted for patients and asymptomatic food handlers after a massive STEC epidemic associated with consumption of white radish sprouts occurred in 1996, and it has been found that the incidence of asymptomatic carriage, estimated as 84.2/100,000 population [Morita-Ishihara T, et al. Emerg Infect Dis 2016], is much higher than that of STEC infection. The molecular epidemiology of STEC isolates from food handlers, however, is not entirely known. Our study revealed that isolates with 5 non-O157 serotypes (O26:H11, O103:H2, O103:H8, O121:H19, and O145:H28) which had as many virulence genes as O157 isolates were prevalent among both patients and food handlers. Apart from O103:H8, these serotypes have been linked to epidemics and serious infections and have been frequently detected among clinical isolates worldwide. In contrast, isolates with remaining 12 serotypes which were only found in food handlers in this study were negative for major virulence genes, eae, tir, espB, espA, espJ, nleA, nleB and nleC.

WGS analysis revealed clonal expansion of highly virulent STEC strains (e.g., O26:H11 strain with stx1a+stx2a) among both patients and food handlers in Miyagi prefecture, Japan. These findings demonstrated the importance of monitoring the genomic characteristics of STEC isolates from asymptomatic food handlers in addition to symptomatic patients.

This manuscript has been edited and rewritten by an experienced scientific editor, who has improved the grammar and stylistic expression of the manuscript.

C2-1 Line 23. our region -> Miyagi prefecture, Japan

R2-1 This change has been made.

C2-2 Line 26. What “n=19, 29%” meant was unclear. There are 20 isolates of O157 and O26 according to Fig. S1.

R2-2 We think the reviewer is mistaken on this point. Perhaps our explanation was not clear enough. There were 20 isolates of O157 and 20 isolates of O26. While the O157 isolates belonged to O157:H7 ST11 and ST2966, the O26 isolates belonged to O26:H11 ST21 and ST1705. Thus, there were 19 isolates of O157:H7 ST11 and 19 isolates of O26:H11 ST21.

C2-3 Line 29. Add “type III” before “secretion system”.

R1-3 We have added it in accordance with this comment.

C2-4 Line 76. remove “During a one-year period”.

R1-4 We have removed this phrase.

C2-5 Line 76. How were strains isolated? Please add the details of isolation.

R2-5 We agree that this point requires clarification, and we have added the following text to the “Material and Methods” section (p. 6, lines 84-88): Isolation of STEC from stool samples was done with sorbitol-MacConkey agar containing cefixime and tellurite, in addition to conventional E. coli isolation agar (e.g., triple sugar iron agar and lysine-indole-motility medium). A latex agglutination test (VTEC-RPLA, Denka Seiken, Japan) and PCR with the EVT-1&2 and EVS-1&2 primers (TaKaRa Biomedicals, Tokyo, Japan) were used to detect Stx and Stx genes, respectively.

C2-6 Are there any epidemiological links between the isolates used in this study? Discuss it in the Discussion section.

R2-6 We thank the reviewer for this comment. All 65 isolates analyzed in this study were not epidemiologically linked. We have added this information to the “Material and Methods” section (p.6, lines 79-81).

C2-7 Were all the strains isolated in Miyagi prefecture during the study period included in this study?

R2-7 All isolates from Miyagi Prefecture during the study period were included in this study. We have added this point on p.6, lines 79-81.

C2-8 In this study, serotypes of the isolates were mentioned. However, it is not mentioned how they were determined. For instance, what did OUT or H- mean. No identical O genotype even in in silico analysis? Did H- mean “no motility”? If so, did you search fliC genotype?

R2-8 We used CGE SerotypeFinder 1.1 to identify serotypes (Please see p.8, lines 115).

Since “O-” was unclear, we have changed O- to OUT (O-serotype untypable) (Please see Fig 1, Fig 2, and S1 Fig).

No identical O serotype was detected in the one serotype OUT isolate in this study (isolate ID: E58) by SerotypeFinder. In this isolate, no O-processing genes (wzx, wzy, wzm, and wzt) were detected by BLAST search (http://blast.ncbi.nlm.nih.gov). This isolate may be assigned to O14 or O57, since O-processing genes for O14 and O57 could not be identified in their genomes [DebRoy C, et al. PLoS One. 2016;11: e0147434]. We have added this information to the “Results” section (p.10, lines 143-147).

O145:H- isolates were assigned to O145:H28 after reanalysis with SerotypeFinder. We have corrected all relevant parts of the manuscript.

C2-9 Line 94. Add reagent kit name.

R2-9 We have mentioned the reagent kit name, NEBNext Ultra DNA Library Prep Kit for Illumina, in the “Material and Methods” section (p.7, lines 99-101).

C2-10 Line 97. It seems that the data is not published yet. Please make the data public when the manuscript is submitted.

R2-10 We have made these data public.

C2-11 Line 106. It is useful to search saa sequence, which is the major adhesin in LEE negative STEC. In Virulence Finder, saa cannot be detected, because almost half of the sequence is repeat sequences.

R2-11 We thank the reviewer for this suggestion. We searched for the major adhesin gene saa, which cannot be detected by VirulenceFinder, using BLAST (http://blast.ncbi.nlm.nih.gov). We have added this information to p.8, lines 117-119. The results of searching for saa showed that only 5 of 65 isolates (7%) possessed this gene. All of these isolates were non-O157 and were from food handlers. We have added these points to P.17, lines 227-228. These findings suggested that saa may not play an independent role in STEC pathogenicity in the isolates we investigated.

C2-12 Line 108. What strain was used as reference?

R2-12 We used the following 6 reference genomes: STEC O157:H7 Sakai (GenBank accession numbers: BA000007), STEC O157:H7 strain EDL933 (AE005174), STEC 157:H7 strain EC4115 (CP001164), STEC O157:H7 strain TW14359 (CP001368), STEC O26:H11 strain 11368 (AP010953), and STEC O103:H2 strain 12009 (AP010958). We have mentioned this point in the “Materials and Methods” section (p.8, lines 120-126).

C2-13 Line 115. Please add the citation of RAxML, the model used, and the number of bootstrap replicates.

Did you remove recombinogenic region?

R2-13 We agree that these points require clarification. We have added the version and citation of RAxML, and the number of boostrap replicates to the “Material and Methods” section (p.9, lines 128-130) as follows: “A phylogenic tree with 1000 boostrap replicates was constructed by using the randomized accelerated maximum likelihood (RAxML) program v 8.2.12 (https://cme.h-its.org/exelixis/web/software/raxml/index.html)”.

Recombinogenic regions were removed using Parsnp v 1.2 (https://harvest.readthedocs.io/en/latest/content/parsnp/quickstart.html#advanced-usage). We have also added the following information to the “Material and Methods” section (p.8, line 120-122).

C2-14 Line 127. What does “new” mean in this sentence? For instance, several strains of ST2836 of O103:H8 can be found in EnteroBase.

R2-14 We are sorry for not providing an explanation. Escherichia coli strains of these 13 sequence types are found in EnteroBase, but these sequence types were new as STEC strains. We have added the following text to the “Results” section (p.10, lines142-144): Seven sequence types have already been reported as STECs causing human disease in Enterobase (http://enterobase.warwick.ac.uk), while the other 13 sequence types, including O103:H8 ST2836, are new as STEC strains.

C2-15 Line 142. According to the surveillance information of Japan, the “non-patient serotypes” in this study have been reported from the patients.

R2-15 In this study, “non-patient serotype” was defined as “serotypes only found in food handlers in this study”. Thus, isolates with non-patient serotypes could be found in patients elsewhere.

Since the terms “patient serotype” and “non-patient serotype” were confusing, we have replaced these terms throughout the paper with “major serotype” and “minor serotype”, respectively. Please see our response to the editor (R1-4 above) regarding the same comment.

C2-16 Line 146. How many SNPs are there in the different allele?

R2-16 We thank the reviewer for this comment. We have added following text to the “Results” section (p.11, lines 169-171): There was only one SNP difference between ST11 and ST2966 in purA, as well as between ST21 and ST1705 in gyrA.

C2-17 Line 158-160. Are they multiple comparison? If so, how was the P values were adjusted?

R2-17 These are not multiple comparisons. We compared the incidence in infants and small children with that in each of the other age groups (one by one) using Fisher’s exact test.

C2-18 Line 199. Please remove “and all 18 isolates had at least one AMR gene”, because it is redundant.

R2-18 We agreed with your assessment, and we have removed this text.

C2-19 Line 214. Is O26:H11 correct, rather than O157:H7?

R2-19 We appreciate this comment. We have changed this sentence as follows: “Within the O157:H7 cluster, isolates positive for stx1a+stx2a formed a subcluster. Isolates positive for microcin genes mcmA, mchB, mchC, and mchF were assigned to a subcluster within the O26:H11 cluster” (p.18, lines 249-251).

C2-20 Line 214. The O26 cluster in unclear in Fig. 1.

R2-20 We agree with your assessment. Accordingly, we have replaced the phylogenetic tree with a new one that has a clearer branch (see revised Fig 1).

C2-21 L230-233. The meaning of these sentences is unclear. The authors should clarify the implication of the information.

R1-21 We agreed with your assessment. We have changed this sentence as follows: “Among these 9 core genomes, 5 were from patients, 3 were from food handlers, and one was isolated from both a patient and a food handler” (p.18, lines 266-268).

C2-22 Line 240. What are major virulence genes? LEE-related genes?

R1-22 We have added “eae, tir, espB, espA, espJ, nleA, nleB, and nleC” to this section (p.20, line 278).

C2-23 Line 241. The etiologic agent of most of the young children patients was O26. Therefore, the serotype should be focused on.

R2-23 We appreciate the reviewer’s concern on this point. Among the isolates from patients, isolates with O103 and O121 were more likely to be from young children (both 1/2, 50%) than isolates with O157 (2/19, 11%), even though the total number of isolates was low. O26 isolates from patients and non-O157 isolates (excluding O26 isolates) from patients harbored an equal number of virulence genes (both had a median of 18 virulence genes per isolate, P=0.84). The distribution of putative virulence genes among was similar O26 isolates from patients and non-O157 isolates (excluding O26 isolates) from patients (Please see Table 1 in "Response to Reviewers").

The incidence of O26 infection was significantly higher in infants and small children (9.0) than in other age groups (all P<0.001), while there were no significant differences in the incidence of O157 infections among age groups (Please see Table 2 in "Response to Reviewers"). Stx gene stx1a, adhesion gene efa1, secretion system genes espF and cif, and fimbrial gene lpfA were significantly more frequent among O26 isolates from patients than among O157 isolates from patients (Please see Table 3 in "Response to Reviewers"). These findings were similar to the results obtained by analysis focusing on all non-O157 serotypes (including O26) from patients. Therefore, we would like to retain the original text in this section.

C2-24 Line 254-256. These are paradoxical statements. The fact “women are more likely to work in food related occupations” do not explain that more than half of the cases of food handlers are women.

R2-24 We thank the reviewer for this comment. According to a survey by the Japanese Ministry of Economy, Trade and Industry (https://www.meti.go.jp/statistics/tyo/syokozi/result-2/h2c5kjaj.html), more than half of food handlers in Japan are women, and we think the predominance of women among STEC carriers may reflect this fact.

Because this point is not important, we have deleted the relevant text.

C2-25 Line 272. This suggestion is unclear. Please specify the recommendation (e.g. isolation method).

In Japan (and other countries), “non-patient serotypes” in this study are responsible for severe diseases, although the incidence rate is not high. How do the authors evaluate them?

R2-25 We thank the reviewer for raising these important points. The current STEC surveillance system for food handlers in Japan is only based on serotyping and detection of Stx. We think STEC surveillance should focus on major virulence genes, such as eae, tir, espB, espA, espJ, nleA, nleB, and nleC. We have modified our suggestion to clarify it (p.22, lines 319-322).

As we mentioned on p.22, lines 305-310, STEC are E. coli strains of different lineages that have acquired virulence genes independently at different time points, and STEC strains from the same O serogroups are polyphyletic since horizontal transfer of the O- antigen gene can occur among different E. coli strains. These points raise the possibility that new serotypes of highly virulent STEC strains may emerge and that highly virulent serotypes could differ from country to country.

C2-26 Line 290. “under certain conditions” should be clarified.

R2-26 We thank the reviewer for this suggestion. We have changed this text to: “Antimicrobial therapy is generally not recommended for STEC infection due to the possible risk of HUS, but it may be beneficial for patients with persistent diarrhea or food handlers with long-term STEC carriage” (p.23, lines 332-333).

C2-27 Line 294-298. Subclusters of O157 is also unclear in Fig. 1. What about clade of O157?

R2-27 Please see R2-20 above and the revised Fig 1.

C2-28 Line 297-301. Virulence of O26:H11 was reported previously (Ishijima et. al. Sci Rep 2017). Please compare them and discuss the relevance.

R2-28 We appreciate this comment. Ishijima et al. reported a newly emerging virulent O26:H11/H- ST29 STEC clade in Japan. However, the O26:H11 strain with stx1a+stx2a in this study was assigned to a different sequence type, ST21. We have added this explanation to the “Discussion” section (p.24, lines 347-349).

C2-29 Line 321. It is vague conclusion.

R2-29 We agree with your assessment. We have added information about specific serotypes, virulence genes, and AMR genes to the conclusion (p.25, lines 364-368).

C2-30 Fig 1: The resolution is too low. Please replace it.

Number of isolates would be better rather than number of cases.

Please mention the definition of “patient serotype” and “non-patient serotype” in the legend.

R2-30 We thank the reviewer for these comments. We have replaced Fig 1 with a clearer one. We have changed “number of cases” to “number of isolates”. Instead of “patient” and “non-patient” serotypes, we have used the terms “major serotypes” and “minor serotypes”, which are defined as “serotypes found in patients in this study” and “serotypes only found in isolates from food handlers in this study”, respectively.

C2-31 Table S1. It seems that this table is identical to Table 1.

R2-31 We thank the reviewer for pointing out this error. We have replaced Table S1.

C2-32 Table S1 and S2. Table title should be included.

R2-32 Please see p.33, lines 497-498 and 502-504.

C2-33 Fig. S1. Stx2e positive O157 would be rare. Did you confirm the result by PCR or other methods?

R2-33 We apologize for this error. This strain was actually Stx2a positive. We have corrected this error throughout the manuscript.

Our responses to the comments from Reviewer 2:

The manuscript, Genomic analysis of Shiga toxin-producing E. coli from patients and asymptomatic food handlers in Japan, written by Baba et al is a good manuscript that will enrich our understanding of molecular epidemiology of STEC and correlation of STEC, virulence factors, and HUS.

C3-1 Page 18 Line 247: please change the word form to from (the isolates form patients).

R3-1 This change has been made.

C3-2 Page 18 Lines 248-249: reword the following sentence because it is not clear what you mean by that: were significantly more often present among frequent among non-O157 isolates than O157.

R3-2 We thank the reviewer for pointing out this error and we have corrected this sentence by removing “often present among” (p.21, line 288).

C3-3 Add bootstrap analysis to the phylogenetic tree.

R3-3 Please see our response to reviewer 1 (R2-13 above) regarding the same comment.

C3-4 In Materials and methods, WGS section: can you add the criteria for accepting a WGS run.

R3-4 We agree with your point and we have added the following text to Material and Methods (p.7, lines 104-105): “The passing filter ranged from 90.88 to 96.74% (mean, 93.19%), and the average Q30 ranged from 78.30 to 87.21% (mean, 83.50%)”.

C3-5 In genetic analysis section: Can you please elaborate on genome assembly. How did you trim the reads, does Platanus assembler do de novo assembly, what kind of software is used for this assembly (spades, shovil, or velvet), did you filter out small contigs? what sizes did you filter out if any?

R3-5 We thank for raising this important question. We have added the following explanation: “Sequence reads were trimmed of adapters and filtered to remove reads shorter than 36 bp using Trimmomatic, followed by assembly using Platanus assembler v 1.2.4.” (p.8, lines 110-111).

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

30 Oct 2019

PONE-D-19-20606R1

Genomic analysis of Shiga toxin-producing Escherichia coli from patients and asymptomatic food handlers in Japan

PLOS ONE

Dear Dr. Baba,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The revised manuscript is improved; however, additional changes that are needed are the following:

Line 20:  comma after disease.

Line 29:  comma after tir.

Line 30: comma after nleB.

Line 44:  change causes to cause.

Line 46:  change to -  …..serotype, severe infections caused by non-O157 serogroups are…..

Line 64:  change to ….pathogenicity and antibiotic resistance….

Line 118: change to ….. (AE005174), STEC O157:H7…..

Line 122: change to ….A phylogenetic tree….

Line 124: change to ……and it was visualized…..

Lines 137-141:  change to …….STEC strains.  An O-group was not detected in the one OUT (O-serogroup untypable) isolate by SerotypeFinder.  In this isolate, no O-processing genes (wzx, wzy, wzm, wzt) were detected by BLAST.  It is possible that this isolate could be assigned to serogroups O14 or O57 since O-processing genes for these serogroups have not been found in their genomes [20], or it could represent a new serogroup.

Line 157:  In Figure 1, O145:H28 is grouped as a minor serotype, not a major serotype.  Please clarify/correct.

Lines 194-197:  This sentence is still not clear as written.  To what do the figures 41/48 and 47/48 refer (which genes or groups of genes)? Please rewrite this sentence based on what is shown in Table 2.

Line 323:  change to ……to provide a better……..

Line 347:  Int J Infect Dis.

Line 351:  italicize gene name.

References #13, 20, 25, 35, 38 – The first letter of the words in the title of the paper should not be capitalized.

References #20 and 35 – italicize Escherichia coli.

Line 394:  Is this written correctly?

Line 417:  italicize stx

Line 418:  Front Cell Infect Microbiol.

Line 435:  Euro Surveill.

We would appreciate receiving your revised manuscript by Dec 14 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Pina Fratamico, Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

31 Oct 2019

Our responses to the comments from the Editors:

The revised manuscript is improved; however, additional changes that are needed are the following:

C-1

Line 20: comma after disease.

Line 29: comma after tir.

Line 30: comma after nleB.

Line 44: change causes to cause.

Line 46: change to - …..serotype, severe infections caused by non-O157 serogroups are…..

Line 64: change to ….pathogenicity and antibiotic resistance….

Line 118: change to ….. (AE005174), STEC O157:H7…..

Line 122: change to ….A phylogenetic tree….

Line 124: change to ……and it was visualized…..

Lines 137-141: change to …….STEC strains. An O-group was not detected in the one OUT (O-serogroup untypable) isolate by SerotypeFinder. In this isolate, no O-processing genes (wzx, wzy, wzm, wzt) were detected by BLAST. It is possible that this isolate could be assigned to serogroups O14 or O57 since O-processing genes for these serogroups have not been found in their genomes [20], or it could represent a new serogroup.

Line 323: change to ……to provide a better……..

Line 347: Int J Infect Dis.

Line 351: italicize gene name.

References #13, 20, 25, 35, 38 – The first letter of the words in the title of the paper should not be capitalized.

References #20 and 35 – italicize Escherichia coli.

Line 417: italicize stx

Line 418: Front Cell Infect Microbiol.

Line 435: Euro Surveill.

R-1 We thank the editor for pointing out grammatical and typographical errors and have corrected these throughout the manuscript.

C-2 Line 157: In Figure 1, O145:H28 is grouped as a minor serotype, not a major serotype. Please clarify/correct.

R-2 We thank the editor for pointing out this error. We grouped O145:H28 as a major serotype in Figure 1 (Please see revised Figure 1).

C-3 Lines 194-197: This sentence is still not clear as written. To what do the figures 41/48 and 47/48 refer (which genes or groups of genes)? Please rewrite this sentence based on what is shown in Table 2.

R-3 We thank the editor for this comment. We have changed this sentence to “Among the 48 isolates with major serotypes, adhesion genes eae, tir, and espB were detected in 47 (98%), 46 (96%), and 45 (94%) isolates respectively, and secretion system genes espA, espJ, nleA, nleB, and nleC were detected in 46 (96%), 46 (96%), 41 (85%), 45 (94%), and 41 (85%) isolates respectively” (p.14, lines 197-200).

C-4 Line 394: Is this written correctly?

R-4 We thank the editor for this comment. We have changed reference #19 to “Stamatakis A. Using RAxML to infer phylogenies. Curr Protoc Bioinformatics. 2015;51: 6.14.1-14.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

4 Nov 2019

Genomic analysis of Shiga toxin-producing Escherichia coli from patients and asymptomatic food handlers in Japan

PONE-D-19-20606R2

Dear Dr. Baba,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Pina Fratamico, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

8 Nov 2019

PONE-D-19-20606R2

Genomic analysis of Shiga toxin-producing Escherichia coli from patients and asymptomatic food handlers in Japan

Dear Dr. Baba:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr Pina Fratamico

Academic Editor

PLOS ONE