Emergence and spread of O16-ST131 and O25b-ST131 clones among faecal CTX-M-producing Escherichia coli in healthy individuals in Hunan Province, China.

OBJECTIVES: The objectives of this study were to determine CTX-M-producing Escherichia coli ST131 strain prevalence in stool specimens from healthy subjects in central China and to molecularly characterize clonal groups.
METHODS: From November 2013 to January 2014, stool specimens from healthy individuals in Hunan Province were screened for ESBL-producing E. coli using chromogenic medium and CTX-M genotypes and phylogenetic groups were determined. ST131 clonal groups were detected by PCR and characterized for antibiotic resistance, fimH, gyrA and parC alleles, plasmid-mediated quinolone resistance determinants, virulence genotypes and PFGE patterns.
RESULTS: Among 563 subjects, 287 (51.0%) exhibited the presence of faecal ESBL-producing E. coli, all of which produced CTX-M enzymes. The most common CTX-M genotypes were CTX-M-14 (48.4%), CTX-M-15 (27.5%) and CTX-M-27 (15.0%). Of the 287 CTX-M-producing isolates, 32 (11.1%) belonged to the ST131 clone. O16-ST131 isolates were dominant (75%) and contained the fimH41 allele. The remaining eight (25%) ST131 isolates were of the O25b subgroup and contained fimH30 or fimH41. Ciprofloxacin resistance was found in 100% of the O25b-ST131 isolates, whereas only 8% of the O16-ST131 isolates were resistant. All of the O25b-ST131 isolates except one showed gyrA1AB and parC1aAB mutations; most of the O16-ST131 isolates had gyrA1A and parC1b mutations. The virulence genotypes of O16-ST131 resembled those of the O25b-ST131 isolates. The 32 ST131 isolates formed one large group at the 64% similarity level. They comprised 15 PFGE groups (defined at ≥85% similarity).
CONCLUSIONS: O16-ST131 isolates have emerged as the predominant type of ST131 isolate in faecal CTX-M-producing E. coli in healthy individuals in China.

Introduction

CTX-M-producing Escherichia coli are the most common type of ESBL-producing E. coli, causing hospital-associated and community infections with extensive dissemination in the environment.[1] There are few data on the community carriage of CTX-M ESBL-producing E. coli in China[2] and two studies describe greatly different carriage rates: 7% and 50.5%.[3,4] ST131 is the predominant extraintestinal pathogenic E. coli (ExPEC) clone and frequently carries CTX-M.[5]

The gut is the main reservoir of ExPEC,[5] although there are only a few studies of colonization by ST131, with differences in carriage rates.[5,6] Studies on ST131 carriage, with one exception,[7] either failed to distinguish the O16-ST131 clade from the more common O25b-ST131[8] or missed the O16 subgroup.[4] Therefore, further studies are necessary to characterize O16-ST131 colonization of the digestive tracts of healthy subjects.

The aim of this study was to determine the prevalence and genotype distribution of CTX-M-producing E. coli and the prevalence of ST131 and the O16/O25b subgroups in healthy subjects across a province in China.

Methods

Participants

A cross-sectional survey was conducted on healthy subjects who had routine health screening at community health service centres (CHSCs) in Hunan Province between November 2013 and January 2014. A total of 11 CHSCs were selected using a multistage stratified sampling method. In the first stage, Hunan Province was divided into four different areas (northern, eastern, western and southern Hunan) according to economic development levels and geographical location. In the second stage, 11 counties were identified, representative by population, and a single CHSC from each was selected for study. Healthy individuals (≥18 years of age) who were examined at the 11 CHSCs from November 2013 to January 2014 were given an attendee list number, which was used to identify participants on the day of visit. Every day, one individual from each centre was selected by matching one computer-generated random number to the attendance list at the centre and was invited to participate in the study. The proportion of refusals was 2.3%. Individuals were excluded if they had digestive tract diseases or had been exposed to a hospital environment or treated with antibiotics within 3 months prior to sample collection. The study was approved by the Ethics Committee of the Xiangya Hospital of Central South University (reference number ECCT 201310096). Participants provided fresh stools and informed consent on the day of visit.

ESBL-producing E. coli identification, CTX-M type and phylogenetic group

Faecal samples were plated within 24 h onto chromogenic ESBL selective medium (ChromID, bioMérieux, Marcy-l'Étoile, France). E. coli were identified using API 20E (bioMérieux). Three to five pure colonies were examined for ESBL production by the double-disc synergy test according to CLSI guidelines.[9]

ESBL-producing E. coli isolates were investigated for the presence of blaCTX-M genes by multiplex PCR and sequencing as previously described.[10,11] The phylogenetic groups of E. coli were determined by quadruplex PCR.[12]

Identification of ST131

The ST131 clonal group and O16/O25b variants were identified as previously described.[13] MLST was used to resolve discrepant results (http://mlst.warwick.ac.uk/mlst/dbs/Ecoli).

Antimicrobial susceptibility testing

For all of the ST131 isolates, three to five pure colonies were tested for antibiotic susceptibility by disc diffusion and the results were interpreted according to the breakpoints recommended by the CLSI.[9] The following antimicrobial agents were tested: ampicillin, ampicillin/sulbactam, cefazolin, ceftriaxone, ceftazidime, gentamicin, ciprofloxacin, piperacillin/tazobactam, imipenem, minocycline, nitrofurantoin, chloramphenicol and trimethoprim/sulfamethoxazole. E. coli ATCC 25922 was used as a quality control strain. Intermediate susceptibility was considered as resistant. The resistance score was calculated as previously described.[7] Isolates that were considered MDR were those resistant to at least three antimicrobial categories.

Molecular characterization of ST131

The alleles of fimH, gyrA and parC, the H30-Rx subclones, plasmid-mediated quinolone resistance (PMQR) determinants [qnrA, qnrB, qnrC, qnrS, aac(6′)-Ib-cr and qepA], virulence factor genes (n = 32) and virotypes were determined as previously described.[14-17]

ST131 isolates were subjected to PFGE analysis using XbaI digestion. A PFGE dendrogram was constructed with BioNumerics software (Applied Maths, Sint-Martens-Latem, Belgium) according to the unweighted pair group method based on Dice coefficients. Isolates with a Dice similarity index ≥85% were considered to belong to the same PFGE group.

Statistical analysis

Comparisons of proportions were performed using Fisher's exact test. Comparisons of antibiotic resistance scores were performed using the Mann–Whitney U-test. A P value <0.05 was considered to be statistically significant.

Results

Prevalence and genotype distribution of faecal CTX-M-producing E. coli

A total of 563 faecal samples, one sample from each participant, were analysed. The age range of the individuals (299 males and 264 females) recruited into the study was 18–83 (mean 44.3) years. Of the 563 samples collected, 287 were identified as ESBL-producing E. coli, with all strains harbouring blaCTX-M-type genes. The prevalence of CTX-M-producing E. coli was thus 51.0% (287/563 samples). Of the 287 isolates, 139 (48.4%) produced CTX-M-14, 79 (27.5%) CTX-M-15, 43 (15.0%) CTX-M-27, 11 (3.8%) CTX-M-65, 10 (3.5%) CTX-M-3, 2 (0.7%) CTX-M-24, 2 (0.7%) CTX-M-104 and 1 (0.3%) CTX-M-101.

Phylogenetic analysis of the 287 CTX-M-producing E. coli strains showed that 101 (35.2%) belonged to phylogroup D, 66 (23.0%) to phylogroup A, 63 (22.0%) to phylogroup B2, 36 (12.5%) to phylogroup B1, 11 (3.8%) to phylogroup F, 7 (2.4%) to phylogroup C, 2 (0.7%) to phylogroup E and 1 (0.3%) to clade I.

E. coli ST131 clonal group and antimicrobial susceptibility

PCR analysis showed that 32 (11.1%) of the 287 CTX-M-producing isolates belonged to the ST131 clonal group, comprising 24 O16-B2-ST131 and 8 O25b-B2-ST131 isolates; only 2 isolates required full MLST.

All of the 32 ST131 isolates were found to be resistant to ceftriaxone, cefazolin and ampicillin, 17 (53.1%) to ceftazidime, 10 (31.3%) to ciprofloxacin, 14 (43.8%) to gentamicin, 28 (87.5%) to ampicillin/sulbactam, 20 (62.5%) to trimethoprim/sulfamethoxazole, 8 (25.0%) to minocycline, 2 (6.3%) to piperacillin/tazobactam and 4 (12.5%) to chloramphenicol. None of the isolates was resistant to imipenem or nitrofurantoin.

E. coli ST131 characterization

Characterization of the 32 E. coli ST131 isolates is shown in Figure 1. The fimH gene was carried by all 24 O16-ST131 isolates, which were H41 lineage. In contrast, only half of the eight O25b-ST131 isolates belonged to the H41 lineage. The remaining four isolates harbouring fimH belonged to the H30 lineage, with none representing the H30-Rx sublineage. All of the O25b-ST131 isolates except one showed gyrA1AB and parC1aAB mutations; most of the O16-ST131 isolates had gyrA1A and parC1b mutations. None of the PMQR determinants was detected in any of the 32 ST131 isolates. Nine PFGE clusters with a similarity index of ≥85% were identified. Unique PFGE profiles were seen in six isolates (Figure 1).

Figure 1.

The XbaI PFGE-based dendrogram for the 32 E. coli ST131 isolates. The dendrogram was generated according to the unweighted pair group method based on Dice coefficients. O types, serogroup; FQ, fluoroquinolone phenotype (R, resistant; S, susceptible); fimH, allele of fimH (type 1 fimbrial adhesin); gyrA/parC, allele of gyrA and parC (encoding DNA gyrase and topoisomerase, respectively); 1/1b, gyrA1 (WT) combined with parC1b (one silent mutation); 1A/1b, gyrA1A (one replacement mutation, Ser83Leu) combined with parC1b; 1AB/3A, gyrA1AB (two replacement mutations, Ser83Leu and Asp87Asn) combined with parC3A (a recombined variant containing one replacement mutation Ser80Ile); 1AB/1aAB, gyrA1AB combined with parC1aAB (parC1a plus the replacement mutations Ser80Ile and Glu84Val). The 32 ST131 isolates formed one large group at the 64% similarity level. They comprised 15 PFGE groups (defined at ≥85% similarity).

The virulence genotypes of O16-ST131 resembled those of the O25b-ST131 isolates (Table S1, available as Supplementary data at ). All of the ST131 isolates were found to belong to virotype C.

Association of O16 and O25b subgroups with antimicrobial resistance and CTX-M type

As indicated in Table 1, the O25b-ST131 isolates had a significantly higher prevalence of resistance to ciprofloxacin compared with O16-ST131. The O16-H41 isolates were more likely to be positive for blaCTX-M-14. All of the O25b-H41 isolates produced CTX-M-15, which was not found in the O25b-H30 isolates.

Table 1.

Antimicrobial resistance and CTX-M genotypes of the 32 E. coli ST131 isolates

	No. (%) of ST131 isolates			P value
	O16-H41 (n = 24)	O25b		O16-H41 versus O25b-H30	O16-H41 versus O25b-H41	O25b-H30 versus O25b-H41
		H30 (n = 4)	H41 (n = 4)
Antimicrobial resistance
ciprofloxacin	2 (8)	4 (100)	4 (100)	0.001	0.001	—
gentamicin	12 (50)	2 (50)	0	1.000	0.113	0.429
minocycline	4 (17)	1 (25)	3 (75)	1.000	0.038	0.486
piperacillin/tazobactam	1 (4)	1 (25)	0	0.270	1.000	1.000
ampicillin/sulbactam	22 (92)	2 (50)	4 (100)	0.086	1.000	0.429
trimethoprim/sulfamethoxazole	16 (67)	4 (100)	0	0.295	0.024	0.029
imipenem	0	0	0	—	—	—
nitrofurantoin	0	0	0	—	—	—
chloramphenicol	1 (4)	0	3 (75)	1.000	0.005	0.143
cefazolin	24 (100)	4 (100)	4 (100)	—	—	—
ceftriaxone	24 (100)	4 (100)	4 (100)	—	—	—
ceftazidime	10 (42)	3 (75)	4 (100)	0.311	0.098	1.000
ampicillin	24 (100)	4 (100)	4 (100)	—	—	—
MDR	11 (46)	4 (100)	3 (75)	0.102	0.596	1.000
Resistance score	3.4 (2–6)a	4.5 (3–6)a	4.5 (3–5)a	0.269	0.055	0.760
CTX-M genotype
CTX-M-14	14 (58)	2 (50)	0	1.000	0.098	0.429
CTX-M-15	5 (21)	0	4 (100)	1.000	0.006	0.029
CTX-M-27	3 (13)	2 (50)	0	0.135	1.000	0.429
other CTX-Mb	2 (8)	0	0	1.000	1.000	—

aThe value denotes the mean and range.

bCTX-M-3 was found in one O16-ST131 isolate and CTX-M-24 was found in one O16-ST131 isolate.

Discussion

We have found the colonization prevalence of CTX-M-producing E. coli in healthy individuals to be 51.0%, which is much higher than that reported from European countries (0.6%–11.3%).[1,18] Although a high prevalence (27.5%) of CTX-M-15 has not been observed in previous studies of community carriage from China,[3,4] our findings suggest that this worldwide dominant genotype is displacing the current genotypes (CTX-M-14 and CTX-M-27) in China. Our observation correlates with a recent report on clinical E. coli isolates across China showing 14.5% CTX-M-15 and 19.7% CTX-M-55 (a single-locus variant of CTX-M-15).[19] This emphasizes the importance of studying the human faecal resistome as many strains causing infection are derived from the gut.

O25b:H4 was considered the dominant ST131 clone serotype in almost all previous studies,[5,6] whereas O16:H5, a recently identified serotype of ST131 isolates, accounted for a small subset (1%–12%) in a few countries.[5-7] However, data from China are lacking and the present study is the first to show the presence of O16-ST131 in this country. Surprisingly, we found that O16-ST131 was the predominant subset (75%) among ST131 isolates, whereas this type only accounted for 8% in France.[7] These disparities may be due to geographical or host population differences. To our knowledge, our study is the first to report that the O16 subset accounts for such a high proportion of the ST131 isolates. This finding implies that O16-ST131 may be poised to emerge as a major type of E. coli ST131 in the community in China.

All of the O16-ST131 isolates were found to belong to the fimH-based H41 subclone, as reported previously.[7,13] In contrast, the O25b-ST131 isolates carried the fimH30 or H41 allele. This is the first study to show the presence of O25b-ST131 with the fimH41 allele. Multiple previous studies have documented that most O25b-ST131 isolates contain the H30 allele, with a minority containing H22 and the remainder containing one of several rarer fimH alleles, but not fimH41.[6,13] The reason for the emergence of the fimH41 allele in O25b-ST131 is not yet clear, but may be due to the homologous recombination of fimH alleles in O25b-ST131 isolates.[20] Further studies are required to elucidate and monitor the evolution of O25b-ST131 with the fimH41 allele.

A limitation was that the small number of O25b-ST131 isolates restricted the broad generalizability of the characterization (especially statistical analyses) of O16-H41, O25b-H30 and O25b-H41 subgroups presented in this study. Thus, studies with larger sample sizes and more broadly distributed populations are required to further explore the characterization of these subgroups in the future.

In conclusion, to our knowledge, this is the first study to report O16-ST131 as the predominant type of clonal group in ST131. Our findings indicate that future studies should focus more on O16-ST131 isolates, especially in high-incidence areas such as China.

Funding

This work was supported by the Fundamental Research Funds for the Central Universities of Central South University (2014zzts352) and by a grant from the Development and Reform Commission of Hunan Province ([2014]658).

Transparency declarations

None to declare.

Supplementary data

Table S1 is available as Supplementary data at