Carriage of colistin-resistant, extended-spectrum β-lactamase-producing Escherichia coli harboring the mcr-1 resistance gene after short-term international travel to Vietnam.

BACKGROUND: Due to increasing colistin usage, the dissemination of the colistin-resistant gene mcr-1 has been increasingly investigated. The aim of this study was to determine whether a traveler on a short-term international trip to a developing country could bring mcr-1 back to their home country.
MATERIALS AND METHODS: Thirty-four travel events from Japan to Vietnam encompassing 19 travelers were assessed. A fecal specimen was collected from each traveler before and after each travel event and was inoculated on CHROMagar containing cefotaxime (CTX). Three to seven colonies exhibiting the characteristics of Escherichia coli were collected. Susceptibility to antibiotics and extended-spectrum β-lactamase (ESBL) production were determined by the disk diffusion method and the double-disk synergy test, respectively. ESBL-encoding genes were genotyped, and phylogenetic groupings were determined by multiplex polymerase chain reaction (PCR). The presence of mcr-1 was also confirmed by PCR and sequencing.
RESULTS: A total of 175 ESBL-producing E. coli isolated before and up to 2 weeks after traveling to Vietnam were analyzed. Genotyping of ESBL-producing isolates showed that blaCTX-M-1/blaTEM (27.7%) and blaCTX-M-9 (45.9%) were the most prevalent genotypes, while the most frequently detected phylogenetic group was D (41.9%) followed by B2 (23.0%). In a significant number of travel events, travelers brought ESBL-producing E. coli back to Japan and three events by three travelers carried mcr-1. ESBL-producing E. coli isolates harboring mcr-1 were identified as those carrying both blaCTX-M-14 or blaCTX-M-55 and mcr-1.
CONCLUSION: Using Vietnam as an example, we have shown that even a short-term trip to some countries may result in ESBL-producing mcr-1-positive E. coli carriage by international travelers.

Introduction

With increasing globalization, people frequently travel to other countries. Under such circumstances, travelers can represent potential reservoirs for the dissemination of extended-spectrum β-lactamase (ESBL)-producing Enterobacteriaceae.1

Our previous study revealed that veterinarians in Vietnam prefer to prescribe colistin-based drugs for use on chicken farms.2 Although colistin is effective for treating multidrug-resistant (MDR) bacteria, including carbapenem-resistant bacteria, this has resulted in the widespread dissemination of colistin-resistant bacteria harboring plasmids encoding the colistin-resistant gene, mcr-1.3 Therefore, the prevalence of mcr-1 requires investigation. Additionally, a recent report showed that colistin-resistant Escherichia coli harboring mcr-1 had been isolated from food animals as well as from farmers and urban persons in Vietnam.4,5 Haenni et al6 reported that the ESBL-related gene blaCTX-M is encoded on mobile plasmids with mcr-1. Recently, transmission of colistin-resistant ESBL-producing E. coli by international travelers from China and India to developed countries has been reported.7,8 Accordingly, we investigated whether travelers were bringing ESBL-producing E. coli harboring mcr-1 back to their home countries after short-term stays in developing countries, such as Vietnam.

Materials and methods

A prospective cohort study was approved by the ethics committees of Osaka University (no eki31) and Osaka Prefectural Institute of Public Health (no 1602-02-2). All participants provided written informed consent. For any participant younger than 18 years, the parents provided written informed consent. A total of 34 travel events from Japan to Vietnam made by 19 travelers (age: 12–61 years, males: 11, length of stay in Vietnam: 2–12 days) between June 2015 and August 2016 were investigated.

An ~0.1 g fecal specimen was collected from each traveler within 6 days before and up to 21 days after each travel event, and 10 mL of sterile phosphate-buffered saline was added to each sample and homogenized. A 100 μL serially diluted homogenized solution was inoculated on CHROMagar™ ECC (Chromagar, Paris, France) containing 1 μg/mL of cefotaxime (CTX) and cultured at 37°C for 24 h. Three to seven colonies exhibiting the characteristics of E. coli were collected from each sample plate.

For E. coli identification, biochemical tests were conducted, including assays in lysine indole motility medium (Nissui, Tokyo, Japan), cellobiose lactose indole β-glucuronidase medium (Kyokuto Pharmaceutical Industrial, Tokyo, Japan), triple sugar iron slants (Nissui), and API 20E (bioMérieux, Marcy l’Etoile, France).

Susceptibility to antibiotics (ampicillin, fosfomycin, cefoxitin, CTX, ceftazidime, meropenem, gentamycin, kanamycin, streptomycin, tetracycline, ciprofloxacin, chloramphenicol, nalidixic acid, and sulfamethoxazole–trimethoprim) was determined by disk diffusion assay. ESBL production was determined by the double-disk synergy test using ceftazidime and CTX with and without clavulanic acid. Both antibiotic susceptibility and ESBL production tests were conducted as recommended by the Clinical and Laboratory Standards Institute.9

Bacterial DNA was extracted by boiling the bacterial suspension in tris(hydroxymethyl)aminomethane–ethylenediaminetetraacetic acid buffer. Genotyping of ESBL-encoding genes and phylogenetic groupings were determined by multiplex polymerase chain reaction (PCR).10 The presence of the colistin-resistant gene mcr-1 was also confirmed by PCR.3 For both ESBL-encoding genes and mcr-1, the PCR products were identified by sequencing.3,10 In mcr-1-positive strains, the colistin minimum inhibitory concentration (MIC) was evaluated by E-test® (bioMérieux).

The phylogenetic grouping of identified E. coli isolates was also determined by triplex PCR using a combination of two genes (chuA and yjaA) and the DNA fragment TSPE4. C2, as previously described.11

The clonal relationships of isolated E. coli strains were assessed by pulsed-field gel electrophoresis (PFGE), as previously described.12

The conjugation assay was performed as previously described.13

Statistical analysis was performed using Student’s t-test. The significance level was set at P<0.05.

Results

A total of 207 CTX-resistant E. coli strains (43 and 164 before and after travel, respectively) were isolated from the fecal specimens of travelers before and up to 2 weeks after traveling to Vietnam. Among these isolates, 175 strains were identified as ESBL-producing E. coli (Table 1).

Table 1

Detection of ESBL-producing Escherichia coli harboring mcr-1 before and after travel

Traveler	Traveling	Number of isolates
		Before travel		After travela
		ESBLb	mcr-1c	ESBL	mcr-1
1	November 2015
2	August 2016			3
3	August 2016			3
4	March 2016			6
	May 2016	3		6
5	May 2016			3
6	May 2016			2
7	February 2016
8	August 2015	3		3
	January 2016			1
9	August 2015			7
10	August 2015			7	1, B-1d
11	September 2015	2		6
12	September 2015			7
13	June 2015	4		7
	September 2015	3		7
	December 2015	3		3
	January 2016	3		6
14	July 2015	3
	November 2015			5
15	July 2015			6
	November 2015			4
16	June 2015			6
	December 2015			3
17	July 2015			3
	September 2015			2
	November 2015			5
18	June 2015			6
	July 2015	3		7	1, A-1
	September 2015	3		5
	November 2015
	January 2016			3
19	September 2015			6	1, C-1
	January 2016	3		4
Total			*
19	34				3

Notes:

Specimens were collected up to 2 weeks after traveling.

Extended-spectrum β-lactamase-producing E. coli.

mcr-1-positive E. coli.

Strain name.

P<0.001.

Abbreviation: ESBL, extended-spectrum β-lactamase.

A significant number of travel events (88.2%, 30/34) by the 19 travelers brought ESBL-producing E. coli back to Japan, while in 32.3% (11/34) of travel events, it was detected before traveling in seven travelers (Table 1). Among these, 42.9% of ESBL-producing E. coli isolates showed the same PFGE pattern as before traveling.

Genotyping of ESBL-producing isolates showed that blaCTX-M-1 group/blaTEM and blaCTX-M-9 group were the most prevalent genotypes, while blaCTX-M-2 and blaCTX-M-8 groups and blaSHV were not detected in travelers. Phylogenetic analysis of ESBL-producing E. coli isolates was also conducted. The most frequently detected phylogenetic group was D (41.9%, 62/142) followed by B2 (23.0%, 34/142), while B1 (11.5%, 17/142) and A (16.2%, 24/142) were the least common.

ESBL-producing E. coli isolates showed 100% resistance to ampicillin and CTX. In the case of quinolones, ESBL-producing E. coli isolates exhibited a rate of resistance of 76.1% (108/142) to nalidixic acid, and >50% of isolates were resistant to sulfamethoxazole–trimethoprim (66.9%, 95/142) and tetracycline (52.1%, 74/142).

Moreover, three events by three travelers brought back mcr-1. The three travelers with mcr-1, who visited Vietnam on business, did not carry mcr-1 before traveling. ESBL-producing E. coli harboring mcr-1 was identified by the co-occurrence of blaCTX-M and mcr-1. Furthermore, the transmissibility of these mcr-1 genes was confirmed by conjugation assay.

Characteristics of these three ESBL-producing mcr-1-positive E. coli isolates are summarized in Table 2. Traveler A, a 38-year-old man, traveled to Nha Trang by the way of Hanoi for 4 days. One day after returning to Japan, a fecal sample was collected and investigated for the presence of ESBL-producing mcr-1-positive E. coli, and three ESBL-producing isolates with different phylogenetic patterns were identified. These belonged to the blaCTX-M-9 group and the phylogenetic group A, the blaCTX-M-1 group/blaTEM and the phylogenetic group D, and the blaCTX-M-14 genotype and the phylogenetic group B2, the latter of which harbored mcr-1. However, no ESBL-producing E. coli harboring mcr-1 was detected in a sample collected 17 days after returning to Japan. Traveler B traveled to Ho Chi Minh City for 12 days. Three days after returning to Japan, a fecal specimen was collected and four types of ESBL-producing E. coli were identified. These belonged to the blaCTX-M-1 group and the phylogenetic group D, the blaCTX-M-9 group and the phylogenetic group A, the blaCTX-M-1 group/blaTEM and the phylogenetic group D, and the blaCTX-M-55/blaTEM genotype and the phylogenetic group B2, the latter of which harbored mcr-1. Traveler C traveled to Can Tho by the way of Ho Chi Minh City for 5 days. Six days after returning to Japan, a fecal specimen was collected and four types of ESBL-producing E. coli were identified. These belonged to the blaCTX-M-9 group and the phylogenetic group A, the blaCTX-M-1 group and the phylogenetic group D, and the blaCTX-M-14 genotype and the phylogenetic group D, the latter of which harbored mcr-1. In travelers B and C, no ESBL-producing E. coli harboring mcr-1 was detected 20 and 19 days, respectively, after returning to Japan.

Table 2

Characteristics of travelers carrying ESBL-producing Escherichia coli isolates harboring mcr-1

Variable	Traveler	A	B	C
Travelers	Age (years)	38	35	33
	Sex	Male	Male	Male
	Travel sites in Vietnam	Hanoi, Nha Trang	Ho Chi Minh City	Can Tho, Ho Chi Minh City
	Travel term	July 2015	August 2015	September 2015
	Length of stay in Vietnam	4 days	12 days	5 days
	Day of fecal sampling after returning to Japan	1 day	3 days	6 days
	Duration of carriage of mcr-1 after returning to Japan	<17 days	<20 days	<19 days
Bacteria	Isolates	A-1	B-1	C-1
	Phylogenetic group	B2	B2	D
	Antibiotic resistancea	AMP, CTX, CHL, TET, GEN, KAN	AMP, CTX, SXT, TET	AMP, CTX, NAL, CHL, SXT, TET, STR
	ESBL CTX-M gene	blaCTX-M-14	blaCTX-M-55	blaCTX-M-14
	Colistin resistance gene	mcr-1	mcr-1	mcr-1
	Colistin MIC (µg/mL)	1.0	3.0	4.0

Note:

Antibiotic resistance was determined by the Clinical & Laboratory Standards Institute guidelines: AMP, CTX, NAL, CHL, SXT, TET, GEN, KAN, and STR.

Abbreviations: AMP, ampicillin; CHL, chloramphenicol; CTX, cefotaxime; ESBL, extended-spectrum b-lactamase; GEN, gentamycin; KAN, kanamycin; MIC, minimum inhibitory concentration; NAL, nalidixic acid; STR, streptomycin; SXT, sulfa/trimethoprim; TET, tetracycline.

Discussion

Widespread dissemination of colistin-resistant MDR bacteria represents a worldwide threat owing to the limited number of effective antibiotics for treatment. In particular, the emergence of the mobile colistin-resistant gene mcr-1 suggests the possibility of the rapid dissemination of MDR bacteria with colistin resistance in a community because of the potential for horizontal gene transfer. In these circumstances, international travelers could act as carriers, transferring mcr-1-positive MDR bacteria to their home countries. Even if only a limited number of travelers carry mcr-1-positive bacteria, the possibility of genetic transfer and expansion of mcr-1 across borders seems to be likely. In this regard, there have been several reports of healthy travelers that carried ESBL-producing E. coli back to their home countries.1,14 However, it is not clear whether a traveler on a short trip can carry mcr-1 back to his/her home country.

In our previous reports, a high prevalence of human ESBL carriage (51%–71%) was demonstrated among Southeast Asian countries, such as Vietnam, Thailand, and Laos.11 In contrast, the dissemination of ESBL-producing E. coli is limited in Japan, where there is a low prevalence of human ESBL carriers.15 To date, only a limited number of bacterial isolates have been reported to harbor mcr-1 in E. coli isolates from healthy animals, but in sick swine, there were high mcr-1-positive bacteria detected in Japan.16,17

It has been reported that in Hanoi, Vietnam, 37.5% of food animals are contaminated with E. coli harboring mcr-1.4 In this study, we revealed that 15.8% of the travelers assessed became the carriers of ESBL-producing E. coli harboring mcr-1 after a short stay in Vietnam. Even though the details of these carriage mechanisms are not clear, our results indicate that travelers may come to resemble Vietnamese residents, a percentage of whom are carriers of mcr-1-positive bacteria (unpublished data), during their stay in Vietnam. The three travelers who became mcr-1 carriers ate breakfast in four-star hotels every morning, and they frequented middle-class restaurants for lunch and dinner. They experienced upper-class conditions during their stays in Vietnam and after returning to Japan. Their genotyping results were similar to those of healthy Vietnamese residents in terms of blaCTX-M genotypes, such as blaCTX-M-1 andblaCTX-M-9 groups.10 Conversely, an abundance of phylogenetic group D was observed in the isolates. However, the reason for this difference from healthy Vietnamese residents10 was not clear. It may be due to differences in intestinal flora between Japanese travelers and Vietnamese residents.

More than 40% of isolates from each traveler before and after travel showed the same PFGE patterns, suggesting low genetic diversity within each traveler. Further study regarding the clonal relationships with residents may provide valuable information on the carriage of MDR bacteria by travelers.

Since the intestinal flora in travelers can be changed by diet and environment factors during travel, it may be responsible for the carriage of ESBL-producing E. coli with mcr-1 after short-term travel.

Our follow-up study showed that the ESBL-producing E. coli harboring mcr-1 disappeared in the three travelers, who were not treated with any antibiotics, within 3 weeks of returning home. Therefore, the stability of the carriage of mcr-1-positive bacteria may be limited if there is no exposure to antibiotics.

The three ESBL-producing E. coli isolates harboring mcr-1 were further characterized, and the colistin MICs of the isolates were relatively low in comparison with those of isolates from food animals (4–8 μg/mL).4 Since antibiotic MICs can be increased by treatment with a high concentration of another antibiotic, the acquisition of high colistin MICs may be possible.

Conclusion

The findings in this study suggest that even a short-term international trip to a developing country may result in ESBL-producing mcr-1-positive E. coli being carried back to the travelers’ home countries. Although the dissemination of E. coli harboring mcr-1 is limited in developing countries, the number of international travelers continues to increase, thereby increasing the risk of spreading E. coli harboring mcr-1 into developed countries.