# CARBAPENEM-RESISTANT ENTEROBACTERIACEAE IN THE ASIA PACIFIC AND BEYOND

EDITED BY : Yi-Wei Tang, Barry N. Kreiswirth and Liang Chen PUBLISHED IN : Frontiers in Microbiology

### Frontiers Copyright Statement

© Copyright 2007-2019 Frontiers Media SA. All rights reserved. All content included on this site, such as text, graphics, logos, button icons, images, video/audio clips, downloads, data compilations and software, is the property of or is licensed to Frontiers Media SA ("Frontiers") or its licensees and/or subcontractors. The copyright in the text of individual articles is the property of their respective authors, subject to a license granted to Frontiers.

The compilation of articles constituting this e-book, wherever published, as well as the compilation of all other content on this site, is the exclusive property of Frontiers. For the conditions for downloading and copying of e-books from Frontiers' website, please see the Terms for Website Use. If purchasing Frontiers e-books from other websites or sources, the conditions of the website concerned apply.

Images and graphics not forming part of user-contributed materials may not be downloaded or copied without permission.

Individual articles may be downloaded and reproduced in accordance with the principles of the CC-BY licence subject to any copyright or other notices. They may not be re-sold as an e-book.

As author or other contributor you grant a CC-BY licence to others to reproduce your articles, including any graphics and third-party materials supplied by you, in accordance with the Conditions for Website Use and subject to any copyright notices which you include in connection with your articles and materials.

All copyright, and all rights therein, are protected by national and international copyright laws.

The above represents a summary only. For the full conditions see the Conditions for Authors and the Conditions for Website Use. ISSN 1664-8714 ISBN 978-2-88963-121-6 DOI 10.3389/978-2-88963-121-6

### About Frontiers

Frontiers is more than just an open-access publisher of scholarly articles: it is a pioneering approach to the world of academia, radically improving the way scholarly research is managed. The grand vision of Frontiers is a world where all people have an equal opportunity to seek, share and generate knowledge. Frontiers provides immediate and permanent online open access to all its publications, but this alone is not enough to realize our grand goals.

### Frontiers Journal Series

The Frontiers Journal Series is a multi-tier and interdisciplinary set of open-access, online journals, promising a paradigm shift from the current review, selection and dissemination processes in academic publishing. All Frontiers journals are driven by researchers for researchers; therefore, they constitute a service to the scholarly community. At the same time, the Frontiers Journal Series operates on a revolutionary invention, the tiered publishing system, initially addressing specific communities of scholars, and gradually climbing up to broader public understanding, thus serving the interests of the lay society, too.

### Dedication to Quality

Each Frontiers article is a landmark of the highest quality, thanks to genuinely collaborative interactions between authors and review editors, who include some of the world's best academicians. Research must be certified by peers before entering a stream of knowledge that may eventually reach the public - and shape society; therefore, Frontiers only applies the most rigorous and unbiased reviews.

Frontiers revolutionizes research publishing by freely delivering the most outstanding research, evaluated with no bias from both the academic and social point of view. By applying the most advanced information technologies, Frontiers is catapulting scholarly publishing into a new generation.

### What are Frontiers Research Topics?

Frontiers Research Topics are very popular trademarks of the Frontiers Journals Series: they are collections of at least ten articles, all centered on a particular subject. With their unique mix of varied contributions from Original Research to Review Articles, Frontiers Research Topics unify the most influential researchers, the latest key findings and historical advances in a hot research area! Find out more on how to host your own Frontiers Research Topic or contribute to one as an author by contacting the Frontiers Editorial Office: researchtopics@frontiersin.org

# CARBAPENEM-RESISTANT ENTEROBACTERIACEAE IN THE ASIA PACIFIC AND BEYOND

Topic Editors:

Yi-Wei Tang, Memorial Sloan Kettering Cancer Center, United States Barry N. Kreiswirth, University of Medicine and Dentistry of New Jersey, United States

Liang Chen, Public Health Research Institute (PHRI), United States

Image: Figure 4 from Cai et al. (2019)

Cai Y, Chen C, Zhao M, Yu X, Lan K, Liao K, Guo P, Zhang W, Ma X, He Y, Zeng J, Chen L, Jia W, Tang Y-W and Huang B (2019) High Prevalence of Metallo-ß-Lactamase-Producing Enterobacter cloacae From Three Tertiary Hospitals in China. *Front. Microbiol*. 10:1610. doi: 10.3389/fmicb.2019.01610

Citation: Tang, Y.-W., Kreiswirth, B. N., Chen, L., eds. (2019). Carbapenem-Resistant Enterobacteriaceae in the Asia Pacific and Beyond. Lausanne: Frontiers Media. doi: 10.3389/978-2-88963-121-6

# Table of Contents


Chendi Zhu, Veranja Liyanapathirana, Carmen Li, Vasanthi Pinto, Mamie Hui, Norman Lo, Kam T. Wong, Nilanthi Dissanayake and Margaret Ip

*20 Genotypic and Phenotypic Characterization of IncX3 Plasmid Carrying*  bla*NDM-7 in* Escherichia coli *Sequence Type 167 Isolated From a Patient With Urinary Tract Infection*

Yingying Hao, Chunhong Shao, Yuanyuan Bai and Yan Jin


Sara M. Przybysz, Carlos Correa-Martinez, Robin Köck, Karsten Becker and Frieder Schaumburg


Chau-Chyun Sheu, Ya-Ting Chang, Shang-Yi Lin, Yen-Hsu Chen and Po-Ren Hsueh


# Molecular Characterization of Carbapenem-Resistant *Enterobacter cloacae* in 11 Chinese Cities

Chunmei Jin1†, Jiangang Zhang2†, Qi Wang<sup>2</sup> , Hongbin Chen<sup>2</sup> , Xiaojuan Wang<sup>2</sup> , Yawei Zhang<sup>2</sup> and Hui Wang<sup>2</sup> \*

<sup>1</sup> Department of Clinical Laboratory, Yanbian University Hospital, Yanji, China, <sup>2</sup> Department of Clinical Laboratory, Peking University People's Hospital, Beijing, China

### *Edited by:*

Yi-Wei Tang, Memorial Sloan Kettering Cancer Center, United States

### *Reviewed by:*

Rong Zhang, Zhejiang University School of Medicine, China Remy A. Bonnin, Université Paris-Saclay, France

### *\*Correspondence:*

Hui Wang wanghui@pkuph.edu.cn; whuibj@163.com

†These authors have contributed equally to this work and are co-first authors.

### *Specialty section:*

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

*Received:* 27 February 2018 *Accepted:* 27 June 2018 *Published:* 17 July 2018

### *Citation:*

Jin C, Zhang J, Wang Q, Chen H, Wang X, Zhang Y and Wang H (2018) Molecular Characterization of Carbapenem-Resistant Enterobacter cloacae in 11 Chinese Cities. Front. Microbiol. 9:1597. doi: 10.3389/fmicb.2018.01597 Carbapenem-resistant Enterobacteriaceae (CRE) are usually resistant to most of antibiotics. Infections caused by such bacteria have a high mortality and pose a serious threat to clinical management and public health. Enterobacter cloacae ranks third among Enterobacteriaceae that cause nosocomial infections. In this study, the molecular characteristics of carbapenem-resistant E. cloacae in China were investigated. From November 2012 to August 2016, 55 non-repetitive strains of carbapenem-resistant E. cloacae were collected from 12 hospitals in 11 Chinese cities. The bacteria were identified with matrix-assisted laser desorption/ionization time of flight mass spectrometry. Antimicrobial susceptibility tests were determined by agar dilution method. Carbapenemase and other β-lactamase genes were detected with PCR and sequencing. Multilocus sequence typing and plasmid conjugation tests were performed. Among the 55 E. cloacae strains, 50 strains were detected to produce 8 types of carbapenemase including NDM-1, NDM-5, IMP-4, IMP-26, IMP-1, KPC-2, and VIM-1. NDM-1 accounted for 68.0% (34/50) among the carbapenemase-producing E. cloacae. A total of 24 sequence types were identified and ST418 was the most common, accounting for 20% (11/55). For further investigation, a pulsed-field gel electrophoresis (PFGE) assay was conducted to identify the PFGE patterns of the strains. These 23 isolates yielded 13 PFGE patterns, which were designated as type A–M. Eight isolates obtained from Shenzhen had the same PFGE pattern (type A) and the remaining 15 isolates belonged to the other 12 PFGE patterns (type B–M). The observation that 8 of the 15 blaNDM−1-positive E. cloacae isolates obtained from Shenzhen with the same PFGE pattern (type A) suggested a transmission outbreak of a common strain. S1-nuclease PFGE and Southern blotting were also conducted to estimate the size of plasmids harbored by blaNDM−1-positive strains. The results showed that the plasmids harboring the blaNDM−<sup>1</sup> gene ranged in size from approximately 52–58 kilobases. Our study indicates that carbapenem-resistant E. cloacae strains that produce NDM carbapenemase have strong resistance. Early detection and monitoring of the prevalence of these strains are urgent.

Keywords: carbapenem-resistance, *Enterobacter cloacae*, carbapenemase, NDM-1, ST418

**5**

# INTRODUCTION

In recent years, the emergence of carbapenem-resistant Enterobacteriaceae (CRE) has become a serious issue both on community-acquired infections and healthcare-associated infections (van Duin and Doi, 2017). As well as other Enterobacteriaceae, Enterobacter cloacae (E. cloacae) is a conditional pathogen found in the intestine. Healthcareassociated infections caused by E. cloacae ranked third among all the Enterobacteriaceae (Dai et al., 2013). Enterobacter cloacae can produce chromosome mediated AmpC β-lactamase and has resistance to ampicillin, amoxicillin/clavulanic, cephamycin and first and second generation cephalosporin. A wide spectrum of antibacterial drugs such as carbapenems may be used in treatment more often. Thus, multidrug resistance has emerged rapidly under antibiotic selection pressure. Carbapenem-resistant E. cloacae infections have been reported in many countries such as Spain, Australia, the United States, India, and China (Kiedrowski et al., 2014; Fernández et al., 2015; Liu et al., 2015; Sidjabat et al., 2015). The emergence of carbapenem-resistant E. cloacae is an enormous challenge to clinical treatment. It is well known that the main mechanism for reduced susceptibility to carbapenems in E. cloacae is the deregulation of ACT (the natural cephalosporinase of E. cloacae), which is associated with a decrease in membrane permeability. In addition to this, producing carbapenemases is another important mechanism of Enterobacteriaceae in carbapenem resistance (Walsh et al., 2005; Nordmann et al., 2009; Tzouvelekis et al., 2012). Also, the mechanism of combinations of either ESBL or AmpC and mutation of porins may hold a certain proportion (Yang et al., 2010).

Up until now, there was a lack of multicenter research on carbapenem-resistant E. cloacae in China. So, we conducted this molecular epidemiological study on carbapenem-resistant E. cloacae to further understand the prevalence of the bacteria in China.

# MATERIALS AND METHODS

### Sample Collection

From November 2012 to August 2016, we collected 55 unrepeated strains of carbapenem-resistant [any carbapenem (imipenem, meropenem, or ertapenem) as determined by standard methods] E. cloacae from 12 hospitals in 11 Chinese cities (Beijing, Chengde, Zunhua, Ji'nan, Xuzhou, Xi'an, Wuhan, Xiamen, Guangzhou, Dongguan, and Shenzhen; **Figure 1**). The participating hospitals include Peking University People's Hospital, Peking Union Medical College Hospital, Affiliated Hospital of Chengde Medical University, People's Hospital of Zunhua, Qilu Hospital of Shandong University, Affiliated Hospital of Xuzhou Medical University, Xijing Hospital, Tongji Hospital, The First Affiliated Hospital of Xiamen University, The First Affiliated Hospital Sun Yat-sen University, Donghua Hospital Sun Yat-sen University, and Shenzhen Second People's Hospital.

### Identification of the Bacterial Strains and Antimicrobial Susceptibility Tests

All isolates were identified with matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS) (Bruker Daltonics, Bremen, Germany). Minimum inhibitory concentrations (MICs) were determined by the agar dilution method according to CLSI guidelines (M100- S27). The tested drugs included ceftriaxone (Roche China, Shanghai, China), cefotaxime, ceftazidime, cefepime, aztreonam, amikacin, levofloxacin, minocycline, fosfomycin (National Institute for Food and Drug Control of China, Beijing, China), piperacillin/tazobactam, tigecycline (Pfizer, NY, USA), imipenem (Merck Sharp & Dohme, Hangzhou, China), meropenem (Sumitomo Pharmaceuticals, Suzhou, China), ciprofloxacin (Bayer, Leverkusen, Germany), and polymyxin B (Amresco, Solon, USA). Strains used in quality control were Escherichia coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853. The results were interpreted according to 2017 CLSI standards (M100-S27). The tigecycline test was performed according to the Food and Drug Administration standards.

### Detection of Antimicrobial Resistance Genes

Phenotypic screening for the resistance genes of carbapenemresistant E. cloacae strains was based on the 2017 CLSI guidelines. Modified Hodge test (MHT), imipenem-EDTA double-disk synergy test (DDST) (Lee et al., 2001), and modified carbapenem inactivation method (mCIM) were used to test carbapenemase production. Polymerase chain reaction (PCR) was used to detect carbapenemase genes (blaNDM, blaKPC, blaIMP, blaIMI, blaNMC, blaGES, blaSME, blaSIM, blaVIM, and blaOXA−48) and other β-lactamase genes (blaCTX−M, blaTEM, blaSHV, blaDHA, and blaCMY) (Lewis et al., 2007; Yang et al., 2010). The products were submitted for sequencing.

# Multilocus Sequence Typing (MLST)

MLST was performed according to a previously described method (https://pubmlst.org/ecloacae/). New alleles and sequence types were submitted to the MLST website and approved. Sequence Type Analysis and Recombinational Tests 2 (START2) (http://pubmlst.org/software/analysis/start2/) software was used to generate the phylogenetic tree (Jolley et al., 2001).

# Plasmid Conjugation Test

The plasmid conjugation test was used to test carbapenemresistant gene transfer. Ten strains were selected for the test. Escherichia coli EC600 (rifampicin resistant) was used as the recipient. Conjugants were screened using China blue lactose agar plates containing rifampicin (300µg/ml) and imipenem (1µg/ml). The donor and the recipient were mixed at a ratio of 1:1 for 24 h. Transconjugants were selected on China blue lactose agar plates (OXOID, Basingstoke Hampshire, UK), supplemented with rifampicin (100µg/ml) and imipenem (1µg/ml). PCR was used to screen for blaNDM−1, blaVIM−1, blaKPC−2, and blaIMP−<sup>1</sup> as previously described (Wang et al., 2014).

### Pulsed-Field Gel Electrophoresis (PFGE)

Enterobacter cloacae isolates were characterized by PFGE according to the previously published protocol by Ribot et al., with modifications (Ribot et al., 2002). We selected 23 blaNDM−1 positive isolates (including the 15 isolates from Shenzhen, 6

TABLE 1 | In vitro activities of antimicrobial agents against carbapenemase-producing Enterobacteriaceae.


isolates involved in the conjugation experiments, and another 2 representative isolates).

Electrophoresis conditions were altered to have an initial switch time of 2.16 s and a final switch time of 54.17 s, and gels were run for 18 h. The resulting PFGE patterns were analyzed in BioNumerics software (Applied Maths, Austin, TX, USA) with dendrograms based on the Dice coefficient with a band position tolerance of 1%. Patterns with no discernible differences were considered indistinguishable and given the same PFGE pattern designation.

### S1-Nuclease PFGE and Southern Blotting

S1-nuclease PFGE and Southern blotting were performed to estimate the size of plasmids harbored by blaNDM−1-positive


ecl, Enterobacter cloacae; cas, Enterobacter asburiae; ICU, intensive care unit; EICU, emergency intensive care unit; ba, broncho-alveolar lavage; bl, blood; ca, catheter; dr, drainage; sp, sputum; ur, urine; wd, wound; EDTA-DDST, EDTA double-disk synergy test.

strains as described previously. We selected 23 blaNDM−1 positive isolates as mentioned above.

The blaNDM−<sup>1</sup> gene was detected by digoxigenin-labeled specific probes (DIG High Prime DNA Labeling and Detection Starter Kit II, Roche Diagnostics, Mannheim, Germany). Salmonella enterica H9812 was used as a size marker.

### Statistical Analyses

WHONET (version 5.6) software (http://www.whonet. org/software.html) and SPSS (version 22.0) software (SPSS Inc., Chicago, IL, USA) were used for statistical analyses.

### Ethical Approval

This study was approved by the research ethics board at Peking University People's Hospital. Informed consent was not needed as this study was retrospective and participants were anonymized. Medical records and patient's information were retrospectively reviewed and collected.

# RESULTS

### Characteristics of Collected Samples

The most common specimens were respiratory tract (20 cases, 36.4%), followed by urine (13 cases, 23.6%), blood (11 cases, 20%), ascitic fluid (7 cases, 12.7%), bile (2 cases, 3.6%), catheter (1 case, 1.8%), and wound (1 case, 1.8%).

### Antimicrobial Susceptibility Tests

Of all the antimicrobials tested, the most susceptible antimicrobial was polymyxin B (100%, 55/55), followed by amikacin (89.1%, 49/55), fosfomycin (80%, 44/55), tigecycline (78.2%, 43/55), minocycline (52.7%, 29/55), levofloxacin (30.9%, 17/55), and ciprofloxacin (25.5%, 14/55). All of the 36 blaNDM-positive strains were resistant to piperacillin-tazobactam, ceftazidime, cefotaxime, ceftriaxone, cefepime, imipenem, and meropenem. It is worth noting that the most susceptible antimicrobials to blaNDMpositive strains was polymyxin B (100%, 36/36), followed by amikacin (86.1%, 31/36), and fosfomycin (86.1%, 31/36). There was some differences between blaNDMpositive strains and strains with other carbapenemases (**Table 1**).

### Genotype Analysis

Among the 55 strains, 50 were confirmed to produce 8 types of carbapenemases including NDM-1, NDM-5, IMP-4, IMP-26, IMP-1, KPC-2, and VIM-1. The corresponding numbers of the strains that produced the foregoing types of carbapenemases were 34, 2, 6, 3, 2, 2, and 1. Other carbapenemase genes were not detected. No strains contained two or more carbapenemase genes. NDM-1-producing carbapenem-resistant E. cloacae was primarily distributed in Shenzhen (**Table 2**). Carbapenemase genes were not detected in the other 5 strains.

NDM-1-producing E. cloacae isolates were mainly collected from Shenzhen (44.1%, 15/34), followed by Xuzhou (20.6%, 7/34), Beijing, Dongguan, Guangzhou, Ji'nan, Xi'an, Xiamen, Wuhan, and Zunhua. These samples were primarily collected from the Intensive Care Unit and the Emergency Intensive Care Unit (41.2%, 14/34), followed by the department of neurosurgery (23.5%, 8/34). NDM-1-producing E. cloacae isolates were most commonly identified in sputum samples (35.3%, 12/34), followed by urine samples (32.4%, 11/34). All strains were positive for MHT, imipenem-EDTA-DDST, and mCIM. Results of the three tests were consistent. In addition, 6 strains also produced the AmpC enzyme DHA-1, 8 strains produced CTX-M-3, and 2 strains produced both CTX-M and DHA-1.

# MLST

The results of the MLST are shown in **Figure 2**. A total of 24 sequence types were detected in the 55 E. cloacae strains. ST418 was the most common (20%, 11/55), followed by ST93 (14.5%, 8/55).


MEM, meropenem; IMP, imipenem; FEP, cefepime; CAZ, ceftazidime; TZP, piperacillin/tazobactam; ATM, aztreonam; AMK, amikacin; CIP, ciprofloxacin; LVX, levofloxacin; PB, polymyxin B; TGC, tigecycline; TC, transconjugant strain; EC600, recipient strain.

# Plasmid Conjugation Test

Plasmids from 9 carbapenemase-producing E. cloacae strains were successfully transferred to E. coli EC600. Drug sensitivity tests showed the MICs of meropenem increased by 5- to 8-fold in the 9 conjugators; for imipenem, cefepime, ceftazidime, and piperacillin/tazobactam, the MICs increased by 3∼6-, 7∼11-, 7∼10-, and 3∼7-fold, respectively (**Table 3**).

### PFGE, S1-Nuclease PFGE and Southern Blotting

When typed by PFGE to determine if they were related, the 23 isolates yielded 13 PFGE patterns, which were designated as type A–M. Eight isolates (ecl768, ecl771, ecl774, ecl776, ecl779, ecl782, ecl784, and ecl786) obtained from Shenzhen had the same PFGE pattern (type A) and the remaining 15 isolates belonged to the other 12 PFGE patterns (type B–M) (**Table 2**). The observation that 8 of the 15 blaNDM−1-positive E. cloacae isolates with the same PFGE pattern (type A) and the same sequence type (ST418) suggested a transmission outbreak of a common strain.

The results of S1-nuclease PFGE and Southern blotting showed that the plasmids harboring the blaNDM−<sup>1</sup> gene ranged in size from approximately 52–58 kilobases, respectively (**Table 2**). The plasmids harboring the blaNDM−<sup>1</sup> gene of the 8 isolates obtained from Shenzhen were the same size (approximately 52 kilobases).

# DISCUSSION

Carbapenemase-producing E. cloacae has been reported in many countries, such as strains producing OXA-48 and VIM-1 have been reported in Spain. In Brazil, Australia and America, strains producing NDM-1, IMP-4, and KPC-3, respectively, have been reported (Kiedrowski et al., 2014; Rozales et al., 2014; Villa et al., 2014; Fernández et al., 2015; Sidjabat et al., 2015), while in Chongqing, Henan and Ningxia of China, strains producing NDM-1 have been identified (Dai et al., 2013; Liu et al., 2015; Shi et al., 2017). Strains that produced other carbapenemases have also been reported in the Sichuan province of China (Huang et al., 2015). In the present study, we found that the E. cloacae prevalent in China mainly produced NDM-1 (68.0%, 34/50) and IMP-4 (12.0%, 6/50). NDM-1 was found in the highest proportion and may represent a significant drug-resistant mechanism of carbapenem-producing Enterobacteriaceae in China.

The plasmid conjugation test was completed with 6 blaNDM−1-positive strains. Conjugants were all detected to have the blaNDM−1. Susceptibility results showed that compared with the receptor bacteria EC600, the conjugants have a higher MIC value on cephalosporins and carbapenems. There were no MIC promotions on polymyxin B and tigecycline between conjugants and EC600. Many studies have demonstrated that the plasmid owned blaNDM also have other resistant genes, such as blaTEM−1, blaCMY, qnrA6, and qnrB1 for quinolone resistance, armA, rmtA, and rmtC for aminoglycoside resistance (Poirel et al., 2011a,b; Kocsis et al., 2016). But all strains in this study have no blaTEM−<sup>1</sup> and blaCMY. Five of the conjugants have no MIC difference on quinolone with the EC600. Maybe the relative plasmid did not harbor the quinolone resistant gene.

MLST showed subtype diversity. A total of 24 sequence types were detected in 55 E. cloacae strains. ST418 was detected the most frequently (11/55, 20%), and the second was ST93 (14.5%, 8/55). Three new sequence types were found, namely ST916, ST919, and ST920. Our study reveals the diversity of carbapenem-resistant E. cloacae and the difference in genetic affinity, which is consistent with the study of Gomez-Simmonds et al. (2016). Our study showed that ST418 is the main epidemic strain in Shenzhen in China; while in America, Central de Asturias of Spain, and the Henan province of China, the main epidemic strains were ST171, ST74, and ST120, respectively (Fernández et al., 2015; Liu et al., 2015; Gomez-Simmonds et al., 2016). We found that ST418 was genetically closer to ST127 and ST755 with START2 analysis. Studies have found that all ST418 strains produced NDM-1 carbapenemase, indicating that there might be a small outbreak of NDM-1-ST418 carbapenem-resistant E. cloacae in Shenzhen and Dongguan City of Guangdong province of China. In this study, ST78-NDM-1 type carbapenem-resistant E. cloacae was also found in Xuzhou, Xi'an, Wuhan, and Xiamen, which should be taken seriously concern.

In conclusion, our study indicates that ST418, which produces NDM-1 carbapenemase, is the main epidemic strain of carbapenem-resistant E. cloacae in Shenzhen and Dongguan City of China. Early detection and monitoring are necessary to prevent the further spread of the bacteria.

# AUTHOR CONTRIBUTIONS

HW conceived and designed the study. CJ and JZ wrote this paper. CJ, QW, and JZ performed the experiments. QW and JZ analyzed the data. HC, XW, and YZ assisted CJ and JZ to finish the experiments. All authors approved the final version.

# FUNDING

This study was supported by National Natural Science Foundation of China (grant No.81625014).

# ACKNOWLEDGMENTS

We thank all the hospitals that provided carbapenem-resistant E. cloacae isolates (Peking University People's Hospital, Peking Union Medical College Hospital, Affiliated Hospital of Chengde Medical University, People's Hospital of Zunhua, Qilu Hospital of Shandong University, Affiliated Hospital of Xuzhou Medical University, Xijing Hospital, Tongji Hospital, The First Affiliated Hospital of Xiamen University, The First Affiliated Hospital Sun Yat-sen University, Donghua Hospital Sun Yat-sen University, and Shenzhen Second People's Hospital).

# REFERENCES


**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2018 Jin, Zhang, Wang, Chen, Wang, Zhang and Wang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

# Characterizing Mobilized Virulence Factors and Multidrug Resistance Genes in Carbapenemase-Producing Klebsiella pneumoniae in a Sri Lankan Hospital

Chendi Zhu<sup>1</sup>† , Veranja Liyanapathirana<sup>2</sup>† , Carmen Li<sup>1</sup> , Vasanthi Pinto<sup>3</sup> , Mamie Hui<sup>1</sup> , Norman Lo<sup>1</sup> , Kam T. Wong<sup>1</sup> , Nilanthi Dissanayake<sup>2</sup> and Margaret Ip<sup>1</sup> \*

<sup>1</sup> Department of Microbiology, The Chinese University of Hong Kong, Shatin, Hong Kong, <sup>2</sup> Department of Microbiology, Faculty of Medicine, University of Peradeniya, Kandy, Sri Lanka, <sup>3</sup> Department of Anesthesiology and Critical Care, Faculty of Medicine, University of Peradeniya, Kandy, Sri Lanka

### Edited by:

Yi-Wei Tang, Memorial Sloan Kettering Cancer Center, United States

### Reviewed by:

Miklos Fuzi, Semmelweis University, Hungary Liang Li, Los Angeles Biomedical Research Institute, United States

### \*Correspondence:

Margaret Ip margaretip@cuhk.edu.hk †These authors have contributed equally to this work

### Specialty section:

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

Received: 14 June 2018 Accepted: 13 August 2018 Published: 31 August 2018

### Citation:

Zhu C, Liyanapathirana V, Li C, Pinto V, Hui M, Lo N, Wong KT, Dissanayake N and Ip M (2018) Characterizing Mobilized Virulence Factors and Multidrug Resistance Genes in Carbapenemase-Producing Klebsiella pneumoniae in a Sri Lankan Hospital. Front. Microbiol. 9:2044. doi: 10.3389/fmicb.2018.02044 Limited data is available on the epidemiology and characteristics of carbapenemresistant Enterobacteriaceae (CRE) and their associated plasmids or virulence determinants from Sri Lanka. Through whole genome sequencing of CREs from the intensive care units of a Sri Lankan teaching hospital, we identified a carbapenemase gene, blaOXA−<sup>181</sup> in 10 carbapenemase-producing Klebsiella pneumoniae isolates (two strains of ST437 and eight strains of ST147) from 379 respiratory specimens. blaOXA−<sup>181</sup> was carried in three variants of ColE-type plasmids. K. pneumoniae strains with ompK36 variants showed high minimum inhibitory concentrations to carbapenem. Furthermore, genes encoding for extended spectrum β-lactamases (ESBL), plasmidmediated quinolone resistance (PMQR) determinants (qnr, aac(6<sup>0</sup> )-Ib-cr, and oqxAB) were present in all 10 strains. Amino acid substitution in chromosomal quinolone resistance-determining regions (QRDRs) gyrA (Ser83Ile) and parC (Ser80Ile) were also observed. All strains had yersiniabactin genes on mobile element ICEkp. Strict infection control practices and judicious use of antibiotics are warranted to prevent further spread of multidrug-resistant K. pneumoniae.

Keywords: OXA-181, quinolone, yersiniabactin, Klebsiella pneumoniae, Sri Lanka

### INTRODUCTION

Carbapenem-resistant Enterobacteriaceae (CRE) is a global threat and infections caused by CRE are associated with high morbidity and mortality (Centers for Disease Control and Prevention, 2013). Among all the resistance mechanisms, plasmid-mediated horizontal transfer of carbapenemase genes is the main route for acquiring resistance in CRE (Nordmann et al., 2012). These mobile elements are capable of transferring resistance between different lineages and thus, pose a potential for dissemination. Three types of carbapenemases (class A: blaKPC; class B: blaNDM, blaIMP, and blaVIM; class D: blaOXA−48−like) can hydrolyze carbapenems at varying levels along with other mechanisms, such as porin mutation and overexpression of efflux pump proteins (Poirel et al., 2012; Zhang et al., 2014; Lunha et al., 2016). Klebsiella pneumoniae clonal complex 258 (CC258)

have been known to associate with epidemic plasmids carrying numerous antimicrobial resistance genes and virulence factors like yersiniabactin. These interactions were hypothesized to provide a survival advantage for these clones (Holt et al., 2015). Sri Lanka sits in the southern tip of the Indian subcontinent. Although the epidemiology and diversity of carbapenemases, have been reported in India, the epidemiology and characteristics of the CRE and resistance determinants in Sri Lanka is lacking. Furthermore, the single study currently available on CRE from Sri Lanka does not contain information on associated plasmids or virulence determinants (Hall et al., 2014). Hence, we sequenced the CRE strains isolated from patient specimens in the intensive care units of one Sri Lankan hospital using whole genome sequencing (WGS) to describe their antimicrobial resistance and genetic profiles so as to provide a more in-depth view of the CRE in Sri Lanka.

# MATERIALS AND METHODS

### Bacterial Isolates

Single patient isolates were obtained from the respiratory specimens received from inpatients admitted to the intensive care units of the Teaching Hospital, Peradeniya, Sri Lanka between February to September 2015. During this period, a total of 379 respiratory specimens were processed and 64 coliforms were obtained. Among these coliforms, 2.6% (10 isolates) were found to be resistant to carbapenems using Stokes sensitivity testing method (Sri Lanka College of Microbiologists, 2011) and were saved for further study. The study was approved by the Institutional Ethical Committee of the Faculty of Medicine, University of Peradeniya.

### Bacterial Identification and Antimicrobial Resistance

Bacterial strains were plated on blood agar and incubated at 37◦C overnight and the identification of all strains were confirmed by matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry at the Department of Microbiology, the Chinese University of Hong Kong. Minimum inhibitory concentrations (MICs) of antimicrobials were determined by the microbroth dilution method according to Clinical and Laboratory Standards Institute (CLSI) guideline (CLSI, 2018). The following antibiotics were tested: amikacin (AK); ceftazidime (CAZ); ciprofloxacin (CIP); gentamicin (GN); colistin (CT); cefotaxime (CTX); ertapenem (ETP); fosfomycin (FOS); imipenem (IPM); meropenem (MEM); and tigecycline (TG).

### Whole Genome Sequencing and Data Analysis

Bacterial DNA was extracted with Wizard genomic DNA purification kit (Promega, Madison, WI, United States). WGS was performed using the Illumina HiSeq 2500 platform, and unique index-tagged libraries were created for each sample to generate 90 bp paired-end reads (Global Biologics, LLC). The libraries gave 100× average coverage for each strain. Quality control of the raw reads was performed by FastQC (Andrews, 2010). Genomes were assembled using SPAdes assembler (version 3.5.0) (Bankevich et al., 2012). Contigs of ≥500 bp from each genome were included in the analyses. Prokka (version 1.9) software was used for genome annotation, including ORF finding and gene function annotation (Seemann, 2014). Raw reads and assembled contigs were used for multilocus sequence typing (MLST) analysis. SRST2 (Version 0.1.5) was used to map raw reads to the following databases: pubMLST database for sequence types; ARG-ANNOT V3 database for resistance genes; PlasmidFinder database (Updated 20170220) for plasmid replicons; and K. pneumoniae BIGSdb virulence gene database for virulence genes at http://bigsdb.web.pasteur.fr (Accession date: 20180313) (Carattoli et al., 2014; Gupta et al., 2014; Inouye et al., 2014). The contig containing carbapenemase for each genome was matched against the public database with NCBI BLAST to find the top hit plasmids of at least 95% identity and 95% coverage and were matched back to the plasmids in our genome dataset to extract possible plasmid contigs (Altschul et al., 1990). Extracted contigs were further aligned to the reference plasmid using Mauve Alignment software to check for similarity and coverage (Darling et al., 2004). Gaps were filled by PCR with primers designed from our genome data. Pan-genome dendrogram was created by Roary (Page et al., 2015). Contig files of all 10 strains in this study were deposited to GenBank BioProject under the accession number PRJNA439172.

# RESULTS

### Antimicrobial Susceptibilities and Molecular Characteristics of the CRE Isolates

Of the 10 CRE isolates included in the study, three were from the subsidiary ICU and seven were from the main ICU. All strains were confirmed K. pneumoniae and the susceptibility testing revealed all strains to be resistant to CIP, CTX, and sensitive to FOS and AK, while MIC for CT was ≤0.5 mg/L and MIC for TG was ≤2 mg/L (**Table 1**). All except one strain (SL50) were resistant to CAZ. All of the strains except SL33, SL49, and SL54 were resistant to ETP or MEM with MICs ≥16 and ≥8 mg/L, respectively, while resistance to IPM varied among isolates. In previous studies, mutations of K. pneumoniae outer membrane proteins ompK35 and ompK36 were found to confer increased MIC to carbapenems (Garcia-Fernandez et al., 2010; Zhang et al., 2014; Lunha et al., 2016). There were neither mutations nor insertions at ompK35 and its promoter in all of our strains when we aligned them to the wild-type gene in public database (GenBank Accession No. AJ011501). OmpK36 was identical in six highly resistant strains (ompK36 sl1). SL33 and SL49 shared the same gene (ompK36-sl2), but SL54 contained a unique variation (ompK36-sl3). The difference of ompK36-sl1 and ompK36-sl2 was the insertion of amino



∗∗∗Length of stay from date of specimen collection.

FIGURE 1 | Pan-genome dendrogram of 10 Sri Lanka strains with annotation. <sup>∗</sup>Red boxes indicate presence and yellow for absence.

acids glycine and aspartate after PEFXG domain within the L3 loop and one mutation in loop L4 and alpha-helix, respectively. The PEFXG domain (porin size determinant) insertion was also observed in several studies that addressed strains with high resistance to carbapenems (Garcia-Fernandez et al., 2010; Hall et al., 2014; Zhang et al., 2014). OmpK36-sl3 was different from the others in the region between loop L3 and loop L6, but there was no insertion interruption in the PEFXG domain.

The K. pneumoniae isolates belonged to ST437 (2 strains) and ST147 (8 strains) (**Table 1**). Capsular genes showed two wzi alleles 64 and 109 in strains ST147 and ST437, respectively (**Figure 1**). Only one carbapenemase gene, blaOXA−<sup>181</sup> was found in all isolates. All isolates harbored β-lactamase genes, including blaTEM−1, and blaOXA−1, blaSHV−11, and to extendedβ-lactamase, blaCTX−M−15. Plasmid-mediated quinolone resistance (PMQR) determinants (qnr, aac(6<sup>0</sup> )-Ib-cr, and oqxAB) were also detected. We also found two amino acid substitutions at gyrA (S83I) and parC (S80I) in all strains, which have been frequently reported in quinolone-resistant K. pneumoniae worldwide (Aldred et al., 2014).

The virulence profiles among the 10 strains were the same and all carried yersiniabactin genes (ybt, irp1, irp2, and fuyA) and kfu, mrk, but were absent of rmpA or rmpA2 genes. Yersiniabactin genes were found on mobile element ICEkp. 4437 genes and 2182 accessory genes were used to establish the dendrogram of their pan-genomes (**Figure 1**). The distinct accessory gene profiles of ST437 and ST147 indicated that ST437 was a distant cluster. The pan-genome dendrogram also revealed a cluster of ST147 (SL36, SL46, and SL68) that contained fewer resistance genes (**Figure 1**) and were sensitive to GN and with lower MIC (8 mg/L) to CIP compared with other ST147 strains (**Table 1**).

# Plasmids Harboring blaOXA−<sup>181</sup>

fmicb-09-02044 August 29, 2018 Time: 20:28 # 5

Three different ColE-type plasmids were identified (**Figure 2**). One plasmid (CUHK\_SL-A) was identical to KP3 (GenBank accession no. JN205800) and was found in the two ST437 strains. CUHK\_SL-A was a short plasmid (7,606 bp) harboring only one resistance gene and was previously well described (Potron et al., 2011a). Another plasmid (CUHK\_SL-B) was found in 7 ST147 strains of this study with a deletion of an insertion sequence (ISEcp1) when compared to CUHK\_SL-A. Several studies from the United States, Germany, and France have reported this plasmid in both K. pneumoniae and Escherichia coli (GenBank Accession No. CP006802, CP016038, and JX423831). The third plasmid (CUHK\_SL-C) was found in strain SL54 with a ∗∗mobile gene deletion (mobC) when compared to CUHK\_SL-B. This plasmid was not reported before.

# DISCUSSION

Carbapenemase blaOXA−<sup>181</sup> was first described in India as a Class D blaOXA−48−like enzyme from clinical samples in year 2006 and 2007 (Castanheira et al., 2011). It was thought to originate from an environmental strain as a chromosomal gene (Potron et al., 2011b). Although it has been detected worldwide, most of the patients have a travel history to the Indian subcontinent, especially India (Poirel et al., 2012). In 2014, blaOXA−<sup>181</sup> and blaNDM were reported in K. pneumoniae in Sri Lanka, and included mainly ST14 and ST147 (Hall et al., 2014). ST437 belongs to CC258 and is a single locus variant of the globally prevalent ST258, carrying blaKPC in Brazil and blaNDM, blaOXA−<sup>245</sup> (plasmid: IncL/M) in Spain (Seki et al., 2011; Oteo et al., 2013). blaOXA−<sup>181</sup> was previously described in plasmid ColE, IncT and in the chromosome, which were all isolated from patients transferred from India, as well as described in plasmid IncX3 from China (Potron et al., 2011a; Dimou et al., 2012; Villa et al., 2013; Kayama et al., 2015; Liu et al., 2015; Partridge et al., 2015). ColE plasmid encoding various β-lactamase genes and blaOXA−<sup>181</sup> is related to transposon Tn2013. Insertion sequence ISEcp1 was considered to be associated with blaOXA−<sup>181</sup> acquisition, and its deletion may stabilize the resistance gene in the plasmid (Potron et al., 2013). mobC gene deletion in the strains may affect the frequency of conjugal plasmid mobilization (Zhang and Meyer, 1997).

In this study, all strains harbored quinolone-resistant determinants with quinolone resistance-determining region (QRDR) mutation on their chromosomes. A recent epidemiology study has shown the correlation of quinolone consumption and CRE in United States military health system, and another case-control study of CRE outbreaks in the Netherlands have determined quinolone use as the only risk factor for the acquisition of blaOXA−48−like producing Enterobacteriaceae compared to other antibiotics use (Lesho et al., 2015; Dautzenberg et al., 2016). Possible mechanisms may be due to co-transfer of two plasmids bearing dual resistance mechanisms or recombination into one plasmid as described before (Rodríguez Martínez et al., 2014; Liu et al., 2015). Furthermore, the impact of different plasmids and fluoroquinolone resistant determinants on fitness cost has been found to vary with strains of different STs. This may have contributed to the spread of this particular clone (Tóth et al., 2014; Johnson et al., 2015). It was suggested that energetically favorable mutations with the double serine mutations in the QRDR region in the current study have been described to favor the fitness and dissemination of such clones (Fuzi et al., 2017). Although all strains were non-hypervirulent K. pneumoniae (negative for rmpA/rmpA2 genes, non K1/K2 capsule serotypes), all strains encoded several yersiniabactin genes. These genes being on the integrative conjugative elements (ICEKp) can contribute to the spread of K. pneumoniae in the population and may serve as a predictor of invasive infection (Lam et al., 2018).

In conclusion, we explored the genetic profile of multidrug resistant CRE in Sri Lanka. blaOXA−<sup>181</sup> (through ColE-type plasmid) and yersiniabactin genes have disseminated to different STs of K. pneumoniae. We recommend active surveillance of high risk inpatients and long term studies to determine possible intra-unit transfer and facilitate infection control, especially as the length of stay in all instances where data was available is >72 h. Judicious antibiotics use especially that of quinolones is also recommended.

# AUTHOR CONTRIBUTIONS

VP and ND conceptualized, collected, preliminarily analyzed, and approved the manuscript. MI, MH, and VL conceptualized, conducted, analyzed, drafted, and approved the manuscript. CZ, NL, KW, and CL contributed to sample storage and preparation, molecular analysis, and approval of the manuscript. CZ prepared the first draft of the manuscript.

# FUNDING

This research was supported by Health and Medical Research Fund, Food and Health Bureau of the Hong Kong Special Administrative Region to PI: MI, Grant no: CU-15-B4 and CU-16-C8.

# ACKNOWLEDGMENTS

We thank the Food and Health Bureau of Hong Kong Special Administrative Region for the support from Health and Medical Research Commissioned Fund.

### REFERENCES

fmicb-09-02044 August 29, 2018 Time: 20:28 # 6


type 131 compared with other fluoroquinolone-resistant E. coli. Antimicrob. Agents Chemother. 59, 4471–4480. doi: 10.1128/AAC.00673-15



resistance in clinical isolates of KPC-2-producing Klebsiella pneumoniae. Antimicrob. Agents Chemother. 58, 1214–1217. doi: 10.1128/AAC.02 045-12

**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2018 Zhu, Liyanapathirana, Li, Pinto, Hui, Lo, Wong, Dissanayake and Ip. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

fmicb-09-02468 October 20, 2018 Time: 14:53 # 1

# Genotypic and Phenotypic Characterization of IncX3 Plasmid Carrying blaNDM-7 in Escherichia coli Sequence Type 167 Isolated From a Patient With Urinary Tract Infection

Yingying Hao, Chunhong Shao, Yuanyuan Bai and Yan Jin\*

Department of Clinical Laboratory, Shandong Provincial Hospital Affiliated to Shandong University, Jinan, China

### Edited by:

Yi-Wei Tang, Memorial Sloan Kettering Cancer Center, United States

### Reviewed by:

Dave Siak-Wei Ow, Bioprocessing Technology Institute (A∗STAR), Singapore Rong Zhang, Zhejiang University School of Medicine, China

> \*Correspondence: Yan Jin

sdslyyjy@163.com

### Specialty section:

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

Received: 08 July 2018 Accepted: 26 September 2018 Published: 23 October 2018

### Citation:

Hao Y, Shao C, Bai Y and Jin Y (2018) Genotypic and Phenotypic Characterization of IncX3 Plasmid Carrying blaNDM-7 in Escherichia coli Sequence Type 167 Isolated From a Patient With Urinary Tract Infection. Front. Microbiol. 9:2468. doi: 10.3389/fmicb.2018.02468 Infections due to New Delhi metallo-beta lactamase (NDM)-7-producing Escherichia coli are infrequent and sporadic. In this study, we report one case of recurrent urinary tract infection caused by blaNDM−7-producing E. coli belonging to phylogenetic group A, sequence type (ST) 167. In this study, we aimed to describe the genotype and phenotype of blaNDM−7-producing E. coli in China. The isolate exhibited resistance to β-lactam antimicrobials, trimethoprim-sulfamethoxazole, quinolones, and aminoglycosides. blaNDM−<sup>7</sup> is located on a conjugative plasmid designated pJN05NDM-7 belonging to type IncX3. pJN05NDM-7 was fully sequenced and compared with all publicly available blaNDM−7-harboring plasmids. pJN05NDM-7 is almost identical to pKpN01-NDM7 and pKW53T, although the plasmids are geographically unrelated. The comparison of IncX3 plasmids harboring blaNDM in China showed high similarity, with genetic differences within insertion fragments. Notably, the differences in plasmids of animal and human origin were insignificant, because only one plasmid showed deletion inside the ISAba125 region compared with pJN05NDM7. Our study demonstrates that E. coli carrying IncX3 plasmids play an important role as a reservoir and in the spread of blaNDM. Further studies should be performed to control the dissemination of blaNDM among food animals.

Keywords: NDM-7, carbapenemase, Escherichia coli, multi-drug resistance, China

# INTRODUCTION

New Delhi metallo-beta-lactamase (NDM)-producing bacteria are spread worldwide and pose a serious threat to public health, and is highly disseminated in China (Zhang et al., 2017; Liu et al., 2018). The surveillance for carbapenem-resistant Enterobacteriaceae (CRE) showed that blaNDM production was the second major mechanism of carbapenem resistance in Escherichia coli, and blaNDM−<sup>1</sup> was the most frequent variant (Khan et al., 2017). Since the first report on NDM-1 fmicb-09-02468 October 20, 2018 Time: 14:53 # 2

in 2009, 20 variants of NDM have been assigned in the Lahey Clinic database (Liu et al., 2018). NDM-7, which differs from NDM-1 by two point mutations corresponding to amino acid substitutions, was described in 2013 with increased carbapenemase activity compared with NDM-1 (Cuzon et al., 2013). blaNDM−<sup>7</sup> is infrequently detected, and sporadic cases of infections due to blaNDM−7-producing enterobacteria have been reported in France, India, the United States of America, and Japan (Cuzon et al., 2013; Chen et al., 2015; Wang et al., 2016; Devanga Ragupathi et al., 2017; Pal et al., 2017; Sugawara et al., 2017; Espinal et al., 2018). In China, blaNDM−7 producing E. coli ST131 was first reported in 2016; however, the genetic content of blaNDM−7-harboring plasmids was not clearly described (Wang et al., 2016). Considering its global distribution, increasing attention should be paid to epidemiological survey of blaNDM−7.

In this study, we detected a blaNDM−7-producing E. coli isolate from a patient without a history of traveling admitted in a Chinese Hospital. To elucidate the molecular epidemiology and evolutionary dynamics involved in the dissemination of blaNDM, the genomic content and in-depth molecular characterization of the strains was determined in this study.

# MATERIALS AND METHODS

### Bacterial Strains

The carbapenem-resistant E. coli strain JN05 was recovered from urine sample of a 61-year-old woman with recurrent urinary tract infection at a teaching hospital in Shandong Province of China in 2015. The patient was diagnosed with vesicovaginal fistula secondary to cervical cancer after chemotherapy and electrocautery surgery in 2009. According to the abdominal ultrasonography, the patient was diagnosed with hydronephrosis and hydroureter of upper segment on admission. After the treatment with multiple antibiotics failed, nephrostomy was performed to improve hydronephrosis. There was no history of traveling abroad. Informed consent was signed by the patient involved in this study. The methods in this study were approved by the Ethics Committee of Shandong Provincial Hospital and were carried out in accordance with the approved guidelines. The strain obtained from the patient was identified as E. coli by using Vitek-2 compact system and confirmed by Vitek-MS system (BioMérieux, France). Phenotypic detection of carbapenemases was performed using carbapenem inactivation method (CIM) and EDTA-modified CIM (eCIM) test.

### Antibiotic Susceptibility Assay

Susceptibility assay of antibiotics was performed on Mueller-Hinton (MH) agar plates using E test strips (**Table 1**). Susceptibility assay results were interpreted by Clinical Laboratory Standards Institute (CLSI) breakpoints (CLSI, 2017), with the exception of tigecycline, polymyxin B, and fosfomycin, which were interpreted by EUCAST breakpoints (EUCAST, 2017).


TABLE 1 | Antibiotic susceptibilities

 of

Escherichia

 coli JN05 and its

transconjugant.

 < =

TZP, LEV,levofloxacin;

 STX,

piperacillin/tazobactam;

 ATM, aztreonam; trimethoprim/sulfamethoxazole;

 FOS, fosfomycin;

 and TGC, tigecycline.

 < =

 < =

 < =

 CZO, cefazolin; CRO, cefatriaxone;

 < =

 < =

 < =

 FEP, cefepime; FOX,cefoxitin;

 < =

 < =

 IPM, imipenem;

 MEM, meropenem;

 ETP, ertapenem;

 AK, amikacin;

 < =

 < =

 < =

 < =

 < =

 < =

CN,gentamicin;

 CIP, ciprfloxacin;

# Molecular Typing

fmicb-09-02468 October 20, 2018 Time: 14:53 # 3

Multilocus sequence typing (MLST) and phylogenetic typing was performed for molecular typing of the isolate as previous described (Wirth et al., 2006; Wang et al., 2016). The virulence factors of extraintestinal pathogenic E. coli (ExPEC)-associated genes were screened by PCR-based assays (Wang et al., 2016).

# Screening of Antibiotic Resistance Genes

Antimicrobial resistance genes were screened by PCR and DNA sequencing as described previously (Zhu et al., 2016). These antimicrobial resistance genes included carbapenemaseencoding genes, extended-spectrum β-lactamase genes, AmpC β-lactamase genes, 16S rRNA methylase genes, fosfomycin resistance genes, quinolone resistance genes, and polymyxin B resistance genes (mcr-1) (Du et al., 2016; Zhu et al., 2016).

# Analysis of blaNDM-Carrying Plasmids

Conjugation test was performed by mixed broth method using E. coli J53Azi<sup>R</sup> as the recipient strain. Transconjugants were selected on MH agar plates containing 6 µg/mL ceftazidime and 100 µg/mL sodium azide. The antimicrobial susceptibility test of the transconjugant was carried out as antibiotic susceptibility assay of clinical strain.

The size and amounts of plasmids carried by the clinical isolate and transconjugant were evaluated by S1-pulsed-field gel electrophoresis (PFGE) as previously described (Liu et al., 2018).

# Plasmid Sequencing

The plasmid pJN05NDM carrying blaNDM−<sup>7</sup> (present in strain JN05) was extracted and sequenced using an Illumina Hiseq platform and assembled by SOAPdenovo at the MajorBio Co (Shanghai, China). The gaps were closed through PCR and Sanger Sequencing at Sangon Biotech (Shanghai, China). The plasmid sequences were annotated by BLAST against the non-redundant protein database. PlasmidFinder was used for detection and typing of the plasmid.

# RESULTS

### Resistance Profile of JN05 Strain

The carbapenem-resistant E. coli isolate JN05 was identified as metallo-beta-lactamase (MBL)-producing strains by eCIM. The JN05 strain was resistant to aztreonam, carbapenems, cephalosporins, quinolones, aminoglycosides, piperacillintazobactam, and trimethoprim-sulfamethoxazole, but was susceptible to fosfomycin, polymyxin B, and tigecycline (**Table 1**).

# Molecular Grouping, Resistance Genotyping, and Virulence Genotyping

The E. coli strain JN05 was assigned to ST167 and belonged to phylogenetic group A. It carried papG II, which may play an important role in the pathogenic process. Multiple antimicrobial resistance genes, including blaNDM−7, blaCTX−M−3, blaCTX−M−14, blaTEM−1, qnrS, armA, and acc(6<sup>0</sup> )- Ib genes, are responsible for the resistance profile of strain JN05.

# Analysis of the Plasmid Harboring NDM

New Delhi metallo-beta-lactamase -harboring plasmid of strain JN05 was successfully transferred into E. coli J53Azi<sup>R</sup> by conjugation experiment. The presence of NDM-7 in the transconjugant was confirmed using PCR, and MLST was used to distinguish the transconjugants from the clinical strain. The transconjugant J05 was susceptible to aztreonam, quinolones, and aminoglycosides, but resistant to carbapenems and cephalosporin. S1-PFGE showed that the clinical strain JN05 harbored six plasmids, and the transconjugant J05 contained a single plasmid, which was approximately 46 Kb (**Supplemental Figure S1**).

pJN05NDM-7 is a 46,161-bp plasmid belonging to the IncX3 incompatibility group. The complete sequence of plasmid pJN05NDM-7 was submitted to GenBank under accession number MH523639. In pJN05NDM-7, blaNDM−<sup>7</sup> was preceded by IS3000-ISAba125-IS5 in the upstream region and followed by ble-trpF-dsbC-IS26-∆umuD in the downstream region. This blaNDM genetic structure was common in Enterobacteriaceae for the horizontal transfer of blaNDM (Pal et al., 2017).

The full published sequences of seven plasmids harboring NDM-7 were downloaded and compared, including pKW53T-NDM (Accession No. KX214669), pEC50-NDM-7 (Accession No. KX470735), pKPN01-NDM-7 (Accession No. NZ\_CP012990), pOM26-1 (Accession No. KP776609), pM110\_X3 (Accession No. AP018141), pABC218-NDM (Accession No. KX214670), and pABC133-NDM (Accession No. KX214671) (Espedido et al., 2015; Pal et al., 2017). Sequence alignments revealed that pJN05NDM-7 was 100% identical to the previously described plasmid pKW53T-NDM of E. coli isolated in Kuwait (KW53T). pJN05NDM-7 plasmid showed high overall nucleotide identity (99%) with pEC50-NDM-7 from China and pKPN01-NDM-7 from Canada. In addition, pJN05NDM-7 was similar to the plasmid pOM26-1 isolated from Oman, and pABC218-NDM and pABC133-NDM from UAE. However, they lacked mobile genetic elements or even the topoisomerase III gene (**Figure 1**).

To explore the geographic distribution of IncX3 plasmids harboring blaNDM in China, 29 plasmids were screened and analyzed, including two plasmids from North China (pEc1929, pNDM5-E6CN), 14 plasmids from East China (NUHL24835, p112298-NDM, pAD-19R, pNDM-5\_IncX3, pNDM5-SSH006, pNDM-20, pNDM-QD28, pNDM-QD29, RJA274, pYE315203, pYQ13500-NDM, pZHDC33, pZHDC40, and pJN05NDM), six plasmids from South China (p112298-NDM, pCREC-A6-NDM, pNDM-HF727, pP785-NDM5, pP788A-NDM5, and pP855- NDM5), four plasmids from Central China (pEC50-NDM-7, pNDM-HN380, pP744T-NDM5, and pP768-NDM-5), and three plasmids from West China (pECNDM101, p3-NDM, and pSCE516-2) (**Supplementary Figure S2**).

fmicb-09-02468 October 20, 2018 Time: 14:53 # 4

the schematic map of plasmid pJN05NDM-7 (Accession No. MH523639). The genes were labeled with different colors according to their functional annotations. The internal three rings represent a comparative analysis of three blaNDM-harboring plasmids with pJN05NDM7, including pOM26 -1 (blue), pABC218 -NDM (purple) and pABC133 -NDM (green) (constructed by BRIG).

Multiple NDM variants were harbored in the plasmids, including NDM-1, NDM-5, NDM-7, NDM-13, NDM-17, and NDM-20. We observed that the IncX3 plasmids carrying blaNDM−<sup>5</sup> originating from different provinces of China showed high similarity, except three plasmids with various lengths of insertion sequences (pP744, pRJA274, and pZHDC40). Six IncX3 plasmids originating from pigs and one plasmid from chicken were identical to pNDM-HN380, thus confirming that this mobile NDM vector is widespread in China (He et al., 2017; Ho et al., 2018).

As obvious differences were observed among sequences of pJN05NDM-7, pP744, pRJA274, and pZHDC40, linear structural comparison of whole genome sequences of pJN05NDM with the above plasmids was performed (**Figure 2**). The backbone of these plasmids showed high degrees of conservation and similarity, with sequence polymorphism at the region of additional insertion around the NDM gene. The plasmids did not carry any resistance genes other than NDM, except pRJA274. pRJA274 is a 53,134-bp circular IncX3 type plasmid haboring two resistance genes including blaNDM−<sup>1</sup> and blaSHV−12. pRJA274 is almost identical to pJN05NDM-7, but the ISAba125 element (935-bp) between IS3000 and IS5 at the nucleotide position 40,753 was missing. In addition, the backbone of pRJA274 shared identity with plasmid pIncX-SHV. Compared with pJN05NDM-7, the inserted sequence of pP744 was different, with a deletion of 543 bp at ISAba125 located downstream of IS5. In pZHDC40, fmicb-09-02468 October 20, 2018 Time: 14:53 # 5

deletion of 816 bp at IS26 downstream of NDM-7 was observed.

# DISCUSSION

In this study, we aimed to evaluate the genotype and phenotype of blaNDM−<sup>7</sup> -producing E. coli in China and found that E. coli carrying IncX3 plasmids play an important role as a reservoir and in the spread of blaNDM.

Although ST131 is the most prevalent strain type of E. coli worldwide, ST167 is considered to be related to clinical infections in China (Yang et al., 2014). In this study, JN05 assigned to ST167 was isolated from a 61-year-old woman with recurrent urinary tract infection. The isolate JN05 belonged to phylogroup A and was positive for papG II, which increased the ability of P-fimbriae adhesin (Wang et al., 2016). In addition, this isolate contained multiple resistance genes, including blaTEM−1,blaCTX−M−3, blaCTX−M−14, blaTEM−1, qnrS, armA, and acc(6<sup>0</sup> )-Ib genes; therefore, showed multidrug resistance and increased resistance to β-lactam drugs.

According to the surveillance of CRE strains in China, blaNDM was mainly responsible for carbapenemase resistance in E. coli, while blaNDM−<sup>7</sup> was relatively uncommon. Since the first report on clinical infection due to blaNDM−<sup>7</sup> in France, this is the first report on fully sequenced plasmid carrying blaNDM−<sup>7</sup> isolated from China.

Escherichia coli isolates carrying blaNDM−<sup>7</sup> belonging to different STs were sporadic reported worldwide (Cuzon et al., 2013; Wang et al., 2016; Devanga Ragupathi et al., 2017; Pal et al., 2017; Espinal et al., 2018). According to previous reports, blaNDM−<sup>7</sup> gene can be carried by several Enterobacter species and multiple types of plasmids including IncX3, IncF, and IncA/C groups, with sizes ranging from 37 to > 100 kb. IncX3, a selfconjugative plasmid, was most frequently observed to be the carrier of blaNDM−7.

Interestingly, the plasmid pJN05 was identical to the plasmid pKW53T-NDM-7 isolated in Kuwait. Geographical contiguity or travel history could not be considered as a cause of resistance gene transmission, suggesting that the plasmids maybe native and not imported. We proved that plasmids harboring blaNDM were hidden in the environment and in the human gut worldwide long before we identified them. It is possible that IncX3 plasmids carrying different variants originated from the same plasmid, but point mutations during transmission and evolution generated the differences. Exposure to carbapenem agents speed up the evolution of plasmids carrying blaNDM variants and enhance enzyme activity toward carbapenems.

Notably, blaNDM-producing isolates of animal origin increased, indicating that food animals have become the reservoir of blaNDM (He et al., 2017; Kong et al., 2017). To understand the geographical distribution and gene polymorphism among the plasmids originating from different region, IncX3 plasmids carrying blaNDM isolated from different areas were compared. Six plasmids harboring blaNDM originating from pigs or chickens showed high similarity (> 99%) to those from patients. Because carbapenems were not approved for use in food animals in China, we assumed that the NDM-producing isolates were introduced to the farm via contaminated feed and water. The food animals contaminated by bacteria harboring blaNDM accelerated the spread of resistance genes among healthy population. Nonetheless, more data are needed to explain the dissemination of blaNDM among animals and humans.

# CONCLUSION

In conclusion, this study identified self-transmissible IncX3 plasmids carrying blaNDM, which were disseminated in geographically segregated areas in China and other countries in the world. This study emphasizes the important role of IncX3 plasmids in transmission of blaNDM in China. Effective measures should be taken to monitor and control the rapid dissemination of blaNDM.

### AUTHOR CONTRIBUTIONS

fmicb-09-02468 October 20, 2018 Time: 14:53 # 6

YH and YJ contributed to experiment conception, design, and wrote the paper. CS and YB performed data analysis.

### FUNDING

The study was supported by grants from Focus on Research and Development Plan in Shandong Province (2016GSF201078) and the Medical and Health Development Plan of Shandong Province (2017WS471).

# REFERENCES


### ACKNOWLEDGMENTS

We thank Professor Zhiming Lu for helpful discussion and continuous encouragement.

### SUPPLEMENTARY MATERIAL

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb. 2018.02468/full#supplementary-material

FIGURE S1 | S1-pulsed-field gel electrophoresis (S1-PFGE) patterns of clinical isolate JN05. (A) PFGE of clinical isolate JN05; (B) PFGE of transconjugant J05; M, marker, Salmonella enterica serotype Braenderup H9812.

FIGURE S2 | Alignment of pJN05NDM7 with 28 IncX3-typed plasmids harboring blaNDM in China identfied in GenBank (To April 2018).


**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2018 Hao, Shao, Bai and Jin. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

# The Simplified Carbapenem Inactivation Method (sCIM) for Simple and Accurate Detection of Carbapenemase-Producing Gram-Negative Bacilli

Xiaopeng Jing<sup>1</sup> , Huan Zhou<sup>2</sup> , Xiaochun Min<sup>1</sup> , Xing Zhang<sup>1</sup> , Qing Yang<sup>3</sup> , Shuaixian Du<sup>4</sup> , Yirong Li<sup>5</sup> , Fangyou Yu<sup>6</sup> , Min Jia<sup>7</sup> , Yu Zhan<sup>8</sup> , Yi Zeng<sup>8</sup> , Bo Yang<sup>8</sup> , Yunjun Pan<sup>9</sup> , Binghuai Lu10, Rong Liu<sup>2</sup> and Ji Zeng<sup>1</sup> \*

<sup>1</sup> Department of Clinical Laboratory, Wuhan Fourth Hospital, Puai Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, <sup>2</sup> Department of Pathophysiology, School of Basic Medicine, Key Laboratory of Ministry of Education of China for Neurological Disorders, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, <sup>3</sup> Department of Clinical Laboratory, First Affiliated Hospital, College of Medicine, Zhejiang University, Hangzhou, China, <sup>4</sup> Department of Clinical Laboratory, Union Hospital, Huazhong University of Science and Technology, Wuhan, China, <sup>5</sup> Department of Clinical Laboratory, Zhongnan Hospital, Wuhan University, Wuhan, China, <sup>6</sup> Department of Clinical Laboratory, Shanghai Pulmonary Hospital, Tongji University School of Medicine, Shanghai, China, <sup>7</sup> Department of Clinical Laboratory, Wuhan First Hospital, Wuhan, China, <sup>8</sup> Department of Clinical Laboratory, The Central Hospital of Wuhan, Wuhan, China, <sup>9</sup> Department of Clinical Laboratory, Renmin Hospital, Hubei University of Medicine, Shiyan, China, <sup>10</sup> Department of Laboratory Medicine, Civil Aviation General Hospital, Peking University Civil Aviation School of Clinical Medicine, Beijing, China

This study reports the simplified carbapenem inactivation method (sCIM) to detect carbapenemase-producing gram-negative bacilli in a simple and accurate manner. This method is based on the modified carbapenem inactivation method (mCIM) with the improvement of experimental procedures. Instead of incubating the antibiotic disk in the organism culture media, the organism to be tested was smeared directly onto the antibiotic disk in the sCIM. For evaluating the sensitivity and specificity of the method, a total of 196 Enterobacteriaceae, 73 Acinetobacter baumannii, and 158 Pseudomonas aeruginosa isolates were collected. Polymerase chain reaction (PCR) was used to detect the carbapenemase genes. Phenotypic evaluations were performed using both the sCIM and the mCIM. PCR results showed that, of the 196 Enterobacteriaceae strains, 147 expressed the carbapenemase genes blaKPC−<sup>2</sup> (58.5%), blaIMP−<sup>4</sup> (21.8%), blaIMP−<sup>2</sup> (2.0%), blaVIM−<sup>1</sup> (6.1%), blaNDM−<sup>1</sup> (10.2%), and blaOXA−<sup>48</sup> (1.4%). sCIM results had high concordance with PCR results (99.5%) and mCIM results (100%) with the exception of one Klebsiella pneumoniae strain, which had an minimal inhibitory concentration (MIC) for imipenem of 0.25 mg/L. PCR demonstrated that 53 of the 73 A. baumannii isolates expressed the carbapenemase genes blaOXA−<sup>23</sup> (98.1%) and blaVIM−<sup>2</sup> (1.8%). sCIM and PCR results corresponded but all A. baumannii isolates were carbapenemase negative by the mCIM. PCR demonstrated that 25 of the 158 P. aeruginosa isolates expressed carbapenemase genes blaVIM−<sup>1</sup> (52%), blaVIM−<sup>2</sup> (8%), blaVIM−<sup>4</sup> (36%), and blaIMP−<sup>4</sup> (4%). sCIM results had high concordance with PCR results (100%) and the mCIM results (99.4%) with the exception of one P. aeruginosa isolate that expressed

### Edited by:

Yi-Wei Tang, Memorial Sloan Kettering Cancer Center, United States

### Reviewed by:

Xian-Zhi Li, Health Canada, Canada Maurizio Sanguinetti, Università Cattolica del Sacro Cuore, Italy

> \*Correspondence: Ji Zeng whzjmicro@163.com

### Specialty section:

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

Received: 24 June 2018 Accepted: 18 September 2018 Published: 30 October 2018

### Citation:

Jing X, Zhou H, Min X, Zhang X, Yang Q, Du S, Li Y, Yu F, Jia M, Zhan Y, Zeng Y, Yang B, Pan Y, Lu B, Liu R and Zeng J (2018) The Simplified Carbapenem Inactivation Method (sCIM) for Simple and Accurate Detection of Carbapenemase-Producing Gram-Negative Bacilli. Front. Microbiol. 9:2391. doi: 10.3389/fmicb.2018.02391

**26**

the blaVIM−<sup>4</sup> gene. The sCIM offers specificity and sensitivity comparable to PCR but has the advantage of being more user-friendly. This method is suitable for routine use in most clinical microbiology laboratories for the detection of carbapenemase-producing gram-negative bacilli.

Keywords: carbapenemase, modified carbapenem inactivation method, gram-negative bacilli, Enterobacteriaceae, simplified carbapenem inactivation method

### INTRODUCTION

fmicb-09-02391 October 27, 2018 Time: 17:22 # 2

Carbapenemases can be divided based on their molecular characteristics into class A, B, or D using the Ambler classification system. Class A and D carbapenemases require serine at their active sites, whereas class B carbapenemases, also called metalloβ-lactamases (MBLs), require zinc for β-lactam hydrolysis (Jean et al., 2015; Bonomo, 2017; Khan et al., 2017). The most common class A carbapenemases are KPC enzymes, while notable transmissible class B carbapenemase include IMP, VIM, and NDM enzymes. Common class D carbapenemases include OXA-23-like, OXA-24-like, OXA-48-like, and OXA-58-like enzymes (Jean et al., 2015).

The prevalence of carbapenemases in gram-negative bacilli, especially carbapenemase-producing Enterobacteriaceae (CPE), Acinetobacter baumannii, and Pseudomonas aeruginosa, has increased markedly in the past 10 years (Jean et al., 2015; Logan and Weinstein, 2017). Genes encoding carbapenemases are often located on plasmids, facilitating the spread of carbapenem resistance between different bacteria (Jean et al., 2015). Carbapenem-resistant strains have caused difficulties in the clinical treatment and prevention of nosocomial infections.

The convenient and accurate detection of carbapenemases are of great clinical importance. CLSI (2010) introduced the modified Hodge test for carbapenemase detection, but this method can only be used for the accurate detection of KPC-type carbapenemases in Enterobacteriaceae. CLSI (2012) recommended the Carba NP test method for the detection of carbapenemases in gram-negative bacilli; however, the preparation of the reagents required for this test is complicated and the solutions cannot be stored for extended periods, limiting its clinical application. van der Zwaluw et al. (2015) designed a new detection method, carbapenem inactivation method (CIM), which is easy to operate and highly sensitive in the detection of carbapenemases. In 2017, based on the CIM method, CLSI recommended the modified carbapenem inactivation method (mCIM). This method is effective at detecting a variety of carbapenemases (CLSI, 2017; Pierce et al., 2017). However, it is a relatively complex method and can only be used to detect carbapenemases in Enterobacteriaceae and P. aeruginosa (CLSI, 2018). In the present study, based on the mCIM, we designed a simplified carbapenem inactivation method (sCIM) for simple and accurate detection of carbapenemases in gram-negative bacilli and compared it with polymerase chain reaction (PCR) and mCIM methods.

### MATERIALS AND METHODS

### Bacteria

To validate the sCIM, we collected 194 Enterobacteriaceae, 73 A. baumannii, and 158 P. aeruginosa clinical isolates from eight hospitals in China during 2017, and two OXA-48 producing Klebsiella pneumoniae and Escherichia coli from clinically conserved strains (Yu et al., 2017). Microorganisms were identified using the Microflex LT system (Bruker Daltonik GmbH, Bremen, Germany) and minimal inhibitory concentration (MIC) of imipenem, meropenem, and ertapenem were determined using the broth microdilution method. The 194 Enterobacteriaceae clinical isolates included 104 strains of K. pneumoniae, 72 strains of E. coli, and 18 strains of Enterobacter cloacae, in which 146 strains were resistant to imipenem and meropenem and five strains were resistant to ertapenem but susceptible to imipenem and meropenem. The two OXA-48 producing K. pneumoniae and E. coli were intermediate to imipenem, meropenem, and ertapenem. Fifty-three strains of A. baumannii and 149 strains of P. aeruginosa were resistant to imipenem and meropenem.

### Modified Carbapenem Inactivation Method

In the mCIM, 1 µL loopfuls of Enterobacteriaceae or 10 µL loopfuls of P. aeruginosa or A. baumannii from blood agar plates was emulsified in 2 mL trypticase soy broth (TSB). A meropenem disk was then immersed in the suspension and incubated for a minimum of 4 h at 35◦C. A 0.5 McFarland suspension of E. coli ATCC 25922 was prepared in saline using the direct colony suspension method. A Mueller–Hinton agar (MHA) plate was inoculated with E. coli ATCC 25922 using the routine disk diffusion procedure. The meropenem disk was removed from the TSB and placed on an MHA plate previously inoculated with the E. coli ATCC 25922 indicator strain. Plates were incubated at 35◦C in ambient air for 18−24 h. An inhibition zone diameter of 6–15 mm or colonies within a 16–18 mm zone was considered to be a positive result, and a zone of inhibition ≥19 mm was considered to be a negative result (CLSI, 2017; Pierce et al., 2017; CLSI, 2018).

### Simplified Carbapenem Inactivation Method

The sCIM is based on the mCIM with improvement of experimental procedures. Instead of incubating the antibiotic disk in the organism culture media for 4 h as in the mCIM, the organism to be tested was smeared directly onto an antibiotic

disk in the sCIM. To perform the sCIM, for Enterobacteriaceae, a 0.5 McFarland standard suspension (using the direct colony suspension method) of E. coli ATCC 25922 was inoculated onto the MHA plate, following the routine disk diffusion procedure; for A. baumannii and P. aeruginosa, a 0.5 McFarland standard suspension (using direct colony suspension method) of E. coli ATCC 25922 was diluted 1:10 in saline and inoculated onto the MHA plate, following the routine disk diffusion procedure. Plates were allowed to dry for 3–10 min. Then, 1–3 overnight colonies of the test organisms grown on blood agar were smeared onto an imipenem disk (10 µg; Oxoid, Hampshire, United Kingdom) (**Figure 1**) to allow one side of the disk was evenly coated with the test bacteria; immediately afterward, the side of the disk having bacteria was placed on the MHA plate previously inoculated with E. coli ATCC 25922. An imipenem disk placed on an MHA plate was used as the control. All plates were incubated at 35◦C for 16–18 h in ambient air. Bacterial strains that produced carbapenemase can hydrolyze imipenem; hence. the susceptible indicator strain grew unchecked. In contrast, the zone of inhibition around the disk shows a diameter of 6– 20 mm (**Figure 2**), or the satellite growth of colonies of E. coli ATCC 25922 around the disk with a zone diameter ≤22 mm (**Figure 3**), indicating that the isolate was capable of producing carbapenemase; a zone of inhibition ≥26 mm was considered to be a negative result; a zone of inhibition of 23–25 mm was considered to be a carbapenemase indeterminate result.

# PCR Detection of Carbapenemase Genes

To perform PCR, primers were designed to detect the blaKPC, blaIMP, blaVIM, blaNDM, blaOXA−48−like, and blaOXA−23−like genes (**Table 1**). PCR was performed according to previously described procedures (Yigit et al., 2001; Woodford et al., 2006; Doyle et al., 2012). Briefly, 25 µL of PCR Master Mix (CWBio, Beijing, China) was mixed with 4 µL of forward and reverse primers and water

FIGURE 1 | The sCIM testing procedure. Bacteria to be tested were grown overnight on a blood agar plate. One to three overnight colonies were smeared onto an imipenem disk and the disk was then placed on the testing plate.

FIGURE 2 | Results of sCIM testing of gram-negative bacilli. The zones of inhibition of the negative isolate and the control were similar, whereas the zones of inhibition of the positive isolates (the left is a K. pneumonia producing KPC-2; the right is an E. coli producing NDM-1) were 6 mm.

FIGURE 3 | Results of sCIM testing of gram-negative bacilli. Positive isolates (the left is a P. aeruginosa producing VIM-4; the right is an A. baumannii producing OXA-23) demonstrate satellite growth within the zone of inhibition around the disk.

to a final volume of 45 µL. Then, 5 µL of sample lysate from each test isolate was added to the mix. For Enterobacteriaceae and P. aeruginosa, the PCR program consisted of an initial denaturation step at 94◦C for 5 min, followed by 35 cycles of denaturation at 94◦C for 10 s, annealing at 56◦C for 30 s, elongation at 72◦C for 90 s, and a final extension at 72◦C for 2 min. For A. baumannii, the PCR program consisted of an initial denaturation step at 94◦C for 5 min, followed by 35 cycles of denaturation at 94◦C for 25 s, annealing at 52◦C for 40 s,



<sup>a</sup>The first and second primers for each gene are forward and reverse primers, respectively.

elongation at 72◦C for 50s, and a final extension at 72◦C for 6 min. PCR products were selected for sequencing and sequences were aligned using the BLAST software tool<sup>1</sup> .

### sCIM Detection on Positive Blood Cultures

To further confirm the sensitivity of the sCIM in clinical detection, in August 2018, 47 gram-negative bacilli of positive blood cultures collected from three hospitals were directly tested with the sCIM and antimicrobial susceptibility tests at the same time. The positive rate of carbapenemases-producing strains was analyzed to assess if the method may shorten the turnaround time (TAT).

### RESULTS

### Sensitivity and Specificity of sCIM

Of the 196 Enterobacteriaceae tested, 148 were shown to produce carbapenemase by the sCIM test, whereas 147 were shown to carry a carbapenemase-encoding gene by PCR. Both the OXA-48-producing K. pneumoniae and E. coli were carbapenemase positive by the sCIM and their MICs for imipenem were 2 mg/L. One K. pneumoniae isolate with an MIC for imipenem of 0.25 mg/L harbored CTX-M-15 and was found to be carbapenemase positive by the sCIM but carbapenemase negative by PCR. One K. pneumoniae isolate with an MIC for imipenem of 16 mg/L was found to be carbapenemase negative by both PCR and the sCIM, suggesting that this isolate may be a carbapenemase negative CRE (**Table 2**).

Of the 73 A. baumannii strains, 53 were resistant to imipenem. These 53 strains were all found to be carbapenemase positive by the sCIM. Fifty-two strains were found to carry the blaOXA−<sup>23</sup> gene and one strain carry the blaVIM−<sup>2</sup> gene by PCR. Twenty imipenem-susceptible A. baumannii strains were negative by both the sCIM and PCR. The concordance rate of the sCIM and PCR for A. baumannii was 100% (**Table 3**).

<sup>1</sup>https://blast.ncbi.nlm.nih.gov/Blast.cgi

Of the 158 P. aeruginosa strains tested, 149 were resistant to imipenem and 25 were found to be carbapenemase positive by the sCIM. The concordance rate of the sCIM and PCR for P. aeruginosa was 100% (**Table 3**).

In our experiment, the zones of inhibition of strains found to be carbapenemase negative by the sCIM ranged in size from 28 to 32 mm. In contrast, the zones of inhibition of strains found to be carbapenemase positive by the sCIM were 6 mm for Enterobacteriaceae isolates, 15 ± 4 mm for A. baumannii isolates, and 7 ± 3 mm for P. aeruginosa isolates (**Figures 2**, **3** and **Table 4**). Many small colonies were observed in the zones ≥12 mm around the imipenem disks, demonstrating the "satellite phenomenon" (**Figure 3**).

### Carbapenemase Enzymes Produced by Tested Strains

In this study, 225 isolates were found to be positive for carbapenemase-encoding genes by PCR. The carbapenemases identified in Enterobacteriaceae isolates were KPC-2 (58.5%), IMP-4 (21.8%), IMP-2 (2.0%), VIM-1 (6.1%), NDM-1 (10.2%), and OXA-48 (1.4%) (**Table 2**). The carbapenemases identified in A. baumannii isolates were OXA-23 (98.1%) and VIM-2 (1.9%). The carbapenemases identified in P. aeruginosa were VIM-1 (52%), VIM-2 (8%), VIM-4 (36%), and IMP-4 (4%) (**Table 3**).

### The Sensitivity and Specificity of the sCIM Is Comparable to the mCIM

The results of tests of 196 Enterobacteriaceae showed that the concordance rate of the sCIM and the mCIM was 100%, including one false positive. Of the 158 P. aeruginosa isolates, 25 were found to be carbapenemase positive by the sCIM and 24 were found to be carbapenemase positive by the mCIM. The results were inconsistent for one VIM-4-producing isolate. Of the 73 A. baumannii isolates, 53 were found to be carbapenemase positive by the sCIM but all were found to be carbapenemase negative by the mCIM (**Tables 2**, **3**). These results showed that the sCIM and the mCIM had similar rates of detection for CPE and P. aeruginosa and that the sCIM may be superior to the mCIM for the detection of carbapenemase-producing A. baumannii.



+, positive; −, negative; sCIM, simplified carbapenem inactivation method; mCIM, modified carbapenem inactivation method.



+, positive; −, negative; sCIM, simplified carbapenem inactivation method; mCIM, modified carbapenem inactivation method.

### sCIM Detection on Positive Blood Cultures

A total of 47 gram-negative bacteria isolated from blood cultures in three hospitals were tested by the sCIM. Eight K. pneumoniae producing KPC-2 and one A. baumannii producing OXA-23 isolates were positive, with a positive rate of 19.1% (9/47). No false-negative strains were found. These data indicate that the sCIM can directly detect blood culture positive strains and report the enzyme-producing strains 1 day earlier than the routine antimicrobial susceptibility test.

### DISCUSSION

The detection principles of the sCIM and the mCIM are similar, based on the fact that carbapenemases can hydrolyze carbapenem (Queenan and Bush, 2007; Jean et al., 2015; van der Zwaluw et al., 2015; Bonomo, 2017; Pierce et al., 2017). However, the strategies of the two methods to hydrolyze carbapenem are different. In the mCIM, the antibiotic disk is put into the TSB containing the test organisms for approximately 4 h, whereas in the sCIM, the test organisms are directly smeared onto the antibiotic disk. The mCIM requires TSB and MHA plates, while the sCIM only requires MHA plates. Thus, the operation in the sCIM experiment has fewer steps and is more convenient than the mCIM. When carbapenemase-producing bacteria is smeared on imipenem disks, the enzyme spreads and hydrolyzes the antibiotics on the paper, leading to a reduction in the size of the zone of inhibition.

Another difference of the sCIM compared with the mCIM is that the imipenem disk is selected. In general, the different types


TABLE 4 | Comparison of sCIM and mCIM results for selected strains.

fmicb-09-02391 October 27, 2018 Time: 17:22 # 6

sCIM, simplified carbapenem inactivation method; mCIM, modified carbapenem inactivation method. <sup>∗</sup>Satellite growth observed around the disk within the zone of inhibition.

of carbapenemases have different hydrolysis rates for different carbapenem (Jean et al., 2015). For most carbapenemases, the Km values for imipenem is much larger than that for meropenem (Queenan and Bush, 2007). Therefore, we compared the efficiency of different carbapenem disks in detecting carbapenemases in the current study. We found that the diameter of the zone of inhibition was 6 mm for all the testing CPEs when using the imipenem disk. When using the meropenem disk, the diameter of the zone of inhibition varied considerably in different CPEs, indicating that imipenem was more rapidly hydrolyzed by carbapenemases compared to meropenem. For an easier interpretation of the experimental results, we chose an imipenem disk for the sCIM test. Because a few of ESBLs or AmpC produced by Enterobacteriaceae might hydrolyze imipenem with low activity (Carvalhaes et al., 2010; Bialek-Davenet et al., 2017; van Boxtel et al., 2017), it may result in few false positives in the sCIM test. In our study, we discovered a false positive result in a K. pneumoniae strain producing CTX-15. but the MIC of imipenem was less than 2 mg/L for this isolate. However, no false-positive isolate was discovered from those with a MIC for imipenem ≥2 mg/L. Therefore, the sCIM may be suitable for the determination of carbapenem-nonsusceptible strains, while the carbapenem-susceptible strains may have few false positives.

The concentration of bacteria in the disk diffusion method can affect the diameter of the zone of inhibition. At the same concentration of the drug, the lower the concentration of the tested bacterial used, the larger the diameter of the zone of inhibition that will be obtained. Correspondingly, when the concentration of bacteria is lower, the change of the zone of inhibition is also greater upon the variation of antibiotics concentration. In general, the hydrolysis rate of carbapenem by carbapenemase-producing non-fermenting bacteria is weaker than that of CPE. Therefore, reducing the concentration of E. coli ATCC 25922 can increase the sensitivity of the sCIM in detecting non-fermenting bacteria. For initial experiments with non-fermenting bacteria, the concentration of E. coli ATCC 25922 was adjusted from the 0.5 McFarland standard suspension by 5-fold and 10-fold dilutions. We found that the results obtained using these two dilutions were similar for A. baumannii. However, only 24 strains of P. aeruginosa were carbapenemase positive at the fivefold dilution, whereas 25 strains were carbapenemase positive at the 10-fold dilution. Therefore, the concordance rate with PCR was higher at the 10-fold dilution than the 5-fold dilution. Based on these results, we selected a 10 fold dilution of the 0.5 McFarland standard suspension of E. coli ATCC25922 for experiments with P. aeruginosa and Acinetobacter spp.

PCR results showed that KPC, VIM, NDM, IMP, and OXA-type enzymes were produced by Enterobacteriaceae, A. baumannii, and P. aeruginosa, and the sCIM can be used for sensitive (100%) and specific (99.6%) detection of carbapenemase-producing gram-negative bacilli. Only one of the 226 sCIM positive isolates, a CTX-15-producing K. pneumoniae strain, caused a false positive in sCIM tests. Pierce et al. (2017) reported that TEM-1 and TEM-52 enzymes might cause false positives in mCIM tests, too. CLSI (2018) recommended that the mCIM be used to detect carbapenemases in Enterobacteriaceae and P. aeruginosa. We found that only one isolate carrying the gene encoding VIM-4 was carbapenemase negative by the mCIM in our tests. Therefore, we concluded that the sensitivity of the sCIM for detecting CPE and P. aeruginosa was similar to that of the mCIM.

The carbapenemases identified in Enterobacteriaceae isolates were mainly KPC-2 and IMP-4. Because mainland China is not an OXA-48-endemic area, no OXA-48 positive strains were found in our collected isolates. These results were similar to the distribution of enzymes reported in mainland China by Zhang et al. (2017). To confirm the capability of the sCIM to detect OXA-48 carbapenemase, we collected two clinically conserved OXA-48-producing strains. Both strains were positive by the sCIM and the mCIM. However, we still cannot completely evaluate the capability of the sCIM to detect OXA-48 because of the small number of tests, and more investigations are needed to further confirm the reliability of the sCIM in detecting OXA-48-expressing isolates.

Our experimental data showed that the majority of CRE was CPE (99.3%), and only one K. pneumoniae isolate with an MIC for imipenem of 16 mg/L was not producing carbapenemase. This isolate was resistant to beta-lactams and susceptible to aminoglycosides and fluoroquinolones. The ratio of carbapenemase-producing isolates was 16.8% in carbapenemresistant P. aeruginosa, and was slightly higher than the data reported by Yin et al. (2018). Carbapenemase-producing isolates are only a small part of carbapenem-resistant P. aeruginosa, and the loss or alteration of OprD is thought to be the most prevalent mechanism for carbapenem resistance in P. aeruginosa

(Li et al., 2012; Yin et al., 2018). Among the 53 carbapenemresistant A. baumannii, 52 produced OXA-23 type enzyme, and one produced VIM-2 type enzyme. The resistance mechanism of carbapenem-resistant A. baumannii is mainly related to carbapenemase (Jean et al., 2015).

Compared to other carbapenemase detection methods, the sCIM has several obvious advantages. First, the sCIM does not require special equipment or reagents, hence is less costly. Second, the sCIM is easy to perform using the conventional paper diffusion method and is not complicated. Third, the results can be easily assessed. Finally, the sCIM has a wide range of detection and can be used to detect CPE, Acinetobacter spp., and P. aeruginosa. The sCIM is suitable for routine use in most clinical microbiology laboratories to detect carbapenemase-producing bacteria and can contribute to the reduction of carbapenemaseproducing gram-negative bacilli in hospitals.

### REFERENCES


# AUTHOR CONTRIBUTIONS

XJ, XM, XZ, QY, SD, YL, MJ, YiZ, YuZ, BY, YP, FY, and BL isolated the bacteria and performed the laboratory measurements. JZ, XJ, and HZ made substantial contributions to conception and design. JZ and RL wrote and revised the manuscript. JZ drafted the manuscript. All authors read and approved the final manuscript.

### FUNDING

This work was partially supported by the Natural Science Foundation of Hubei Province, China (Grant No. 2015CKC897), the Foundation of the Hubei Health and Family Planning Commission (Grant No. WJ2015MB152), and the National Natural Science Foundation of China (Grant No. 81471304).


**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2018 Jing, Zhou, Min, Zhang, Yang, Du, Li, Yu, Jia, Zhan, Zeng, Yang, Pan, Lu, Liu and Zeng. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

# High Prevalence of blaNDM Variants Among Carbapenem-Resistant Escherichia coli in Northern Jiangsu Province, China

Ruru Bi 1†, Ziyan Kong<sup>1</sup> , Huimin Qian<sup>2</sup> , Fei Jiang<sup>3</sup> , Haiquan Kang<sup>3</sup> , Bing Gu1,3 \* and Ping Ma1,3 \*

### Edited by:

*Liang Chen, Public Health Research Institute (PHRI), United States*

### Reviewed by:

*Jian-Hua Liu, South China Agricultural University, China Kalyan Chavda, Rutgers, The State University of New Jersey, Newark, United States*

### \*Correspondence:

*Bing Gu gb20031129@163.com Ping Ma 672443193@qq.com*

### †Present Address:

*Ruru Bi, Department of Laboratory Medicine, Suzhou Science and Technology Town Hospital, Suzhou, China*

### Specialty section:

*This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology*

Received: *26 June 2018* Accepted: *23 October 2018* Published: *13 November 2018*

### Citation:

*Bi R, Kong Z, Qian H, Jiang F, Kang H, Gu B and Ma P (2018) High Prevalence of blaNDM Variants Among Carbapenem-Resistant Escherichia coli in Northern Jiangsu Province, China. Front. Microbiol. 9:2704. doi: 10.3389/fmicb.2018.02704* *<sup>1</sup> Medical Technology School, Xuzhou Medical University, Xuzhou, China, <sup>2</sup> Jiangsu Provincial Center for Disease Control and Prevention, Nanjing, China, <sup>3</sup> Department of Laboratory Medicine, Affiliated Hospital of Xuzhou Medical University, Xuzhou, China*

The continuous emergence of carbapenem-resistant *Escherichia coli* (CRECO) presents a great challenge to public health. New Delhi metallo-lactamase (NDM) variants are widely disseminated in China, so the research on the prevalence and transmission of diverse *bla*NDM variants is urgently needed. In the present study, 54 CRECO isolates were collected from 1,185 *Escherichia coli* isolates in five hospitals in Northern Jiangsu Province, China from September 2015 to August 2016. Antimicrobial susceptibility tests, PCR detection of resistance determinants, multi-locus sequence typing (MLST) and pulsed-field gel electrophoresis (PFGE) were performed to characterize these strains. Plasmid conjugation experiments were carried out to determine the transferability of resistant genes from selected isolates. PCR-based replicon typing (PBRT), S1 nuclease-PFGE, and Southern blotting were conducted for plasmid profiling. Carbapenemase genes were detectable in all CRECO isolates, among which thirty-one CRECO isolates were found to carry *bla*NDM−<sup>5</sup> (54.7%), while, *bla*NDM−1, *bla*NDM−7, *bla*NDM−4, *bla*NDM−9, and *bla*KPC−<sup>2</sup> were identified in 14, five, two, one, and one isolates, respectively. MLST results revealed 15 different STs and four new STs were first reported to be linked with NDM-producing isolates. PFGE typing showed that no more than two isolates with the same ST appeared to the same band pattern except three ST410 isolates. Twenty-six selected NDM-producing isolates were successfully transferred to *E. coli* J53 by conjugation experiments. Notably, 50.0% (13/26) of *bl*aNDM variants were found to be carried by ∼55 kb IncX3 plasmid. Our study reported a high prevalence of *bl*aNDM variants, especially *bla*NDM−5, in Northern Jiangsu province, China. Diverse *bla*NDM variants were mainly carried by ∼55 kb IncX3 plasmids, suggesting that the fast evolution and high transferability of this kind of plasmid promote the high prevalence of *bla*NDM variants. Therefore, large-scale surveillance and effective infection control measures are also urgently needed to prevent diverse *bla*NDM variants from becoming epidemic in the future.

Keywords: carbapenem, Escherichia coli, blaNDM variants, diversity, plasmid

# INTRODUCTION

Carbapenem, a β-lactam that is highly potent against Gramnegative bacteria, has been recognized as a last resort for treating of infections caused by multidrug-resistant bacteria. However, the increasing number of carbapenem-resistant Enterobacteriaceae (CRE) is unexpected despite infection control efforts, and it poses a great challenge to clinic (Zilberberg and Shorr, 2013). Carbapenem resistance is predominantly attributed to the presence of carbapenemases, among which Class A (blaKPC), Class B (blaNDM, blaVIM, blaIMP), and Class D (blaOXA−48) types are most common for Enterobacteriaceae (Walsh, 2010; Albiger et al., 2015). An emerging carbapenemase, New Delhi metallo-lactamase (NDM), was first reported in a Swedish patient with a hospitalization history in India, it exhibited resistant to all β-lactams except for monobactams, and has great potential to cause global health crisis (Yong et al., 2009). Initially, blaNDM gene was endemic to India subcontinent and NDM-producing isolates tested worldwide have geographical links with these high prevalence areas. However, an increasing number of regions worldwide have reported that patients with blaNDM−positive isolates have never been abroad, indicating that blaNDM genes are also associated with some special clones (Leverstein-Van et al., 2010).

In China, since the first report of blaNDM gene in four carbapenem-resistant Acinetobacter baumannii isolates (Chen et al., 2011), increasing Enterobacteriaceae have been identified as carriers of the blaNDM gene. Escherichia coli, an important member of Enterobacteriaceae, are often spread globally through some epidemiological lineages. Although the prevalence of NDM-producing CRE strains is low, outbreaks caused by blaNDM-positive isolates have been identified in several regions of China, indicating high transferability of the blaNDM gene and the severity of infections caused by blaNDM−1-positive organisms (Wang et al., 2014; Jin et al., 2015; Yu et al., 2016). Furthermore, Kaase et al. (2011) first reported a novel blaNDM variant, blaNDM−2, which differs by one amino acid substitution (Pro28Ala) from blaNDM−1, and the subsequent discovery of other blaNDM variants highlights the rapid evolution of this multi-drug resistance gene. In 2012, the NDM enzyme reservoir, India, first reported diverse blaNDM variants among Enterobacteriaceae and blaNDM variants exhibited higher minimum inhibitory concentration (MIC) levels of carbapenem compared with blaNDM−<sup>1</sup> (Rahman et al., 2014). Although the blaNDM gene is continuously recoverable in China, data on the prevalence and characteristics of blaNDM variants among Enterobacteriaceae are still needed for preventing its transmission. Notably, a study conducted by Hu et al. (2017) have discovered that various species of bacteria harbored several kinds of blaNDM variants in China, which were mainly carried by diverse plasmids with different sizes. In the present study, we reported a high prevalence of blaNDM variants among E. coli from five hospitals in Northern Jiangsu Province, China. Moreover, these diverse blaNDM variants were mainly located on the same plasmid.

# MATERIALS AND METHODS

### Study Design

From September 2015 to July 2016, five hospitals (two in Xuzhou, two in Suqian, and one in Lianyungang) in Northern Jiangsu Province of China collected 1,185 E. coli isolates to examine the prevalence and molecular epidemiology of carbapenemresistance isolates. Initial species identification and antimicrobial susceptibility testing was performed by the Vitek 2 system (bioMe'rieux, France) and MALDI-TOF MS (Bruker Microflex LT, Bruker Daltonik GmbH, Bremen, Germany) according to the manufacturer's instructions.

# Antimicrobial Susceptibility Testing

Initial susceptibility testing was examined by Vitek 2 system. Further MICtesting was conducted by agar dilution method for cefoxitin, ceftriaxone, ceftazidime, cefepime, aztreonam, amikacin, ciprofloxacin, tigecycline, and piperacillin/tazobactam. The MICs of imipenem, meropenem, and ertapenem were determined by E-tests. For colistin, MIC values were tested by broth microdilution method. The agar dilution method and Etest were performed according to the standard Clinical and Laboratory Standards Institute guideline (M100-S26) (CLSI, 2017). The breakpoints of Food and Drug Administration (FDA) and European Committee on Antimicrobial Susceptibility Testing were used for tigecycline and polymyxin, respectively.

### Molecular Detection of Resistance Genes

DNA templates were prepared by alkaline lysis method using the kit (MoBio, USA). Carbapenemase genes (blaKPC blaNDM, blaSME, blaGES,blaVIM, blaIMP, and blaOXA−48) (Senda et al., 1996; Queenan et al., 2000; Poirel et al., 2004; Endimiani et al., 2008; Yang et al., 2012; Pereira et al., 2015; Al-Agamy et al., 2017), extended spectrumβ-lactamase genes (blaSHV, blaTEM, blaCTX−M−1group, blaCTX−M−2group, blaCTX−M−8group, and blaCTX−M−9group) (Schmitt et al., 2007; Yu et al., 2007), and plasmid-mediated AmpC genes(blaACC, blaFOX, blaMOX, blaDHA, blaCIT/SPM, and blaEBC) (Pérez-Pérez and Hanson, 2002) were examined by PCR. E. coli ATCC25922 was used for the quality control. Positive amplifications were subject to Sanger sequencing (GENEWIZ Company, Suzhou, China).

# PFGE and MLST

Molecular typing of 54 NDM-producing E. coli isolates was performed by pulsed-field gel electrophoresis (PFGE). The plugs containing genomic DNA were prepared according to the procedure described by Pereira et al. (2015). The DNA fragments digested with restriction endonuclease XbaI (TaKaRa Biotechnology, Dalian, China) were separated by PFGE on 1% SeaKem Gold agarose (Lonza, Rockland, ME, USA) using the CHEF Mapper XA PFGE system (Bio-Rad, USA) for 18 h at 14◦C. The electrophoretic switch times were 6.8–35.4 s. Salmonella H9812 was used as reference marker. Dice coefficients was used to calculate the similarity of PFGE patterns. Dendrograms were constructed by the unweighted pair group method with arithmetic averages (UPGMA) using BioNumerics software version 5.10. Isolates were categorized to be of the same cluster when their dice similarity index was ≥85%. Multi-locus sequence typing of E. coli was conducted by PCR as previously described (Wirth et al., 2006). The allelic profiles and sequence types were identified by amplifying and sequencing the seven housekeeping genes (adk, fumC, gyrB, icd, mdh, purA, recA) according to the reference website (https://enterobase.warwick.ac.uk/species/ index/ecoli). A minimum spanning tree of 54 blaNDM-positive isolates was also constructed by BioNumerics software version 5.10.

### Conjugation Assay

The conjugation experiment was implemented by mix broth mating among 26 selected isolates. The donor (clinical strains harboring the blaNDM gene) and recipient (sodium azideresistant E. coli J53) were mixed and cultured in broth at 37◦C overnight. The transconjugants were selected on MH agar with sodium-azide (180µg/mL) and imipenem (1µg/mL).. Initial species identification was conducted by Vitek MS system. Transformants were regarded as transconjugants when it exhibited resistance to carbapenem and harbored the blaNDM gene.

### Plasmid Analysis

Twenty-six selected isolates, including blaNDM-<sup>5</sup> and blaNDM-<sup>1</sup> gene with different STs from the aforementioned five different hospitals and all blaNDM-<sup>4</sup> (one isolates was lost in transit), blaNDM-7, and blaNDM-9-positive isolates, were subjected to further plasmid analysis. Incompatibility groups of plasmids extracted from transconjugants were determined by PCR-based replicon typing as described previously (Carattoli et al., 2005; Johnson et al., 2012). S1-PFGE and Southern blotting were conducted to isolate and locate resistance plasmids. Briefly, the gel plugs embedded with blaNDM-positive isolates were digested with S1 nuclease (TaKaRa Biotechnology, Dalian, China) and linear plasmids were separated by CHEF-Mapper XA PFGE system (Bio-Rad) as described above. The universal primers (F: GAAGCTGAGCACCGCATTAG; R: GGGCCGTATGAGTGATTGC) were used for probe synthesis. The plasmid DNA were transferred to positive-charged nylon membranes (Millipore, USA), and DIG-labeled blaNDM−specific probe served to hybridize plasmids according to the instructions of the DIG High Prime DNA Labeling and Detection Starter Kit (Roche, USA).

# RESULTS

# Clinical Data and Prevalence of Carbapenemase Genes Among CRECO

A total of 54 (4.56%, 54/1185) non-duplicate E. coli isolates that exhibited resistance to imipenem or meropenem were obtained from five hospitals in Northern Jiangsu Province, China. The Affiliated Hospital of Xuzhou Medical University (Hosptial A, n = 18), the Children's Hospital of Xuzhou (Hosptial B, n = 13), the People's Hospital of Suqian (Hosptial C, n = 11), the First People's Hospital of Suqian (Hosptial D, n = 6), and the Second People's Hospital of Lianyungang (Hosptial E, n = 6) were included. Among 54 CRECO isolates, 53 (98.1%, 53/54) were found to be blaNDM-positive and 1 was blaKPC−2-positive. Interestingly, five different blaNDM variants were identified in this collection (**Figure 1**). Among them, the blaNDM−<sup>5</sup> was the prevailing variant, accounting for 58.5% (31/53) of blaNDMpositive isolates, followed by blaNDM−<sup>1</sup> (26.4%, 14/53). Moreover, blaNDM−7, blaNDM−4, and blaNDM−<sup>9</sup> genes were also identified in 5, 2, and 1 isolates, respectively. NDM variations in amino acid substitutions at various positions are shown in **Table 1**. The distribution of the blaNDM variants in hospitals is depicted in **Figure 1**.

### Antimicrobial Susceptibility Patterns and Prevalence of Additional Resistance Genes of NDM-Producing CRECO

Fifty-three NDM-producing CRECO isolates were resistant to all cephalosporins (cefoxitin, ceftriaxone, ceftazidime, and cefepime) and enzyme inhibitors (piperacillin/tazobactam) but remained susceptible to colistin and tigecycline. The resistance rates to aztreonam, amikacin, and ciprofloxacin were 84.9, 22.6, and 92.4%, respectively. As shown in **Table 2**, among 31 blaNDM−5-positive isolates, 93.5% were resistant to ciprofloxacin, 74.2% to aztreonam, and 32.2% to amikacin. Resistance to aztreonam and ciprofloxacin were 85.7 and 92.9% among 14 blaNDM−1-producing isolates and were susceptible to amikacin. As for blaNDM−7-positive isolates, susceptibility was only found for amikacin. The blaNDM−<sup>4</sup> and blaNDM−<sup>9</sup> isolates were resistant to all antibiotics tested in this study except for colistin and tigecycline. One blaKPC−2-positive isolate was resistant to amikacin, colistin, and tigecycline. Molecular features revealed that most CRECO isolates carried the ESBLs gene, AmpC gene, or both. Overall, blaCTX−M, blaSHV, and blaTEM were identified in 42, 11, and 32 isolates, respectively. blaCTX−M-<sup>15</sup> was the most common ESBLs gene in our study, accounting for 26.4% followed by blaCTX−M-<sup>55</sup> (n = 9), blaCTX−M-<sup>65</sup> (n = 8), blaCTX−M-<sup>14</sup> (n = 8). blaCTX−M-<sup>64</sup> (n = 1), blaCTX−M-<sup>90</sup> (n = 1), and blaCTX−M-<sup>123</sup> (n = 1). All blaTEM positive isolates were identified as blaTEM−1. Moreover, 6 blaCMY−2, 5 blaCMY−42, and 1 blaDHA−<sup>1</sup> were also identified.

# Molecular Typing of CRECO

A total of 15 STs were identified in 54 CRECO isolates (**Figure 2**). Among 53 blaNDM-producing isolates, ST167 was the most prevalent, accounting for 35.8% (19/53), followed by ST410 (16.9%, 9/53), ST617 (13.2%, 7/53), ST405 (5.6%, 3/53), ST155 (3.8%, 2/53), ST156 (3.8%, 2/53), ST361 (3.8%, 2/53), and ST2659 (3.8%, 2/53). ST90, ST224, ST46, ST648, ST2376, and ST2083 were identified in one isolate. ST167 and ST617 are different by one allele and both correspond to clonal complex CC10. Moreover, ST410 and ST90 belong to clonal complex CC23. One KPC-2-producing isolate belong to ST131. PFGE typing revealed that no more than two isolates with the same ST type appeared to the same band pattern except for three ST410 isolates from hospital B (**Figure 3**). Surprisingly, the two isolates (E11 and E28) from hospital A and B shared the same patterns, indicating the occurrence of cross-transmission between hospitals.

FIGURE 1 | The distribution of *bla*NDM variants among different hospitals. Hospital A, the Affiliated Hospital of Xuzhou Medical University; Hospital B, the Children's Hospital of Xuzhou; Hospital C, the People's Hospital of Suqian; Hospital D, the First People's Hospital of Suqian; Hospital E, the Second People's Hospital of Lianyungang.

TABLE 1 | Amino acid substitutions of initially reported NDM variants.


### Characteristic of Plasmids Harboring the blaNDM Gene

All of plasmids haboring blaNDM gene from 26 selected CRECO isolates were successfully transferred to E. coli J53, and transconjugants exhibited resistance to carbapenem, cephalosporins and enzyme inhibitors (**Table 3**). As shown in **Figure 4**, S1-PFGE and Southern blotting revealed that all blaNDM−<sup>5</sup> genes were located on the same size (∼55 kb) plasmids, which was associated with IncX3 (n = 5), IncFI (n = 3), IncFII (n = 1), and untypeable replicon (n = 3). The blaNDM−<sup>1</sup> genes were carried by 55∼210 kb plasmids, including IncX3 (n = 5), IncFI (n = 1), and IncFII (n = 1) replicon types. Among five blaNDM−<sup>7</sup> positive isolates, four harbored ∼55 kb IncX3 plasmids with DIG-labeled blaNDM−7, and the remaining one was carried by ∼110 kb IncFI plasmid. The blaNDM−<sup>4</sup> and blaNDM−<sup>9</sup> genes were located on ∼55 kb IncX3 and ∼110 kb IncI1 plasmids, respectively. Surprisingly, E5 isolates harbored three blaNDM−5−positive plasmids of ∼55, ∼105, and ∼320 kb in size, suggesting high insertion efficiency of the blaNDM gene. Further plasmid sequencing from E5 isolates revealed that the respective GenBank accession numbers for ∼55 and ∼105 kb plasmids were NC\_022740.1 and AP018144.1, however, the nucleotide sequences of ∼320 kb plasmid could not be completely obtained.

# DISCUSSION

Metallo-lactamase, NDM, is an emerging carbapenem-resistant β-lactamase that is of major public concern due to its high medical and economic burden (Otter et al., 2017), especially for developing countries such as India, Pakistan, and the Balkan countries. As the most populous country in the world, there are major difficulties in preventing the dissemination of multidrug resistant genes in China. Therefore, comprehensive, extensive studies on diverse blaNDM variant-positive E. coli are needed to provide clear information to optimize antibiotic policy in endemic areas.

Generally, the prevalence of the blaNDM gene has continuously increased worldwide. As of now, the NDM enzyme has been identified in almost all of the world, including many countries in Asia, Africa, Europe, the Americas, and Australia (Berrazeg et al., 2014). A study from India also analyzed the occurrence of the blaNDM gene among carbapenem resistant isolates, and it accounted for 45.4% of them (Rahman et al., 2018). Recently,



*IMP, imipenem; ETP, ertapenem; MEP, meropenem; PB, colistin; TGC, tigecycline; ATM, aztreonam; FOX, cefoxitin; FEP, cefepime; CA, ceftazidim; CRO, cefatriaxone; YZP, piperacillin/tazobactam; AMK, amikacin; CIP, ciprofloxacin.*

a survey from the French National Reference Center revealed that among 140 carbapenem-resistant isolates 21% were NDM producer (Gauthier et al., 2018). In 2017, a nationwide study of clinical CRE strains in China demonstrated that 49% were NDM producer among carbapenem-resistant E. coli (Zhang et al., 2017). Furthermore, a multicenter study of the China CRE network revealed that among 39 carbapenem-resistant E. coli isolates, 74.4% were NDM producer, suggesting that there is a serious challenge in combating infections caused by this "superbug" in China (Zhang et al., 2018). In the present study, we identified 53 blaNDM-carrying isolates among 54 CRECO, which is much higher than in any other region of China (Wang S. et al., 2016; Hu et al., 2017; Liang et al., 2017). To the best of our knowledge, this is also the first report on the blaNDM gene in Northern Jiangsu Province. Moreover, the emergence of such a high prevalence of blaNDM variants indicates that the blaNDM gene is increasing in this area.

Since its first identification in 2009, the blaNDM gene has evolved at a fast pace during the past 10 years. Twenty-one blaNDM variants have been identified in different countries, all of which are archived at http://www.lahey.org/Studies/other.asp. Khan et al. (2017) reported that the Asian continent, especially China and India, was a reservoir of NDM producers, in which about a 58.2% abundance of the blaNDM−<sup>1</sup> variants was found. Among these blaNDM variants, the blaNDM−<sup>1</sup> gene has been reported as the most prevalent type worldwide (Nordmann and Poirel, 2014). In our study, five blaNDM variants were identified as responsible for MBL production, with the blaNDM−<sup>5</sup>


Isolates were categorized to be of the same cluster when their dice similarity index was ≥ 85%.

gene being the most prevalent. Compared with NDM-1, NDM-5 producers exhibited higher hydrolytic activity and toxicity toward carbapenem and cephalosporin (Mei et al., 2017). The blaNDM−<sup>5</sup> gene is an emerging blaNDM variants and has tended to surpass blaNDM−<sup>1</sup> recently, and it differs from the NDM-1 enzyme at two amino acid substitutions, exhibited increased carbapenem resistance (Zhang et al., 2016). The blaNDM−<sup>4</sup> and blaNDM−<sup>7</sup> genes have been discovered among E. coli in China with relatively low prevalence, involving in 4 and 6 patients, respectively. Moreover, clinical blaNDM−9-positive E. coli has only been found in Taiwan (Lai et al., 2017), with this being the first report on the mainland China. Moreover, not merely NDM-5, amino acid substitutions in NDM-4 (M154L), NDM-7 (D130A), and NDM-9 (E152A) could also result in high levels of carbapenem resistance (Düzgün, 2018). Stewart et al. (2017) reported that the substitution in M154L, which is found in most blaNDM variants, could enhance resistance to ampicillin at low zinc(II) concentrations relevant to infection sites.

blaNDM variant-positive isolates exhibited multi-drug resistance. Variable resistances to aztreonam, amikacin, and ciprofloxacin were identified in NDM-1 and NDM-5-producing E. coli. The NDM-7, NDM-4, and NDM-9-producing isolates exhibited pan-resistant phenotypes, showing resistance to almost all commonly used clinical antibiotics except for colistin and tigecycline. In view of this, the expert recommended polymyxins was regarded as the last resort for NDM-producing isolates (Yamamoto and Pop-Vicas, 2014).

ST167 was confirmed as significant carrier in the present study, and is reported to be associated with the production of the NDM enzyme, especially NDM-5 (Chen et al., 2016). Notably, four blaNDM-positive ST617 strains were identified in the world, but we identified 6 isolates in the present study. Moreover, a novel new variant of blaNDM, blaNDM−21, belonged to ST617, which


Frontiers in Microbiology | www.frontiersin.org



TABLE

3


Continued

*bGenBank accession no. AP018144.1.*

should deserve more attention (Liu et al., 2018). Furthermore, four new STs ST361, ST46, ST2376, and ST2083 were firstly reported to be linked with NDM-producing isolates.

The high efficiency of transfer renders the NDM producer ubiquitous throughout the world. Torres-González et al. (2015) reported an outbreak caused by blaNDM−1-carrying plasmid, which is easily transferred between E. coli and Klebsiella pneumoniae. Similarly, in the present study, the plasmids of 26 selected CRECO isolates were also tested for transfer. All selected blaNDM−<sup>5</sup> genes were identified to be located on 55 kb plasmid, among which IncX3 was the most common replicon type. Notably, blaNDM−<sup>5</sup> gene was frequently reported to be carried by the IncX3 plasmid of 55 kb in size, which was widely reported in China (Yang et al., 2014), Inidia (Krishnaraju et al., 2015), Australia (Wailan et al., 2015), and Damark (Hammerum et al., 2015). Moreover, Li et al. (2018) reported that IncX3 type plasmids play an important role in the transmission of the blaNDM−<sup>5</sup> gene in Enterobacteriaceae and this kind of plasmid occurred in different species. Further illustrated the challenge of preventing the dissemination of the blaNDM−<sup>5</sup> gene. The widespread blaNDM−1-carrying plasmid has been found to be associated with multiple replicon types, including IncX3, IncF, and IncA/C etc. In agreement with previous studies (Göttig et al., 2013; Wang L. H. et al., 2016), four blaNDM−7-positive plasmids were identified as ∼55 kb IncX3 type, while the blaNDM−<sup>7</sup> gene was also detected in the IncFI plasmid, which have been identified in India (Rahman et al., 2014). As for blaNDM−9, to the best of our knowledge, this is the first report on ∼105 kb IncI1 type plasmid harboring blaNDM−<sup>9</sup> gene. Overall, most plasmids harboring blaNDM variants were identified as 55 kb IncX3 types, hinting that amino acid mutations might occur in the process of plasmid transfer, resulting in the emergence of blaNDM variants. The high prevalence of the blaNDM genes due in part to plasmid transfer, meanwhile, the fast evolution of this multidrug resistance gene also favors the persistence of such bacteria harboring it.

In summary, the present study reported high prevalence of blaNDM variants, especially blaNDM−5, among carbapenemresistant E. coli in Northern Jiangsu Province. The presence of five different variants further increases the threat to public health because of the limited treatment options. Notably, diverse blaNDM variants were mainly located on ∼55 kb IncX3 plasmids, indicating that the fast evolution and high transferability of this kind of plasmid has led to the high prevalence of blaNDM variants. Timely detection of NDM enzyme and antimicrobial susceptibility testing are necessary so that infections caused by NDM producers receive appropriate and effective therapy. Similarly, large-scale surveillance and effective infection control measures are also urgently needed to prevent diverse blaNDM variants from becoming epidemics in the future.

### AUTHOR CONTRIBUTIONS

The laboratory measurements were performed by RB, ZK, and HQ. BG and PM participated in experimental design and manuscript revision. Data analysis were implemented by RB, ZK, HQ, FJ, and HK.

### ACKNOWLEDGMENTS

This research was supported by the National Natural Science Foundation of China (81471994), Jiangsu Provincial Natural Science Foundation (BK20151154), Jiangsu Provincial Medical

### REFERENCES


Talent (ZDRCA2016053), Six Talent Peaks Project of Jiangsu Province (WSN-135), Advanced Health Talent of Six-One Project of Jiangsu Province (LGY2016042), and Jiangsu Provincial Commission of Health and Family Planning Research Project (H201631).


in an intensive care unit in China. Clin. Microbiol. Infect. 18, E506–E513. doi: 10.1111/1469-0691.12035


**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2018 Bi, Kong, Qian, Jiang, Kang, Gu and Ma. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

# SuperPolymyxinTM Medium for the Screening of Colistin-Resistant Gram-Negative Bacteria in Stool Samples

Sara M. Przybysz <sup>1</sup> , Carlos Correa-Martinez <sup>2</sup> , Robin Köck 1,2,3, Karsten Becker <sup>1</sup> and Frieder Schaumburg<sup>1</sup> \*

1 Institute of Medical Microbiology, University Hospital Münster, Münster, Germany, <sup>2</sup> Institute of Hygiene, University Hospital Münster, Münster, Germany, <sup>3</sup> Institute of Hygiene, DRK Kliniken Berlin, Berlin, Germany

Colistin is one of the last resort antimicrobials for the treatment of infections caused

### Edited by:

Liang Chen, Public Health Research Institute (PHRI), United States

### Reviewed by:

Yi-Tsung Lin, Taipei Veterans General Hospital, Taiwan Fangyou Yu, Shanghai Pulmonary Hospital, China

\*Correspondence:

Frieder Schaumburg frieder.schaumburg@ukmuenster.de

### Specialty section:

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

Received: 25 June 2018 Accepted: 31 October 2018 Published: 21 November 2018

### Citation:

Przybysz SM, Correa-Martinez C, Köck R, Becker K and Schaumburg F (2018) SuperPolymyxinTM Medium for the Screening of Colistin-Resistant Gram-Negative Bacteria in Stool Samples. Front. Microbiol. 9:2809. doi: 10.3389/fmicb.2018.02809

by multidrug-resistant Gram-negative bacteria. After the emergence of transferable colistin resistance genes (mcr-1–5), a reliable culture-based screening method to detect colonization with colistin-resistant Gram-negative bacteria (CRGN) is needed. The objective of this study was to test the performance of SuperPolymyxinTM medium to screen for CRGN in stool samples and to compare different methods for the confirmation of colistin resistance (e.g., Etest, broth microdilution [BMD], and the Rapid PolymyxinTM NP test). Colonization with CRGN was analyzed in a prospective cohort study among travelers. Stool samples (Fecal TranswabTM) taken before, during and after travel were cultured on SuperPolymyxinTM agar. Every phenotypically different colony was subcultured for species identification using MALDI-TOF mass spectrometry. Susceptibility to colistin was tested using Etest and confirmed by BMD and the Rapid PolymyxinTM NP test. In total, 128 participants provided 1,495 stool samples. After culture on SuperPolymyxinTM medium (37◦C, 24–48 h), 1,851 phenotypically different colonies were isolated. Isolates belonging to intrinsically colistin-resistant genera (e.g., Morganella, Providencia, Proteus) or Stenotrophomonas maltophilia were excluded from further analysis (n = 421). Among the remaining 1,430 isolates, colistin resistance was confirmed in 279 by Etest® (19.5%) and 218 by BMD (15.3%). The Rapid PolymyxinTM NP test was compared with BMD (reference) to detect colistin resistance (specificity: 88.6%, sensitivity 71.1%). SuperPolymyxinTM medium is suitable to screen for fecal colonization with CRGN. The high proportion of colistin-susceptible isolates growing on SuperPolymyxinTM medium caused a high workload. The confirmation of CRGN with the Rapid PolymyxinTM NP Test could be a less labor-intensive alternative to BMD.

Keywords: Colistin, Resistance, Screening, Stool, mcr-1

# INTRODUCTION

The emergence and spread of antimicrobial resistance is currently one of the biggest challenges in health care worldwide. This is particularly true for Gram-negative bacteria. In the past decades, carbapenems were considered as safe and efficient antimicrobials for the treatment of infections with extended-spectrum β-lactamase (ESBL)-producing Enterobacterales or non-fermenting bacteria (e.g., Pseudomonas aeruginosa, Acinetobacter baumannii). Today, the increase and global spread of carbapenem-resistant bacteria raise serious concerns and physicians can find themselves "beamed back" to the preantimicrobial era as only very few compounds are available to treat infections with these multidrug-resistant microorganisms. For example, 13 of 38 European countries reported in 2015 an inter-regional dissemination or even an endemic occurrence of carbapenemase-producing Enterobacterales (Magiorakos et al., 2013; Albiger et al., 2015). In addition, carbapenemaseproducing Enterobacterales were also detected in livestock, companion animals, seafood and wildlife (Köck et al., 2018).

Although parenteral colistin (Polymyxin E) was not considered safe (e.g., nephro- and neurotoxicity) and effective in the past, it experiences a revival in human medicine nowadays due to the lack of alternative antimicrobial agents to treat infections due to carbapenem-resistant bacteria. Moreover, colistin is widely used in veterinary medicine to treat diarrhea in poultry and pig production systems. In 2012, the European Union estimated that the use of polymyxins in food-producing animals was 600-times higher than in humans (Skov and Monnet, 2016).

Colistin is a decapeptide with a poor resorption after oral administration due to its hydrophilic properties. For the treatment of multidrug-resistant Gram-negative bacteria, it is given intravenously as the prodrug colistin methanesulfonate (colistimethate) (Grégoire et al., 2017). Colistin interacts with lipopolysaccharides on the outer membrane of Gram-negative bacteria and causes membrane damage leading to bacterial death (Grégoire et al., 2017). Some genera are intrinsically resistant to colistin (e.g., Burkholderia, Hafnia, Morganella, Proteus, Providencia, Serratia) (Leclercq et al., 2013; Jayol et al., 2017). However, acquired colistin resistance has been reported, which is either encoded on the bacterial chromosome (e.g., mutations in lpxA, lpxC, lpxD, pmrA, pmrB, mgrB), or on transferable plasmids (e.g., mcr-1–5) (Nordmann et al., 2016a; Grégoire et al., 2017). Recently, Enterobacterales harboring mcr genes have emerged in livestock (Irrgang et al., 2016; Kieffer et al., 2017; Wang et al., 2017). Hence, mcr genes (mainly mcr-1) are currently the main mechanism of colistin resistance in livestock farming and mcr positive colistin-resistant Gram-negative bacteria (CRGN) can now be found in humans, breeding animals (e.g., pigs, poultry) or even "filth flies"(Guenther et al., 2017; Wang et al., 2017; Onwugamba et al., 2018). International travel also contributes to the spread of colistin-resistant bacteria, particularly through tourists from South-East Asia (Arcilla et al., 2016).

Effective screening media are needed to detect individuals colonized with CRGN. Recently, a colistin resistance screening agar was developed by Nordmann et al. (SuperPolymyxinTM medium). To the best of our knowledge this is currently the only available selective agar for the screening of colistinresistant bacteria (Nordmann et al., 2016a). SuperPolymyxinTM medium is based on eosin methylene blue, which is selective for Gram-negative bacteria. It also contains colistin, daptomycin and amphotericin B. Presumptive lactose-fermenting species grow in darkblue-brown colonies, while lactose-non-fermenters grow colorless or light lavender. In addition, the medium allows for the differentiation of lactose-fermenters as Escherichia coli colonies have a characteristic metallic green sheen and colonies of Enterobacter spp. and Klebsiella spp. are brown, dark-centered and mucoid. This SuperPolymyxinTM medium was already challenged with a selection of well-characterized colistinsusceptible and -resistant isolates (Nordmann et al., 2016a; Abdul Momin et al., 2017; Jayol et al., 2018).

Several test are available for the detection of colistin resistance but they are either labor intensive (broth microdilution [BMD]) or have a high rate of very major errors (Etest <sup>R</sup> ) (Matuschek et al., 2018). Recently, the Rapid PolymyxinTM NP test was developed by Nordmann et al., which can detect colistin resistance within 2 h (Nordmann et al., 2016b). Briefly, the test is based on the metabolism of glucose by viable cells leading to a decrease of the pH-value. The change in pH is indicated by phenol-red. In case of colistin resistance, bacteria can grow in the presence of colistin (3.75 mg/L) and the test suspension turns from red to yellow (Nordmann et al., 2016b). This Rapid PolymyxinTM NP test could be a suitable alternative to BMD or Etest <sup>R</sup> .

The objective of this study was now to assess the test performance and applicability of SuperPolymyxinTM medium in routine screening of fecal samples and to compare different methods for the confirmation of colistin resistance (e.g. Etest <sup>R</sup> , BMD, and the Rapid PolymyxinTM NP test).

### MATERIALS AND METHODS

### Ethics

Ethical clearance was obtained from the ethical committee of the medical faculty of "Westfälische Wilhelms-Universität," Münster, Germany (approval number: 2014-013-f-S). All participants signed a written informed consent prior to any study-related procedures.

### Fecal Samples and Media Characteristics

Fecal samples were self-collected by international travelers who took part in a prospective cohort study on the import of antimicrobial-resistant bacteria from abroad (Münster, Germany, April 2016–April 2018). Participants were trained to collect samples strictly avoiding cross-contamination from the environment by either sticking the whole swab in the stool or by swabbing the stool from the anal region before cleaning. The participants provided stool samples in Cary-Blair medium (Fecal TranswabTM, MWE medical wire, Corsham, England) once before, during and after the trip (Arcilla et al., 2017). The fecal swabs were stored at room temperature until being cultured. For that purpose, 10 µl of the inoculated Cary-Blair medium were streaked on SuperPolymyxinTM medium (three-phase streak technique) and cultured for 24–48 h at 37◦C under aerobic conditions (Nordmann et al., 2016a).

### Bacterial Cultures

One of each phenotypically different colony growing on SuperPolymyxinTM medium was subcultured on Columbia blood agar for species identification using matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (Bruker, Bremen, Germany) applying the database MBT Compass 4.1. All intrinsically resistant species were removed from further susceptibility testing (e.g., Hafnia alvei, Morganella morganii, Proteus hauseri, Proteus vulgaris, Proteus mirabilis, Providencia rettgeri, Providencia stuarti, Providencia alcalifaciens, Serratia liquefaciens, Serratia marcescens, Serratia ureolytica) (Magiorakos et al., 2012; Leclercq et al., 2013; Jayol et al., 2017; Saly et al., 2017) (**Figure 1**). Stenotrophomonas maltophilia was excluded due to missing EUCAST breakpoints for colistin.

### Susceptibility Testing

All isolates being not intrinsically resistant to colistin were further tested using colistin Etest <sup>R</sup> (bioMérieux, Marcy l'Étoile, France) on Mueller-Hinton agar (BD Diagnostics, Heidelberg, Germany) according to the manufacturer's instruction. Minimum inhibitory concentrations (MIC) were interpreted according to EUCAST clinical breakpoints (Version 8.0; colistin-susceptible: MIC ≤2 mg/L, colistin-resistant: MIC >2 mg/L). MICs from Etest <sup>R</sup> were rounded to the next doubling step for comparison with MIC from broth microdilution (Schaumburg et al., 2017).

Colistin resistance was confirmed by the BMD reference method (SensititreTM FRCOL, Thermo Fisher Scientific, Wesel, Germany) according to the manufacturer's instruction. Briefly, bacterial suspensions (0.5 McFarland) were prepared in Mueller-Hinton broth with TES buffer and transferred to the microtitre plates (range of colistin concentration: 0.12–128 mg/L). The results were read after 18–24 h at 35◦C. Test results with "skipped wells" (i.e., no growth in one well but growth in a well with higher colistin concentration) were considered non-interpretable (Poirel et al., 2017).

The Rapid PolymyxinTM NP Test was assessed as a rapid and cheap alternative to BMD to confirm colistin resistance (Nordmann et al., 2016b). The test was performed in triplicate. In case of inconsistent results, the final interpretation of the test was based on the result as suggested by the majority of the three test runs.

E. coli NCTC 13846 (DSMZ 105182, colistin-resistant) and P. aeruginosa ATCC 27853 (colistin-susceptible) were used as controls.

### mcr-1 Screening

All isolates having a MIC >2–64 mg/L as tested by BMD were screened for the presence of mcr-1 by isothermal amplification (eazyplex <sup>R</sup> SuperBug mcr-1, AmplexDiagnostics GmbH, Gars-Bahnhof, Germany). This range was chosen as almost all mcr positive CRGN showed a colistin MIC ≤ 64 mg/L (Nordmann et al., 2016a; Eiamphungporn et al., 2018; Poirel et al., 2018).

### Statistics

Statistical analyses were performed with "R" (package "epiDisplay" and "caret"). The 95% confidence interval (95% CI) of error rates (major and very major errors) and the categorical agreement between BMD and Etest <sup>R</sup> or Rapid PolymyxinTM NP Test was calculated with Wilson procedure without a correction for continuity.

### RESULTS

In total, 128 participants were included who provided 1,495 fecal swabs (median number of swabs per participant: 12, range: 2–23). Of these, 1,258 (84.2%) showed growth on SuperPolymyxinTM medium. No growth was detected on 237 plates (15.8%, **Figure 1**).

The median number of phenotypically different colonies growing on SuperPolymyxinTM medium was one colony (range: 1–3). In total, 1,851 phenotypically different colonies from SuperPolymyxinTM medium were subcultured for species identification (MALDI-TOF MS). We identified 47 different species.

In total, 421 isolates (22.7%) belonging to the following species were considered intrinsically colistin resistant and were excluded from further testing: P. rettgeri (n = 116), M. morganii (n = 98), S. marcescens (n = 66), P. alcalifaciens (n = 52), P. mirabilis/vulgaris (n = 33), H. alvei (n = 8), P. hauseri (n = 3), S. ureolytica (n = 3), P. stuarti (n = 2) and S. liquefaciens (n = 2, **Figure 1**). S. maltophilia (n = 38) was excluded due to missing EUCAST breakpoints.

Overall, 77.3% of the isolates from SuperPolymyxinTM agar (n = 1,430) were Gram-negative species that are normally susceptible to colistin, such as Klebsiella pneumoniae (n = 363), E. coli (n = 251), Enterobacter cloacae (n = 188), Klebsiella variicola (n = 168), Enterobacter asburiae (n = 147), Pseudomonas aeruginosa (n = 72), Enterobacter kobei (n = 46), Raoultella ornithinolytica (n = 39), Enterobacter aerogenes (n = 38), Citrobacter freundii (n = 34), Comamonas testosteroni (n = 16) and others (n = 68). Only one Gram-positive species (Enterococcus sp.) grew on the medium.

The non-intrinsically resistant species (n = 1,430) were screened for colistin resistance by Etest <sup>R</sup> . Colistin resistance was detected in 19.5% (279/1,430) of the isolates. Although Etest <sup>R</sup> is a convenient method to measure MICs, it is not recommended for colistin susceptibility testing due to false susceptible results (Matuschek et al., 2018). We confirmed colistin resistance with BMD in all isolates being tested resistant by Etest <sup>R</sup> (n = 279). BMD confirmed colistin resistance in 78.1% (218/279). Thus, 15.3% (218/1,430) of all non-intrinsically colistin resistant species growing on SuperPolymyxinTM medium were colistin resistant.

To assess the proportion of mcr-1 mediated colistin resistance, all isolates exhibiting a colistin MIC between >2–64 mg/L (n = 124) were screened for mcr-1 by PCR. This comprised Acinetobacter junii (n = 1), Delftia acidovorans (n = 1), E. coli (n = 39), E. cloacae (n = 39), E. asburiae (n = 17), R. ornithinolytica (n = 5), E. kobei (n = 13), C. testosteroni (n = 6), Ochrobactrum sp. (n = 1), Pseudomonas protegens (n = 1), and Yokenella regensburgei (n = 1). A total of 26.6% (n = 34) were mcr-1 positive (33 E. coli, 1 E. asburiae).

Since Etest <sup>R</sup> can underreport colistin resistance, we tested the Etest <sup>R</sup> performance in our setting using BMD as reference. For that purpose, we selected 130 consecutively collected species from our samples. Twelve isolates were excluded due to noninterpretable results in BMD (skipped wells, **Table 1**).

The essential agreement between Etest <sup>R</sup> and BMD was 59.3% (70/118, 95%CI: 50.3%−67.7%) and the category agreement was



<sup>a</sup>12 isolates were excluded due to non-interpretable results in BMD (skipped wells).

89.8% (106/118, 95%CI: 83.1–94.1%). Wrong Etest <sup>R</sup> results were either due to major errors (false resistant: 3/118, 2.5%, 95%CI: 0.9–7.2%) or very major errors (false susceptible: 9/118, 7.6%, 95%CI: 4.1–13.9%).

The specificity and sensitivity of Etest <sup>R</sup> (reference: BMD) was 95.8% and 80.4%, respectively. The negative and positive predictive values for the detection of colistin resistance using Etest <sup>R</sup> were 88.5 and 92.5%, respectively (accuracy: 89.8%).

BMD is recommended for colistin susceptibility testing, but it is labor intensive and time consuming. Therefore, we tested, if the Rapid PolymyxinTM NP Test can be applied to rapidly confirm or rule out colistin resistance. For that purpose, the same set of consecutively selected isolates (n = 130) was tested with Rapid PolymyxinTM NP Test (merged test results of triplicate runs) and BMD (reference).

Test results of the Rapid PolymyxinTM NP Test were not evaluable in three subsequent test runs for P. alcalifaciens (n = 4), C. testosteroni (n = 2), Acinetobacter junii (n = 1), Pseudomonas fluorescens (n = 1), E. kobei and P. aeruginosa (n = 1) due to no growth in the growth control. These isolates were excluded for the calculation of the test performance. Additionally, 12 isolates were excluded due to skipped wells in BMD.

The category agreement between Rapid PolymyxinTM NP Test and BMD was 82.1% (89/108, 95%CI: 74.1–88.4%). Disagreement was either due to major errors (false resistant: 8/108, 7.4%, 95%CI: 3.8–13.9%) or very major errors (false susceptible: 11/108, 10.2%, 95%CI: 5.8–17.3%).

The specificity and sensitivity of Rapid PolymyxinTM NP Test compared to BMD were 88.6 and 71.1%, respectively. The negative and positive predictive values to detect colistin resistance using Rapid Polymyxin NP Test were 84.9 and 77.1%, respectively (accuracy: 91%).

The test performance was impaired, if only the first test run of Rapid PolymyxinTM NP Test and not the merged results of the triplicate testing were considered (specificity: 85.7%, sensitivity: 63.4%, negative predictive value: 80.0%, positive predictive value: 72.2%).

### DISCUSSION

We tested the applicability of SuperPolymyxinTM medium and found that it is suitable to detect colistin resistance (including mcr-1 positive isolates) in human fecal samples but it was associated with high workload.

The high proportion of positive SuperPolymyxinTM medium (growth after 48 h, 84.2%) in our study is in contrast to a recent report on hospitalized patients where bacterial growth was found on 17/41 agars (41.5%) after inoculation from rectal swabs (Jayol et al., 2018). In our study, the participants stored the samples at room temperature during their travel and bacteria might have multiplied during this time. The different rates of growth on SuperPolymyxinTM medium (84.2 vs. 41.5%) might be due to an inoculum effect as colistin-susceptible isolates were shown to grow on SuperPolymyxinTM medium at ≥10<sup>6</sup> CFU/mL (Nordmann et al., 2016a). If this played a role in our study, one should expect a higher proportion of colistin susceptible bacteria in early compared to late samples due to longer storage times. However, the proportion of susceptible isolates (according to Etest <sup>R</sup> ) was similar throughout the study fluctuating between 16 and 27% (data not shown).

In total, 279 of 1,430 isolates were found to be colistinresistant by Etest <sup>R</sup> , which was confirmed in 218 isolates (78.1%) by BMD. This is in line with a recent comparison of different Etest <sup>R</sup> manufacturers (reference: BMD) where the categorical agreement was 79–85% (Matuschek et al., 2018).

To detect the 218 isolates with acquired colistin resistance, we had to perform 1,851 MALDI-TOF MS tests (to rule out intrinsically resistant isolates) and 1,430 Etest <sup>R</sup> cultures, thus experiencing a relevant workload. Most likely, this might be due to an inoculum effect due to bacterial growth during storage at room temperature (Nordmann et al., 2016a). A low quality of in-house SuperPolymyxinTM medium is unlikely since we strictly followed the recommendations (e.g., 7 days shelf life, storage of colistin stock solution in glass tubes) (Nordmann et al., 2016a). Others recently showed that SuperPolymyxinTM medium could even be used for up to 4 weeks (Abdul Momin et al., 2017).

Several studies evaluating SuperPolymyxinTM medium did not use clinical samples but monocultures or defined mixed cultures of different species (n: 2–9) and therefore cannot be compared with our study as we used fecal samples (Nordmann et al., 2016a; Abdul Momin et al., 2017). Other reports analyzed rectal colonization rates with mcr-1 positive isolates in pigs (98%) and hospitalized patients (0%) using SuperPolymyxinTM medium (Kieffer et al., 2017; Saly et al., 2017; Jayol et al., 2018). These studies did not focus in detail on the proportion of colistinsusceptible isolates growing on SuperPolymyxinTM medium. Notably the high number of colistin-susceptible isolates growing on SuperPolymyxinTM medium caused the major workload (**Figure 1**).

In a subset of 130 consecutively collected isolates, we studied the test performance of Etest <sup>R</sup> (reference: BMD) in our setting. Here, the categorical agreement (89.9%) was even slightly better compared to a recent study with 75 isolates (79–85%) depending on the manufacturer of the gradient diffusion test (**Table 1**) (Matuschek et al., 2018). Since the very major error rate (i.e., false susceptible isolates) of Etest <sup>R</sup> in our setting was 7.6%, we might have missed approximately 8% of non-intrinsically colistin-resistant isolates using our approach (i.e., screening for colistin resistance with Etest <sup>R</sup> and confirming with BMD, **Figure 1**). We decided to use Etest <sup>R</sup> as a screening method and BMD for confirmation as BMD was considered to cause TABLE 2 | Comparison of the Rapid PolymyxinTM NP Test and broth microdilution to detect colistin resistance.


<sup>a</sup>10 isolates were excluded the due to non-evaluable results in Rapid PolymyxinTM NP Test (no growth in the growth control) and 12 isolates were excluded due to non-interpretable results in broth microdilution (skipped wells).

an inapplicable workload due to the high number of isolates. However, an alternative to confirm/rule-out colistin resistance is the Rapid PolymyxinTM NP Test. (**Table 2**). However, Compared to Etest <sup>R</sup> , Rapid PolymyxinTM NP Test had a lower specificity (95.8 vs. 88.6%) and sensitivity (80.4 vs. 71.1%) to detect colistin resistance (**Tables 1**, **2**).

Although our study provides valuable data for the screening of colistin resistance in fecal samples, some limitations need to be addressed. First, we did not apply broth microdilution to confirm colistin resistance in all isolates growing on SuperPolymyxinTM medium but only used BMD for confirmation of colistin-resistant isolates according to Etest <sup>R</sup> . Since Etest <sup>R</sup> can have high falsesusceptible rates, we might have missed additional colistinresistant isolates. Second, we did not screen for newer mcrmembers (i.e., mcr-2–mcr-5). These members are still very rare and the European Center for Disease Prevention and Control currently recommend only a screening for mcr-1 (European Centre for Disease Prevention Control, 2016; Sun et al., 2018). Third, since we only tested fecal samples from international travelers, we are unable to draw any conclusions on the test

### REFERENCES


performance of SuperPolymyxin mediumTM when screening hospitalized patients, animals or environmental samples for colistin-resistant isolates.

In conclusion, SuperPolymyxinTM medium is suitable to screen for fecal colonization with colistin-resistant isolates but is associated with a high workload due to a high proportion of colistin-susceptible isolates growing on SuperPolymyxinTM medium. The confirmation of colistin resistance with the Rapid PolymyxinTM NP Test is a cheap and but less-accurate alternative to BMD.

### AUTHOR CONTRIBUTIONS

KB, RK, SP, and FS designed the study. SP, CC-M, and FS performed sampling and microbiological analyses. SP and FS analyzed the data. SP and FS drafted the manuscript. All authors reviewed and agreed on the final version of the manuscript. All authors have commented and agreed on the final version of the manuscript.

### FUNDING

This work was supported by the INTERREG Va program of the European Union (project EurHealth-1Health, grant number EU/INTERREG VA-681377) and in part by the German Federal Ministry for Education and Research (BMBF) within the Research Network Zoonotic Infectious Diseases (project #1Health-PREVENT, grant number 01KI1727A). The funders had no role in developing the study protocol.

### ACKNOWLEDGMENTS

We thank Martina Schulte, Anja Hassing and Barbara Grünastel for technical assistance.

isolated from humans in Thailand. J. Global Antimicrob. Res. 15, 32–35. doi: 10.1016/j.jgar.2018.06.007


**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2018 Przybysz, Correa-Martinez, Köck, Becker and Schaumburg. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

# Complete Genome Sequence of blaIMP–6-Positive Metakosakonia sp. MRY16-398 Isolate From the Ascites of a Diverticulitis Patient

Tsuyoshi Sekizuka<sup>1</sup> , Mari Matsui<sup>2</sup> , Tomiyo Takahashi<sup>3</sup> , Michiko Hayashi<sup>2</sup> , Satowa Suzuki<sup>2</sup> , Akihiko Tokaji<sup>3</sup> and Makoto Kuroda<sup>1</sup> \*

<sup>1</sup> Pathogen Genomics Center, National Institute of Infectious Diseases, Tokyo, Japan, <sup>2</sup> Antimicrobial Resistance Research Center, National Institute of Infectious Diseases, Higashimurayama, Japan, <sup>3</sup> The Public Health Institute of Kochi Prefecture, Kochi City, Japan ¯

A novel species of carbapenemase-producing Enterobacteriaceae (CPE) was isolated from a patient diagnosed with sigmoid colon diverticulitis. At first, laboratory testing suggested it was Klebsiella oxytoca or Pantoea sp.; however, a complete genome sequence of the isolate, MRY16-398, revealed that it could be novel species, most similar to [Kluyvera] intestini, of which taxonomic nomenclature is still under discussion. Orthologous conserved gene analysis among 42 related bacterial strains indicated that MRY16-398 was classified as the newly proposed genus Metakosakonia. Further, MRY16-398 was found to harbor the blaIMP−<sup>6</sup> gene-positive class 1 integron (In722) in plasmid pMRY16-398\_2 (IncN replicon, 47.4 kb in size). This finding implies that rare and opportunistic bacteria could be potential infectious agents. In conclusion, our results highlight the need for continuous monitoring for CPE even in nonpathogenic bacteria in the nosocomial environment.

Keywords: Kluyvera, blaIMP−6, carbapenemase, IncN, Metakosakonia

# INTRODUCTION

Antimicrobial resistance (AMR) is a global issue linked to increased and often unrestricted antibiotic use in the clinical settings, which leads to the dissemination of carbapenemresistant Enterobacteriaceae (CRE) in health care facilities (World Health Organization, 2017). Carbapenemases comprise three of the four Ambler classes as follows: Class A (Klebsiella pneumoniae carbapenemases, KPC, some variants of Guiana extended-spectrum β-lactamases, GES), Class B (metallo-β-lactamases, MBL including New Delhi metallo-β-lactamases, NDM, Verona integron-encoded metallo-β-lactamases, VIM, and imipenemase, IMP), and Class D (OXA-48-like carbapenemases) (Logan and Weinstein, 2017). These carbapenemase-producing Enterobacteriaceae (CPE) have the potential to facilitate the widespread transmission of antimicrobial resistance genes (ARGs) via mobile genetic elements through processes including natural competence, transformation, and plasmid transconjugation that can occur in any environment (Kelly et al., 2017; Rozwandowicz et al., 2018).

The widespread detection of CPE is an emerging issue with potentially serious public health implications; further, the distribution of the most common carbapenemase genes in Enterobacteriaceae occurs in a country- and region-specific manner (Logan and Weinstein, 2017). In Japan, IMP is the most predominant type of carbapenemase among clinical CPE isolates (Koyano et al., 2013; Ohno et al., 2017; Yamamoto et al., 2017). VIM, OXA-48-like, KPC, and NDM carbapenemases are detected at low frequencies in Japan, whereas KPC and NDM are predominant

### Edited by:

Yi-Wei Tang, Memorial Sloan Kettering Cancer Center, United States

### Reviewed by:

Ghassan M. Matar, American University of Beirut, Lebanon Filipa Grosso, Universidade do Porto, Portugal

### \*Correspondence:

Makoto Kuroda makokuro@niid.go.jp

### Specialty section:

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

Received: 31 July 2018 Accepted: 06 November 2018 Published: 22 November 2018

### Citation:

Sekizuka T, Matsui M, Takahashi T, Hayashi M, Suzuki S, Tokaji A and Kuroda M (2018) Complete Genome Sequence of blaIMP–6-Positive Metakosakonia sp. MRY16-398 Isolate From the Ascites of a Diverticulitis Patient. Front. Microbiol. 9:2853. doi: 10.3389/fmicb.2018.02853

**51**

in China and OXA-48-like and KPC are the predominant types of carbapenemases in Europe and United States, respectively.

Most CRE/CPE infections occur in hospitals, with major outbreaks at long-term care facilities and affecting patients with severe medical conditions under long stays for clinical management (Grabowski et al., 2017). There are a number of factors that predispose individuals to infections by CRE and other multi-drug resistant Enterobacteriaceae, including extendedspectrum β-lactamase (ESBL)-producers. Indeed, healthy carriers of CTX-M-type ESBL-harboring bacteria represent major public health concerns, because the carriage rates are on the rise, particularly in South East Asia and Eastern Mediterranean regions. Further, carriers from these regions could potentially spread these bacteria to other communities (Woerther et al., 2013).

Exposure to AMR bacteria can cause serious infections in patients with risk factors such as empirical antimicrobials, advanced age, immune-suppression, admission to the intensive care unit, mechanical ventilation, transplantation, and prolonged hospital stay (Gasink et al., 2009). Early intervention, through the administration of effective antimicrobials to such high-risk group patients, must be achieved to prevent death. In addition, a recent systematic review identified a prevalence of 0–29.5% for community-associated CRE, suggesting that the early detection of CRE-carriers among hospitalized patients upon admission to long-term care facilities might help to prevent nosocomial outbreaks and control the limited distribution of such emerging public health threats (Kelly et al., 2017).

Generally, Klebsiella, Escherichia coli, Enterobacter, and Citrobacter have been reported as the main contributors to the nosocomial transmission of CPE (Hrabak et al., 2014; Goodman et al., 2016; Kwong et al., 2018). Other opportunistic pathogens among Enterobacteriaceae species can acquire carbapenemase genes through plasmid transmission from main CPE contributors. Kluyvera is a group of gram-negative rodshaped bacteria and is a member of the Enterobacteriaceae family; the genus contains four species, namely Kluyvera ascorbata, Kluyvera cryocrescens, Kluyvera georgiana, and Kluyvera intermedia, which have all been found in humans (Farmer et al., 1981). K. ascorbata and K. cryocrescens were reported as potential pathogens that are associated with sepsis, bacteremia, catheter-associated urinary tract infections, pyelonephritis, and intraabdominal symptoms in immunocompromised hosts (Karadag Oncel et al., 2015; Yoshino et al., 2016).

Here, we report a novel IMP-6-producing isolate of Metakosakonia sp., namely strain MRY16-398, from a clinical specimen (ascites), and determined the genomic features of this carbapenemase-producing species.

### MATERIALS AND METHODS

### Ethics Approval

The study protocol was approved by the ethics committee of the National Institute of Infectious Diseases in Japan (Approval No. 642, 11/Dec/2015). It was conducted according to the principles of the Declaration of Helsinki, in compliance with the Law Concerning the Prevention of Infections and Medical Care for Patients of Infections of Japan; the ethical committee waived the need for written consent regarding the research of bacteria isolates; the personal data related to the clinical information were anonymized, and we do not request written consent for all patients suffering from bacterial infections.

### Bacterial Strains and Identification

Upon admitting a patient complaining of acute abdominal pain, abdominal computed tomography (CT) scanning showed a diverticulum in the descending and sigmoid colon, resulting in the diagnosis of sigmoid colon diverticulitis. A summary of laboratory data for the patient is shown in **Supplementary Table S1**. Empiric antimicrobial treatment with cefmetazole (0.5 g twice per day) was administered, and the volume of ascites was decreasing at 5 days from onset.

The Metakosakonia sp. MRY16-398 strain was isolated from the ascites of a patient with sigmoid colon diverticulitis in 2015 in Japan. The isolate was identified as Klebsiella oxytoca at the hospital microbiology laboratory using BD Phoenix (Becton Dickinson) with low reliability. Further phenotypic tests were performed using API 20E (bioMérieux) and Lysine-Indole-Motility Medium (Nissui, Tokyo Japan). Matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS)-based identification was conducted with a MicroFlex LT mass spectrometer (Bruker Daltonik), and analyzed using MALDI Biotyper software (Bruker Daltonik).

### Antimicrobial Susceptibility and CPE Screening Tests

Antimicrobial susceptibility was investigated by broth microdilution using the MicroScan Neg MIC 1J panel (Beckman Coulter) and Etest (bioMérieux) according to manufacturers' instructions (Clsi, 2018). Boronic acid, clavulanic acid, and sodium mercaptoacetic acid (SMA) were used as inhibitors for double-disk synergy tests (DDSTs) to identify AmpC-types and KPC-types, as well as extended-spectrum and metalloβ-lactamases, respectively. Carbapenemase production was assessed by performing a Carba NP test, as described previously (Nordmann et al., 2012). PCR testing was subsequently performed for potential CPE using a specific primer-pair for the following types of β-lactamase-encoding genes: blaIMP (Shibata et al., 2003), blaVIM (Shibata et al., 2003), blaOXA−48−like (Poirel et al., 2004), blaKPC (Bradford et al., 2004), and blaNDM (Segawa et al., 2017).

### Plasmid and Chromosome DNA Analysis With Short-Read Sequencing

Plasmid DNA was separated from chromosomal DNA by S1 nuclease-digestion followed by pulsed-field gel electrophoresis. Visible plasmid DNA and chromosomal DNA bands were extracted from the gel and purified using the ZR-96 Zymoclean gel DNA recovery kit (Zymo Research, Irvine, CA, United States). A DNA sequencing library was prepared using the Nextera XT DNA sample preparation kit (Illumina, San Diego, CA, United States) and was sequenced using an Illumina MiSeq

and NextSeq 500 for plasmids and chromosomes, respectively. Sequencing reads (plasmid: 2 × 300-mer, 140 × median coverage; chromosome: 2 × 150-mer, 99 × median coverage) were assembled into contigs using the A5-MiSeq pipeline (Coil et al., 2015). Plasmid replicon typing was performed using the curated PlasmidFinder database at the CGE website<sup>1</sup> (Carattoli et al., 2014).

### Whole-Genome Sequence (WGS) Analysis With Long-Read Sequencing

Genomic DNA from the isolated strain was purified by collecting cells from a 5-mL overnight culture grown in TSB broth. The cell pellet was resuspended in 500 µL of TE10 [10 mM tris (pH 8.0) and 10 mM EDTA] supplemented with 500 µL phenol/chloroform, and the cells were subsequently lysed by bead-beating for 10 min in ZR BashingBead lysis tubes (Zymo Research, Irvine, CA, United States) attached to a vortex adapter (MO BIO Laboratories, QIAGEN, Carlsbad, CA, United States). After centrifugation at 10,000 × g for 5 min; the upper phase was further purified using a Qiagen DNA purification kit (Qiagen, Germany).

The complete genome sequence of the strain was determined using a PacBio RSII sequencer for long-read sequencing (SMRT cell v3 using P6C4 chemistry; insert size, approximately 10 kb). Purified genomic DNA (∼2.0 µg) was used to prepare a SMRTbell library using a SMRTbell template prep kit 1.0 (PacBio, Menlo Park, CA, United States) according to manufacturer's instructions. The obtained raw polymerase reads were analyzed using the HGAP v3.0 pipeline based on Celera de novo assembler and Quiver polishing scripts (Chin et al., 2013).

De novo assembly was performed using HGAP 4 of SMRT Link Analysis v. 4.0.0.190159 and circulator version 1.5.3 (Chin et al., 2013). Error correction of tentative complete circular sequences was performed using Pilon version 1.18 with Illumina short reads (Walker et al., 2014). Annotation was performed using DFAST (Tanizawa et al., 2018) with databases as follows: DFAST default database, ResFinder database (Zankari et al., 2012), Bacterial Antimicrobial Resistance Reference Gene Database (PRJNA313047), and Virulence Factors Database (Chen et al., 2012). Circular representations of complete plasmid sequences were visualized using GView server (Petkau et al., 2010).

### Comparative Genome Sequence Analysis

All publicly available draft genome sequences were searched based on 16S rRNA gene homology, comparing them to that of the MRY16-398 strain, and 41 entries were retrieved from the NCBI genome database (see **Supplementary Table S2**). Among those 42 strains, orthologous core-gene sets were extracted using a nucleotide homology search with a threshold ≥80%, resulting in the identification of 479 core-gene sets (see **Supplementary Table S2**). Using these core-gene sets, core-gene phylogeny was generated using the maximum-likelihood phylogenetic method with FastTree v2.1.10 (Price et al., 2010). To construct a

<sup>1</sup>https://cge.cbs.dtu.dk//services/PlasmidFinder/

pairwise amino acid homology distance matrix, all amino acid sequences were compared pairwise, against each other, for each genome using the BLASTP program, which was followed by the calculation of average identity scores and standard deviations (**Supplementary Table S3**).

### Nucleotide Sequence Accession Numbers

The complete, annotated genomic sequence of Metakosakonia sp. MRY16-398 was deposited in a public database (accession numbers: chromosome, AP018756; pMRY16-398\_1, AP018757; pMRY16-398\_2, AP018758). The short- and long-read sequences for DNA-Seq were deposited in the DNA Data Bank of Japan (BioProject PRJDB7098, BioSample SSUB009772, DRA accession DRA007011).

# RESULTS

### Bacterial Identification of Metakosakonia sp. MRY16-398

A potential CPE, designated as strain MRY16-398, was isolated from the ascites after abdominocentesis. The isolate was identified as Klebsiella oxytoca at the hospital laboratory using BD Phoenix (Becton Dickinson) with low reliability, whereas API20E testing indicated the isolate should be a Pantoea sp. instead of K. oxytoca. This isolate was negative for lysine decarboxylate activity and showed weak motility, which indicated that the isolate was not Klebsiella. MALDI-TOF MS did not result in any bacterial species with a score higher than 2.000, which secures genus and probable species identification. The highest score value was 1.885 for Klebsiella aerogenes, followed by 1.789 for K. oxytoca.

Generally, 16S-rRNA gene sequencing is one of first tools used to determine the correct bacterial species of such novel CPE isolates, and thus we considered that WGS would be a more straightforward approach to characterize the species and plasmids involved in AMR.

The MRY16-398 isolate was observed to harbor blaIMP−6, exhibited resistance to meropenem, and was positive based on the Carba NP test and DDST using SMA. Further antimicrobial susceptibility testing indicated that MRY16-398 exhibited remarkably reduced susceptibility to most β-lactam antibiotics (**Table 1**).

### Whole-Genome Sequence Analysis of Metakosakonia sp. MRY16-398

Basic information regarding the complete chromosome and plasmid sequences of Metakosakonia sp. MRY16-398 is shown in **Table 2**. The strain possessed two plasmids, and pMRY16- 398\_2 was determined to be an IncN replicon plasmid, harboring multiple AMR-encoding genes including the blaIMP−<sup>6</sup> carbapenemase-encoding gene (**Table 2**). The IMP-6 metalloβ-lactamase is an IMP variant with a S214G amino acid substitution in the catalytic domain of IMP-1, resulting in significantly diminished enzymatic activity toward imipenem

Frontiers in Microbiology | www.frontiersin.org

### TABLE 1 | Antimicrobial susceptibility testing.

fmicb-09-02853 November 20, 2018 Time: 15:8 # 4


<sup>a</sup>CLSI guideline for MIC breakpoints of Enterobacteriaceae. S, Susceptible; I, Intermediate; R, Resistant.



NA, not available; ND, not detected.

but not meropenem (Oelschlaeger et al., 2005). Thus, MRY16- 398 showed susceptibility to imipenem, but resistance to other carbapenems (**Table 1**). The aacA4<sup>0</sup> -3 gene encoding aminoglycoside-300-adenylyltransferase, and the aadA2 gene encoding streptomycin 300-adenylyltransferase are involved in resistance to aminoglycosides. The tet(A) gene could be involved in reduced susceptibility to minocycline. Comparative analyses of the MRY16-398 genome sequence including the pMRY16-398\_2 plasmid are described in the following section.

### Orthologous Gene Phylogenetic Analysis of Metakosakonia sp. MRY16-398

To determine the potential bacterial species of the MRY16-398 strain, we performed orthologous gene phylogenetic analysis using 41 publicly available bacterial genome sequences (on 2017/03/14), including draft genomes (see the strain list in **Supplementary Table S2**). Among those 42 strains including MRY16-398, 479 orthologous gene sets were extracted at ≥80% nucleotide identity, and phylogeny and matrix distance clearly suggested that MRY16-398 was closely related to the bacterial species [Kluyvera] intestini str. GT-16 (Tetz et al., 2017), which was isolated from the stomach of a patient with gastric cancer (**Figure 1**). Recently, the taxonomic nomenclature for [Kluyvera] intestini str. GT-16 has been re-classified into a new proposed genera, namely Metakosakonia, which includes M. massiliensis JC163 (Alnajar and Gupta, 2017). As well as the proposal, this study demonstrated GT-16 strain shows the closest lineage to Metakosakonia, and having a distinct lineage from the main Kluyvera species (K. georgiana and K. intermedia) (**Figure 1**). Thus, MRY16-398 was found to be a clearly distinct lineage from Kluyvera, Pantoea, and other well-characterized genera of the Enterobacteriaceae family (**Figure 1**), indicating that this clinical isolate is a novel species. Here, we tentatively classified MRY16-398 as Metakosakonia sp.

### Structural Comparison of pMRY16-398\_2-Associated IncN Plasmids

S1-PFGE suggested that MRY16-398 carries two plasmids (**Figure 2A**), and the size of both plasmids corresponded to sequencing results as well as whole genome sequencing (**Table 2**). An analysis of conserved genes in the pMRY16- 398\_2 plasmid indicated that horizontally acquired AMR genes [class 1 integron, blaCTX−M−2, and tet(A)] are variable in each plasmid, although IncN backbone regions remained well conserved (**Figure 2B**). The class 1 integron has been classified as In722 (intI1, aacA4'-3, blaIMP−6, aadA2, and

FIGURE 1 | Pairwise homology sequence analysis using 479 core-gene sets from 42 strains related to MRY16-398. Maximum-likelihood phylogenetic analysis with 1,000× bootstrapping and the generation of a pairwise homology matrix were performed for the core-gene set listed in Supplementary Table S2. The MRY16-398 strain is highlighted with a bright purple background. The phylogeny and matrix distance clearly suggested that MRY16-398 is closely related to the bacterial species [Kluyvera] intestini str. GT-16 (Tetz et al., 2017), and distinct from other Enterobacteriaceae. Detailed homology % values for the matrix distances can be seen in Supplementary Table S3.

gcISKpn22) in the INTEGRALL database<sup>2</sup> (Moura et al., 2009). pMRY16-398\_2 shared an identical ARG profile and organization with pKPI-6 from Klebsiella pneumoniae KPI-6 (**Figure 2B**).

Pairwise alignment clearly showed that some genes involved in the conjugal transfer system have been excised and replaced with arsenical resistance proteins (Ars system) (Diorio et al., 1995) via an IS6100-mediated homologous recombination event (**Figure 3**). A mating transconjugation experiment to recipient E. coli showed negative plasmid transmission with pMRY16- 398\_2, although positive transmission was observed with a certain IncN plasmid harboring a full set of tra genes (data not shown). In addition to multiple AMR genes, pMRY16-398\_2 appears to have lost its self-conjugation transfer ability to other bacteria, whereas it acquired arsenic resistance. This likely led to an increase in the persistence and fitness of the novel bacterial

<sup>2</sup>http://integrall.bio.ua.pt/

species, which is an opportunistic pathogen, in the presence of high concentrations of disinfectants in the hospital environment.

### Additional Potential AMR Genes

fmicb-09-02853 November 20, 2018 Time: 15:8 # 6

A search for ARGs revealed an additional potential class A β-lactamase (MRY16398\_50310), with 76% amino acid similarity to the TEM-1A variant, in the chromosomal DNA of MRY16-398 (**Figure 4**). A maximum-likelihood phylogeny among TEM-1Arelated class A β-lactamases suggested that MRY16398\_50310 is closely related to those of Kluyvera spp.

# DISCUSSION

In this study, we isolated an IMP-6-producing novel Enterobacteriaceae species, Metakosakonia sp. strain MRY16- 398, from the ascites of a diverticulitis patient. Nosocomial CPE outbreaks are generally caused by virulent pathogens; however, avirulent bacteria can cause opportunistic infection as apparent pathogens upon acquiring a notable resistance determinant. Such rare cases of avirulent or novel bacteria species have are not often documented as case reports, because conventional testing for bacterial identification are not always correct for novel species, as shown in this study. WGS comprised a straightforward approach to characterize the overall features of this isolate and its plasmids involved in AMR, and this genome sequence will be helpful for further characterization of infections caused by Metakosakonia sp.

A few studies related to the Metakosakonia genus have been reported thus far, and the most genetically related genus Kluyvera represents an informative reference for further discussion. Kluyvera spp. strains have been reported as potential pathogens in immunocompromised hosts; in addition, the Kluyvera genus is one source of CTX-M genes, which are thought to be the most common and important extended-spectrum β-lactamaseencoding genes (Humeniuk et al., 2002). For example, KLUAproducing Kluyvera ascorbata can survive for a long time in environments such as sewage and the human gut and promote

drug resistance-associated gene transfer (Farmer et al., 1981). Based on a report of AMR in Kluyvera-related species, blaGES−<sup>5</sup> positive, carbapenem-resistant K. intermedia were isolated from a hospital environment (Ribeiro et al., 2014). Further, KPC-2 producing K. ascorbata has been reported in a case of biliary tract infection (Wang et al., 2018). A K. ascorbata isolate positive for the colistin resistance gene, mcr-1, was identified from hospital sewage in China (Zhao and Zong, 2016). Such opportunistic pathogens including Kluyvera represent important multi-drug resistant bacteria in clinical settings and other environmental sources.

Here, we isolated a novel species, Metakosakonia sp. MRY16- 398, and revealed the horizontal acquisition of the blaIMP−<sup>6</sup> plasmid in this novel species that is rarely associated with the clinical settings. Such novel opportunistic pathogens might act as a potential reservoir/source of clinically relevant antibiotic

resistance genes. In conclusion, these findings highlight the fact that bacterial identification is a crucial primary step when an isolate exhibits markedly reduced susceptibility as CPE. Moreover, continuous and comprehensive monitoring including WGS should be conducted for the detection of CPE even in nonpathogenic bacteria isolated from the clinical settings.

### DATA AVAILABILITY STATEMENT

The complete, annotated genomic sequence of [Kluyvera] intestini MRY16-398 was deposited in a public database (Accession Nos. chromosome, AP018756; pMRY16-398\_1, AP018757; pMRY16-398\_2, AP018758). The short- and longread sequences for DNA-Seq were deposited in the DNA Data Bank of Japan (BioProject PRJDB7098, BioSample SSUB009772, DRA accession DRA007011).

### AUTHOR CONTRIBUTIONS

TT and AT contributed to the isolation of the IMP-6 positive Metakosakonia sp. strain MRY16-398. MM performed S1-PFGE analysis to detect individual plasmids. MM and SS performed antimicrobial susceptibility testing and DNA preparation for whole genome sequencing. TS and MK performed genome sequencing and the comparative genome

### REFERENCES


analysis. MH performed mating transconjugation experiments. MK wrote the manuscript.

### FUNDING

This work was supported by the Research Program on Emerging and Re-emerging Infectious Diseases from the Japan Agency for Medical Research and Development, AMED (Grant Nos. JP17fk0108121, JP18fk0108048, and JP18fk0108019). The funding agencies had no role in the study design, data collection and analysis, decision to publish, or manuscript preparation.

### ACKNOWLEDGMENTS

We would like to thank Kumiko Kai for technical assistance with S1-PFGE analysis, and also Miho Nishio and Norimi Kishi for technical assistance with NGS experiments. We would also like to thank Editage (www.editage.com) for English language editing.

### SUPPLEMENTARY MATERIAL

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb. 2018.02853/full#supplementary-material


evidence from a complex multi-institutional KPC outbreak. PeerJ 6:e4210. doi: 10.7717/peerj.4210


**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2018 Sekizuka, Matsui, Takahashi, Hayashi, Suzuki, Tokaji and Kuroda. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

# Carbapenem-Resistant Enterobacteriaceae Infections: Taiwan Aspects

Shio-Shin Jean1,2, Nan-Yao Lee<sup>3</sup> , Hung-Jen Tang4,5, Min-Chi Lu<sup>6</sup> , Wen-Chien Ko<sup>3</sup> and Po-Ren Hsueh7,8 \*

<sup>1</sup> Department of Emergency, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan, <sup>2</sup> Department of Emergency Medicine and Department of Emergency and Critical Care Medicine, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan, <sup>3</sup> Department of Internal Medicine, National Cheng Kung University Medical College and Hospital, Tainan, Taiwan, <sup>4</sup> Department of Medicine, Chi Mei Medical Center, Tainan, Taiwan, <sup>5</sup> Department of Health and Nutrition, Chia Nan University of Pharmacy and Science, Tainan, Taiwan, <sup>6</sup> Department of Microbiology and Immunology, School of Medicine, China Medical University, Taichung, Taiwan, <sup>7</sup> Department of Laboratory Medicine, National Taiwan University Hospital, National Taiwan University College of Medicine, Taipei, Taiwan, <sup>8</sup> Department Internal Medicine, National Taiwan University Hospital, National Taiwan University College of Medicine, Taipei, Taiwan

### Edited by:

Yi-Wei Tang, Memorial Sloan Kettering Cancer Center, United States

### Reviewed by:

Hasan Ejaz, Al Jouf University, Saudi Arabia Zhiyong Zong, Sichuan University, China

> \*Correspondence: Po-Ren Hsueh hsporen@ntu.edu.tw

### Specialty section:

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

Received: 05 October 2018 Accepted: 12 November 2018 Published: 27 November 2018

### Citation:

Jean S-S, Lee N-Y, Tang H-J, Lu M-C, Ko W-C and Hsueh P-R (2018) Carbapenem-Resistant Enterobacteriaceae Infections: Taiwan Aspects. Front. Microbiol. 9:2888. doi: 10.3389/fmicb.2018.02888 Carbapenem-resistant Enterobacteriaceae (CRE), a major resistance concern emerging during the last decade because of significantly compromising the efficacy of carbapenem agents, has currently become an important focus of infection control. Many investigations have shown a high association of CRE infections with high case-fatality rates. In Taiwan, a few surveys observed that a significant proportion (29–47%) of the CR-Klebsiella pneumoniae isolates harbored a plasmidic allele encoding K. pneumoniae carbapenemases (KPC, especially KPC-2). A significant increase in the number of oxacillinase (OXA)-48-like carbapenemases among CR-K. pneumoniae isolates was observed between 2012 and 2015. By striking contrast, isolates of CR-Escherichia coli and CR-Enterobacter species in Taiwan had a much lower percentage of carbapenemase production than CR-K. pneumoniae isolates. This differs from isolates found in China as well as in the India subcontinent. Apart from the hospital setting, CRE was also cultured from the inpatients from communities or long-term care facilities (LTCF). Therefore, implementation of regular CRE screening of LTCF residents, strict disinfectant use in nursing homes and hospital settings, and appropriate control of antibiotic prescriptions is suggested to alleviate the spread of clinical CRE isolates in Taiwan. Although there are some promising new antibiotics against CRE, such as ceftazidime-avibactam, meropenem-vaborbactam, aztreonam-avibactam and cefiderocol, these agents are not available in Taiwan currently. Therefore, in order to effectively decrease case-fatality rates among patients with the infections owing to carbapenemase-producing CRE isolates, combination antibiotic schemes, including colistin (or amikacin) and/or tigecycline in combination with an anti-pseudomonal carbapenem agent, remain the mainstay for treating clinical CRE infections.

Keywords: carbapenem-resistant Enterobacteriaceae, carbapenemase, Klebsiella pneumoniae, Escherichia coli, long-term care facility, tigecycline, colistin, avibactam

# INTRODUCTION

fmicb-09-02888 November 24, 2018 Time: 16:19 # 2

Carbapenem-resistant Enterobacteriaceae (CRE), drawing great attention because of their serious resistance spectra and outbreak episodes in the northeastern United States since about two decades ago, have shown a high potential of rapid disseminations (Bratu et al., 2005a). Infections caused by CRE with production of various carbapenemases have become a major global worrisome concern due to their association with high (>30%) case-fatality rates in the current antibiotic pipeline (Tumbarello et al., 2012; Tzouvelekis et al., 2012; Navarro-San Francisco et al., 2013; Jean et al., 2015). In addition, McConville et al. conducted an observational study investigating CRE colonization (by rectal swabs) with subsequent impact on clinical outcomes for patients requiring admission to an intensive care unit (ICU) (McConville et al., 2017). By multivariate analysis, they found that CRE colonization also independently predicted development of a further systemic CRE infection at 30 days [adjusted odds ratio (aOR), 10.8; 95% confidence interval (CI), 2.8–41.9, P = 0.0006]. In 2017, the World Health Organization ranked CRE one of the first antibiotic-resistant "critical priority pathogens" (World Health Organization [WHO], 2017). Nevertheless, among these CRE isolates, the carbapenem non-susceptibility phenotypes are attributed to production of carbapenemase(s), or more likely, production of extended-spectrum β-lactamase (ESBL) plus AmpC β-lactamase with dysfunctional entry routes (i.e., porin change) of carbapenems, integrons and insertion sequence common region 1 (ISCR1) carrying various resistance genes, and/or efflux pumps, etc. (Doumith et al., 2009; Yang et al., 2012; Rui et al., 2018). To alleviate the resistance burden, close monitoring of CRE prevalence rates, as well as investigation of carbapenemase types, has become the main foci for infection control in most countries (Gupta et al., 2011).

Between 2004 and 2013, the prevalence rates of CRE among all clinical Enterobacteriaceae isolates have gradually risen (up to about 7%) in medical centers and major regional teaching hospitals throughout Taiwan. However, official surveillance data, conducted by Taiwan Centers for Disease Control (Taiwan CDC), clearly showed the annual prevalence rates of overall CRE isolates collected from ICUs rose from 3.7% in 2008 to 15.3% in 2017 (data until the third quarter) (Taiwan Nosocomial Infection Surveillance [TNIS], 2017). In addition, among Escherichia coli isolates acquired at ICUs, CRE rates rose from 1.2% in 2008 to 4.0% in 2017 at medical centers (bed number, >1000) (until the third quarter), while CRE rates in major regional teaching hospitals (bed number, 500–1000) rose from 1.0% in 2008 to 2.8% in 2017. These data might reflect the differences in CRE burden between hospitals of various care levels. By striking contrast, during the same period, significantly higher CRE rates for ICUacquired Klebsiella pneumoniae isolates were observed in medical centers (from 6.1 to 28.2%) and in major regional teaching hospitals (from 3.7 to 24.8%) than those of ICU-acquired E. coli (Taiwan Nosocomial Infection Surveillance [TNIS], 2017). It was determined by multivariate logistic regression analyses that clinicians in Taiwan (aOR, 1.73; 95% CI, 1.03–2.92) and in the US (aOR, 1.89; 95% CI, 1.05–3.39) are more likely to prescribe carbapenem antibiotics to treat the potential ESBL-producing Enterobacteriaceae infections instead of β-lactam combination agents compared to other countries (Harris et al., 2017). Clinical overuse of carbapenem agents unavoidably result in increased drug resistance (Tumbarello et al., 2012; Sheu et al., 2018).

# STUDY DESIGNS

We searched and reviewed literature from the 2002–2018 PubMed database containing important keywords, including "carbapenem-resistant Enterobacteriaceae," "carbapenemaseproducing Enterobacteriaceae," "prevalence rates," "mortality," "case-fatality," "Taiwan," "Centers for Disease Control," "Klebsiella pneumoniae," "Klebsiella pneumoniae carbapenemase (KPC)," "New Delhi metallo-β-lactamase (NDM)," "oxacillinase-48 (OXA-48)-like," "Escherichia coli," "Enterobacter species," "resistance mechanisms," "community-acquired," "long-term care facility," "Clinical and Laboratory Standards Institute," "European Committee on Antimicrobial Susceptibility Testing," "monotherapy," "combination therapy," "tigecycline," "colistin," and "novel antibiotics."

### PREVALENCE AND MORTALITY RATES OF CARBAPENEMASE-PRODUCING ENTEROBACTERIACEAE (CPE) AMONG CRE

In Pakistan, spread of special sequence types (STs; ST15 and ST48) of multidrug-resistant (MDR) Klebsiella species (most were ESBL-producing K. quasipneumoniae) isolates was observed since 2010 at one tertiary-care hospital (Ejaz et al., 2017). Similarly, among 83 carbapenem-resistant K. pneumoniae isolates studied in Brazil, a clinical isolate of KPC-2 and OKP-B-6 β-lactamase-producing K. quasipneumoniae subsp. similipneumoniae was also reported. (Nicolás et al., 2018). By stark contrast, global dissemination of CPE (especially K. pneumoniae isolates) in fact has been occurring at an alarming pace for many years (Logan and Weinstein, 2017). In Taiwan, most CP-K. pneumoniae isolates are the KPC producers (Jean et al., 2015). The blaKPC genes mostly reside on transferable plasmids which contain transposase, resolvase, and mobile transposons (Tn4401), thereby posing a formidable threat to infection control (Nordmann et al., 2009). As compared to other Enterobacteriaceae species, it is noteworthy that these mobile transposons are only detected among few epidemic clones of KPC producers of K. pneumoniae with distinct STs (predominantly ST258 in US and Israel, while ST11 in China and Taiwan) (Naas et al., 2008). The ST11 clone was the most prevalent ESBLproducing clone (particularly CTX-M-15, CTX-M-14, and SHV-5 types) of K. pneumoniae isolates in many Asian countries (Lee et al., 2011; Ma et al., 2013a).

Regarding the molecular analyses of carbapenemase types in CRE in Taiwan, a multicenter study first examined imipenemresistant K. pneumoniae isolates collected from 13 Taiwanese hospitals from 2002 through 2009 (Ma et al., 2013a). This study revealed that a majority (84.6%) of K. pneumoniae strains (mainly

ST11 clone) with > 2 mg/L minimum inhibitory concentrations (MICs) to imipenem harbored blaIMP−8, plus various genes encoding ESBLs in combination with loss of porins (mainly OmpK35) (Ma et al., 2013a). In 2012, a small-scale Taiwanese study, conducted by Chang et al., investigated 66 patients with CRE (comprising K. pneumoniae and E. coli) infections (n = 46) or colonization (n = 20). It showed that the CPE prevalence rate was 28.8% (n = 19; 14 of which harbored blaKPC−2), and a 30-day mortality rate was 50% (23/46) among patients with CRE infections (Chang et al., 2015). The CPE prevalence of Chang's study was different from that [11.5%; most were metalloβ-lactamase (MβL) producers] of the study conducted by Javed et al. (2016) in Pakistan during 2013–2014. In addition, this mortality rate was higher compared to that (40.8%) of the other CRE study (susceptibility evaluated by the disk diffusion method, and data interpretation in accordance with the 2009 criteria) at a medical center located in the middle-western part of Taiwan during 2010–2011 (Huang et al., 2014). In Chang's study, nearly one-half [47.8% (22/46)] of the infections originated in the lower respiratory tract. Furthermore, patients with a co-morbidity of diabetes mellitus, initially presenting with shock or high scores of the Acute Physiology and Chronic Health Evaluation (>23 points), or receiving non-susceptible antibiotic regimen therapy (regardless of single or combination drug therapy) for > 48 h


CRE, carbapenem-resistant Enterobacteriaceae. NA, non-applicable.

had significantly higher case-fatality rates (P < 0.05) than the other factors by the univariate analysis (Chang et al., 2015). Of note, patients who acquired CRE infections owing to KPC (dominant carbapenemase type)-producing Enterobacteriaceae also had a trend toward more fatal outcomes than those without KPC (P = 0.14) (Chang et al., 2015). An additional CRE study conducted at a single medical center of northern Taiwan during 2012–2013 showed that a CRE isolate with an imipenem MIC ≥ 16 mg/L independently predicted 14-day mortality among patients regardless if the isolate was from infection or colonization (Wu et al., 2016). In the survey published by Wu et al., KPC-2 was the dominant (87.1%) carbapenemase among CPE (29.5% of overall CRE), and the in-hospital mortality rate among patients with CR-K. pneumoniae was 43.8% (Wu et al., 2016). Wang et al. examined 1135 clinical isolates of various Enterobacteriaceae species collected from four major hospitals in Taiwan between 2010 and 2012. Fifty-seven isolates were carbapenemase-producing Enterobacteriaceae (CPE) isolates (5%), 54.4% of the CPE isolates co-harbored blaESBL alleles. Furthermore, Wang et al. found that Enterobacter cloacae isolates predominantly harbored the blaIMP−<sup>8</sup> allele (26/27), while isolates of CP-K. pneumoniae mainly harbored blaKPC−<sup>2</sup> alleles (16/17; ST11 clone predominated) (Wang et al., 2015). In addition, among Taiwanese CRE isolates surveyed, approximately one-half (45.8%) harbored various blaESBL genes (80% were the blaCTX−<sup>M</sup> types), while 62.0% harbored alleles encoding various plasmid-mediated AmpC β-lactamases (70.6% were blaDHA and 22.3% were blaCMY) (Wang et al., 2015). The high rates of K. pneumoniae isolates co-harboring ESBL and/or AmpC-encoding alleles among CRE corresponded well with those of an Asia-Pacific study with respect to Enterobacteriaceae isolates responsible for complicated intra-abdominal infections

(cIAI) and complicated urinary tract infections (cUTI) from 2008 through 2014 (Jean et al., 2017). According to the above findings, instead of carbapenemase production, ESBL with concomitant AmpC production and membrane impermeability caused by porin loss obviously more likely confers in vitro non-susceptibility to carbapenem agents (especially ertapenem) among CRE in Taiwan and the Asia-Pacific region. These results are similar to that found in Hong Kong (Ho et al., 2016).

Despite non-CPE isolates account for a majority of CRE in Taiwan, an up-surging CPE trend [three-fourths (74.5%) of CPE harboring blaKPC−<sup>2</sup> and were ST11 clone, followed by CPE harboring blaVIM−<sup>1</sup> allele (12.7%), etc.] was observed by Chiu et al. (2013). The CPE rates have increased from 0% in 2010 to 22.3% in 2012 among Taiwanese CR-K. pneumoniae isolates (showing > 1 mg/L of MICs to imipenem or meropenem) (Chiu et al., 2013). Furthermore, it is noteworthy that the prevalence rates of CPE [36.4%; mostly related to various blaKPC (75.9%), followed by blaOXA−48-like (8.8%), blaVIM−<sup>1</sup> (7.9%), and blaIMP−<sup>8</sup> (5.7%), etc] among the studied CR-K. pneumoniae isolates (approximately 99% were non-susceptible to imipenem, and 85.8% were collected from medical centers) were estimated to have 1.5-fold increase from 2012 through 2015. Most significantly, an increasing trend was observed in the middle-western part of Taiwan (Chiu et al., 2018). Data regarding important CRE studies in Taiwan and other countries are shown in **Table 1**. There were big variations in the rates of CPE among clinical CRE isolates in Taiwan as compared to those from other countries (Lorenzoni et al., 2017; Woodford et al., 2017; Laolerd et al., 2018). It is notable that persistently high KPC-2 rates (73.9% in 2012, and 61.1% in 2015) and emergence of KPC-17 as well as KPC-34 since 2014 were observed (Chiu et al., 2018). As the K. pneumoniae of ST11 clone accounted for

86.9% of overall CP-K. pneumoniae isolates (Chiu et al., 2018), heightened infection control maneuvers are absolutely warranted for Taiwanese medical centers. **Figure 1** illustrates the annual percentages of overall carbapenemase producers, overall KPCs and KPC-2 producers, as well as OXA-48-like producers among clinical CR-K. pneumoniae isolates (mainly non-susceptible to imipenem and ertapenem) collected between 2012 and 2015 in Taiwan (Chiu et al., 2018; Jean et al., 2018b).

Carbapenem-resistant Enterobacteriaceae that only harbor blaOXA−48-like allele theoretically exhibit non-susceptibility to most carbapenem agents while sparing most β-lactam agents. However, most of the blaOXA−48-like-harboring CRE (especially K. pneumoniae) isolates are actually also the ESBL and/or AmpC co-producers (Saïdani et al., 2012; Girlich et al., 2014). Consequently, many OXA-48-like-producing CRE strains show in vitro extensively drug-resistant phenotypes. In 2014, an OXA-48-like carbapenemase-producing E. coli strain, which concomitantly harbored blaCTX−M−1, was cultured from pus sample of a Cambodian patient who resided in Taiwan (Jao et al., 2017). An additional 2010 study from India regarding CRE, some of which were CPE harboring blaNDM, blaVIM or blaOXA−181-like alleles, revealed that the case-fatality rate was 56.7% (Mariappan et al., 2017). A sharp rise in the annual prevalence rate of OXA-48-like production among Taiwanese CPE isolates between 2012 and 2015 was also observed (Chiu et al., 2018) (**Figure 1**). Spread of blaNDM alleles occurred rapidly in the Indian subcontinent during the last decade (Kumarasamy et al., 2010), and also in southeastern Asia (Vietnam and the Philippines) (Jean et al., 2017). Nevertheless, reports regarding the NDM-producing Enterobacteriaceae colonization in one patient (3 K. pneumoniae isolates harboring blaNDM−1, probably acquired from India) (Wu et al., 2010) and infection in the other patient (1 Klebsiella oxytoca isolate harboring blaNDM−1, acquired from China) (Lai et al., 2011) in Taiwan were rare (Jean et al., 2017).

### COMMUNITY AND LONG-TERM CARE FACILITY (LTCF)-ACQUIRED CRE

Few studies have addressed the percentage of community acquisition of clinical CRE infections in the PubMed database. However, a few investigations in Taiwan have addressed this important topic. A clinical study, conducted at a medical center located in southern Taiwan in 2015, showed approximately 30% (23/78) of cases, with predominance of elderly female patients and infections originating in the urologic system, acquired CRE in a community setting (Tang et al., 2016). The percentage of community-acquired CRE was higher than that [12% (14/117)] of another Taiwanese study conducted at a large medical center in 2010 (P = 0.002) (Lai et al., 2013), and also higher than that [5.6% (17/305)] of an US-CRE study (Thaden et al., 2014). In the latter study, 91% of overall CRE isolates were K. pneumoniae, while 59% were cultured from urine between 2008 and 2012 (P < 0.0001) (Thaden et al., 2014). In addition, it was observed that many residents from the LTCF in Taiwan were also the carriers of CRE, most (24/27) were K. pneumoniae (Chuang, 2015; Mao et al., 2018). This was similar to the study conducted by Lee C.M. et al. (2017) who observed that a significant proportion of K. pneumoniae strains cultured from rectal swabs of LTCF residents were CRE (38.9 and 88.9% showed in vitro non-susceptibility to ertapenem and imipenem, respectively). Therefore, in Taiwan, regular screening for CRE carriage (urine, sputum, and rectal swabs) among the LTCF residents to implement subsequent contact isolation precaution, hand hygiene, strict implementation of environmental disinfection (including equipment of LTCF), and adequate education to caregivers and family about the importance of infection control are beneficial to decrease CRE carriage rates of LTCF residents. All of these measures are crucial for preventing CRE spread at LTCF (Lee C.M. et al. (2017); Lai et al., 2018; Liu et al., 2018).

### SURVEILLANCE DATA OF ANTIMICROBIAL RESISTANCE IN TAIWAN IN 2017

To fully explore the evolutionary trends of the annual resistance burden, the Taiwan CDC conducted a consecutive 4-year survey of antimicrobial resistance (AMR) regarding important clinical bacterial pathogens since 2017 (Jean et al., 2018b). Among 673 bloodstream K. pneumoniae isolates collected from 16 major teaching hospitals throughout Taiwan in 2017, 71 (10.5%) displayed non-susceptibility to at least one carbapenem agent. In addition, a total of 92 (13.8%) K. pneumoniae isolates exhibited ESBL phenotypes that were unrelated to KPC and/or NDM production [with a 92% non-susceptible (NS) rate to piperacillin-tazobactam, 100% to ceftazidime, and 74% to cefepime]. Phenotypic ESBL isolates had high NS rates to ertapenem (40.2%), tigecycline [7.6% based on the US Food and Drug Administration (FDA) criteria, and 16.3% based on the criteria recommended by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) in 2018 (European Committee on Antimicrobial Susceptibility Testing [EUCAST], 2018)], amikacin (13.0%), as well as ceftolozanetazobactam [70.7% based on criteria recommended by Clinical, and Laboratory Standards Institute [CLSI], 2018, and 80.4% based on the EUCAST 2018 criteria]. Of note, among K. pneumoniae blood isolates validated as CPE by multiplex PCR tests, 24 (3.6%) harbored blaKPC allele (1 isolate showed susceptibility to all tested carbapenem agents), eight (33.3%) of the KPC-producing K. pneumoniae isolates were acquired from a community setting. In addition, 7 (1.0%) K. pneumoniae isolates harbored blaOXA−48-like alleles (1 isolate showed susceptibility to all tested carbapenem agents, while 6 exhibited ESBL phenotypes), 1 (0.15%) and 4 (0.6%) isolates harbored blaNDM and mcr-1 gene, respectively. The rate of KPC-producing isolates [32.4% (23/71)] among Taiwanese CR-K. pneumoniae isolates in 2017 was significantly higher than that (2.8%) in the 2010-2012 Taiwanese study (Wang et al., 2015). Furthermore, out of the total KPC-producing K. pneumoniae isolates (n = 24), there were two (8.3%) showing non-susceptibility to ceftazidimeavibactam, 5 (20.8%) showing non-susceptibility to colistin, and 4 (16.7%) displaying non-susceptibility to tigecycline based on the EUCAST 2018 criteria (Jean et al., 2018b, in Infect Drug Resist).

### CARBAPENEM RESISTANCE MECHANISMS AMONG CR-E. coli ISOLATES

fmicb-09-02888 November 24, 2018 Time: 16:19 # 6

As compared to a much higher CPE rate among clinical isolates of K. pneumoniae, two Taiwanese studies concluded that coexistence of a plasmidic AmpC β-lactamase (DHA-1, CMY-2) in combination with loss of an outer membrane porin (OmpC/F) is the main mechanism responsible for non-susceptibility to carbapenems for CR-E. coli (Chia et al., 2009; Ma et al., 2013b). This finding differs from that in Mainland China [89% of CR-E. coli isolates (n = 164) harbored blaNDM (n = 81) or blaKPC (n = 65) alleles] (Zhang et al., 2017), and in India (most CR-E. coli harbored blaNDM with or without blaOXA−48-like alleles) from 2011 through 2013 (Mohanty et al., 2017). Surveillance data from the Taiwan CDC-AMR 2017 showed a total of 686 E. coli blood isolates were collected. Among them, only 14 (2.0%) isolates showed in vitro non-susceptibility to ertapenem. Of note,


KPC, Klebsiella pneumoniae carbapenemase. CPE, carbapenemase-producing Enterobacteriaceae. MβL, metallo-β-lactamase. VIM, Verona integron-encoded metalloβ-lactamases.

5 (0.7%) isolates harbored the mcr-1 gene, similar to the one found in Mainland China (Quan et al., 2017), while the other two harbored a blaKPC and blaNDM−<sup>1</sup> allele, respectively.

# CARBAPENEM RESISTANCE MECHANISMS AMONG CR-Enterobacter spp. ISOLATES

Fewer studies have examined mechanisms of non-susceptibility to carbapenem agents among isolates of Enterobacter species than those for K. pneumoniae in the PubMed database. A study focusing on ertapenem-resistant Enterobacter cloacae isolates (MIC > 2 mg/L, in accordance with the CLSI 2009 criteria) collected in 2007 was conducted by Yang et al. (2012) in Taiwan. Analysis of porin expression, detection of efflux pumps and β-lactamase(s), as well as susceptibility tests were performed. Only a few isolates [3/53 (5.7%)] harbored blaIMP−8, while porin change (30–40%) and efflux pump(s) (≥70%) in combination with ESBL or AmpC significantly contributed to ertapenem non-susceptibility for Taiwanese ertapenem-resistant E. cloacae isolates. Because of geographic variations, production of different types of carbapenemases among the carbapenem (imipenem, ertapenem)-NS E. cloacae isolates was observed in India (VIM-2, VIM-6, and NDM-1) (Khajuria et al., 2014) and in Israel (KPC-2) (Marchaim et al., 2008).

# MONOTHERAPY VS. COMBINATION THERAPY FOR CRE INFECTION

Patients with CPE infections are undoubtedly at an extremely high risk for inappropriate antibiotic therapy (Tumbarello et al., 2012; Chang et al., 2015). As stated earlier, tigecycline and colistin showed good in vitro susceptibility results against most CRE isolates and KPC or MβL-producing CPE isolates (Bratu et al., 2005b; Castanheira et al., 2008). As stated in the Taiwan CDC-AMR 2017 data, the tigecycline/colistin MIC values of one E. coli and one K. pneumoniae isolate harboring the blaNDM allele were 0.06/0.25 mg/L and 0.25/0.25 mg/L, respectively. Nevertheless, to maximize the clinical effectiveness of colistin against impending resistant Gram-negative bacteria (GNB; i.e., MIC of colistin is 2 mg/L), a study showed that both the loading, as well as maintenance dose should be increased (Garonzik et al., 2011). In addition, tigecycline was validated as a bacteriostatic agent with suboptimal concentrations in bloodstream (<2 mg/L) after administration of the standarddose regimen (Yahav et al., 2011). Consequently, monotherapy with tigecycline or colistin is not recommended in the treatment of severe GNB infections owing to high clinical failure and superinfection rates. Bass et al. found that combination therapy of multiple agents that have appropriate in vitro activities for ≥ 48 h was associated with improved survival rates (OR, 0.19; 95% CI, 0.06 to 0.56; P < 0.01) for critically ill patients with CR-GNB bacteremia (regardless of Enterobacteriaceae species or non-fermentative GNB) (Bass et al., 2015; Lee C.H et al., 2017). Many combination therapies were also been shown to result in favorable outcomes in patients with CP-GNB. For example, a combination of doripenem with colistin was reported to have an excellent in vitro synergistic effect against CP-K. pneumoniae isolates (Jernigan et al., 2012; Rodríguez-Baño et al., 2018). In addition, a tigecyclinecolistin combination was also shown to be superior to colistin monotherapy in decreasing future resistance to colistin (Lee et al., 2009).

In Taiwan, data comparing therapeutic efficacy between different antibiotic regimens against CRE or CPE were scarce (Lee et al., 2011; Ku et al., 2017). Although the rate of tigecycline non-susceptibility among Taiwanese CRE (E. coli plus K. pneumoniae) blood isolates in 2017 was not high [3.5% (3/85) based on FDA criteria and 12.9% (11/85) based on the EUCAST 2018 criteria], a Taiwanese study aiming specifically at 16 tigecycline-resistant K. pneumoniae strains elucidated that ramR deficiency and/or widespread mutated tet(A) are the main mechanisms conferring non-susceptibility to the tetracyclineclass agents for these K. pneumoniae isolates (Chiu et al., 2017). In addition, acquisition of fosfomycin-resistant genes, fosA subtypes and fosC2, which were mainly transmitted by plasmids and/or transposons, was found in many of the Enterobacteriaceae species in Eastern Asian countries, including Taiwan (Yang et al., 2017). In addition, although amikacin, gentamicin and fosfomycin exhibited acceptable in vitro susceptibility rates against some CPE species (KPC and VIM) (Endimiani et al., 2010; Lorenzoni et al., 2017), these agents are recommended as mere adjunctive options against critical CP-GNB infections because of rapid induction of resistance after monotherapy (Tseng et al., 2017; Yang et al., 2017). By contrast, against the non-CP-CR-K. pneumoniae strains (with changes of outmembrane porins) collected from a medical center of southern Taiwan during 2008–2012, Ku et al. used the sub-inhibitory concentrations (1/2 × MIC) of antibiotics under evaluation and time-kill studies to analyze in vitro synergism between different combination regimens. As compared to tigecycline plus fosfomycin and colistin plus fosfomycin, Ku et al. observed that tigecycline in combination with colistin showed much better in vitro synergism, as well as bactericidal efficacy (Ku et al., 2017).

### CLINICAL EVIDENCE OF COMBINED ANTIMICROBIAL SCHEMES AGAINST CPE INFECTIONS

Although the above in vitro synergy data supported the use of combination regimens, well-designed clinical randomized control trials that strictly investigated which antimicrobial schemes are effective against KPC-producing Enterobacteriaceae isolates were few. Nevertheless, there were many retrospective studies in favor of the use of various combination therapy regimens against clinical CPE infections. None of these studies were investigated in Taiwan. The recommended antibiotic regimens included anti-pseudomonal carbapenem plus colistin or tigecycline, polymyxin plus tigecycline or

aminoglycoside, or meropenem in combination with tigecycline and colistin. Important data regarding therapeutic efficacy of different antibiotic regimens (various combination schemes vs. monotherapy) for patients with CPE infections are illustrated in **Table 2** (Lee and Burgess, 2012; Qureshi et al., 2012; Tumbarello et al., 2012; Tzouvelekis et al., 2012, 2014; Daikos et al., 2014; Food and Drug Administration [FDA], 2018; Rodríguez-Baño et al., 2018).

### OTHER ANTIMICROBIAL AGENTS AGAINST CPE INFECTIONS

Apart from tigecycline and colistin against CPE, previous studies showed that avibactam, a bridged diazabicyclo octanone, exhibited good in vitro activity against GNB with serine β-lactamases (especially against the KPC producers) when combined with ceftazidime, but it is inactive against class B β-lactamases (Bonnefoy et al., 2004; Tzouvelekis et al., 2012; van Duin and Bonomo, 2016). In the Taiwan CDC-AMR 2017 survey, the MICs for ceftazidime-avibactam for one E. coli and one K. pneumoniae isolate that harbored the blaNDM allele were both > 64 mg/L. Consequently, the Taiwanese 2017 data were consistent with previous studies. Ceftazidime-avibactam has been approved by the US-FDA in treatment of CRErelated cIAI and cUTI. In addition, meropenem-vaborbactam has an equivalent potency to ceftazidime-avibactam against class A carbapenemases (typified by KPC), but also has limited activity against class B along with oxacillinase carbapenemases (Castanheira et al., 2017; Chew et al., 2018). This drug was also approved in treatment of CRE-related cUTI recently. By contrast, aztreonam-avibactam displays good in vitro susceptibility in dual-carbapenemase (KPC, NDM)-producing CRE isolates (Chew et al., 2018). In addition, cefiderocol, a new siderophore cephalosporin, also shows potentially good in vitro activity against both KPC, as well as NDM-producing Enterobacteriaceae isolates (Hackel et al., 2018). The clinical trials comparing clinical efficacy of aztreonam-avibactam and cefiderocol with other best available antibiotics are undergone. The other future promising agents against CPE include imipenem-relebactam, plazomicin, and eravacycline (Rodríguez-Baño et al., 2018).

### ANTIBIOTIC TREATMENT FOR NON-CARBAPENEMASE-PRODUCING-CRE INFECTIONS

Compared to ample databases regarding treatment of CPE, limited data were available regarding treatment

### REFERENCES

Bass, S. N., Bauer, S. R., Neuner, E. A., and Lam, S. W. (2015). Impact of combination antimicrobial therapy on mortality risk for critically ill patients with carbapenem-resistant bacteremia. Antimicrob. Agents Chemother. 59, 3748–3753. doi: 10.1128/AAC.00091-15

recommendations for clinical non-CP-CRE isolates. Nevertheless, Tamma et al. observed MIC values for imipenem and meropenem in the subgroup of CPE isolates (>90% were KPC producers) were significantly higher than those of the non-CP-CRE subgroup. Furthermore, the overall 14-day mortality rate was 4-fold higher among patients with CPE bacteremia than those of the non-CP-CRE group (Tamma et al., 2017). Recently, we investigated detection of CPE predictors among ertapenem-NS Enterobacteriaceae isolates causing IAI collected from patients hospitalized in Asia-Pacific countries during 2008–2014 (Jean et al.,2017). This Asia-Pacific IAI-CRE study showed that imipenem non-susceptibility (i.e., MIC > 2 mg/L) and culturing ertapenem-NS Enterobacteriaceae isolates from the peritoneal space were highly likely as CPE (dominated by the blaNDM and blaIMP-harboring isolates, rather than KPC producers; Jean et al., 2018a, in Infect Drug Resist). Moreover, the non-CP-ertapenem-NS IAI Enterobacteriaceae isolates were more likely to have cefepime MICs of > 8 mg/L, which might be valuable in distinguishing it from the CPE. However, the clinical application of the dose adjustment of an anti-pseudomonal carbapenem agent to effectively treat the non-CP ertapenem-NS Enterobacteriaceae isolates needs further investigation in clinical scenarios.

### SUMMARY

In Taiwan, the main research gap is that there are no sufficient clinical studies exploring the significant risk factors with respect to acquisition of CRE isolates among hospitalized or LTCF patients. Of note, the annual CPE proportion has been on a sharp rise among clinical CRE isolates since 2012, particularly for CR-K. pneumoniae isolates. To effectively limit the spread of CPE (especially ST11 K. pneumoniae clone) in clinical settings, close monitoring of this worrisome MDR trend is warranted at all major teaching hospitals, as well as nursing homes. In addition, although ceftazidime-avibactam and meropenemvaborbactam show excellent efficacy against some CPE, it is not currently available in Taiwan. Combination therapy schemes, such as colistin and/or tigecycline plus an anti-pseudomonal carbapenem agent, are still the preferred treatment for CRE and CPE infections in Taiwan.

### AUTHOR CONTRIBUTIONS

S-SJ, N-YL, and P-RH collected and analyzed the data. S-SJ participated in the writing of the manuscript. S-SJ, N-YL, H-JT, M-CL, W-CK, and P-RH read and approved the final version of the manuscript.


City: a new threat to our antibiotic armamentarium. Arch. Intern. Med. 165, 1430–1435. doi: 10.1001/archinte.165.12.1430



coli in Taiwan: first identification of KPC-2 or NDM-1-producing E. coli in Taiwan. BMC Infect. Dis. 13:599. doi: 10.1186/1471-2334-13-599



in the UK: results of the UK infection-Carbapenem Resistance Evaluation Surveillance Trial (iCREST-UK). J. Antimicrob. Chemother. doi: 10.1093/jac/ dkx471 [Epub ahead of print].


**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2018 Jean, Lee, Tang, Lu, Ko and Hsueh. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

# Genetic Diversity of Carbapenem-Resistant Enterobacteriaceae (CRE) Clinical Isolates From a Tertiary Hospital in Eastern China

Edited by:

Miklos Fuzi, Semmelweis University, Hungary

### Reviewed by:

Hong Du<sup>1</sup>

\* and Liang Chen<sup>5</sup>

Krisztina M. Papp-Wallace, Louis Stokes Cleveland VA Medical Center, United States Hasan Ejaz, Al Jouf University, Saudi Arabia Michael Bachman, University of Michigan, United States

### \*Correspondence:

Hong Du hong\_du@126.com; hongdu@suda.edu.cn Haifang Zhang haifangzhang@sina.com Xia Zhang xiazhangsz@qq.com †These authors have contributed

equally to this work

### Specialty section:

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

Received: 08 October 2018 Accepted: 27 December 2018 Published: 15 January 2019

### Citation:

Miao M, Wen H, Xu P, Niu S, Lv J, Xie X, Mediavilla JR, Tang Y-W, Kreiswirth BN, Zhang X, Zhang H, Du H and Chen L (2019) Genetic Diversity of Carbapenem-Resistant Enterobacteriaceae (CRE) Clinical Isolates From a Tertiary Hospital in Eastern China. Front. Microbiol. 9:3341. doi: 10.3389/fmicb.2018.03341 Minhui Miao1,2† , Huiyan Wen<sup>1</sup>† , Ping Xu<sup>3</sup>† , Siqiang Niu<sup>4</sup> , Jingnan Lv<sup>1</sup> , Xiaofang Xie<sup>1</sup> , José R. Mediavilla<sup>5</sup> , Yi-Wei Tang<sup>6</sup> , Barry N. Kreiswirth<sup>5</sup> , Xia Zhang<sup>7</sup> \*, Haifang Zhang<sup>1</sup> \*,

<sup>1</sup> Department of Clinical Laboratory, The Second Affiliated Hospital of Soochow University, Suzhou, China, <sup>2</sup> Department of Clinical Laboratory, Jiangyin Hospital of Traditional Chinese Medicine, Jiangyin, China, <sup>3</sup> Department of Clinical Laboratory, The Fifth People's Hospital of Suzhou, Suzhou, China, <sup>4</sup> Department of Laboratory Medicine, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China, <sup>5</sup> Public Health Research Institute Tuberculosis Center, New Jersey Medical School, Rutgers University, Newark, NJ, United States, <sup>6</sup> Memorial Sloan Kettering Cancer Center, New York, NY, United States, <sup>7</sup> Department of Clinical Laboratory, The North District of Affiliated Suzhou Hospital, Nanjing Medical University, Suzhou, China

The prevalence of carbapenem-resistant Enterobacteriaceae (CRE) is increasing globally, with different molecular mechanisms described. Here we studied the molecular mechanisms of carbapenem resistance, including clonal and plasmid dissemination, of 67 CRE isolates collected between 2012 and 2016 from a tertiary hospital in Eastern China, an CRE endemic region. Species identification and susceptibility testing were performed using the BD Phoenix Automated Microbiology System. Isolates were characterized by PCR (for carbapenemases, ESBLs, AmpC and porin genes), multilocus sequence typing (MLST), pulsed-field gel electrophoresis (PFGE), and conjugation transfer experiments. Selected blaKPC−<sup>2</sup> -harboring plasmids were subjected to nextgeneration sequencing using the Illumina Miseq platform. Among the 67 CRE isolates, 42 Klebsiella pneumoniae, 10 Serratia marcescens, 6 Enterobacter cloacae, 2 Raoultella ornithinolytica, 2 K. oxytoca, 1 K. aerogenes, and 4 Escherichia coli isolates were identified. Six different carbapenemases were detected, including blaKPC−<sup>2</sup> (n = 45), blaKPC−<sup>3</sup> (n = 1), blaNDM−<sup>1</sup> (n = 6), blaNDM−<sup>5</sup> (n = 1), blaIMP−<sup>4</sup> (n = 2), and blaVIM−<sup>1</sup> (n = 2); blaOXA−48-like genes were not detected. One E. cloacae strain possessed both blaNDM−<sup>1</sup> and blaKPC−3, while two E. cloacae isolates harbored blaNDM−<sup>1</sup> and blaVIM−1. ESBLs (CTX-M, SHV, and TEM) and/or AmpC (CMY, DHA, and ACT/MIR) genes were also identified in 59 isolates, including 13 strains that lacked carbapenemases. Several insertions or stop codon mutations were found within porin genes of K. pneumoniae, E. coli and S. marcescens isolates, both with and without carbapenemases. The 42 K. pneumoniae isolates belonged to 12 different sequence types (ST), with ST11 being the most common, while the 6 E. cloacae isolates comprised 4 different STs. The 10 S. marcescens all shared the same PFGE pulsotype, suggestive of clonal

spread. Complete plasmid sequencing and PCR screening revealed both intra-strain and inter-species spread of a common blaKPC−2-harboring plasmid in our hospital. Taken together, our study revealed extensive genetic diversity among CRE isolates form a single Chinese hospital. CRE isolates circulating in the hospital differ significantly in their species, STs, porin genes, carbapenemase genes, and their plasmid content, highlighting the complex dissemination of CRE in this endemic region.

Keywords: carbapenem-resistant Enterobacteriaceae, carbapenemase, resistance mechanism, genetic diversity, plasmid

### INTRODUCTION

Enterobacteriaceae are among the most common pathogenic Gram-negative bacteria (GNB), causing various community- and healthcare-acquired infections. Nowadays, multidrug resistant GNB (MDR-GNB) are increasingly described in clinical settings, and carbapenems are regarded as the most effective antibiotic therapy for infections caused by MDR-GNB. However, as a result of clinical use of carbapenems since the late 1980s, the occurrence of carbapenem-resistant Enterobacteriaceae (CRE) has been increasingly reported worldwide, including in China (Gupta et al., 2011; Nordmann and Poirel, 2014). According to reports by the China Antimicrobial Resistance Surveillance System (CARSS), the detection rate of CRE in China increased from 2005 to 2014, demonstrating a continuous upward trend and suggesting a worsening situation (Hu et al., 2016).

Carbapenem resistance in Enterobacteriaceae can arise through distinct molecular mechanisms, mainly via the production of carbapenemases, but also as a consequence of outer membrane porin dysfunction coupled with hyperproduction of AmpC cephalosporinases or extended-spectrum β-lactamases (ESBLs) (Bush and Jacoby, 2010; Bush and Fisher, 2011). Carbapenemases are a group of β-lactamases that are capable of hydrolyzing carbapenem antibiotics, in addition to cephalosporins and other β-lactam antimicrobials. Three major class of carbapenemases are widespread globally in clinical CRE isolates, including class A (mainly KPC), class B (VIM, NDM, and IMP) and class D (OXA-48 and its variants, OXA-162 and OXA-181, etc.) (Ambler, 1980; Hall et al., 2003; Nordmann and Poirel, 2014). Notably, carbapenemase genes are primarily carried by large conjugative plasmids, thereby facilitating horizontal transfer of carbapenem resistance among different bacterial strains and species. As mentioned above, another common mechanism of carbapenem resistance involves the combination of porin dysfunction with hyper-production of AmpC (e.g., CMY, DHA, and ACT) or ESBLs (e.g., TEM, SHV, and CTX-M) (Logan and Weinstein, 2017). The lack of the production of porins can preclude diffusion of antibiotics through bacterial membranes, along with the action of ESBL and AmpC enzymes, thereby producing the phenotype of carbapenem resistance in Enterobacteriaceae (Paterson and Bonomo, 2005; Jacoby, 2009; Bush and Jacoby, 2010). Unlike carbapenemases, porin dysfunction-associated resistance is not able to spread through horizontal transfer, but may disseminate via clonal expansion.

Dissemination of CRE usually demonstrates geographical and temporal variation in specific global regions. Here we characterized CRE clinical strains collected from a tertiary hospital in eastern China between August 2012 and August 2016. The genetic relatedness, antimicrobial susceptibility, and carbapenem-resistance mechanisms of these CRE isolates were examined in detail.

# MATERIALS AND METHODS

### Identification of Carbapenem-Resistant Enterobacteriaceae Isolates

Sixty-seven unique (one isolate per patient) CRE clinical isolates were retrospectively collected from the Second Affiliated Hospital of Soochow University between August 2012 and August 2016. In this study, carbapenem resistance was defined as resistance to meropenem or imipenem based on 2016 Clinical and Laboratory Standards Institute (CLSI) guidelines (CLSI, 2016). The isolates were collected from various sources, including sputum (n = 48), urine (n = 14), blood (n = 2), catheter (n = 1), ascites (n = 1), and drainage fluid (n = 1). Species identification was performed using the Phoenix 100 Automated Microbiology System (Becton-Dickinson, United States), and confirmed by 16S rRNA sequencing (Weisburg et al., 1991). This study was approved by the institutional review board (IRB) of The Second Affiliated Hospital of Soochow University. The clinical isolates were retrospectively collected, and patient data were not included in this study.

### Antimicrobial Susceptibility Testing

The minimal inhibitory concentrations (MICs) of the CRE strains were performed using the Phoenix 100 Automated Microbiology System and interpreted according to CLSI criteria (CLSI, 2016). A total of 18 antibiotics belonging to eight classes of antimicrobials were tested, including carbapenems (imipenem and meropenem), penicillins (ampicillin), β-lactam/β-lactamase inhibitor complexes (amoxicillinclavulanate, ampicillin-sulbactam, and piperacillin-tazobactam), cephalosporins (cefazolin, cefuroxime, ceftazidime, and cefepime), monocyclic β-lactams (aztreonam), aminoglycosides (gentamicin and amikacin), fluoroquinolones (ciprofloxacin and levofloxacin), folate metabolic pathway inhibitors (trimethoprim-sulfamethoxazole), colistin and tigecycline.

# Detection of Carbapenemases, ESBLs, AmpC, and Porin Genes

Polymerase chain reaction (PCR) was performed to investigate the presence of carbapenemase-encoding genes, including blaKPC, blaNDM, blaVIM, blaIMP and blaOXA−48. Simultaneously, we examined ESBLs (CTX-M, SHV, and TEM), AmpC cephalosporinases (CMY, ACT, and DHA), and mutation in porin encoding genes (OmpK35/OmpF, OmpK36/OmpC), using PCR followed by Sanger sequencing. Oligonucleotide primers used for screening the above genes have been reported previously (Mammeri et al., 2010; Bokaeian et al., 2015; Candan and Aksoz, 2015; Sugawara et al., 2016).

# Multilocus Sequence Typing (MLST)

Multilocus sequence typing (MLST) was conducted to investigate the genetic relationships of different CRE isolates. PCR followed by Sanger sequencing was used to detect conserved housekeeping genes in distinct species, including Klebsiella spp. (gapA, infB, mdh, pgi, phoE, rpoB, and tonB), E. coli (adk, fumC, gyrB, icd, mdh, purA, and recA), and E. cloacae (dnaA, fusA, gyrB, leuS, pyrG, rplB, and rpoB). Allelic profiles and sequence types (STs) were determined according to species-specific MLST databases<sup>1</sup> . Primers used for MLST were described in previous reports (Diancourt et al., 2005; Wirth et al., 2006; Miyoshi-Akiyama et al., 2013; Herzog et al., 2014).

### Pulsed-Field Gel Electrophoresis (PFGE)

Raoultella ornithinolytica and Serratia marcescens strains (for which no MLST schemes Are available) Were further investigated by PFGE using a CHEF Mapper Power Module instrument (Bio-Rad, United States). In brief, genomic DNA Was digested With Xba I, and then electrophoresed Under the following conditions: voltage 6 V/cm, running time 18–19 h, temperature 14◦C, and pulse times of 5–40 s (R. ornithinolytica) and 5–20 s (S. marcescens). Salmonella strain H9812 Was used as a control strain and size marker. Clonal relatedness Between strains Was evaluated based on the criteria proposed by Tenover et al. (1995).

### Plasmid Sequencing and Screening

Conjugation transfer experiments were performed with selected blaKPC−2-harboring strains and rifampicin-resistant E. coli EC600. Experiments were carried out using mixed broth culture method as described previously (Chen et al., 2014a). Transconjugants were identified by detecting resistance genes using PCR. Plasmid DNA from E. coli EC600 transconjugants harboring single plasmids was extracted using a Qiagen Plasmid Midi Kit (Qiagen, Valencia, CA, United States), and sequenced using the Illumina Miseq system (Illumina, United States) (Du et al., 2016). Sequencing reads were assembled de novo into contigs using SPAdes (Bankevich et al., 2012), then manually inspected using Geneious 9.1<sup>2</sup> ; and gaps were closed by PCR and Sanger sequencing.

A PCR mapping strategy was developed to detect a common blaKPC−2-harboring pSZF\_KPC/p628-KPC-like plasmid

<sup>2</sup>http://www.geneious.com/

sequenced in this study. The scheme includes six individual PCR reactions. PCR-I was designed to target the region spanning IncFII replicon gene repA and its downstream DNA methylase gene, while PCR-V was designed to target the junction of second replicon gene repB and its upstream parA. PCR-II, III, and PCR-IV were designed to target the blaKPC−<sup>2</sup> neighboring regions of 1rep-klcA, klcA-blaKPC−2, and blaKPC−2-IS26, respectively (**Figure 1**). PCR-VI was used to detect the traX-finO junction in pSZF\_KPC/p628- KPC-like plasmids. The oligonucleotide primer target regions for identification of pSZF\_KPC/p628-KPC-like plasmids are shown in **Figure 1**, and primer sequences are listed in **Table 1**.

# RESULTS

### Species and Antimicrobial Susceptibility

A total of 67 non-duplicate CRE isolates were collected from our hospital from August 2012 to August 2016, consisting of 42 K. pneumoniae, 2 K. oxytoca, 1 K. aerogenes, 10 S. marcescens, 6 E. cloacae, 2 R. ornithinolytica, and 4 E. coli. The results of antimicrobial susceptibility testing are shown in **Table 2**. All isolates were resistant to ampicillin, cefazolin, cefuroxime, imipenem, meropenem, and amoxicillin/clavulanate, and exhibited high resistance rates to most of the other β-lactam antibiotics tested. The most active compounds against all isolates were colistin (97.0% susceptible), tigecycline (94.0% susceptible), trimethoprim/sulfamethoxazole (56.7% susceptible) and amikacin (50.7% susceptible).

### Detection of Carbapenemase Genes

In this study, 54 (80.6%) of the 67 CREs were found to harbor at least one carbapenemase gene (**Table 3**). However, the distribution of carbapenemases among different species varied significantly, while the frequencies of carbapenemase-producing Enterobacteriaceae (CPE) in different species were 100%, 88.1%, 50.0%, and 50.0% in E. coli, K. pneumoniae, E. cloacae and S. marcescens, respectively.

Various carbapenemases were identified among the 54 CPE isolates, including blaKPC−<sup>2</sup> (n = 45), blaKPC−<sup>3</sup> (n = 1), blaNDM−<sup>1</sup> (n = 6), blaNDM−<sup>5</sup> (n = 1), blaIMP−<sup>4</sup> (n = 2), and blaVIM−<sup>1</sup> (n = 2). No strains were found to carry blaOXA−48-like genes. Among these, KPC was the most predominant carbapenemase (80.7%), and was primarily found in K. pneumoniae (37/42, 88.1%). Three E. cloacae strains were found to co-harbor two carbapenemaseencoding genes, with one strain harboring both blaKPC−<sup>3</sup> and blaNDM−1, while the other two harboring both blaNDM−<sup>1</sup> and blaVIM−1.

### Other Mechanisms Associated With Carbapenem Resistance

As described above, PCR failed to identify any carbapenemases among 13 out of 67 CRE strains, including 5 K. pneumoniae, 3 E. cloacae and 5 S. marcescens, suggesting that other mechanisms may have contributed to the phenotypic carbapenem resistance

<sup>1</sup>http://pubmlst.org/

TABLE 1 | Oligonucleotide primers used to screen pSZF\_KPC/p628-KPC-like plasmids.


<sup>a</sup>The primer locations are illustrated in Figure 1.

among these isolates. We therefore examined ESBL-encoding genes (blaTEM, blaSHV, and blaCTX−M), AmpC-encoding genes (blaCMY, blaDHA, and blaACT) and outer membrane porin genes. ESBL and AmpC genes were tested in all isolates, while outer membrane porin genes were examined in K. pneumoniae (ompK35 and ompK36), as well as in E. coli, E. cloacae, and S. marcescens (ompF and ompC).

Fifty-nine isolates were found to carry at least one ESBL and/or AmpC gene, including 13 non-carbapenemase-producing strains. Specifically, 51 (76.1%) strains were found to carry blaCTX−<sup>M</sup> genes, including blaCTX−M−<sup>65</sup> (n = 19), blaCTX−M−<sup>14</sup> (n = 19), blaCTX−M−<sup>9</sup> (n = 7), blaCTX−M−<sup>15</sup> (n = 3), and blaCTX−M−<sup>3</sup> (n = 3). These were found in several species, including K. pneumoniae (n = 32), S. marcescens (n = 10), E. cloacae (n = 6), E. coli (n = 2), and R. ornighinolytica (n = 1). In addition, 30 strains were positive for blaSHV−12, most of which were K. pneumoniae (26/30, 86.7%), while 21 K. pneumoniae harbored blaDHA−1, 6 E. cloacae possessed blaACT, and two E. coli were positive for blaCMY−2.

Outer membrane porin gene sequence analysis showed that 27 K. pneumoniae isolates harbored ompK35 and/or ompK36 mutations. ompK35 mutations (n = 26) were exclusively due to premature stop codons, while ompK36 mutations included glycine-aspartic acid (GD) insertions at amino acid positions 134–135 (n = 23), IS10 insertions (n = 2), and stop codons (n = 1). Five non-carbapenemase-producing K. pneumoniae CRE contained at least one outer membrane porin gene mutant (ompK35 or ompK36) while also harboring ESBL genes blaCTX−<sup>M</sup> or blaSHV−12, which likely explains the carbapenem resistance among these isolates. Sequence analysis of ompF and ompC genes in S. marcescens showed that they all possess mutated ompF, with premature stop codons at amino acid position 72. One noncarbapenemase-producing E. cloacae isolate was also found to carry an ompF mutation (stop codon), in addition to blaACT

and blaCTX−M−9. However, two non-carbapenemase-producing E. cloacae CRE isolates did not display mutations in either ompF or ompC, although they were found to harbor ESBL genes (either blaCTX−M−<sup>14</sup> or blaCTX−M−3). We suspect additional mechanisms, such as efflux pumps or penicillin-binding protein modifications, may contribute to carbapenem resistance in the latter two isolates. Therefore, carbapenem resistance among 11 out of 13 non-carbapenemase-producing CRE isolates may be

TABLE 2 | Susceptibility of CRE isolates against different antimicrobial agents.


∗ IPM, imipenem; MEM, meropenem; AMP, ampicillin; AMC, amoxicillin-clavulanate; SAM, ampicillin-sulbactam; TZP, piperacillin-tazobactam; CZO, cefazolin; CXM, cefuroxime; CAZ, ceftazidime; FEP, cefepime; ATM, aztreonam; GEN, gentamicin; AMK, amikacin; CIP, ciprofloxacin; LEV, levofloxacin; SXT, trimethoprimsulfamethoxazole; TGC, tigecycline; CL, colistin. n (%), n = number of isolates that were susceptible; % = percentage of isolates susceptible.

TABLE 3 | Molecular characteristics of CRE clinical isolates.


<sup>∗</sup>−, negative; N/A, not available or not performed.

explained by the combination of porin gene mutants and the presence of ESBL or AmpC-encoding genes, while the resistance mechanisms in two non-carbapenemase-producing E. cloacae CRE isolates remain to be determined. Meanwhile, among the 54 CPE isolates, 28 (51.9%) also carry at least one outer membrane porin mutant.

# Distribution of MLST Sequence Types and PFGE Patterns

Multilocus sequence typing results showed that 42 K. pneumoniae belonged to 12 different STs, with ST11 being the most common (25/42, 59.5%). All ST11 isolates possessed the same OmpK35 stop codon, while 23 of them harbored the 134–135 GD OmpK36 mutant. The two K. oxytoca isolates belonged to ST135 and ST180, while the 4 E. coli strains were assigned to ST1488, ST3234, ST167, and ST354. The six E. cloacae also comprised 4 STs, including ST231 (n = 3), ST120 (n = 1), ST421 (n = 1), and ST97 (n = 1). Two R. ornithinolytica and 10 S. marcescens strains were further analyzed by Xba I-PFGE. The results showed that all 10 S. marcescens strains shared the same PFGE pattern, suggesting clonal spread (**Supplementary Figure 1**). In addition, the ten S. marcescens isolates were collected within 7 months in 2013 from three closed wards (respiratory, neurology, and renal wards) and mostly from respiratory samples (n = 9), suggestive of the likelihood of a small S. marcescens outbreak. As aforementioned, all S. marcescens harbored the same OmpF mutant, and among them, 5 were also found to carry blaKPC−2. By contrast, the two R. ornithinolytica isolates displayed unrelated PFGE patterns (data not shown).

# Sequencing and Screening of blaKPC−2-Harboring Plasmids

Since KPC-2 is the most common carbapenemase found in our hospital, we sought to determine the complete sequences of blaKPC−<sup>2</sup> -harboring plasmids spreading through our institution. We selected 4 blaKPC−2-harboring strains, including two K. pneumoniae (ST8 and ST11), one E. coli (ST3234), and one S. marcescens for plasmid conjugation and complete plasmid sequencing.

The blaKPC−2-harboring plasmid (subsequently named pSZN\_KPC) isolated from K. pneumoniae ST8 belonged to incompatibility group N. Plasmid pSZN\_KPC is 65,604 bp in length, with an average G+C content of 53.3 %, and contains 88 predicted open reading frames (ORFs) (**Figure 1**). A BLAST search of the pSZN\_KPC plasmid sequence against the GenBank database<sup>3</sup> showed that pSZN\_KPC exhibits a high degree of identity to the previously published IncN plasmid pECN580 (accession no. KF914891) from an E. coli strain found in China (Chen et al., 2014a), with 100% query coverage and overall 99% nucleotide identity. The major difference is that the region harboring antirestriction protein gene ardA in plasmid pSZN\_KPC was located downstream of gene ccgCD (**Figure 1**), likely as a consequence of recombination.

The blaKPC−2-harboring plasmids from K. pneumoniae ST11 strain Kp715, E. coli ST3234 strain Ec732, and S. marcescens strain Sm703 were highly similar (each differed by only 4 SNPs), and belonged to the same IncFII incompatibility group. This plasmid (subsequently named pSZF\_KPC) is 106,201 bp in length with a G+C content of 53.2 %, and harbors 122 predicted ORFs. Comparative sequence analysis for plasmid pSZF\_KPC showed that it is highly similar (100% query coverage and >99% nucleotide identity) to p628-KPC (accession no. KP987218) from a K. pneumoniae isolate collected in our hospital in 2010 (Wang et al., 2015). Compared to p628-KPC, the main difference in pSZF\_KPC is that there is an additional ISKpn18 element, with IS4321 located downstream of the mcr operon (**Figure 1**). The sequences of pSZN\_KPC and pSZF\_KPC have been deposited in GenBank under the accession numbers MH917122 and MH917123, respectively.

The plasmid sequence results described above suggest that pSZF\_KPC/p628-KPC-like plasmids have been spreading in our hospital since at least 2010, undergoing horizontal transmission into different species. We therefore developed a set of PCR assays to screen for the presence of pSZF\_KPC/p628-KPClike plasmids among our 67 CRE isolates. The PCR results showed that 27 (64.2%) of the 42 blaKPC−<sup>2</sup> positive strains harbored pSZF\_KPC/p628-KPC-like plasmids. The 27 strains encompassed different species and STs, including K. pneumoniae ST11 (n = 14), ST774 (n = 3), ST1107 (n = 2), ST211 (n = 1), ST655 (n = 1), and ST218 (n = 1); S. marcescens (n = 4); and E. coli ST3234 (n = 1). These pSZF\_KPC/p628-KPC-like plasmid-harboring isolates were collected in 8 different wards, spanning from 2012 to 2016, indicating the frequent horizontal transfer of this common plasmid among Enterobacteriaceae in our hospital.

# DISCUSSION

At the present time, KPC, NDM and OXA-48 are the most common carbapenemases worldwide (Nordmann and Poirel, 2014; Sugawara et al., 2016). KPCs are most frequently identified in K. pneumoniae from the United States, China, Colombia, Israel, Greece, and Italy, while NDMs are primarily found in K. pneumoniae, E. coli and Enterobacter spp. from the Indian subcontinent, and OXA-48-like carbapenemases in K. pneumoniae and E. coli from North Africa and Turkey (Nordmann and Poirel, 2014). In addition, the spread of CPEs has been associated with several high-risk clones. One notable example involves the global spread of KPCs, which has been largely associated with K. pneumoniae clonal group 258 (CG258) strains, of which ST258 is the most predominant KPC-producing K. pneumoniae clone in North America, while ST11 is most common in East Asia, especially China (Patel and Bonomo, 2013; Chen et al., 2014b; Zhang et al., 2017).

China, in particular eastern China, is regarded as one of the primary global endemic regions for CRE (Zhang et al., 2017). In this study, we phenotypically and genetically characterized the CRE isolates collected from an eastern Chinese hospital, and investigated the molecular mechanisms underlying carbapenem resistance. Our study revealed several interesting findings.

<sup>3</sup>http://blast.ncbi.nlm.nih.gov/Blast.cgi

Firstly, the CRE isolates were recovered from seven different Enterobacteriaceae species. Although carbapenem resistance has been frequently identified in K. pneumoniae, Enterobacter spp., and E. coli, it is fairly uncommon in other species such as R. ornithinolytica and S. marcescens. R. ornithinolytica is mostly recovered from the environment and rarely causes severe infections in humans; nevertheless several reports have described the emergence of carbapenem-resistant R. ornithinolytica in China (Zhou et al., 2013; Qin et al., 2014; Yang et al., 2018). Notably, in this study we identified two carbapenem-resistant R. ornithinolytica, harboring blaKPC−<sup>2</sup> and blaIMP−4, respectively, suggesting that different carbapenemase plasmids have spread into R. ornithinolytica. In contrast, carbapenem resistance in S. marcescens has been historically associated with a specific group of carbapenemases, SMEs. However, in this study, none of the ten S. marcescens strains were found to carry blaSME, and the observed carbapenem resistance was likely due to the mutations of OmpK35 porin encoding genes, as well as production of CTX-M-14 and KPC-2. Interestingly, only five of the ten S. marcescens strains were found to harbor blaKPC−2, although PFGE results suggested that the spread of carbapenem-resistant S. marcescens was largely clonal. We suspect it is likely that the S. marcescens OmpK35 mutant further acquired a blaKPC−<sup>2</sup> plasmid (e.g., pSZF\_KPC/p628- KPC-like).

In this study, K. pneumoniae was the most common CRE species, accounting for 62.7% (42/67) of all CRE isolates. Among these, ST11, a member of the epidemic CG258 clone, was the predominant ST (59.5%, 25/42), which is consistent with the molecular epidemiology described in other regions of China (Zhou et al., 2013; Zhang et al., 2017; Yang et al., 2018). However, in our study carbapenem-resistant K. pneumoniae were found in 12 different STs, including some STs rarely associated with carbapenem resistance (e.g., ST774 and ST1107). Our plasmid screening results showed that the diversity of STs was largely due to the frequent transfer of a common blaKPC−<sup>2</sup> vector into different K. pneumoniae genetic backgrounds.

Secondly, our study revealed diverse molecular mechanisms of carbapenem resistance within our hospital, albeit with carbapenemase production the primary cause. Diverse types of carbapenemases, including KPC-2, KPC-3, NDM-1, NDM-5, IMP-4, and VIM-1, were identified, whereas in most Chinese hospitals, usually only KPC and/or NDM carbapenemases are prevalent (Hu et al., 2014). It is noteworthy that four different classes of carbapenemases were found in our hospital, suggesting that different carbapenemase producing plasmids/strains are spreading locally. Notably, however, ∼20% strains were noncarbapenemase producers. It is therefore likely that besides carbapenemases, other mechanisms such as porins defects and production of ESBLs (TEM, SHV, and CTX-M) or AmpC β-lactamases (CMY, DHA, and ACT) significantly contributed to carbapenem resistance in our hospital. Meanwhile, it is worth noting that 51.9% (28/54) of carbapenemase-producing isolates also harbored additional porin gene mutations, potentially rendering antimicrobial treatment more challenging in comparison to strains without porin mutations (Clancy et al., 2013).

Lastly, we identified a widespread blaKPC−2-harboring plasmid vector within our institution. Several pSZF\_KPC/p628- KPC-like plasmids were identified within distinct K. pneumoniae clones (STs), as well as in different species (K. pneumoniae, S. marcescens and E. coli). Our PCR-based screening of blaKPC−2-positive K. pneumoniae isolates revealed that this plasmid is widely disseminated in our hospital and was found in nearly two third (22/37) of KPC-2 positive isolates, highlighting the importance of plasmid horizontal transfer in the dissemination of KPC. More importantly, this plasmid was identified in different K. pneumoniae STs, as well as other species, suggesting that intra-strain and inter-species transfer of this plasmid have significantly contributed to the spread of carbapenem resistance in our hospital.

In conclusion, our study revealed extensive genetic diversity among CRE isolates from a single Chinese hospital. Both clonal expansion (e.g., K. pneumoniae ST11 and S. marcescens) and plasmid horizontal transfer (i.e., pSZF\_KPC/p628-KPC-like plasmids) were identified. Different carbapenemases classes and outer membrane porin defects were found in several species. CRE isolates circulating in our hospital differ significantly in their species, STs, porin genes, carbapenemase genes, and plasmid content, highlighting complex dissemination of CRE within our hospital. Further studies are required to understand the factors underlying the genetic diversity of CRE in our hospital, in order to effectively control further spread of carbapenem resistance.

# AUTHOR CONTRIBUTIONS

MM, HW, and PX contributed to work, data analysis, and manuscript preparation. SN, JL, JM, Y-WT, and BK prepared the manuscript. XX contributed to work and data analysis. XZ and HZ analyzed the data. HD and LC contributed to study design, data analysis, and manuscript preparation.

# FUNDING

This study was supported by the National Natural Science Foundation of China (81572032), Six Talent Peaks Project in Jiangsu Province (2016-WSN-112), Key Research and Development Project of Jiangsu Provincial Science and Technology Department (BE2017654), Gusu Key Health Talent of Suzhou, Jiangsu Youth Medical Talents Program (QN-866, 867), and the Science and Technology Program of Suzhou (SZS201715, SYS201619, and SYSD2018145). This work was in part supported by the following grants from the National Institutes of Health (R01AI090155, R21AI117338, and P30CA008748).

# SUPPLEMENTARY MATERIAL

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb.2018. 03341/full#supplementary-material

### REFERENCES

fmicb-09-03341 January 11, 2019 Time: 17:5 # 8


**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2019 Miao, Wen, Xu, Niu, Lv, Xie, Mediavilla, Tang, Kreiswirth, Zhang, Zhang, Du and Chen. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

# Infections Caused by Carbapenem-Resistant Enterobacteriaceae: An Update on Therapeutic Options

Chau-Chyun Sheu1,2, Ya-Ting Chang2,3, Shang-Yi Lin2,3, Yen-Hsu Chen2,4,5 \* and Po-Ren Hsueh6,7 \*

<sup>1</sup> Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Kaohsiung Medical University Hospital, Kaohsiung, Taiwan, <sup>2</sup> School of Medicine, Sepsis Research Institute, Graduate Institute of Medicine, Graduate Institute of Clinical Medicine, Center of Dengue Fever Control and Research, College of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan, <sup>3</sup> Division of Infectious Disease, Department of Internal Medicine, Kaohsiung Medical University Hospital, Kaohsiung, Taiwan, <sup>4</sup> Department of Internal Medicine, Kaohsiung Municipal Ta-Tung Hospital, Kaohsiung, Taiwan, <sup>5</sup> Department of Biological Science and Technology, College of Biological Science and Technology, National Chiao Tung University, Hsinchu, Taiwan, <sup>6</sup> Department of Laboratory Medicine, National Taiwan University Hospital, College of Medicine, National Taiwan University, Taipei, Taiwan, <sup>7</sup> Department of Internal Medicine, National Taiwan University Hospital, College of Medicine, National Taiwan University, Taipei, Taiwan

### Edited by:

Liang Chen, Public Health Research Institute (PHRI), United States

### Reviewed by:

Elizabeth Diago Navarro, Stony Brook University, United States Zackery Bulman, University of Illinois at Chicago, United States

### \*Correspondence:

Yen-Hsu Chen infchen@gmail.com Po-Ren Hsueh hsporen@ntu.edu.tw

### Specialty section:

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

Received: 05 November 2018 Accepted: 15 January 2019 Published: 30 January 2019

### Citation:

Sheu C-C, Chang Y-T, Lin S-Y, Chen Y-H and Hsueh P-R (2019) Infections Caused by Carbapenem-Resistant Enterobacteriaceae: An Update on Therapeutic Options. Front. Microbiol. 10:80. doi: 10.3389/fmicb.2019.00080 Carbapenems are considered as last-resort antibiotics for the treatment of infections caused by multidrug-resistant Gram-negative bacteria. With the increasing use of carbapenems in clinical practice, the emergence of carbapenem-resistant pathogens now poses a great threat to human health. Currently, antibiotic options for the treatment of carbapenem-resistant Enterobacteriaceae (CRE) are very limited, with polymyxins, tigecycline, fosfomycin, and aminoglycosides as the mainstays of therapy. The need for new and effective anti-CRE therapies is urgent. Here, we describe the current understanding of issues related to CRE and review combination therapeutic strategies for CRE infections, including high-dose tigecycline, high-dose prolongedinfusion of carbapenem, and double carbapenem therapy. We also review the newly available antibiotics which have potential in the future treatment of CRE infections: ceftazidime/avibactam, which is active against KPC and OXA-48 producers; meropenem/vaborbactam, which is active against KPC producers; plazomicin, which is a next-generation aminoglycoside with in vitro activity against CRE; and eravacycline, which is a tetracycline class antibacterial with in vitro activity against CRE. Although direct evidence for CRE treatment is still lacking and the development of resistance is a concern, these new antibiotics provide additional therapeutic options for CRE infections. Finally, we review other potential anti-CRE antibiotics in development: imipenem/relebactam and cefiderocol. Currently, high-dose and combination strategies that may include the new β-lactam/β-lactamase inhibitors should be considered in severe CRE infections to maximize treatment success. In the future, when more treatment options are available, therapy for CRE infections should be individualized and based on molecular phenotypes of resistance, susceptibility profiles, disease severity, and patient characteristics. More high-quality studies are needed to guide effective treatment for infections caused by CRE.

Keywords: avibactam, carbapenems, carbapenemase, carbapenem-resistant Enterobacteriaceae, combination therapy, relebactam, vaborbactam

# INTRODUCTION

fmicb-10-00080 January 28, 2019 Time: 18:40 # 2

The increasing prevalence of bacterial resistance to antibiotics is a critical public health problem. Infections caused by antibioticresistant bacteria are associated with significant morbidity and mortality worldwide. Many previous efforts to combat multidrug-resistant (MDR) bacteria were focused on methicillinresistant Staphylococcus aureus (MRSA). In recent years, several new therapeutic options for MRSA have become available (David et al., 2017). Currently the major threat of antibiotic-resistant bacteria is from MDR Gram-negative organisms, particularly those which have developed resistance to carbapenem. Along with carbapenem-resistant Acinetobacter baumannii (CRAB) and carbapenem-resistant Pseudomonas aeruginosa (CRPA), carbapenem-resistant Enterobacteriaceae (CRE) are among the top tier of the WHO list of antibiotic-resistant "priority pathogens" that pose the greatest threat to human health (Willyard, 2017).

Enterobacteriaceae are common pathogens causing a variety of severe infections, including bloodstream infections (BSIs), community-acquired pneumonia (CAP), hospitalacquired pneumonia (HAP), ventilator-associated pneumonia (VAP), complicated urinary tract infections (cUTIs), and complicated intra-abdominal infections (cIAIs). Therefore, antibiotic resistance in these bacteria has significant clinical and socioeconomic impacts (Lee et al., 2017; Rodriguez-Bano et al., 2018; Ting et al., 2018). As the prevalence of infections caused by extended-spectrum β-lactamase (ESBL)-producing Enterobacteriaceae is increasing, the medical community has been forced to use carbapenem as a first-line empirical treatment. The increasing use of carbapenem for possible ESBL infections has led to a more serious problem: the emergence of carbapenemase-producing Enterobacteriaceae (CPE) (Sheu et al., 2018).

Carbapenem is a β-lactam antibiotic which inhibits transpeptidases (penicillin-binding proteins) and prevents peptidoglycan synthesis, leading to lytic cell death (Kohanski et al., 2010). The resistance of CRE to carbapenems is generally based on two mechanisms: carbapenemase production or the combination of structural mutations with the production of other β-lactamases, such as AmpC cephalosporinase (AmpC) and ESBL (Tzouvelekis et al., 2012; Munoz-Price et al., 2013; Goodman et al., 2016; Tamma and Simner, 2018). The classification and characteristics of major carbapenemases in CRE are summarized in **Table 1**. The three major classes of carbapenemases are Ambler Class A Klebsiella pneumoniae carbapenemase (KPC); Class B metallo-β-lactamases (MBLs) such as New Delhi MBL (NDM), Verona integrin-encoded MBL (VIM), and imipenemase (IMP); and Class D oxacillinases (OXA)-type enzymes such as OXA-48-like carbapenemases. These carbapenemases exhibit variable levels of carbapenem resistance through their carbapenemhydrolyzing activity. For instance, a certain proportion of VIM and IMP-producing K. pneumoniae have been observed to have low carbapenem minimum inhibitory concentrations (MICs) in studied isolates (Yan et al., 2001; Psichogiou et al., 2008; Daikos et al., 2009). On the other hand, NDM carbapenemase seemed to exhibit higher carbapenem MICs (Kumarasamy et al., 2010) while KPC-producing isolates demonstrated wide variations in carbapenem MICs in different geographic regions (Endimiani et al., 2009a; Daikos and Markogiannakis, 2011; Qi et al., 2011). Some carbapenem-producing Enterobacteriaceae (CPE) are even susceptible to carbapenems themselves and this is particularly observed in OXA-48 producers (Dautzenberg et al., 2014; Navarro-San Francisco et al., 2013). Some CPE may also coproduce AmpC or ESBLs. The impact of the co-production of these enzymes on treatment and outcomes remains unclear.

To date, the treatment options for CRE infections remain very limited. Polymyxins (colistin or polymyxin B) and tigecycline have been historically considered as drugs of choice for infections caused by CRE. However, resistance to these antibiotics is increasing (Capone et al., 2013; Giacobbe et al., 2015; Yang et al., 2018). In addition to polymyxins and tigecycline, fosfomycin and aminoglycosides are occasionally used (Tang et al., 2016; Tseng et al., 2017; Rodriguez-Bano et al., 2018). Carbapenems still play a role in the treatment of CRE infections, particularly when used in the treatment of CRE with lower MICs, either in higher doses, in combination with other active anti-CRE agents, or through double-carbapenem therapy (DCT). Older antibiotics such as minocycline, doxycycline, trimethoprim/sulfamethoxazole, and chloramphenicol may be effective for some CRE isolates (Falagas et al., 2011; Livermore et al., 2011). Recently, novel β-lactamase inhibitor combinations have provided new therapeutic options for CRE infections. However, these new β-lactamase inhibitors are not active against all carbapenemases. Avibactam inhibits both Class A KPC and Class D OXA-48 (Zasowski et al., 2015), while vaborbactam and relebactam inhibits only Class A KPC (Petty et al., 2018; Zhanel et al., 2018) (**Table 1**).

**Table 2** summarizes the currently available antimicrobial agents and their recommended doses for treatment of CRE infections. Because the therapeutic options are limited, all potentially active drugs should be tested in vitro. It is recommended to select anti-CRE agents according to in vitro susceptibility data, clinical severity, and all other available information. Higher doses may be necessary for severe infections, including pneumonia and septic shock (Rodriguez-Bano et al., 2018). Many anti-CRE agents have recently been reviewed (Ni et al., 2016; Thaden et al., 2017; Trecarichi and Tumbarello, 2017; Zavascki et al., 2017; Rodriguez-Bano et al., 2018). We therefore focus the present review on several potential combination therapeutic strategies and new antibiotics (**Table 3**).

# COMBINATION THERAPEUTIC STRATEGIES

Many studies have investigated the benefits of various combinations of antimicrobial agents for the treatment of CRE, in vitro or in vivo (Ku et al., 2017). However, it has been difficult to draw conclusions due to the diversity of study designs and resistance mechanisms. Among the studies that found combination therapy to contribute to lower mortality rates than monotherapy (Qureshi et al., 2012; Tumbarello et al., 2012, 2015; Tofas et al., 2016; Trecarichi et al., 2016; Gutierrez-Gutierrez et al., 2017; Machuca et al., 2017; Papadimitriou-Olivgeris


KPC, Klebsiella pneumoniae carbapenemase; MBL, metallo-β-lactamase; NDM, New Delhi metallo-β-lactamase; VIM, Verona integrin-encoded metallo-β-lactamase; IMP, imipenemase; OXA, oxacillinase.

TABLE 2 | Antimicrobial agents used for carbapenem-resistant Enterobacteriaceae infections.


cIAI, complicated intraabdominal infection; cUTI, complicated urinary tract infection; ESBL, extended-spectrum β–lactamase; KPC, Klebsiella pneumoniae carbapenemase; MBL, metallo-β-lactamase; MIC, minimum inhibitory concentration; OXA, oxacillinase; TDM, therapeutic drug monitoring. Adapted from Rodriguez-Bano et al. (2018).

<sup>a</sup>For patients with normal renal function.

et al., 2017), the two largest retrospective studies to date concordantly identified the protective effects to be significant in populations with high disease severity (Tumbarello et al., 2015; Gutierrez-Gutierrez et al., 2017). In the multicenter Italian cohort, with 661 episodes of BSI and non-BSI caused by KPC-producing K. pneumoniae, Tumbarello et al. (2015) compared clinical outcomes between 307 patients receiving monotherapy (colistin in 121, tigecycline in 116, gentamicin in 70) and 354 patients receiving combination therapy (receiving 2 or more in vitro-active drugs, with meropenem in all cases). They found significantly decreased mortality rates with combination therapy among patients with BSI (OR, 0.45; 95% CI, 0.29–0.68), lower respiratory tract infections (OR, 0.35; 95% CI, 0.11–0.99), high APACHE III scores (OR, 0.55; 95% CI, 0.37–0.80), septic shock (OR, 0.18; 95% CI, 0.05–0.53), and among isolates with a meropenem MIC of ≤8 mg/L (OR, 0.57; 95% CI, 0.32–1.03) (Tumbarello et al., 2015). The INCREMENT project, which included monomicrobial BSIs due to CPE from a total of 437 patients worldwide (26 tertiary hospitals in 10 countries), is by far the largest retrospective international cohort study. The TABLE 3 | Potential combination therapeutic strategies and new antibiotics for the treatment of carbapenem-resistant Enterobacteriaceae infections.


fmicb-10-00080 January 28, 2019 Time: 18:40 # 4


subgroup analysis concluded that combination therapy, defined as receiving more than one in vitro-active antimicrobial, did not improve survival except in patients with a high mortality score (Gutierrez-Gutierrez et al., 2017).

The highly heterogeneous methodologies between studies and the fact that most data were derived from isolates of CRKP, preclude optimal synthesis using the currently available evidence. Several systematic reviews have proposed viewpoints regarding combination therapy. Falagas et al. (2014) reviewed 20 nonrandomized studies, comprising 692 patients, and proposed that combination therapy may be considered for severely ill patients. Tzouvelekis et al. (2012) performed a systematic review that included 34 studies and suggested that carbapenem-containing combinations contribute to higher treatment success rates. Polymyxin-based and tigecycline-based combination therapies were reported to significantly decrease 30-day mortality when compared with respective monotherapy by systematic reviews (Ni et al., 2015, 2016). Zusman et al. (2017) performed a metaanalysis to compare polymyxin-based combination therapy and monotherapy. The subgroup analysis for CRE comprised of K. pneumoniae BSI and included seven studies with a total of 285 patients. The meta-analysis favored combination therapy (potentially double-coverage) and demonstrated an OR of 2.09 (95% CI, 1.21–3.6; I <sup>2</sup> = 0%), but with low-quality evidence (Zusman et al., 2017). The most recent meta-analysis, performed by Martin et al. (2018), included 22 studies describing CRE infections. Seven studies were extracted for comparison between combination therapy and monotherapy. Four of the studies included patients with BSIs and three with mixed infections. The results showed a significantly higher risk of overall mortality among patients treated with monotherapy (OR, 2.19; 95% CI, 1.00–4.80), with a high heterogeneity (I <sup>2</sup> = 84.2%; QP = 0.003) (Martin et al., 2018).

The first and only RCT for the treatment of carbapenemresistant Gram-negative bacteria was recently published, and is therefore not included in any of the above-mentioned systematic reviews or meta-analyses. The open-label RCT compared the outcomes of colistin monotherapy vs. combination therapy with high-dose and prolonged infusion meropenem (2 g every 8 h, infused over 3 h) (Paul et al., 2018). A total of 406 patients were enrolled, with pneumonia and bacteremia comprising 87% of the infections. Most infections were caused by A. baumannii (77%), while Enterobacteriaceae only contributed to 18% (73/406) of all infections. Most Enterobacteriaceae infections were BSIs (77%), with K. pneumoniae being the main pathogen (89%). In the post hoc subgroup analysis of Enterobacteriaceae infections, there were no significant differences in clinical outcomes between colistin monotherapy and combination therapy with meropenem. However, combination therapy seemed to be associated with a lower clinical failure rate (46% vs. 68%, P = 0.185) and a lower 28-day mortality (21% vs. 35%, P = 0.235). There is another ongoing RCT (NCT01597973) investigating colistin monotherapy vs. combination with carbapenem in the treatment of bacteremia or pneumonia caused by extensively drug-resistant Gram-negative bacteria. The trial is estimated to be completed in 2021.

Because of the suboptimal quality of the available data, it is not yet possible to make solid recommendations regarding combination therapy in CRE infections. However, there is a growing body of evidence supporting the use of combination therapy, particularly in critically ill patients.

### High-Dose Tigecycline

Besides carbapenems, the most commonly studied high-dose regimen is tigecycline, owing to its non-nephrotoxic nature compared to other potentially active antimicrobial agents for CRE, such as polymyxins and aminoglycosides. A high-dose colistin regimen has been more extensively investigated for its efficacy against carbapenem-resistant A. baumannii infections and seldom for CRE (Gibson et al., 2016; Cheng et al., 2018).

A high-dose tigecycline regimen consists of a 200 mg loading dose and a maintenance dose of 100 mg every 12 h, while a standard-dose regimen consists of a loading dose of 100 mg and a maintenance dose of 50 mg every 12 h. One study assessed the efficacy of tigecycline for carbapenem-producing K. pneumoniae (CPKP) by using 164 non-duplicate clinical strains of CPKP isolated from HAP and incorporating a Monte Carlo simulation into a pharmacokinetic/pharmacodynamics (PK/PD) model. The study revealed that a higher cumulative fraction of response, indicating better clinical efficacy, can be gained by doubling the tigecycline dose (90.2% vs. 71.2%) (Trecarichi et al., 2016).

Two small retrospective studies conducted by Sbrana et al. (2013) and Balandin Moreno et al. (2014) included 26 episodes of KPC-producing K. pneumoniae and 16 episodes of VIM-1-producing K. pneumoniae infections from a trauma-referral ICU and a multidisciplinary ICU, respectively. In the study by Sbrana et al. (2013), high-dose tigecycline was administered in 25/26 infection episodes in combination with gentamicin (19/26) and colistin (12/26). Fosfomycin was used as a third antibiotic in 13/26 episodes. In the study by Balandin Moreno et al. (2014), high-dose tigecycline was administered in 10/16 infection episodes and standard-dose regimen in 6/16 episodes. Fourteen (14/16) episodes were treated with combination therapy, including colistin in 8/16, carbapenem in 5/16, ciprofloxacin in 2/16, piperacillin/tazobactam in 1/16, and amikacin in 1/16. Sbrana et al. (2013) suggested a favorable outcome by the doubleor triple-combination with high-dose tigecycline, with a 30-day

crude mortality rate of 14%. On the other hand, Balandin Moreno et al. (2014) found no significant differences in the mortality rates between high-dose and standard-dose tigecycline. Di Carlo et al. (2013) compared standard-dose to high-dose tigecycline with the combination of colistin in 30 postoperative abdominal surgery ICU patients who had at least two positive blood cultures for KPC-producing K. pneumoniae. They observed a significantly lower mortality rate in the high-dose tigecycline group. In terms of infection sources, De Pascale et al. (2014) found high-dose tigecycline to be the only independent predictor of clinical cure in the VAP subgroup of critically ill patients with MDR bacterial infections (total 63 patients, 28 isolates of CRAB and 27 isolates of carbapenem-resistant K. pneumoniae [CRKP]). For BSIs, a retrospective cohort study of 40 patients with nosocomial CPKP BSI showed no significant differences in in-hospital mortality between 23 patients undergoing high-dose tigecycline-based combination therapy and 17 patients undergoing standard-dose tigecycline therapy (52.2% vs. 76.5%, P = 0.117) (Geng et al., 2018). One systemic review encompassed 25 studies reporting the efficacy and/or safety of tigecycline-based regimens for treating CRE infections, while the subgroup meta-analysis found a much lower ICU mortality with high-dose tigecycline than standarddose tigecycline (OR, 12.48; 95% CI, 2.06–75.43; P = 0.006) (Ni et al., 2016).

In summary, high-dose tigecycline-based combination therapy may be considered in critically ill patients with CRE infections and limited treatment options, either from the PK/PD viewpoint or clinical observations. However, the infection source should be cautiously evaluated since better outcomes were observed in trauma (Sbrana et al., 2013) or postoperative abdominal surgery (Di Carlo et al., 2013) patients, but not in patients with BSI (Geng et al., 2018), which is compatible with the common consensus that tigecycline is extensively distributed beyond the plasma volume and concentrates into tissues.

# High-Dose and Prolonged-Infusion of Carbapenems

The fact that wide disparities of carbapenem MICs exist, even among CPE isolates, complicates the discourse for the role of carbapenems in the treatment of CRE or CPE. Several studies have investigated the efficacy of carbapenems against CPKP in animal models and suggested that with a higher dose of carbapenems it is possible to attain reliable reductions in bacterial density in isolates with lower carbapenem MICs (Daikos et al., 2007; Bulik and Nicolau, 2010b; Bulik et al., 2010a; Souli et al., 2011). Daikos and Markogiannakis (2011) proposed, based on several animal infection model studies, that highdose, prolonged-infusion carbapenems can achieve bactericidal effects in immunocompetent animals infected by KPC-producing K. pneumoniae isolates with MICs up to 8 mg/L. In addition, Daikos and Markogiannakis (2011) also analyzed 22 clinical studies (mostly case series) and found: (1) the therapeutic efficacy of carbapenems increases from 29% for an MIC of >8 mg/L, to 69% for an MIC ≤ 4 mg/L, which is similar to patients infected with non-CPKP (73%); (2) among the 138 patients treated by combination regimens, the mortality rate was lowest in patients who received carbapenem-containing combinations and were infected with isolates of MIC ≤ 4 mg/L (OR, 5.3; 95% CI, 1.5–18.9). More recent retrospective cohort studies echoed the observation that a combination regimen containing carbapenem is associated with significantly higher survival rates in CPKP BSI isolates (Daikos et al., 2014) and KPC-producing K. pneumoniae isolates with a meropenem MIC ≤ 8 mg/L (Tumbarello et al., 2015).

In addition to carbapenem-containing combinations, the strategy of high-dose (2 g every 8 h) carbapenem with prolonged infusion (over 3 h) was also found to be associated with better outcomes in CPKP infections (Tumbarello et al., 2012, 2015; Daikos et al., 2014). Moreover, Giannella et al. (2018) evaluated the efficacy of high-dose carbapenem-based combination therapy among 595 patients with CRKP BSI and studied the benefits of high-dose carbapenem in strains with meropenem MIC ≥ 16 mg/L, which comprised 77% of all isolates in the study. These clinical observations are in line with PK/PD studies showing that a high-dose prolonged-infusion carbapenem regimen can reach attainable targets in isolates with meropenem MICs up to 32–64 mg/L, though failure was observed with very high MICs of 256–1024 mg/L (Del Bono et al., 2017; Giannella et al., 2018).

Based on the current evidence, using a high-dose prolongedinfusion carbapenem-containing combination regimen for the treatment of CRE isolates with carbapenem MIC ≤ 8 mg/L may be considered when there are no other treatment options. It should also be noted that since most conclusions from the abovementioned studies were derived from CPKP or CRKP isolates, the extrapolation of high-dose prolonged-infusion of carbapenemcontaining combination therapy to other CRE with different resistance mechanisms requires further investigation.

# Double-Carbapenem Therapy

The most well-investigated DCT is the combination of ertapenem (with a standard infusion time of 30–60 min) prior to a prolonged infusion of meropenem or doripenem over 3–4 h, with high-dose meropenem of 2 g every 8 h being most commonly applied. This regimen originated from the revolutionary approach proposed by Bulik and Nicolau (2011), as a salvage option for CPKP. The study validated enhanced activities for both ertapenem and doripenem in combination, using an in vitro chemostat and in vivo murine thigh infection model (Bulik and Nicolau, 2011). The rationale for this combination came from the hypothesis that ertapenem might play a sacrificial role, being preferentially hydrolyzed due to its greater affinity to KPC (Anderson et al., 2007), permitting the concomitant administration of carbapenem to sustain a high concentration. Some in vitro studies show a beneficial effect of lower MICs of meropenem (MIC ≤ 128 mg/L) (Oliva et al., 2017b) or doripenem (MIC ≤ 16 mg/L) (Wiskirchen et al., 2013) with regard to ertapenem-based DCT. Nevertheless, ertapenembased DCT is not the only combination that demonstrates synergism (Poirel et al., 2016; Fredborg et al., 2017). One in vitro study reported variable synergistic patterns among KPC and OXA-48 producers, while no synergism was observed for the NDM-producing strain (Poirel et al., 2016).

Despite the diversity of the above-mentioned observations, nearly all reported cases and clinical studies adopted an ertapenem-based DCT with recommended or high-dose

doripenem/meropenem. The short stability of the intravenous imipenem preparation that hinders prolonged infusion may be one of the reasons for such a phenomenon (Mashni et al., 2018). Souli et al. (2017) conducted the largest observational cohort study to date in terms of exclusive DCT as a salvage therapy. The study included 27 patients with CPKP infections, mostly with cUTI (59.3%) and BSI (48.2%) and reported a high clinical success rate of 77.8%. It is noteworthy that the subgroup of pandrug-resistant infections also had a successful clinical and microbiological outcome of 78.5% (11/14). Among critically ill patients with severe sepsis or septic shock, a successful outcome was noted in 81.8% (9/11) (Souli et al., 2017). Oliva et al. (2017a) and Venugopalan et al. (2017), both conducted observational comparator studies to compare the efficacy of DCT. Venugopalan et al. (2017) enrolled 36 patients with CRKP bacteremia, including 18 patients receiving doripenem and ertapenem (DCT group), and 18 patients receiving doripenem and colistin (control group). They found the DCT group had a significantly improved clinical cure rate of 72% (13/18, vs. control group of 39%, 7/18; P = 0.049) and a lower 30-day mortality of 31%, with a trend toward statistical significance (vs. 61%, P = 0.087) (Venugopalan et al., 2017). Oliva et al. (2017a) enrolled 32 patients with CRKP infections, including 18 patients receiving ertapenem and meropenem, and 14 patients receiving ertapenem, meropenem, and colistin. They found the combination of colistin and DCT obtained rapid bactericidal activity up to 24 h in the in vitro analysis. However, there were no significant differences regarding the early response or 60-day mortality between the two groups. Therefore, the author concluded that the addition of colistin to DCT should be considered in severe cases with septic shock at presentation, then withdrawn after clinical stabilization with a stable switch to the less nephrotoxic regimen of DCT (Oliva et al., 2017a). A randomized controlled trial (RCT) addressing this topic has not yet been performed. The only matched case-control study to date included 48 patients with a DCT-containing combination (daily dose of meropenem and ertapenem up to 6 and 2 g, respectively), matched with 96 controls of DCT-sparing regimens. The other concomitant antibiotics comprised colistin (9 MU every 12 h), gentamicin (5–7 mg/kg daily) and high dose tigecycline. The 28-day mortality was significantly higher in the DCT-sparing arm (47.9% vs. 29.2%, P = 0.04). In multivariate analysis, the DCT-containing regimen was associated with a reduction in 28-day mortality (OR, 0.43; 95% CI, 0.23–0.79) (De Pascale et al., 2017). Mashni et al. (2018) performed a critical review of the current studies investigating DCT for CPKP infections, which contained eight case reports and six clinical studies (a total of 171 patients). Most patients were critically ill, and all were treated with ertapenem followed by a prolonged infusion of meropenem or doripenem. Clinical and microbiological successes were reported in approximately 70% of the patients and mortality in 24%. Adverse events, most frequently seizures, sodium disorders, and gastrointestinal symptoms, were reported in 16 patients, without the requirement for treatment interruption (Mashni et al., 2018).

Double-carbapenem therapy seems promising based on current reports, though the majority of infections involved in the studies were caused by Class A carbapenemase or KPCproducing K. pneumoniae. The efficacy of DCT against MBLs, such as NDM, would require further investigation.

# NEW ANTIBIOTICS

# Ceftazidime/Avibactam

Ceftazidime/avibactam (CAZ/AVI, AvyCaz <sup>R</sup> , Allergan Inc., Jersey City, NJ, United States) is a new β-lactam/β-lactamase inhibitor combination recently approved for the treatment of cIAIs and cUTIs in the United States in February 2015 (Kaye and Pogue, 2015), and for the treatment of HAP and VAP in January 2018. Unlike most β-lactamase inhibitors, avibactam is not a β-lactam. Avibactam is a novel synthetic non-β-lactam (diazabicyclooctane)/β-lactamase inhibitor that inhibits a wide range of β-lactamases, including Ambler Class A (GEM, SHV, CTX-M, and KPC), Class C (AmpC), and some Class D (OXA-48) β-lactamases (de Jonge et al., 2016). It does not inhibit Class B MBLs (IMP, VIM, VEB, and NDM) (Syue et al., 2016; Wong and van Duin, 2017). The addition of avibactam restores ceftazidime activity against various Enterobacteriaceae and P. aeruginosa, therefore expanding the activity spectrum of ceftazidime to MDR Gram-negative bacteria.

In vitro studies of CAZ/AVI showed adequate efficiency against CRE isolates (Castanheira et al., 2015; Dupont et al., 2016). However, clinical data on the efficacy of CAZ/AVI in severe infections caused by CRE are still lacking. At the time of writing this review, only one prospective, multicenter, observational study (van Duin et al., 2018) and a few case series and cohort studies describing the use of CAZ/AVI for treating CRE infections had been published (Shields et al., 2016, 2017b; Castón et al., 2017; King et al., 2017; Krapp et al., 2017; Temkin et al., 2017). Three studies have compared treatment with CAZ/AVI and other agents for CRE infections (**Table 4**) (Castón et al., 2017; Shields et al., 2017b; van Duin et al., 2018). The first study enrolled hematologic patients with CRE bacteremia in Spain and Israel (Castón et al., 2017). Compared to patients treated with other agents (n = 23), the patients receiving CAZ/AVI (n = 8) had a higher 14-day clinical cure rate in univariate analysis (85.7% [6/8] vs. 34.8% [8/23], P = 0.031). However, there was a lack of statistical significance in multivariate analysis due to the small number of cases. Another study of patients with CRE bacteremia showed that 13 patients treated with CAZ/AVI had a higher clinical success rates, compared to 96 patients treated with other regimens (85% [11/13] vs. 40.6% [39/96], P = 0.003) (Shields et al., 2017b). Although a better clinical response was consistent in multivariate analysis, these results were limited by small case numbers with CAZ/AVI treatment and a potential bias in the selection of therapy. Finally, the first prospective cohort study to compare the clinical outcomes for patients with CRE infections was recently published (van Duin et al., 2018). This observational study compared 38 patients treated with CAZ/AVI to 99 patients treated with colistin for KPC-producing CRE infections; combination therapy was used in 63 and 94% of patients treated with CAZ/AVI and colistin, respectively. Primary BSI was the most common infection foci (46%), followed by


HAP (22%). Adjusted all-cause mortality was significantly lower in the CAZ/AVI group (absolute difference 23%; 95% CI, 9% to 35%; P = 0.001). A multicenter, retrospective study reviewed 60 patients receiving CAZ/AVI for CRE infections (King et al., 2017). The authors reported an overall in-hospital mortality rate of 32%, a microbiological cure rate of 53%, and a clinical success rate of 65%. There was no significant difference in the in-hospital mortality rate between patients receiving CAZ/AVI monotherapy vs. CAZ/AVI combination therapy (30% [10/33] vs. 33% [9/27], P = 1.0), and between patients with bacteremia vs. those without (39% [9/23] vs. 27% [10/37], P = 0.397). Shields et al. (2016) reported a case series in a single center including 37 cases of patients with CRE infections who received treatment with CAZ/AVI. The survival rate at 30 days was 76% (28/37). Clinical success was observed as 59% (22/37), with no significant difference between patients receiving monotherapy vs. combination therapy (58% [15/26] vs. 64% [7/11]). However, they also reported a CRE infection recurrence rate of 23% (5/22) among patients who had displayed clinical success, and an overall microbiologic failure rate of 27% (10/37). Temkin et al. (2017) reported a case series of 38 patients with CRE (n = 36) and P. aeruginosa (n = 2) infections treated with CAZ/AVI. Of these, 65.8% (25/38) of patients concurrently received other regimens. The overall clinical and/or microbiological cure rate was 73.7% (28/38), with 69.2% (9/13) in the monotherapy group and 76.0% (19/25) in the combination therapy group.

As described earlier, avibactam inhibits Ambler Class A (KPC) and Class D (OXA-48) but did not inhibit Class B MBLs (NDM, VIM, and IMP). Therefore, CAZ/AVI is not active against all CRE isolates (Falcone et al., 2018). In contrast, the monobactam antibiotic aztreonam (ATM) is stable against MBLs but is hydrolyzed by many other β-lactamases (ESBL, AmpC, and cephalosporinases) frequently co-produced by the MBLproducing strains (Marshall et al., 2017). The combination of CAZ/AVI and ATM has been proposed as a potential therapeutic strategy against infections caused by MBL-producing bacteria. Crandon and Nicolau (2013) tested the efficacy of CAZ/AVI and ATM using the neutropenic-mouse thigh infection model, and concluded that this combination represents an attractive treatment option for infections caused by MBL-producing strains that co-produce ESBLs or AmpC. Although studies have demonstrated good in vitro activity of CAZ/AVI and ATM against MBL-producing Enterobacteriaceae (Marshall et al., 2017; Wenzler et al., 2017), clinical data are still lacking.

These observational studies are subject to selection bias. Further RCTs are warranted to evaluate the effectiveness of CAZ/AVI, as well as CAZ/AVI and ATM combination, for the treatment of CRE infections. Whether CAZ/AVI combination therapy is more effective than monotherapy for deep-seat infections or high-risk patients also needs further evaluation. Moreover, the emergence of CAZ/AVI resistant strains during treatment has been reported. Shields et al. detected CAZ/AVI resistance due to mutations in the blarmKPC−<sup>3</sup> gene in 3 of 10 microbiologic failures, following CAZ/AVI treatment for 10– 19 days (Shields et al., 2016, 2017a). Clinicians should be aware of the possible emergence of resistance following treatment with CAZ/AVI.

TABLE 4


Efficacy of

ceftazidime/avibactam

 for treatment of

carbapenem-resistant

Enterobacteriaceae

infections.

### Meropenem/Vaborbactam

fmicb-10-00080 January 28, 2019 Time: 18:40 # 8

Vaborbactam is a novel boron-containing serine-β lactamase inhibitor which confers activity against certain meropenemresistant bacteria by inhibiting Ambler Class A and C serine carbapenemases, such as KPC (Castanheira et al., 2017). However, it has no in vitro activity against Class B metalloβ-lactamases producers (NDM or VIM) or Class D OXA-48 β-lactamases (Castanheira et al., 2016; Nelson et al., 2017). Meropenem/vaborbactam (MER/VAB) (Vabomere <sup>R</sup> , The Medicines Company, Parsippany, NJ, United States) is a novel carbapenem/β-lactamase inhibitor antimicrobial agent approved in August 2017 by the United States Food and Drug Administration (FDA) for the treatment of cUTIs, including pyelonephritis.

MER/VAB exhibited effective in vitro antibacterial activity against CRE isolates (Castanheira et al., 2017; Pfaller et al., 2018), with susceptibility rates ranging from 66.2 to 100% (Dhillon, 2018). The clinical data supporting MER/VAB for CRE infections is from the multicenter, randomized, openlabel Tango II trial comparing the efficacy and safety of MER/VAB with the best available therapy (BAT) for the treatment of serious CRE infections (Wunderink et al., 2018). Of the 77 enrolled patients, 47 had confirmed CRE infections, including 22 BSI, 16 cUTI/pyelonephritis, 5 HAP/VAP, and 4 cIAI. The results showed statistical significance in favor of MER/VAB over BAT for a clinical cure (65.6% [21/32] vs. 33.3% [5/15], P = 0.03) and 28-day mortality (15.6% [5/32] vs. 33.3% [5/15], P = 0.03). Furthermore, MER/VAB was associated with decreased nephrotoxicity as compared with BAT (4.0% vs. 24%) (Wunderink et al., 2018). Data for MER/VAB on CRE infections are still accumulating. A study of MER/VAB vs. piperacillin/tazobactam on HAP/VAP (TANGO III, NCT02168946) is ongoing.

# Plazomicin

Plazomicin is a next-generation aminoglycoside synthetically derived from sisomicin, which retains activity against bacteria containing aminoglycoside-modifying enzymes (Landman et al., 2010; Castanheira et al., 2018a). Plazomicin (ZemdriTM, Achaogen, Inc., San Francisco, CA, United States) was approved in June 2018 by the FDA for the treatment of adults with cUTI including pyelonephritis who have limited or no alternative treatment options, with a recommended dose of 15 mg/kg every 24 h for normal renal function. The approval was based on two phase 3 clinical trials comparing the efficacy and safety of plazomicin with meropenem (NCT02486627, not published yet) and levofloxacin (NCT01096849, not published yet) for the treatment of cUTI and acute pyelonephritis. Plazomicin demonstrates a broad-spectrum activity against Gram-positive cocci and Gram-negative bacilli, including ESBL producers and CRE (Walkty et al., 2014; Karaiskos et al., 2015). Studies have shown that plazomicin is more potent than other aminoglycosides against KPC-producing Enterobacteriaceae (Endimiani et al., 2009b). A recent study evaluating the activity of plazomicin and comparators against clinical isolates showed that isolates carried 16S rRNA methyltransferases were resistant to all available aminoglycosides and had elevated plazomicin MICs (Castanheira et al., 2018b). The methyltransferase enzymes commonly found in MBL producers still render bacteria resistant to plazomicin.

The clinical data supporting plazomicin for the treatment of serious infections due to CRE is from a multicenter, randomized, open-label study comparing the efficacy and safety of plazomicin vs. colistin (both in combination with tigecycline or meropenem) (CARE trial, NCT01970371). Although the full report of this trial is not yet available, preliminary analysis shows that the plazomicin group had a significantly lower 28-day mortality compared to the colistin group (7.1% [1/14] vs. 40.0% [6/15]; difference, −32.9%; 95% CI, −60.1% to −4.0%) (McKinnell et al., 2017). In addition to a lower mortality rate, it was welltolerated and was associated with a lower incidence of serum creatinine elevations. However, these data should be interpreted with caution due to the small sample size.

### Eravacycline

Eravacycline, a synthetic fluorocycline antibacterial agent of the tetracycline class, has broad-spectrum antimicrobial activity against Gram-positive, Gram-negative, and anaerobic bacteria, exception for P. aeruginosa (Zhanel et al., 2016). Eravacycline (XeravaTM, Tetraphase Pharmaceuticals, Inc., Watertown, MA, United States) was approved by the FDA in August 2018 for the treatment of cIAIs, based on two phase 3 clinical trials which demonstrated statistical non-inferiority of eravacycline to two commonly used comparators: ertapenem (IGNITE1 trial) (Solomkin et al., 2017) and meropenem (IGNITE4 trial, full data not published yet). The recommended dosage is 1 mg/kg intravenous infusion every 12 h. Of note, eravacycline is not indicated for the treatment of cUTI because clinical trials (NCT01978938 and NCT03032510) did not demonstrate the efficacy for the combined endpoints of clinical cure and microbiological success. Eravacycline has been shown to have effective in vitro activity against MDR pathogens (Sutcliffe et al., 2013; Livermore et al., 2016; Zhang et al., 2016), with twofold higher activity than tigecycline against CRE (Livermore et al., 2016). However, there is limited clinical data for the treatment efficacy of eravacycline in CRE infections.

### POTENTIAL ANTIBIOTICS IN DEVELOPMENT

### Imipenem/Cilastatin and Relebactam

Imipenem/cilastatin and relebactam (IMI/REL, Merck & Co. Inc., Kenilworth, NJ, United States) combines an approved carbapenem with a novel β-lactamase inhibitor. The chemical structure of relebactam is similar to avibactam (Watkins et al., 2013). Like avibactam, relebactam contains a diazabicyclooctane core, which covalently and reversibly binds Class A and C β-lactamases in vitro, with an inhibitory mechanism similar to that of avibactam (Blizzard et al., 2014). However, relebactam cannot inhibit Class D OXA-48 like avibactam (Petty et al., 2018; Zhanel et al., 2018).

Relebactam potentiates imipenem activity against imipenemnon-susceptible Enterobacteriaceae (Karlowsky et al., 2018). The clinical efficacy and safety of IMI/REL has been shown in the phase 2 studies for cUTIs (NCT01505634) (Sims et al., 2017) and cIAIs (NCT01506271) (Lucasti et al., 2016).

The RESTORE-IMI 1 study (NCT02452047) is a multicenter, randomized, double-blind, comparator-controlled trial, comparing the efficacy and safety of IMI/REL vs. colistin plus imipenem/cilastatin (COL + IMI) in patients with imipenemnon-susceptible bacterial infections. Patients with HAP/VAP, cIAI, or cUTI due to imipenem-non-susceptible pathogens, were randomized 2:1 to receive IMI/REL or COL + IMI. In this study, 31 of 47 randomized and treated patients met mMITT criteria. Favorable clinical responses at Day 28 were comparable in the IMI/REL and the COL + IMI arms (71.4% [15/21] vs. 70.0% [7/10]), and the 28-day all-cause mortality was lower in the IMI/REL arm compared to the COL + IMI arm (9.5% [2/21] vs. 30.0% [3/10]) (presented by Motsch et al. at the European Congress of Clinical Microbiology and Infectious Diseases; April 22, 2018; Madrid, Spain). In addition, treatmentemergent nephrotoxicity was lower with IMI/REL compared to COL + IMI (10.3% [3/29] vs. 56.3% [9/16], P = 0.002) (presented by Brown et al. at IDWeek, October 6, 2018; San Francisco, CA, United States).

### Cefiderocol

Cefiderocol is the first siderophore antibiotic to advance into late-stage development. Cefiderocol, a novel siderophore cephalosporin, exhibits potent in vitro and in vivo activity against a variety of Gram-negative bacteria, including CRE (Saisho et al., 2018). Cefiderocol has a unique antibacterial mechanism in which its catechol side chain of binds to ferric acid, with the complex then being actively transported into bacteria via bacterial iron transporters (Ito et al., 2016). In addition, cefiderocol is also highly active against carbapenemase hydrolysis (Wright et al., 2017). Cefiderocol has demonstrated potent in vitro activity against CRE isolates, with 97.0% (991/1,022) of isolates demonstrating cefiderocol MICs of ≤4 mg/L (Hackel et al., 2018). A multicenter, randomized, open-label CREDIBLE-CR trial is ongoing to assess the efficacy of cefiderocol, compared to BAT, for severe infections caused by carbapenem-resistant Gram-negative pathogens (NCT02714595).

# CONCLUSION

The increasing prevalence of CRE infections represents a major threat to human health. Effective antibiotics against CRE remain very limited, with polymyxins, tigecycline, fosfomycin, and aminoglycoside being the mainstays of anti-CRE therapy. With the high mortality of CRE infections and increasing resistance to available antibiotics, it is urgent for the medical community to develop new and effective therapeutic strategies. The first potential strategy is to increase the doses of anti-CRE agents. High-dose carbapenem, colistin, and tigecycline have been associated with better clinical outcomes in CRE infections. The second potential strategy is to combine these anti-CRE agents. Although large-scale observational studies and systematic reviews suggest that combination therapy may be beneficial to patients with severe CRE infections, the quality of evidence is low due to substantial heterogeneity in study design and patient population. A recent RCT showed no benefit of colistin combination therapy with meropenem when compared to colistin monotherapy. The debate on whether or not these agents should be used in combination will continue. DCT with ertapenem infusion prior to a high-dose meropenem or doripenem infusion has been adopted as a salvage therapy for critically ill patients with CRE infections. Studies suggest that a combination of colistin or tigecycline with DCT might be a reasonable strategy for severe CRE infections. However, the target patients and molecular phenotypes of CRE that would benefit from DCT require further investigation. The recent introduction of the novel β-lactamase inhibitors avibactam, vaborbactam, and relebactam, has provided additional therapeutic options for CRE infections. Of note, these new β-lactamase inhibitors are not active against all major carbapenemases. Avibactam inhibits KPC and OXA-48, while vaborbactam and relebactam inhibit only KPC. Expectantly, these novel β-lactamase inhibitor combinations will also provide carbapenem-sparing options for the treatment of MDR bacteria and help control the rapid emergence of carbapenemase-producing bacteria. The next-generation aminoglycoside plazomicin and the fully synthetic tetracycline antibiotic eravacycline, both recently approved by the FDA, may provide us alternative therapeutic options for CRE infections. The activity of plazomicin and eravacycline against CPE has been demonstrated in vitro. However, we need more clinical data to support their use in the treatment of serious infections due to CRE. Many other new antibiotics with potential anti-CRE activity are in various stages of development, including the novel β-lactamase inhibitor combination IMI/REL and the novel siderophore cephalosporin cefiderocol. Currently, high-dose and combination strategies that may include the new β-lactam/β-lactamase inhibitors should be considered for severe CRE infections to maximize treatment success.

# PERSPECTIVE

In the future when more treatment options are available, anti-CRE therapy should be individualized and based on molecular phenotypes of resistance, susceptibility profiles, disease severity, and patient characteristics. More clinical trials are needed to provide high-quality evidence and guide the selection of effective anti-CRE strategies.

# AUTHOR CONTRIBUTIONS

C-CS, Y-TC, and S-YL collected and analyzed the data, and wrote the manuscript. Y-HC and P-RH revised and edited the manuscript. C-CS, Y-TC, S-YL, Y-HC, and P-RH read and approved the final version of the manuscript.

# REFERENCES

fmicb-10-00080 January 28, 2019 Time: 18:40 # 10



cephalosporin for Gram-negative bacteria, in healthy subjects. Antimicrob. Agents Chemother. 62:e02163-17. doi: 10.1128/AAC.02163-17


fmicb-10-00080 January 28, 2019 Time: 18:40 # 12

retrospective propensity-matched case control study. J. Microbiol. Immunol. Infect. 51, 621–628. doi: 10.1016/j.jmii.2016.08.022


**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2019 Sheu, Chang, Lin, Chen and Hsueh. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

# Dynamic Colonization of Klebsiella pneumoniae Isolates in Gastrointestinal Tract of Intensive Care Patients

Qiao-ling Sun<sup>1</sup> , Danxia Gu<sup>2</sup> , Qi Wang<sup>1</sup> , Yanyan Hu<sup>1</sup> , Lingbin Shu<sup>1</sup> , Jie Hu<sup>1</sup> , Rong Zhang<sup>1</sup> \* and Gong-Xiang Chen<sup>1</sup> \*

<sup>1</sup> Department of Clinical Laboratory, Second Affiliated Hospital of Zhejiang University, School of Medicine, Hangzhou, China, <sup>2</sup> Key Laboratory of Tumor Molecular Diagnosis and Individualized Medicine of Zhejiang Province, Clinical Research Institute, Zhejiang Provincial People's Hospital, People's Hospital of Hangzhou Medical College, Hangzhou, China

### Edited by:

Yi-Wei Tang, Memorial Sloan Kettering Cancer Center, United States

### Reviewed by:

Michael Bachman, University of Michigan, United States Claudia Monari, University of Perugia, Italy

### \*Correspondence:

Rong Zhang zhang-rong@zju.edu.cn Gong-Xiang Chen chengongxiang@zju.edu.cn

### Specialty section:

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

Received: 17 October 2018 Accepted: 28 January 2019 Published: 11 February 2019

### Citation:

Sun Q, Gu D, Wang Q, Hu Y, Shu L, Hu J, Zhang R and Chen G-X (2019) Dynamic Colonization of Klebsiella pneumoniae Isolates in Gastrointestinal Tract of Intensive Care Patients. Front. Microbiol. 10:230. doi: 10.3389/fmicb.2019.00230 Gastrointestinal carriage is regarded as a major reservoir of K. pneumoniae infections, especially in intensive care patients. A total of 101 (95.3%) KPC-producing carbapenemresistant K. pneumoniae (CRKP) isolates were identified among 106 CRKP isolates collected from stool samples of inpatients performing active rectal screening for carbapenem-resistant Enterobacteriaceae during hospitalization in the ICUs of a tertiary hospital between 2016 and 2017. Among them, six KPC-producing CRKP isolates from three patients (two isolates for each patient) were identified with distinct antibacterial susceptibility. Our findings showed that: (1) blaKPC−<sup>2</sup> gene is predominant in CRKP strains isolated from the intensive care patients and can be incorporated into various plasmids that are transmissible among multiple bacterial hosts in the human gastrointestinal tract; (2) the human gastrointestinal tract has a capacity to dynamically colonize multiple clones of CRKP strains with varied plasmids, diverse antimicrobial resistance genes and virulence genes. K. pneumoniae colonization is an important step in progression to extraintestinal infection, which provides the rationale for establishing intervention measures to prevent subsequent infection. Thus, close surveillance on CRKP colonization, together with effective infection prevention and control measures, should be put into practice.

Keywords: CRKP, KPC-2, gastrointestinal carriage, dynamical colonization, multiple clones

# INTRODUCTION

Klebsiella pneumoniae is a major opportunistic pathogen that can cause invasive hospital-acquired infections among immune-compromised patients especially for those from ICU ward with critical illness. Carbapenem is a first line therapy for the treatment of infections caused by multidrugresistant K. pneumoniae. However, carbapenem-resistant K. pneumoniae (CRKP) has emerged as a public threat to cause serious infections with high mortality up to 33.24–50.06% (Xu et al., 2017). Recent data from the CHINET surveillance program showed that the prevalence of CRKP had dramatically increased from 3 to 17.9% since 2005 (Hu et al., 2016, 2017). A previous study on nationwide surveillance of clinical carbapenem-resistant Enterobacteriaceae (CRE) strains has

**91**

revealed that blaKPC−<sup>2</sup> was the most prevalent genotype in China, accounting for 73% of CRKP isolates (Zhang et al., 2017).

Carbapenem-resistant Hypervirulent K. pneumoniae (CR-HvKP), especially ST11 KPC-producing strains, have been increasingly reported in recent years (Zhang R. et al., 2015; Zhang Y. et al., 2015; Gu et al., 2018). They often cause invasive, even life-threatening infections with high mortality among young and healthy populations. In a previous study, we reported an outbreak of ST11 CR-HvKP in the ICU ward; further study indicated that the acquisition of a virulence plasmid carrying rmpA2 and aerobactin biosynthesis genes by classic ST11 CRKP strains played a critical role (Gu et al., 2018).

Gastrointestinal carriage has been regarded as a major reservoir of K. pneumoniae infections, especially in intensive care patients (Gorrie et al., 2017). A prospective study in 1971 indicated that 18.5% patients colonized with multidrug-resistant K. pneumoniae after hospital admission had higher risk to develop subsequent infection caused by identical bacteria within 21 days compared to those who did not become intestinal carriers (45% vs. 11%) (Martin and Bachman, 2018). A 2016 study reported similar colonization rates (23%) and increased risk of infection following colonization (5.2% in colonized vs. 1.3% in noncolonized) (Martin et al., 2016). The carriage rate of CRKP is reported to range from 8 to 9% in medical and surgical departments to 5% in intensive care units (Wiener-Well et al., 2010). CRKP colonization can persist and spread silently for years, even trigger a clonal outbreak in long-term care facilities while newly colonized patients can develop fatal infections (Martin and Bachman, 2018). Our previous study showed the evolution of tigecycline- and colistin-resistant CRKP has occurred in vivo under the antibiotic selection and CRKP could persistently colonize in the human gastrointestinal tract for 3 years even without antibiotic selection pressure (Zhang et al., 2018). Previous reports mainly focused on epidemiological data of CRKP carriage and associated risk factors (Papadimitriou-Olivgeris et al., 2012; Feldman et al., 2013; Papadimitriou-Olivgeris et al., 2013), in this study, we are going to investigate the molecular characterization of multiple clones of CRKP strains colonized in the gastrointestinal tract of intensive care inpatients, aiming at illustrating certain discipline underlying the colonization of CRKP.

# MATERIALS AND METHODS

### The Rectal CRKP Isolates From the Intensive Care Inpatients

A total of 106 K. pneumoniae isolates that exhibited carbapenem resistance phenotype (MIC value of meropenem ≥4 µg/ml) were identified from stool samples of inpatients performing active rectal screening for CRE during hospitalization in the ICUs of Second Affiliated Hospital of Zhejiang University, School of Medicine (Hangzhou, China) between 2016 and 2017. Identification of species was confirmed via a matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (Bruker Daltonik GmbH, Bremen, Germany). The presence of carbapenem-resistant genes (blaNDM, blaKPC, blaIMP, blaVIM, and blaOXA−48) was screened using PCR (Zhang et al., 2017). Positive products were validated with Sanger DNA sequencing. A total of 101 (95.3%) KPC-2-producing CRKP isolates were identified among 106 rectal CRKP isolates. Among them, six KPC-2-producing CRKP isolates from three patients (two isolates for each patient) were identified with distinct antibacterial susceptibility. In order to investigate the origin of the intestinal CRKP isolates and further look into the relationship between the rectal colonization and extraintestinal infections, we reviewed medical history of the three inpatients in detail and subjected the KPC-2-producing CRKP isolates collected during peri-hospitalization period to phenotypic and genotypic characterization.

### Phenotypic Characterization

The antimicrobial susceptibilities of the isolates were determined using a broth microdilution procedure and the interpretations were in accordance with the guideline document M100-S26 established by Clinical and Laboratory Standards Institute (CLSI, 2016).

We did pulsed-field gel electrophoresis (PFGE), S1-PFGE, and Southern hybridization as previously reported (Huang et al., 2016). A dendrogram was generated from the homology matrix with a coefficient of 0.5% using the unweighted pair-group method using arithmetic averages ("UPGMA") to describe the relationships among PFGE profiles. Isolates were considered to belong to the same PFGE group if their Dice similarity index was ≥85%.

The virulence gene rmpA2, encoded on the virulence plasmid, was identified by PCR as previously described (Gu et al., 2018). All isolates were performed with the string test to identify the hypermucoviscous phenotype as described previously (Gu et al., 2018). As a test of virulence, we quantified virulence with Galleria mellonella larva models. The G. mellonella larva was injected into the hemocoel of each caterpillar via the last left proleg with 10 µl suspensions of a K. pneumoniae strain containing a final concentration of 10<sup>6</sup> CFU/mL, incubated at 37◦C and observed every 12 h for 3 days. The effect of 1 × 10<sup>6</sup> CFU of each K. pneumoniae strain on survival was assessed in G. mellonella. HvKP strain K. pneumoniae 4 and strain PC K. pneumoniae 4, reported in a previous study (Gu et al., 2018), were treated as controls. HvKP strain K. pneumoniae 4 was a ST11 KPC-producing hypermucoviscous strain harboring various virulence genes (rmpA2, iutA, iroE, kpn, ycfM, iucABCD, mrkABCDF, and fimA–H), located on the virulence plasmid pLVPK. Strain PC K. pneumoniae 4 is a mutant strain of K. pneumoniae 4, of which the virulence plasmid has been removed in plasmid curing experiments. Strain PC K. pneumoniae 4 was negative for string test and demonstrated reduced virulence in G. mellonella models. Two control groups were performed: the first group was inoculated with PBS to monitor for killing due to physical trauma and attrition while the second received no injection. Eight larvae from each group were examined and all the experiments were performed in triplicates. Kaplan-Meier survival curves were plotted using GraphPad Prism version 7.00, the log rank (Mantel-Cox) test was used to analyze whether significant differences (P < 0.05) in the survival rates of the infected G. mellonella larvae were observed.

# Whole Genome Sequencing and Bioinformatics Analysis

fmicb-10-00230 February 7, 2019 Time: 18:22 # 3

Genomic DNA was extracted from overnight cultures by using the PureLink Genomic DNA Mini Kit (Invitrogen, Carlsbad, CA, United States) and was subjected to whole genome sequencing using 150 bp pair-end strategies with the Illumina HiSeq X10 platform. Raw reads were trimmed and assembled to contigs using SPAdes v3.11.1 (Bankevich et al., 2012). Assembled genome sequences were submitted to the NCBI database with accession number QMKA00000000, QMKB00000000, QMKC00000000, QMKD00000000, QMKE00000000, and QMKF00000000. Genome sequences were annotated with the RAST tool (Overbeek et al., 2013) and Prokka (Seemann, 2014). Multilocus sequence types (MLSTs), virulence-associated genes encoding yersiniabactin, aerobactin, salmochelin and the regulators of mucoid phenotype were determined with Kleborate (Pichler et al., 2017). Serotyping was performed using Kaptive (Wyres et al., 2016). Acquired antibiotic resistance genes were identified with ResFinder 2.1 (Zankari et al., 2012). Heatmap of the antimicrobial resistance genes was generated using an in-house script.

### Ethics Statement

The study was approved by the Ethics Committee of Second Affiliated Hospital of Zhejiang University, School of Medicine (2017-099). All subjects gave written informed consent in accordance with the Declaration of Helsinki.

### Biosafety Statement

All concerns related to the safe and appropriate use of human-derived materials, infectious agents, or genetically modified organisms were approved by the Institutional Biosafety Committee of Second Affiliated Hospital of Zhejiang University, School of Medicine. All experiments were conducted under the guidelines from the Biological Agent Reference Sheet.

# RESULTS

Three elderly patients aged 61–70 years were admitted to the ICU between August 2016 and February 2017. The three patients all developed diseases including bacterial pneumonia and bloodstream infections and underwent surgery, followed by antimicrobial treatment and mechanical ventilation, but they responded differently to antibiotic treatment. Only two patients, Patient 2 and 3, succeeded in infection control while Patient 1 experienced a persistent fever and pulmonary infection till against-advice discharge. Detailed information about the patients is available in the **Figures 1**, **2**. Active rectal screening for CRE was conducted for all the patients during hospital admission and peri-hospitalization period, and all the three patients were found to carry the KPC-producing CRKP strains in the gastrointestinal tract on admission and during stays at healthcare settings. Thereinto, two CRKP isolates from stool samples of each patient presented distinct antimicrobial susceptibility profiles. In order to study the origin of the intestinal CRKP carriage and the relationship between the rectal colonization and extraintestinal infections, three, five, and five more KPC-producing CRKP isolates from various types of specimens including stool were successively identified in Patient 1, Patient 2, and Patient 3 during peri-hospitalization period, respectively (**Figures 1**, **2**).

# Acquisition of KPC-Producing CRKP Isolates From Stool Samples

Patient 1, a 71-year old woman, had previously been admitted to a local hospital for acute ventricular hemorrhage and was transferred to the Neurological intensive care unit (NICU) of our hospital on Aug 12, 2016. Three KPC-2-producing CRKP isolates (namely JZX2078, JZX2124, and JZX11-E8) were isolated. Isolate JZX2078 was recovered from stool sample after 1 week of admission, belonging to ST11 with unidentified serotype. It harbored three plasmids with sizes of ∼230, ∼210, and ∼80 Kb. The ∼210 Kb plasmid was a pLVPK-like virulence plasmid which bears iucABCD, iroBCD, and rmpA2 genes. The ∼230 Kb plasmid was a pKP04VIM (KU318421.1)-like plasmid, while the ∼80 Kb one was a blaKPC−2-bearing p69- 2 (CP025458.1)-like plasmid (**Table 1**). The second isolate JZX2124 was also isolated from stool a week before discharge. It was identified as a K19, ST1 strain which harbored two multidrug resistance plasmids, the ∼130 Kb p11219-CTXM (MF133442.1)-like plasmid and the ∼170 Kb, blaKPC−2-bearing pIT-06C07 (LT009688)-like plasmid (**Table 1**). Compared to isolate JZX2078 with 9 antibiotic resistance genes, isolate JZX2124 carried 12 resistance genes including tet(A), oqxAB genes and two different genes encoded for extended spectrum β-lactamases (ESBLs) as shown in **Figure 3**. And the third isolate JZX11-E8 was recovered from blood. With completely different antibacterial susceptibility profiles and distinct PFGE patterns (**Figure 2**), these three isolates were proved to originate from different clones.

Patient 2, a 66-year old man, was admitted to the Surgical Intensive Care Unit (SICU) for severe acute pancreatitis with acute respiratory distress syndrome on January 12th, 2017. Two weeks after admission, he was transferred to Surgical Ward preparing for a surgery of cholecystectomy and peritoneal lavage and drainage, immediately followed by transferring to the SICU and experiencing one more surgery of intraperitoneal hemostasis and peritoneal lavage and drainage. Two CRKP isolates, WJQ129 and WJQ196, were isolated from stool and sputum samples, respectively, during his staying in Surgical Ward, while three CRKP isolates (WJQ205, WJQ250, and WJQ298) were collected from stool, sputum, and drainage fluid samples, respectively, during his staying in SICU. With similar antimicrobial susceptibility profiles and PFGE patterns, isolates WJQ129 and WJQ196 shared high homology. The same situation was presented for isolates WJQ205, WJQ250, and WJQ298. However, as shown in **Figure 2**, isolates recovered from the patient during his earlier staying in Surgical Ward were completely distinct from those recovered during his later staying in SICU. It may be suggested that Patient 2

FIGURE 2 | Antimicrobial susceptibility and PFGE profiles of 13 CRKP isolates collected from three inpatients. IPM, Imipenem; MEM, Meropenem; ETP, Ertapenem; CAV, Ceftazidime-avibactam; CAZ, Ceftazidime; CTX, Cefotaxime; FEP, Cefepime; TZP, Piperacillin-tazobactam; SCF, Cefoperazone-sulbactam; ATM, Aztreonam; PE, colistin; TGC, Tigecycline; CIP, Ciprofloxacin; AK, Amikacin; CVC, central venous catheter.

TABLE 1 | Sequence information for rectal CRKP isolates from the three inpatients.


was successively infected by two different strains of KPCproducing K. pneumoniae isolated from different ICU wards. CRKP colonized in the gastrointestinal tract would easily cause extraintestinal infections once host defense system is weakened. Isolate WJQ129 was an ST290, K21 strain which harbored 9 antibiotic resistance genes (**Table 1**, **Figure 3**) and two plasmids with sizes of ∼230 and ∼90 Kb. The ∼230 Kb K. pneumoniae strain FDAARGOS\_443 plasmid unnamed1 (CP023937.1)-like plasmid carries resistance genes tet(A), qnrS, blaCTX−M−14, and dfrA1. The ∼90 Kb p628-KPC (KP987218.1) like plasmid carries blaKPC−2. Isolate WJQ205 belonged to ST11, and identified as K64, carrying three plasmids (∼240, ∼210, and ∼80 Kb) (**Table 1**). The ∼210 Kb plasmid is a pLVPK-like virulence plasmid, carrying iucABCD, iroBCD, and rmpA2 genes. The ∼240 and ∼80 Kb plasmids are pA324- IMP (MF344566)- and blaKPC−2-bearing p69-2 (CP025458.1) like plasmids, respectively.

Patient 3, a 61-year old woman, was admitted to our NICU for disorder of consciousness after a surgery of eliminating intracranial hematoma and bilateral external ventricular drainage in a local hospital. Five CRKP isolates were recovered from different samples belonging to two highly similar PFGE patterns. Interestingly, two homologous CRKP clones ZWH170 and ZWH183, collected from stool samples at short intervals, were both ST11, K64 serotype and harbored highly similar resistance genes, but showed different antibacterial susceptibility profiles (**Table 1** and **Figures 2**, **3**). Isolate ZWH170 carried plasmids with sizes of ∼138 and ∼78 Kb, while isolate ZWH183 carried one more plasmid which is around ∼105 Kb. The ∼138 Kb plasmid was a pKPC-CR-HvKP4-like resistance plasmid carrying blaKPC−<sup>2</sup> and blaCTX−M−<sup>65</sup> genes. The ∼78 Kb plasmid was a p675920-2-like multidrug resistance plasmid carrying blaLAP−2, qnrS1, tet(A), and sul2 genes. Additionally, the ∼105 Kb plasmid was a pLVPK-like virulence plasmid.

After half-year rehabilitation therapy in another hospital, the patient agreed to provide the stool sample for follow-up investigation. Surprisingly, CRKP was no longer detected from her stool samples.

# Characterization of Virulence in CRKP Isolates

Six KPC-producing K. pneumoniae isolates (JZX2078, JZX2124, WJQ129, WJQ205, ZWH170, and ZWH183) from stool samples of three patients were selected for further virulence characterization. According to the presence of virulence plasmid, the six isolates were divided into two groups: rmpA2-positive strains (JZX2078, WJQ205, and ZWH183) and rmpA2-negative strains (JZX2124, WJQ129, and ZWH170). S1-PFGE and Southern hybridization of the marker gene of the virulence plasmid rmpA2, which was hybridized to the roughly 210 Kb and 105 Kb plasmids, confirmed the presence of the virulence plasmid in three of the ST11 KPC-producing CRKP isolates (JZX2078, WJQ205, and ZWH183). The ∼210 Kb pLVPK-like virulence plasmid, harboring rmpA2, iroBCD, and iucABCD genes, were identified in two ST11 CRKP strains isolated from different ICU wards. The ∼105 Kb virulence plasmid in isolate ZWH183, carrying the rmpA2, iucABCD genes was absent in isolate ZWH170, suggesting the attenuated virulence potential. The results of S1-PFGE and Southern hybridization were described in **Figure 4**.

All the six CRKP isolates from stool samples were negative for string test. Moreover, in G. mellonella larva models, the survival rate of larva infected with rmpA2-carrying CRKP strains (JZX2078, WJQ205, and ZWH183) ranged from 20 to 40% at 16 h after infection; as for rmpA2-negative CRKP strains (JZX2124, WJQ129, and ZWH170), 60–80% of the larva survived at 20 h (**Figure 5**). The control group infected with HvKP strain K. pneumoniae 4 resulted in 0% survival by 16 h, whilst 70% survival was reached in larva infected with PC K. pneumoniae 4 after 16 h. The survival rate of G. mellonella larvae infected with rmpA2-carrying CRKP strains (WJQ205 and ZHW183) was significantly lower than that infected with the counterpart rmpA2-negative CRKP strains (WJQ129 and ZWH170) (P < 0.05). However, there was no difference on the survival rate of G. mellonella larvae between the CRKP strains JZX2078 and JZX2124 (P > 0.05). The survival rate of G. mellonella larvae infected with rmpA2-carrying CRKP strains (WJQ205, ZHW183, and JZX2078) and rmpA2-negative CRKP strains (JZX2124, WJQ129, and ZWH170) showed

statistical significance with that infected with HvKP (P < 0.05). Interestingly, the survival rate of G. mellonella larvae infected with all the rmpA2-carrying CRKP strains was significantly lower than that PC K. pneumoniae 4 (P < 0.05), while rmpA2-negative CRKP strains all exhibited no difference on the survival rate of G. mellonella larvae with PC K. pneumoniae 4 (P > 0.05).

### DISCUSSION

Gastrointestinal colonization is regarded as a common and significant reservoir of K. pneumoniae in terms of risk of transmission and infection (Martin and Bachman, 2018). Previous studies have found that K. pneumoniae gastrointestinal colonization was significantly associated with subsequent infections in inpatients (odds ratio >4), and 5% of colonized patients developed infections. In addition, 80% concordance was showed between infecting and colonizing isolates of K. pneumoniae within infected patients (Martin et al., 2016; Gorrie et al., 2017; Martin and Bachman, 2018). In our study, each of Patient 2 and Patient 3 had a homologous strain in the gut and an invasive site according to the **Figure 2**, suggesting that CRKP in the gastrointestinal tract would readily cause extraintestinal infections once the host defense system is weakened. Thus, CRKP colonization in the gastrointestinal tract is of significant importance.

As was widely reported (Holt et al., 2015; Lam et al., 2018), K. pneumoniae isolates are often volatile with a wide spectrum of diversity. The human gut is always considered as a reservoir for antibiotic resistance genes, with various species and abundant genes; horizontal transfer of resistance genes is extremely active. CRKP isolates colonized in the gut can be even more diversified. In the current study, three KPC-producing CRKP isolates from Patient 1 were highly heterogeneous belonging to different clones. Interestingly, the two from her intestine demonstrated completely different antibacterial susceptibility, PFGE and plasmid profiles. Patient 2 was successively infected by two different strains of CRKP isolated from different ICU wards, which indicated that hospitalacquired CRKP strains among the ICU ward is responsible for the colonizing and infecting strains in inpatients, and multiple clones of CRKP strains might spread in the clinical setting. blaKPC−<sup>2</sup> gene was the predominant genotype in 95.3% CRKP isolates and two blaKPC−2-bearing plasmids belonged to two major types of ∼138 Kb and ∼80 Kb in size, suggesting that the blaKPC−<sup>2</sup> gene can be incorporated into various extrachromosomal elements which are capable of horizontal transfer among multiple bacterial hosts in the human gastrointestinal tract.

According to a previous report (Feldman et al., 2013), 74% of the patients were identified with gastrointestinal carriage of CRKP 30 days after discharge from hospital, and when it came to

6 months, the proportion declined to <30%. Persistent carriage of CRKP was associated with several risk factors, including catheter use, long-term care facilities, recent acquisition (<4 months), and a low functional status (Feldman et al., 2013). Additionally, a hypothesis was posed that CRKP could colonize in host cells to circumvent phagocytosis by immune cells, and survival of CRKP within host cells might serve as a reservoir to protect from antibiotic treatments and enable long-term coexistence with the host (Yang et al., 2018). In this study, all the three inpatients underwent mechanical ventilation, carried CRKP isolates in 2 weeks after hospital admission and had suffered from bloodstream infection caused by CRKP strains during hospitalization. The persistent carriage of CRKP in gastrointestinal tract could last for 1–6 months in our study, inferring that long-term hospital stay and prolonged antibiotic usage could acquire CRKP strains from the settings and enrich them in the host gastrointestinal tract, and CRKP might survive intracellularly to persist in the gastrointestinal colonization. Interestingly, after the half-year rehabilitation therapy in another hospital, CRKP was no longer detected from the stool samples of the Patient 3.

ST11 has been one of the most prevalent MLST in the clinical CRKP strains among different parts of the world. Previous study reported an outbreak of ST11 CR-HvKP in the ICU of our hospital, whose emergence was the result of acquisition of a ∼170 Kb virulence plasmid pLVPK carrying rmpA2, iroBCD, and iucABCD genes by classic ST11 CRKP strains (Gu et al., 2018). A ∼210 Kb pLVPK-like virulence plasmid carrying rmpA2, iroBCD, and iucABCD genes was identified in two ST11 KPC-producing CRKP isolates in the present study, which expressed lower virulence compared to that of the ST11 CR-HvKP previously described but higher virulence than that of the CRKP isolates without virulence plasmids. Notably, a comparison of the two homologous isolates ZHW170 and ZHW183 revealed that ZWH183 carried an additional virulence plasmid (∼105 Kb, co-carrying the rmpA2 and iucABCD genes), which may be a glimpse of dynamic exchange of mobile elements in the gastrointestinal tract. Moreover, as was previously reported, CR-HvKP was found to account for only 3% of infections caused by the ST11 CRKP strains across China (Gu et al., 2018). Consistently, most of CRKP isolates in this study are identified with low virulence. K. pneumoniae has a wider ecological distribution, significantly more varied DNA

### REFERENCES


composition, greater antimicrobial resistance gene diversity and a higher plasmid load than other Gram negative opportunists, which means more opportunity to survive within and transfer between multiple environmental and animal-associated hosts; to capture plasmids from environmental microbial communities; to maintain antimicrobial resistance gene-carrying plasmids for long periods; and to transfer plasmids to other clinically important Gram negative bacteria (Wyres and Holt, 2018). Our finding showed that ST11 rmpA2-positive CRKP strains harbored three plasmids, of which two carried antimicrobial resistance genes and the third one was a virulence plasmid; rmpA2-negative CRKP strains only had two multidrug resistance encoding plasmids without virulence plasmid. It is suggested that the human gastrointestinal tract has a great capacity to colonize and enrich multiple clones of CRKP strains with varied plasmids as well as diverse antimicrobial resistance genes and virulence genes, which presumably due to frequent episodes of antibiotic treatment.

Colonization is believed as an important step in progression to extraintestinal infection which provides the rationale for establishing intervention measures to prevent subsequent infection by identifying the colonized patients. It is urgent matter to take the surveillance of rectal CRE carriage into routine test after admission.

### AUTHOR CONTRIBUTIONS

QS did strain characterization and participated in manuscript writing. DG did the whole-genome sequencing and participated in manuscript writing. QW, YH, and JH participated in collecting the clinical data and strain characterization. LS did the Galleria mellonella infection experiments. RZ participated in the research design, data interpretation, and manuscript writing. G-XC designed and supervised the study, interpreted the data and wrote the manuscript.

### FUNDING

This work was supported by the grants from National Natural Science Foundation of China (81601815, 81772250, and 81871705).


pneumoniae in hospitalised patients during a national outbreak. J. Hosp. Infect. 74, 344–349. doi: 10.1016/j.jhin.2009.07.022


**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2019 Sun, Gu, Wang, Hu, Shu, Hu, Zhang and Chen. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

fmicb-10-00230 February 7, 2019 Time: 18:22 # 9

# Antimicrobial Activity of Lactobacillus Species Against Carbapenem-Resistant Enterobacteriaceae

Chi-Chung Chen1,2, Chih-Cheng Lai<sup>3</sup> , Hui-Ling Huang<sup>4</sup> , Wen-Yu Huang<sup>4</sup> , Han-Siong Toh<sup>5</sup> , Tzu-Chieh Weng<sup>5</sup> , Yin-Ching Chuang1,6, Ying-Chen Lu<sup>2</sup> \* and Hung-Jen Tang4,6 \*

### Edited by:

Benjamin Andrew Evans, University of East Anglia, United Kingdom

### Reviewed by:

Alberto Finamore, Council for Agricultural Research and Economics, Italy Lisa Solieri, University of Modena and Reggio Emilia, Italy

### \*Correspondence:

Ying-Chen Lu biolyc2016@gmail.com Hung-Jen Tang 8409d1@gmail.com

### Specialty section:

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

Received: 28 November 2018 Accepted: 27 March 2019 Published: 18 April 2019

### Citation:

Chen C-C, Lai C-C, Huang H-L, Huang W-Y, Toh H-S, Weng T-C, Chuang Y-C, Lu Y-C and Tang H-J (2019) Antimicrobial Activity of Lactobacillus Species Against Carbapenem-Resistant Enterobacteriaceae. Front. Microbiol. 10:789. doi: 10.3389/fmicb.2019.00789 <sup>1</sup> Department of Medical Research, Chi Mei Medical Center, Tainan, Taiwan, <sup>2</sup> Department of Food Science, National Chiayi University, Chiayi, Taiwan, <sup>3</sup> Department of Intensive Care Medicine, Chi Mei Hospital, Tainan, Taiwan, <sup>4</sup> Department of Health and Nutrition, Chia Nan University of Pharmacy and Science, Tainan, Taiwan, <sup>5</sup> Department of Intensive Care Medicine, Chi Mei Medical Center, Tainan, Taiwan, <sup>6</sup> Department of Medicine, Chi Mei Medical Center, Tainan, Taiwan

Objective: This study aims to identify suitable lactobacilli that have anti-carbapenemresistant Enterobacteriaceae (CRE) activity with in vitro tolerance to pepsin and bile salts.

Methods: Fifty-seven Lactobacillus spp. strains encompassing nine species were collected for investigation. Their viabilities in the presence of pepsin and bile salts were tested using tolerance tests. Their anti-CRE effects were assessed by agar well diffusion and broth microdilution assay, as well as time-kill test.

Results: Of the 57 Lactobacillus isolates collected, 31 had a less than 2-log reduction in their viability in both pepsin and bile salt tolerance tests. Of these 31 isolates, 5 (LUC0180, LUC0219, LYC0289, LYC0413, and LYC1031) displayed the greatest anti-CRE activity with a CRE zone of inhibition greater than 15 mm in agar well diffusion assays. The minimal inhibitory percentages of supernatants from these five strains against CREs ranged from 10 to 30%. With the exception of LUC0180, which had a minimal bactericidal percentage ≥ 40%, the bactericidal percentage of all the strains ranged from 20 to 40%. The inhibitory effect of the cell-free culture supernatants from these Lactobacillus strains did not change after heating but was abolished as the pH changed to 7.0. After a 24-h incubation, five of the Lactobacillus strains at a concentration of 10<sup>8</sup> CFU/ml totally inhibited the growth of carbapenem-resistant Escherichia coli (CRE316) and Klebsiella pneumoniae (CRE632). After a 48-h incubation, the growth of CRE316 was completely inhibited under each concentration of lactobacilli based on time-kill test. Furthermore, when the concentration of lactobacilli was at 10<sup>8</sup> CFU/ml, the decline in pH was faster than at other concentrations.

Conclusion: Some Lactobacillus strains exhibit anti-CRE activity, which suggests potential applications for controlling or preventing CRE colonization or infection.

Keywords: Lactobacillus, carbapenem-resistant, Enterobacteriaceae, in vitro – antibacterial, activity

# INTRODUCTION

fmicb-10-00789 April 16, 2019 Time: 17:58 # 2

Although Enterobacteriaceae are normal flora of the human intestinal system, they are also common pathogens causing human infections in the setting of both community-acquired and healthcare-associated infections (Hsueh et al., 2010; Toh et al., 2012; Lai et al., 2014; Jean et al., 2016). In this era of widespread antibiotic resistance, Enterobacteriaceae are no exception. Recently, the emergence of carbapenem-resistant Enterobacteriaceae (CRE) has become a more critical issue due to the limited therapeutic options available for these pathogens and the significant morbidity and mortality associated with CRE infections (Tang et al., 2016a; Rodriguez-Bano et al., 2018). Therefore, there is an urgent need for new treatments for these critical CRE-associated conditions.

Lactobacillus is one of a number of probiotics considered to be biological therapeutics and host immune-modulating biologicals that are generally recognized as safe (GRAS). Recent studies demonstrated several antimicrobial mechanisms of Lactobacillus such as nutrient competition, production of inhibitory compounds, immune-stimulation and competition for binding sites. In addition, Lactobacillus can produce lactic acid, acetic acid, formic acid and other acids to reduce intestinal pH, which may be the most important mechanism. These bacteria can also secrete certain antimicrobial molecules, such as ethanol, fatty acid, hydrogen peroxide and bacteriocins to exert the antimicrobial activity (Georgieva et al., 2015; Inglin et al., 2015). Through these mechanisms, Lactobacillus has demonstrated its ability to inhibit several bacterial pathogens, including Clostridium difficile (McFarland, 2015), Escherichia coli (Kumar et al., 2016), Shigella spp. (Mirnejad et al., 2013), Streptococcus mutans (Ahn et al., 2018), Pseudomonas aeruginosa (Jamalifar et al., 2011), and Staphylococcus aureus (Kang et al., 2017). However, no previous studies have assessed the antimicrobial activity of Lactobacillus against CRE. Thus, we conducted this study to identify suitable lactobacilli that have anti-CRE activity with in vitro tolerance to pepsin and bile salts.

# MATERIALS AND METHODS

### Bacterial Strains and Culture Conditions

Fifty-seven Lactobacillus spp. strains encompassing nine species were obtained from Department of Food Science at the National Chiayi University in Chiayi, Taiwan. Species confirmation was performed by 16S rDNA sequencing. A fragment of the 16S rDNA was amplified by PCR. After amplification, the amplicons were separated by gel electrophoresis and sequenced. Sequences were compared with the NCBI GenBank database using the BLAST search tool to find the closest matches. The basic growth media for LAB were Man-Rogosa-Sharpe (MRS; Oxoid Inc., Ogdensburg, NY, United States).

Twenty clinical strains, including 10 different pulse field gel electrophoresis (PFGE) genotyped carbapenem-resistant Escherichia coli (CREC) and Klebsiella pneumoniae (CRKP) strains were isolated from Chi Mei Medical Center (Lai et al., 2016; Tang et al., 2016b). Species confirmations were performed using the VITEK 2 automated system (bioMérieux, Marcy l'Etoile, France) with standard biochemical methods. Mueller Hinton (MH) broth (Difco Laboratories, Detroit, MI, United States) were used for bacterial pathogens. The isolates were stored at −80◦C in Protect Bacterial Preservers (Technical Service Consultants Limited, Heywood, United Kingdom) before use.

Carbapenem susceptibility testing was performed using the disk diffusion method according to the Clinical and Laboratory Standards Institute (CLSI) guidelines (Clinical and Laboratory Standards Institute [CLSI], 2015). In brief, a 0.5 McFarland turbidity standard inoculum from overnight cultures was followed by incubation of Mueller-Hinton agar plates at 35◦C. The antibiotic disks were placed on the agar surface. After 16–18 h of incubation at 35◦C, results were interpreted as either sensitive, intermediate, or resistant according to the inhibitory zone diameters around the disks using CLSI breakpoints. Carbapenem resistance was defined as resistance to imipenem, meropenem, doripenem, or ertapenem.

### Pulsed-Field Gel Electrophoresis

Pulsed-field gel electrophoresis (PFGE) for the Escherichia coli and K. pneumoniae isolates were performed as described previously (Lai et al., 2016) with a CHEF DR II apparatus (Bio-Rad Laboratories, Hercules, CA, United States). Briefly, bacterial chromosomal DNAs were digested using XbaI (New England Biolabs, Beverly, MA, United States). Electrophoresis was carried out for 22 h at 14 µC, with pulse times ranging from 2 to 40 s at 6 V/cm, using a Bio-Rad CHEF MAPPER apparatus (Bio-Rad Laboratories, Richmond, CA, United States). A dendrogram based on the unweighted pair group was generated using methods previously described. The PFGE patterns were visually examined and interpreted according to the

TABLE 1 | Name and number of each species among the 57 total Lactobacillus isolates.



(1) SXT, only the MIC of trimethoprim is shown. (2) AMP: ampicillin; CHL, chloramphenicol; CLI, clindamycin; ERY, erythromycin; FA, fusidic acid; GM, gentamicin; KAN, kanamycin; LNZ, linezolid; STR, streptomycin; TEC, teicoplanin; SXT, trimethoprim/sulfamethoxazole (1/19); VAN, vancomycin.

TABLE 3 | Pepsin tolerance test results for 57 Lactobacillus isolates at pH 2.5 with 3.5 mg/ml pepsin.

TABLE 3 | Continued


(Continued)

profiles of each strain were compared using a Dice coefficient at 1.0% of tolerance and 1.0% of optimization. Isolates that

TABLE 4 | Bile salt tolerance test results for 57 Lactobacillus isolates at pH 7.3 with 0.3% Oxgall.


TABLE 4 | Continued


had < 80% similarity on the PFGE profiles were considered different types (**Figure 1**).

### Antimicrobial Susceptibility Test

The procedure for broth microdilution was adapted from the CLSI protocol for antimicrobial susceptibility testing (Clinical and Laboratory Standards Institute [CLSI], 2009). Antibiotic solutions were prepared in LAB susceptibility media (90% LSM and 10% MRS broth, adjusted to pH 6.7) (Klare et al., 2005). The following antibiotics were tested: ampicillin, chloramphenicol, clindamycin, erythromycin, fusidic acid, gentamycin, kanamycin, linezolid, streptomycin, teicoplanin, trimethoprim/sulfamethoxazole (1/19), and vancomycin (Sigma-Aldrich, St. Louis, MO, United States). The antibiotic susceptibility tests and their interpretations were carried out according to the CLSI guidelines (British Society for Antimicrobial Chemotherapy [BSAC], 2014; Clinical and Laboratory Standards Institute [CLSI], 2016).

### Pepsin Tolerance

A pepsin solution was prepared by suspending 3 mg/ml pepsin (Sigma-Aldrich, St. Louis, MO, United States) in a 0.85% sterile saline solution adjusted to pH 2.5. To test bacterial viability in the presence of pepsin, the 3 mg/ml pepsin solution, or saline at a pH of 7.0 as control, was inoculated with 10<sup>6</sup> CFU/ml of Lactobacillus spp. and then incubated at 37◦C for 4 h. The viable cell population was determined using the spread plate method. Each experiment was conducted in triplicate (Tokatli et al., 2015).

### Bile Salt Tolerance

The bile salt tolerance assay was performed as previously described (Jacobsen et al., 1999). In brief, each strain was adjusted to 1 × 106 CFU/ml in 0.3% Oxgall (Sigma-Aldrich, St. Louis, MO, United States) in PBS pH 7.3, or in PBS pH 7.3 alone as a control. Survival was tested at 37◦C after 24 h by the spread plate method. Each experiment was conducted in triplicate.

# Cell-Free Supernatant (CFS) Preparation

Lactobacilli cell-free supernatants were cultured in MRS broth at 37◦C for 24 h. The cultures were then centrifuged at 4000 rpm

(Continued)

at 4◦C for 30 min. The supernatants were sterilized by filtration through a 0.22 µm filter (Millipore, Billerica, MA, United States).

### Well Diffusion Assay

The agar well diffusion method (Tagg and McGiven, 1971) was modified to detect antimicrobial activities of supernatants isolated from Lactobacillus strains. First, MH agar plates were swabbed on the surface with CRE bacterial cultures. Then, 6 mm diameter wells were prepared and cell-free supernatants from isolated lactobacilli were loaded in the wells (100 µl/well). Following a 24-h incubation at 37◦C, inhibition zones were recorded.

### Broth Microdilution Assay

Broth microdilution assay was carried out as previously described (Arena et al., 2016) with some modifications. Overnight cultures of pathogenic bacteria were inoculated into fresh MHB media and seeded into 96-well plates (BD Discovery Labware, Bedford, MA, United States). The CFS of each Lactobacillus culture was separated into three aliquots: The first aliquot received no treatment, the second aliquot was heated at 80◦C for 10 min, and the third aliquot was neutralized to pH 7.0 with 1N NaOH. A 200 µl volume of test solution, consisting of 100 µl of the pathogenic bacterial culture (final inoculum was approximately 10<sup>6</sup> CFU/ml) and 100 µl of one of the CFS aliquots, was mixed into the wells. The CFSs were diluted with MRS broth and used at different percentages (i.e., 10, 20, 30, 40, 50%) in the final 200 µl volume. The minimum inhibitory percentage (MIP), defined as the lowest percentage of supernatant that can inhibit the growth of pathogen, was monitored by measuring optical density (OD600 nm). The minimum bactericidal percentage (MBP) was defined as the lowest percentage of CFS that can kill all the pathogenic bacteria, as detected by subculturing treated samples onto MH agar. All tests were done in triplication.

### Time-Kill Test in Co-cultures

Carbapenem-resistant Enterobacteriaceae and Lactobacillus strains were individually cultured in their respective broth

medium at 37◦C for 24 h. The cultures were centrifuged at 6000 rpm, 22◦C for 10 min to collect the cell pellet. Then, pathogenic bacteria were inoculated at 1 × 10<sup>6</sup> CFU/ml and lactobacilli at 1 × 10<sup>5</sup> , 1 × 10<sup>6</sup> , 1 × 10<sup>7</sup> , or 1 × 10<sup>8</sup> CFU/ml into mono-cultures or 1 × 10<sup>6</sup> CFU/ml pathogenic bacteria co-culture with 1 × 10<sup>5</sup> , 1 × 10<sup>6</sup> , 1 × 10<sup>7</sup> , or 1 × 10<sup>8</sup> CFU/ml

lactobacilli in tubes containing 10 ml of MRS-MH broth (1:1) (Drago et al., 1997). Mono-cultures and co-cultures were incubated at 37◦C for 48 h. Samples were collected at 0, 2, 4, 8, 24, and 48 h for the determination of viable cell count and pH measurements. A 1 ml aliquot of each sample was used to prepare serial dilutions that were poured onto the appropriate


TABLE 5 | The resulting pH and MIPs/MBPs of five Lactobacillus strain supernatants (%) against carbapenem-resistant E. coli (CREC) and K. pneumoniae (CRKP).

agar plates; MRS agar was used for Lactobacillus spp., while MacConkey Agar was used for Enterobacteriaceae. Plates were incubated at 37◦C for 24 h and colonies were counted. The assay detection limit was 100 CFU/ml (Shah et al., 2016). All tests were done in triplication.

### Statistical Analysis

The paired t-test was used for statistical analysis. The level of significance for all analysis was p < 0.05.

# RESULTS

### Microbiological Characteristics of Lactobacillus Isolates

A total of 57 Lactobacillus isolates, including L. plantarum (n = 17), L. paracasei (n = 16), L. rhamnosus (n = 8), L. fermentum (n = 6), L. casei (n = 4), L. brevis (n = 2), L. futsaii (n = 2), L. furfuricola (n = 1), and L. sakei (n = 1) were collected for this study (**Table 1**). Based on the findings from MIC testing of these Lactobacillus isolates, we determined that these isolates were highly susceptible to chloramphenicol, clindamycin, erythromycin, linezolid and streptomycin with susceptibility rates ranging from 91.2 to 100%. The resistance rates of these isolates against fusidic acid, kanamycin, trimethoprim/sulfamethoxazole, teicoplanin, and vancomycin ranged from 89.5 to 100% (**Table 2**).

### In vitro Viability of Lactobacillus Isolates

**Table 3** displays the results of the pepsin tolerance tests. The viability of 32 of the isolates was reduced by <2 log. Another 19 isolates were totally inhibited in these simulated conditions. **Table 4** shows the results of the bile salt tolerance tests. With the exception of LYC1164, which had a 2.1-log reduction in growth, all of the isolates had a less than a 2-log reduction in cell viability after incubation in this simulated intestinal condition for 24 h. Based on the findings of these in vitro viability tests, we chose 31 Lactobacillus isolates with less than 2-log reductions in viability in both pepsin and bile salt tolerance tests. Because of their ability to survive in these simulated gastric and intestinal environments, we selected these strains for the assessment their antibacterial activity.

### The Results of the Agar Well Diffusion Method

The 21 isolates displayed the greatest activity against carbapenem-resistant E. coli with zones of inhibition greater than 15 mm (**Figure 2A**). The greatest activity against carbapenem-resistant K. pneumoniae isolates, with zones of inhibition greater than or equal to 15 mm, were observed in the nine isolates (**Figure 2B**). Overall, we chose five Lactobacillus strains (LUC0180, LUC0219, LYC0289, LYC0413, and LYC1031), which had the significantly better antibacterial activities than the most of the other strains (p < 0.05) as determined by the agar well diffusion and time-kill assay for further tests of antibacterial activity.

### The Results of MIP/MBP Tests

**Table 5** shows the MIP and MBP for the five most potent Lactobacillus supernatants against CRE. The MIP of these five strains ranged from 10 to 30%. Except for LUC0180, which had an MBP of ≥40%, all of the strains had MBPs ranging from 20 – 40%. The inhibitory effects of these Lactobacillusstrains did not change after heating up to 80◦C for 10 min. However, the inhibitory effect disappeared once the pH increased to 7.0 (data not show). Additionally, the pH in the presence of LUC0180 seemed to be higher than the other four isolates.

### The Results of Time-Kill Test

**Figure 3** shows the results from time-killing test and assessment of the association between pH and antibacterial effects. After a 24-h incubation, there was no significant change between mono-cultures of lactobacilli and co-culture of lactobacilli and carbapenem-resistant E. coli (CRE316) in term of the concentration of lactobacilli (all p > 0.05) (**Figure 3A**). In contrast, the growth of CRE 316 was significantly inhibited after co-culture with different concentration lactobacilli after 24 h (all p < 0.05) and even 48 h (p < 0.0001) when comparing with mono-culture of CRE (**Figure 3B**). Furthermore, the decreases in pH were observed in the mono-cultures of lactobacilli,

FIGURE 3 | (A,E) The CFU change of five different concentration (10<sup>5</sup> to 10<sup>8</sup> CFU/ml) Lactobacillus isolates after mono-culture or co-culture with CRE316/ 632 for 48 h. (B,F) The CFU change of CRE316/ 632 after mono-culture or co-culture with five different concentration (10<sup>5</sup> to 10<sup>8</sup> CFU/ml) Lactobacillus isolates for 48 h and comparison between mono-culture and co-culture. (C,G) The pH changes of five different concentration (10<sup>5</sup> to 10<sup>8</sup> CFU/ml) Lactobacillus isolates mono-culture for 48 h. (D,H) The pH changes of CRE316/ 632 after mono-culture or co-culture with five different concentration (10<sup>5</sup> to 10<sup>8</sup> CFU/ml) Lactobacillus isolates and comparison between mono-culture and co-culture. (<sup>a</sup> Means the comparisons between 10<sup>8</sup> and 105,6,<sup>7</sup> CFU/ml lactobacillus isolates. <sup>b</sup> Means the comparisons between 10<sup>7</sup> and 105,<sup>6</sup> CFU/ml Lactobacillus isolates. ∗P-value < 0.05, #P-value < 0.001, &P-value < 0.0001).

especially at the concentration of 10<sup>8</sup> and 10<sup>7</sup> CFU/ml (both p < 0.05) (**Figure 3C**). The decrease in pH remains significant while in the co-cultures with CRE316 compared with CRE316 monoculture (all p < 0.05, **Figure 3D**). The similar findings were also noted for the co-culture with carbapenem-resistant K. pneumonia (CR632) (**Figures 3E–H**).

### DISCUSSION

In this study, five Lactobacillus strains, including three L. plantarum strains (LUC0219, LYC0289, LYC1031) and one each of L. paracaseri (LUC0180) and L. rhamnosus (LYC0413), exhibited good antibacterial activity against CREs. These inhibitory effects were demonstrated in through various tests including the agar well diffusion method, broth microdilution assay, and time-kill test. Lactobacillus spp. are non-pathogenic Gram-positive rods that are also recognized as normal human flora. Recently, several studies have reported that Lactobacillus strains can exhibit antibacterial activity through several mechanisms such as the production of antimicrobial substances or metabolites (hydrogen peroxide, lactic acid, bacteriocin), competition for nutrients, inhibition of bacterial adhesion to the mucosa, and enhancement of the immune response (Gill et al., 2001; Reid and Burton, 2002; Servin, 2004; Saulnier et al., 2009). Some in vitro studies have also shown that Lactobacillus strains can exhibit antimicrobial activity against C. difficile, E. coli, Shigella spp., S. mutans, P. aeruginosa, and S. aureus (Jamalifar et al., 2011; Mirnejad et al., 2013; Kumar et al., 2016; Kang et al., 2017; Ahn et al., 2018). To the best of our knowledge, this is the first study to document evidence of Lactobacillus antimicrobial activity against CRE. Although in vitro activity cannot be directly translated to in vivo effect, our findings suggest a promising role for Lactobacillus strains in the prevention and treatment of CRE colonization or infection. However, further animal studies are warranted to clarify this issue.

In the time-kill studies, the co-cultures with CREs did not influence the growth of lactobacilli. Additionally, effective inhibitory activity was generally observed at a pH of less than 4.2. These results were comparable with what we observed in our MIP/MBP experiments. By contrast, the antibacterial activity of Lactobacillus strains was dependent on an acidic environment and the inhibitory effect disappeared once the pH became greater than 6.5 during the MIP/MBP tests. Importantly, antimicrobial activity was not influenced by heating. This indicates that the inhibitory effect is mostly due to the acidic conditions, and not the production of bacteriocin-like substances. Similar findings were noted in the time-kill test. Previous reports have shown that the production of organic

### REFERENCES

Ahn, K. B., Baik, J. E., Park, O. J., Yun, C. H., and Han, S. H. (2018). Lactobacillus plantarum lipoteichoic acid inhibits biofilm formation of Streptococcus mutans. PLoS One 13:e0192694. doi: 10.1371/journal.pone.019 2694

acids by probiotic organisms and the resulting decrease in culture pH is considered to be the principal antimicrobial effect (Zhang et al., 2011; Tejero-Sarinena et al., 2012). In addition to lowering the culture pH, Alakomi et al. (2000) reported that organic acids could also function as outer membrane permeabilizers of some gram-negative bacteria and enhance the activity of other antimicrobial metabolites. These findings suggest that acidic pH and/or the presence of organic acids may be essential for the antibacterial activity observed in the present study.

In this study, only 31 of the 57 Lactobacillus strains that were initially collected for assessment were found to be able to survive in the simulated gastric and intestinal environment. For clinical application, these characteristics are extremely important. To be functional in the lower enteric tract, the tolerance of lactobacilli to pepsin, low pH and bile salts is essential.

This study had several limitations. First, in addition to the effect of acid environment, we did not investigate the detail mechanism or molecular effectors of these function traits. Second, we used separated tests instead of a unique trait by using in vitro GIT system to evaluate the GIT resistance in this study. Further study is warranted to clarify these issues.

### CONCLUSION

Several Lactobacillus strains exhibit antibacterial activity against CRE. We suggest that this effect may have potential applications through the use of Lactobacillus strains as starter cultures in fermented foods or as food preservatives for controlling or preventing CRE infections. Additional studies are required to determine the effects of complex nutrients on the synthesis of the antibacterial substance, as well as to elucidate the mechanisms and genetic basis of the bactericidal activity. Further animal models are also necessary to document the in vivo survival of these acid, pepsin and bile salt tolerant lactobacilli in the lower enteric tract. If the same inhibitory effect can be documented in vivo, further studies including vancomycin-resistant Enterococcus should be performed in the future.

### AUTHOR CONTRIBUTIONS

H-JT and Y-CL are the guarantors of this manuscript. C-CC, C-CL, H-LH, W-YH, H-ST, T-CW, and Y-CC contributed to the conception and design of the study. C-CC and H-JT analyzed and interpreted the data. C-CC, C-CL, and H-JT drafted the manuscript. All authors reviewed the manuscript.


food-borne pathogenic microorganisms. Front. Microbiol. 7:464. doi: 10.3389/ fmicb.2016.00464


**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2019 Chen, Lai, Huang, Huang, Toh, Weng, Chuang, Lu and Tang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

# High Prevalence of Metallo-β-Lactamase-Producing Enterobacter cloacae From Three Tertiary Hospitals in China

Yimei Cai 1†, Cha Chen2,3†, Mei Zhao4†, Xuegao Yu1†, Kai Lan2,3, Kang Liao<sup>1</sup> , Penghao Guo<sup>1</sup> , Weizheng Zhang2,3, Xingyan Ma<sup>1</sup> , Yuting He<sup>1</sup> , Jianming Zeng2,3 , Liang Chen<sup>5</sup> , Wei Jia<sup>4</sup> \*, Yi-Wei Tang6,7 \* and Bin Huang<sup>1</sup> \*

### Edited by:

Benjamin Andrew Evans, University of East Anglia, United Kingdom

### Reviewed by:

Andres Felipe Opazo-Capurro, Universidad de Concepción, Chile Wejdene Mansour, University of Sousse, Tunisia Abdelaziz Touati, University of Béjaïa, Algeria

### \*Correspondence:

Wei Jia 13519299090@126.com Yi-Wei Tang tangy@mskcc.org Bin Huang huangb3@mail.sysu.edu.cn

†These authors have contributed equally to this work

### Specialty section:

This article was submitted to Antimicrobials, Resistance and Chemotherapy, a section of the journal Frontiers in Microbiology

Received: 01 December 2018 Accepted: 27 June 2019 Published: 09 August 2019

### Citation:

Cai Y, Chen C, Zhao M, Yu X, Lan K, Liao K, Guo P, Zhang W, Ma X, He Y, Zeng J, Chen L, Jia W, Tang Y-W and Huang B (2019) High Prevalence of Metallo-β-Lactamase-Producing Enterobacter cloacae From Three Tertiary Hospitals in China. Front. Microbiol. 10:1610. doi: 10.3389/fmicb.2019.01610 <sup>1</sup> Department of Laboratory Medicine, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China, <sup>2</sup> Department of Laboratory Medicine, The Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, China, <sup>3</sup> Department of Laboratory Medicine, Guangdong Provincial Hospital of Chinese Medicine, Guangzhou, China, <sup>4</sup> Department of Laboratory Medicine, Ningxia Hospital of Ningxia Medical University, Yinchuan, China, <sup>5</sup> Public Health Research Institute Tuberculosis Center, New Jersey Medical School, Rutgers University, Newark, NJ, United States, <sup>6</sup> Department of Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, United States, <sup>7</sup> Department of Pathology and Laboratory Medicine, Weill Medical College of Cornell University, New York, NY, United States

Enterobacter cloacae has recently emerged as one of the most common carbapenem-resistant Enterobacteriaceae. The emergence and spread of metallo-β-lactamase-producing E. cloacae have posed an immediate threat globally. Here, we investigated the molecular characteristics of 84 carbapenem-resistant Enterobacter cloacae (CREL) collected from three tertiary hospitals in China between 2012 and 2016. Species identification and antimicrobial susceptibility testing were performed using a VITEK-2 system. Carbapenems, polymyxins B, and tigecycline were tested by broth microdilution method. The carbapenem in activation method (CIM) and cefoxitin three-dimensional test were used to detect carbapenemase and AmpC β-lactamase, respectively. Isolates were screened for β-lactam resistance genes by PCR, and expression of ompC and ompF was determined by qRT-PCR. Genetic relatedness was performed by pulsed-field gel electrophoresis (PFGE) and multilocus sequence typing (MLST), while selected isolates were subjected to whole-genome sequencing. Among the 84 CREL isolates, 50 (59.5%) were detected as carbapenemase producers. NDM-1 was the dominant carbapenemase (80.0%), followed by IMP-26 (8.0%) and IMP-4 (6.0%). Notably, we identified the first NDM-1 and IMP-1 co-producing E. cloacae, carrying plasmids of several incompatibility (Inc) groups, including IncHI2, IncHI2A, and IncN. Most isolates showed decreased expression of ompC and/or ompF, and contained a broad distribution of ESBLs and AmpC β-lactamases. These findings suggested that different molecular mechanisms, including carbapenemase, ESBL and/or AmpC plus loss of porins, have contributed to carbapenem resistance. The blaNDM−1-harboring plasmids contained highly conserved gene environment around blaNDM−<sup>1</sup> (blaNDM−1-bleMBL-trpF-dsbD-cutA1-groES-groEL), which could be associated with the potential dissemination of blaNDM−1. IMP-type MBL was located within a variety of integrons and usually contained various gene cassettes encoding multidrug resistance. These isolates produced 54 different pulsotypes, and were classified into 42 STs by MLST. Nineteen blaNDM−1-positive E. cloacae isolates obtained from Ningxia had the same pulsotype (PFGE type 1), belonging to ST78 within clonal complex 74 (CC74). The plasmid-based replicon typing indicated that IncX3 plasmids mediated the dissemination of blaNDM−<sup>1</sup> among these homologous strains. This is the first report on the outbreak of NDM-1-producing E. cloacae ST78 with contribution of IncX3 plasmids in Northwestern China. There's an immediate need to intensify surveillance attentively to prevent and control the further spread of NDM-1 in China.

Keywords: carbapenem-resistant, Enterobacter cloacae, outbreak investigation, NDM-1, IncX3 plasmids, ST78

### INTRODUCTION

Enterobacter cloacae, an opportunistic pathogen ranking the third among all Enterobacteriaceae in healthcare-associated infections, may cause various nosocomial infections involving urinary tract, lower respiratory tract, skin and soft tissues, biliary tract, intravenous catheters, and central nervous system (Davin-Regli and Pagès, 2015). E. cloacae is intrinsically resistant to ampicillin, amoxicillin/clavulanic, cephamycin, and the 1st and 2nd generation cephalosporins owing to chromosomally encoded AmpC β-lactamase (Jacoby, 2009). Recently, severe comorbid conditions, extensive invasive procedures, and heavy exposure to antibiotics have involved in the global emergence of carbapenem-resistant E. cloacae (CREL). Carbapenems have increasingly been used as a common drug for the treatment of nosocomial infections. According to CHINET, one of the largest antimicrobial resistance surveillance networks in China, carbapenem resistance rates among E. cloacae were <1.0% in 2007. Whereas, in the year of 2017, the resistance rates to imipenem, meropenem, and ertapenem have rapidly increased to 6.9, 7.0, and 8.2%, respectively. Infections caused by CREL usually result in higher mortality rates, longer hospitalization and higher costs of treatment, which has posed an immediate threat to public health (Kelly et al., 2017).

Resistance to carbapenems is associated with several mechanisms. Carbapenemase, which is largely responsible for carbapenem resistance in Enterobacteriaceae, has been classified into three functional groups: class A (mostly KPC, GES), class B metallo-β-lactamase (MBL, mostly IMP, VIM and NDM) as well as class D OXA-type β-lactamase (mostly OXA-48-like) (Tzouvelekis et al., 2012). Genes encoding MBLs are most commonly identified in E. cloacae and often carried on mobile genetic elements such as plasmids and transposons. Integrons located on such mobile elements play a crucial role in the horizontal transfer of MBL genes between bacteria (Villa et al., 2014; Lee et al., 2017a). In China, carbapenemase production is largely attributed to MBLs such as VIM-1 (Yang et al., 2014), IMP-4 (Wang et al., 2017), IMP-8 (Yan et al., 2002), and NDM-1 (Dai et al., 2013), among which NDM-1 was even more worrisome for its sharp increase. Since the first case of carbapenem-resistant E. cloacae harboring NDM-1 was detected in Chongqing (Wang et al., 2017), NDM-1-producing E. cloacae have emerged in various regions across the country (Jia et al., 2018; Jin et al., 2018). Moreover, moderate- to high-level carbapenem resistance in most isolates was closely related to an additional mechanism of resistance, e.g., decreased porin permeability (Majewski et al., 2016). Alteration or loss of nonspecific porins, coupled with ESBL and/or AmpC overexpression is considered as one of the main mechanisms of resistance (Wozniak et al., 2012).

However, there has been little epidemiological data on carbapenem-resistant E. cloacae (CREL) in certain regions in China, such as Southern (e.g., Guangdong) and Northwestern (e.g., Ningxia) China. To gain insights into the evolution of CREL isolates in these two regions, we conducted a molecular epidemiological study to describe the resistance mechanism to carbapenems, clonal relatedness, and the genetic environment of carbapenemases (NDM-1 or IMP) -encoding plasmids among CREL isolates.

### MATERIALS AND METHODS

### Bacterial Strains and Characterization

From 2012 to 2016, a total of 84 non-duplicate strains from patients infected by CREL were collected from three tertiary hospitals including hospital A (First Affiliated Hospital of Sun Yat-sen University, Guangdong, China, 25 strains), hospital B (Guangdong Provincial Hospital of Chinese Medicine, Guangdong, China, 18 strains), and hospital C (Ningxia Hospital of Ningxia Medical University, Ningxia, China, 41 strains). Species identification and initial antibiotic susceptibility were analyzed by a VITEK-2 system (bioMerieux, France). Mueller-Hinton broth (Oxoid, UK) supplemented with calcium and magnesium (25.0 mg/L Ca2<sup>+</sup> and 12.5 mg/L Mg2+) was used to test five antibiotics, including imipenem (ApexBio, USA), meropenem (ApexBio, USA), ertapenem (Menlunbio, China), polymyxin B (Sigma, USA), and tigecycline (ApexBio, USA). Minimum inhibitory concentrations (MICs) were performed according to the criteria of the Clinical and Laboratory Standards Institute (CLSI) (M100-S27). Isolates with an MIC of 4µg/mL were considered resistant to polymyxin B. A susceptibility breakpoint of <2µg/mL for tigecycline was interpreted in accordance with the European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines. Carbapenemase production was assessed by the carbapenem in activation method (CIM) (Aguirre-Quiñonero et al., 2017), while AmpC βlactamase production was examined by three-dimensional test. Strains used in quality control were Escherichia coli ATCC 25922 and Enterobacter cloacae ATCC 700323.

### Genotype Analysis

Crude DNA extracts prepared by boiling method were used as template in polymerase chain reactions (PCR). Various β-lactam resistance genes were examined, including carbapenemase genes (blaKPC, blaNDM, blaIMP, blaVIM, and blaOXA−48), ESBL-related genes (blaTEM, blaCTX−M, and blaSHV) and AmpC-β-lactamase genes (blaEBC, blaMOX, blaCIT, and blaDHA). Moreover, mcr-1 colistin resistance gene and tetX tigecycline resistance gene were also determined by using primers as described previously (Pérez-Pérez and Hanson, 2002; Pagani et al., 2003; Biendo et al., 2008; Poirel et al., 2011; Liu et al., 2016). Positive PCR products were sequenced and analyzed by nucleotide homology comparison against GenBank database by BLAST (www.ncbi.nlm.nih.gov/ blast/). Details about oligonucleotides and thermal conditions are presented in **Supplementary Table 1**.

### Outer Membrane Protein Gene Expressions

Expression of genes encoding outer membrane proteins (OmpF and OmpC porins) was determined using real-time reverse transcription PCR. The total RNA was extracted using the Trizol method and reversely transcribed into cDNA using the PrimeScriptTM RT reagent Kit with gDNA Eraser (TAKARA, China). Real-time quantitative PCR was performed using SYBR <sup>R</sup> Green I assay with analysis of dissociation curve (2× SYBR Green q-PCR Master Mix, Biomake, USA) on Applied Biosystems ViiATM 7 Dx (Life Technologies, USA). Each experiment was performed in triplicate. The expression of porin-encoding genes, ompF and ompC, relative to rpoB was determined using specific primers and conditions described previously (Majewski et al., 2016). Relative expression was derived from the 2-1(1CT) formula, in which 1CT represented the difference of cycle threshold (CT) between target genes and rpoB, while 1(1CT) represented the difference of 1CT between CREL strains and a standard strain, Enterobacter cloacae ATCC 700323.

### Genetic Homologeity Analysis

All isolates were analyzed by PFGE to determine their genetic relatedness. In brief, the bacterial suspension mixed with equal volumes of low-melting-temperature agarose was cleaved with 1% sodium dodecyl sulfate and proteinase K (Sigma, USA) at 54◦C overnight, and then digested with restriction endonuclease Xba I (Takara, Dalian, China) at 37◦C for 8 h. DNA separation was performed in 0.5× TBE buffer using a pulsed-field electrophoresis system (CHEF MAPPER; Bio-Rad Laboratories, California, USA) under the following conditions: temperature 14◦C; voltage 6.0 V/cm; switch angle, 120◦ ; and switch ramp of 2.16–54.17 s for 18 h. Comparison of pulsotypes was performed with Bionumerics software version 6.6. Isolates were allocated into genetic similarity clusters using an 80% cutoff value. MLST was performed as described previously (https:// pubmlst.org/ecloacae/). New alleles and sequence types were submitted to the MLST website and approved (https://pubmlst. org/ecloacae/). EnteroBase together with GrapeTree was used to analyze population evolutionary relationship (https://enterobase. readthedocs.io/en/latest/index.html).

### Whole-Genome Sequencing

Six representative strains producing single NDM-1, six strains producing single IMP (according to clonal relatedness), one strain co-producing NDM-1 and IMP-1, and eight strains negative for all β-lactamase genes were selected for wholegenome sequencing. Genomic DNA was extracted using a MiniBEST Bacteria Genomic DNA Extraction Kit (TaKaRa, Dalian, China). DNA library was prepared using a QIAseq FX DNA Library Kit (Qiagen Inc., Valencia, CA) following the manufacturer's recommendations. The quantity and quality of the libraries were assessed with a Qubit dsDNA HS Assay Kit (Life technologies, USA) and LabChip GX Analyzer (Perkin Elmer; Waltham, MA). All barcoded libraries were pooled together in equimolar amounts and sequenced on NextSeq 500 platform (illumina Inc., San Diego, CA). Sequencing raw reads were processed for library adapter removal and filtering using FASTQ preprocessor Fastp v0.12.5 (Chen et al., 2018), followed by de novo assembly with SPAdes v3.13.0 (Bankevich et al., 2012) and annotating using Prokka annotation pipeline (Seemann, 2014). Antimicrobial resistance genes and plasmidtyping identification were mined using ABRicate program (https://github.com/tseemann/abricate, v0.8.2), and the contig containing carbapenemase genes, such as blaNDM and blaIMP, were extracted with an in-house Python script, followed by BLAST against NCBI sequence database. The sequence containing blaIMP of the variable regions of integrons were analyzed followed by the Integron Database (http://integrall. bio.ua.pt/). Genetic organization analysis was visualized using EasyFig v2.2.3 (Sullivan et al., 2011).

### Statistical Analysis

All the statistical analyses were performed by SPSS 18.0 (IBM Corp., Armonk, USA). The chi-square test was applied to evaluate difference in antibiotic resistance between carbapenemase-positive and carbapenemase-negative subpopulations. Relative changes in gene expression were indicated with median and extremum. All the relevant data were analyzed by non-parametric rank sum test, and was considered statistically significant if P < 0.05.

### Ethical Considerations

This study was approved by Institutional Review Board of Second Affiliated Hospital of Soochow University. The study was retrospective and patients were not identified during data collection. Informed consent was not needed for this study.

# RESULT

### Characteristics of Clinical CREL Isolates

Eighty-four clinical CREL isolates were found to be resistant to at least one carbapenem. These isolates were recovered from different sources: body fluids (n = 40), sputum (n = 19), urine (n = 8), wound secretion (n = 5), blood (n = 5), pus (n = 2), catheter (n = 2), cannula (n = 1), semen (n = 1), and tissue (n = 1). These CREL isolates were obtained from patients admitted to the hepatobiliary surgery (n = 25, 29.8%), intensive care unit (n = 17, 20.2%), general surgery (n = 11, 13.1%), neurosurgery


S, susceptible; I, intermediate; R, resistant. IPM, imipenem; MEM, meropenem; ETP, ertapenem; CAO, cefatriaxone; CAZ, cefepime; TZP, piperacillin/tazobactam; FEP, cefepime; ATM, aztreonam; CIP, ciprofloxacin; LEV, levofloxacin; GEN, gentamicin; TOB, tobramycin; AMK, amikacin; SXT, sulfamethoxazolen; NIT, furazolidin; PB, polymyxin B; TGC, tigecycline. Hospital A, First Affiliated Hospital of Sun Yat-sen University, Guangdong, China; Hospital B, Guangdong Provincial Hospital of Chinese Medicine, Guangdong, China; Hospital C, Ningxia Hospital of Ningxia Medical University, Ningxia, China. \*Two isolates obtained from Hospital B were suspected of heteroresistance to polymyxin B.

TABLE 2 | In vitro activities of carbapenems to 84 CREL isolates with or without carbapenemase.


IPM, imipenem; MEM, meropenem; ETP, ertapenem. R, resistant. \*P-value for comparisons of resistance rates between carbapenemase-positive and carbapenemase-negative groups.

(n = 7, 8.3%), pediatrics (n = 4, 4.8%), burns surgery (n = 5, 6.0%), respiratory department (n = 3, 3.6%), gastroenterological surgery (n = 2, 2.4%), cardiology department (n = 2, 2.4%), department of orthopedics (n = 2, 2.4%), nephrology department (n = 1, 1.2%), and other surgery wards (n = 5, 6.0%). Among 84 CREL isolates, 85.7% (72/84) of the isolates were classified as MDR as they were resistant to three or more classes of antibiotics. Seventy-two (85.7%) isolates were resistant to ertapenem, followed by 57 (67.9%) and 54 (64.3%) resistant to imipenem and meropenem, respectively. The strains showed high resistance rates to cefatriaxone (83.3%), ceftazidime (75.0%), cefepime (72.6%), and aztreonam (76.2%), while 43 (51.2%), 51 (60.7%), and 27 (32.1%) strains were resistant to fluoroquinolones, sulfanilamide, and aminoglycosides, respectively. In contrast, 94.0 and 86.9% were susceptible to amikacin and tigecycline. Thirteen E. cloacae isolates were resistant to polymyxin B and two isolates were suspected of heteroresistance. Two E. cloacae isolates (ECL-ZY07 and ECL-ZY14) with resistance to polymyxin B and tigecycline showed promising in-vitro activity against amikacin. Carbapenemase producers presented a dramatically higher resistance rate than negative ones for carbapenemase (P < 0.01). Descriptive statistics on antimicrobial susceptibility tests were presented in **Tables 1**, **2**.

### Phenotype and Genotype Analysis

Fifty-four isolates were considered positive in the CIM test according to Tijet et al. (2016), but only 50 (59.5%) were demonstrated as carbapenemase producers by PCR analysis. All KPC, NDM, VIM, or IMP producing isolates were unequivocally detected with the CIM test. Four isolates non- producing carbapenemase (one with blaTEM−1, blaSHV−12, blaCTX−M, and blaDHA−1, three hyperproducing AmpC) were positive by the test. The sensitivity and specificity of the detection assay compared with molecular methods were 100 and 89.5%. The blaNDM−<sup>1</sup> gene was found the most prevalent carbapenemase gene (40/50), followed by blaIMP−<sup>26</sup> (4/50), blaIMP−<sup>4</sup> (3/50), blaIMP−<sup>1</sup> (1/50), blaVIM−<sup>4</sup> (1/50), and blaKPC−<sup>2</sup> (1/50). We observed one E. cloacae isolate co-harboring blaNDM−<sup>1</sup> and blaIMP−1, which resulted in strongly high MICs of all carbapenems (MICs = 256 µg/L). Remarkably, 97.6% (40/41) of the isolates obtained from Ningxia were MBL producers, but only 20.9% (9/43) produced MBL in Guangdong. No carbapenemase gene was detected in hospital B (Guangdong Provincial Hospital of Chinese Medicine, Guangdong, China). Moreover, 83.3% of the isolates continuously produced high AmpC β-lactamase, and 10.7% were pAmpC producers (eight with blaDHA−1, and one with blaLAP−2). All carried chromosomal-mediated ACT-type βlactamase. ESBL producers were found in 36 (42.9%) isolates, among which 25 isolates overproduced AmpC. 19.0% (16/84), 15.5% (13/84), 26.2% (22/84), and 23.8% (20/84) of the isolates were detected as positive for blaTEM−30, blaTEM−1, blaCTX−M, and blaSHV−12, respectively. Meanwhile, 31.0% (26/84) of the isolates co-expressed carbapenemase and ESBL, and 7.1% (6/84) produced carbapenemase, ESBL and pAmpC simultaneously. None of the isolates were PCR-positive for blaOXA−48, mcr-1 or tetX.

### Expression of ompC and ompF Genes

Relative expression of ompF gene fell within a range of 0.01–1.58, with a median value of 0.12, while expression of ompC gene ranged from 0.01 to 36.36, with a median value of 2.80. Thirty-two isolates lost or had lower expression of both major porins, while 52 isolates had decreased expression of one porin, especially for OmpF porin (50/52). Isolates with a combination of carbapenemase and porin loss showed extensively high carbapenem MICs. All 34 non-carbapenemaseproducing isolates had lost at least one porin, among which 22 isolates had reduced expression of both major porins. Additionally, 7 isolates produced both ESBL and AmpC, and 24 isolates produced only one type of enzyme. The remaining 3 isolates were negative for ESBLs (TEM, SHV, or CTX-M) and AmpC. Porin loss together with AmpC overexpression typically resulted in insensitivity to one or two carbapenems. Compared with isolates non-susceptible to three carbapenems and isolates only non-susceptible to one or two carbapenems, the latter had significantly lower expression of ompC (P < 0.01), but there was no difference in ompF expression among two groups (P = 0.398). Further, lower expression of ompC was significantly associated with increasing MIC of imipenem (P < 0.01). All details about antibiotic susceptibilities and porin expression of CREL isolates were given in **Table 3**.

### PFGE Analysis

Pulsed-field gel electrophoresis patterns of the 84 CREL isolates were identified into 54 pulsotypes (clusters 1–54), and of these, PFGE types 2, and 3 could be classified into 3 subtypes, while types 4, 5, and 7 were classified into 2 subtypes (**Figure 1**). PFGE revealed genetic diversity, whereas PFGE types 1, 2, 3, 4, 5, 6, and 7 were observed in 19, 6, 3, 2, 2, 2, and 2 different patients, respectively. Notably, all the 19 strains with a single dominant pulsotype (PFGE type 1) were isolated from patients in hepatobiliary surgery (17/19), intensive care unit (1/19), and vascular surgery (1/19) during a detection peak period (from Jan. 2016 to Sep. 2016) in hospital C. As shown in **Figure 2**, the strain ECL-NX38 which was isolated from a 78-year-old patient in hepatobiliary surgery on January 4, 2016 could be the first strain of these outbreak cases. Besides, three strains of type 3 were isolated from patients in two different wards in hospital C during 1-year period. All the six strains of PFGE type 2 and two strains of type 6 were distributed sporadically in different wards in hospital C. All details about clinical characteristics and PFGE patterns of 41 CREL isolates from hospital C was shown in **Table 4**. However, all pulsotypes were single isolates in hospital B. Two strains of PFGE type 4 were isolated from patients in ICU in hospital A within 1 year; two strains of type 5 were isolated from patients in two different wards within 2 months; while two strains of type 7 were isolated from patients in hepatobiliary surgery within 1 month (data not shown).

### MLST Analysis

MLST distinguished 42 STs including 22 novel STs (ST1211- ST1229, ST1231, ST1232, and ST1234). The most prevalent ST was ST78 (n = 27, 32.1%), followed by ST51 (n = 7, 8.3%), ST1226 (n = 5, 6.0%), ST114 (n = 4, 4.8%), ST93 (n = 2, 2.4%), ST346 (n = 2, 2.4%), and ST1044 (n = 2, 2.4%). The remaining STs contained one isolate for each. MLST showed subtype diversity, which was relatively consistent with PFGE. As expected, PFGE was more discriminatory than MLST. Most isolates belonging to different subtypes of the same pulsotype had an identical ST. The only exceptions were one single-locus variant pairs (SLVs) (ST114 and ST1229) and five SLVs (ST51 and ST778), among which isolates of one ST were clustered into one of the pulsotypes of the other ST. Of the 42 STs, 33 (78.6%) were singletons, 6 (14.3%) had single SLVs and 3 (7.1%) clustered into a bigger group within CC114. The ST114 centered group encompassed 6 isolates of 3 STs including ST114 (n = 4), ST1229 (n = 1), and ST1231 (n = 1), and isolates were all obtained from hospital A. Twenty-eight isolates from hospital A and C belonged to ST78 (n = 27) and ST1221 (n = 1) within CC74, while eight isolates represented ST51 (n = 7) and ST50 (n = 1) within CC234. Other single SLVs consisted of ST1044 and ST1228, indicating direct clonal evolution between the SLVs. All of the STs were analyzed by Grapetree in order to reveal their relationships (**Figure 3**). The ST78 clones of eight PFGE pulsotypes exhibited various β-lactamase profiles. The most prevailing type harboring NDM-1 belonged to ST78 and carried plasmids of two or three Inc types, including IncFIB, IncX3, and IncX4 (**Supplementary Table 2**). Plasmidbased replicon typing revealed that the IncX3 type epidemic plasmid carrying blaNDM−<sup>1</sup> caused an outbreak of E. cloacae. These dominated clones carried several other β-lactamase genes, including blaACT−<sup>5</sup> and blaOXA−1. However, ST78 E.cloacae isolates were not always associated with NDM-1. Three ST78 strains from hospital A produced different ESBLs but not NDM-1, and one ST78 E.cloacae from hospital C produced IMP-4. Each of these comprised various Inc types of plasmids. Six ST51 isolates with highly similar pulsotypes produced NDM-1 type MBL, and four isolates belonged to ST1226 diversified into three pulsotypes, were frequently related to IMP-26. Due to the

### TABLE 3 | Drug-resistance characteristics of 84 CREL isolates.


(Continued)

### TABLE 3 | Continued


IPM, imipenem; MEM, meropenem; ETP, ertapenem. Arrows indicated the relative variation of expression in ompC and ompF genes. AmpC β-lactamase was examined by cefoxitin in three-dimensional test. The plus and minus were used to describe the positive or negative results of β-lactamases. Arrows represents up- and down- regulation of OM expression.

deficiency of clinic case and limited data, all these observations were still to be further validated.

### Genetic Environments of NDM-1 or IMP Carrying Contigs

As shown in **Figure 4**, seven NDM-1-harboring plasmids had a common genetic structure, which was highly similar to some previous plasmids, such as pNDM-ECN49 (GenBank Accession No. KP765744). Compared with another epidemic plasmid pNDM-BJ01 in A. lwoffii (GenBank Accession No. JQ001791), however, a complete or partial IS5 supplanted the ISAba125 upstream of the blaNDM−<sup>1</sup> gene. The blaNDM−1-like-carrying plasmids could be divided into two different types. Type A plasmids mainly differed by the insertion of a Tn3-like transposon downstream of groES. The Tn3-like transposon consisted of tnpA, tnpR, and two ORFs encoding hypothetical proteins, flanked by a pair of 38 bp inverted repeats (IR). The inverted repeat motifs of target site 6 bp at the boundaries of the Tn3-like element indicated insertion by transposition. However, IS91-like element, which probably transpose via a rolling-circle replication mechanism, was located downstream of groES among two plasmids of Type B.

All of the blaIMP genes were situated within class 1 integrons. The blaIMP−<sup>8</sup> gene was identified in In655 containing


FIGURE 1 | Dendrogram of the PFGE profiles of CREL isolates. The Dice coefficient was used to identify different types with a cutoff of 80.0% similarity. Hospital A, First Affiliated Hospital of Sun Yat-sen University, Guangdong, China; Hospital B, Guangdong Provincial Hospital of Chinese Medicine, Guangdong, China; Hospital C, Ningxia Hospital of Ningxia Medical University, Ningxia, China.

blaIMP−8-aacA4′ -3, which was previously detected in E. cloacae (GenBank Accession No. MF344574). Three genetic mutations in coding region of intl1 (A38978C, G38958C, and T38953C) were observed in the integron mediated IMP-8. The

integron harboring blaIMP−<sup>26</sup> contained the gene cassette array: blaIMP−26-ereA2–qacE11/sul1, and was detected in 3 isolates. The integron had a structurally similar cassette to pIMP26 in E. cloacae (GenBank Accession No. MH399264), but the difference in ereA2 gene insertion resulted in erythromycin resistance. Integrons of blaIMP−<sup>4</sup> gene had a similar genetic structure to pAR0161 (GenBank Accession No. MF344574), and 3′ -CS qacE11 -sul1 were also observed downstream. However, we were unable to sequence the complete integron-related gene cassettes due to the limitations of short-read sequencing. The blaIMP−<sup>1</sup> gene was determined in a shorter contig accompanied with a partial aacA4 gene downstream. The sequence alignment revealed that one synonymous mutation (T39C) were observed, compared with the widely reported sequence of blaIMP−<sup>1</sup> gene (GenBank Accession No. MG287118). This resultant sequence was almost completely consistent with the partial published sequence in K. oxytoca p7121-IMP (GenBank Accession No. KX784502.1).

### DISCUSSION

E. cloacae is frequently implicated in serious healthcareassociated infections, with a high proportion of carbapenemase

TABLE 4 | PFGE patterns and clinical characteristics of 41 CREL isolates obtained from Hospital C.


bi, bile; su, drainage; sp, sputum; ur, urine; se, secretion; ps, pus; ca, catheter; bl, blood.

producers among carbapenem-resistant Enterobacteriaceae (Davin-Regli and Pagès, 2015). E. cloacae has been reported to acquire carbapenemases including KPC-2 (Andrade et al., 2018), NDM-1 (Torres-González et al., 2015), IMP-4 (Lee et al., 2017b), IMP-26 (Dai et al., 2013), IMP-8 (Pang et al., 2016), IMP-1 (Aoki et al., 2018), and VIM-4 (Sonnevend et al., 2017). Our study demonstrated the high prevalence of carbapenemase among carbapenem-resistant E. cloacae (CREL) in south and northwest of China (59.5%). MBLs, especially NDM-1 was, the dominant mechanism of resistance to carbapenems in E. cloacae in China. Most cases of infections or colonizations of NDM-1-producing bacteria originated in southern Asia, most commonly in India (Walsh and Toleman, 2011). The different prevalence of NDM-1 between countries might be explained by its high genetic transfer rate among unrelated bacterial species. Meanwhile, IMP-type MBL has been identified as the second common carbapenemase (10.7%). Compared to the worldwide distribution of IMP-4 (Leung et al., 2013; Lee et al., 2017b) and IMP-8 (Wang et al., 2015; Pang et al., 2016), IMP-26 and IMP-1 were also detected in our study. Although KPC or VIM-producing E. cloacae has been reported to spread rapidly in the last decade (Kanamori et al., 2017; Sonnevend et al., 2017; Daniels et al., 2018), the prevalence remained relatively low in our study. Moreover, the carriage rate of ESBL-related genes among CREL isolates was high (42.9%), though no dominant ESBL type was identified. Our study showed that 83.3% of the isolates continuously produced high AmpC β-lactamase. ACT-type inducible AmpC enzyme remained conservative. DHA-type, however which has become the most prevalent plasmid-mediated AmpC β-lactamase, was less frequently described in the literature with regard to E. cloacae. These additional β-lactamases might explain a slight increase of carbapenem resistance among CREL isolates (Souna et al., 2014). Of note, two MBLs, NDM-1 and IMP-1, were firstly identified to be co-expressed in E. cloacae, which was isolated

from a patient after cardiac surgery in Ningxia. This XDR E. cloacae isolate co-harbored blaSHV−12, blaTEM−30, blaLAP−2, blaACT−5, qnrS1, aac(6')-IIc, aadA6, aph(3')-Ia, aph(6)-Id, catA2, sul1, and dfrA10. Resistance genes existed in multiple incompatibility groups, including IncN, IncHI2A, and IncHI2, which could further aggravate the dissemination under the selective pressure of antibiotics (**Supplementary Table 2**).

Besides, the absence or reduced expression of porins coupled with ESBL and/or AmpC overexpression, could lead to increased carbapenem MICs (Babouee Flury et al., 2016). In our study, most isolates had decreased membrane permeability caused by low expression of ompC or ompF, or both. Most isolates exhibited down-regulation of ompF, but up-regulation of ompC. These results were consistent with the previous reports and indicated that the decreased expression of ompF might lead to the potential role of OmpC-directed OM protein polarization (Majewski et al., 2016). Combinations of either ESBL or AmpC and porin loss played a pivotal role especially among carbapenemase-negative isolates. These isolates had lost at least one porin, among which 91.2% (31/34) of the ioslates produced different types of β-lactamase. Analysis of ompC expression showed a significant decrease among isolates only non-susceptible to imipenem. The decrease or loss of OmpC, a smaller pore size membrane porin compared to that of OmpF, might be correlated with the slightly increase in imipenem MIC (Lavigne et al., 2012). Hence, porin loss combined with ESBL and/or AmpC overexpression was one of the major mechanisms of drug resistance. Carbapenemase, ESBL and/or AmpC hyperproduction plus porin loss usually resulted in extensively high carbapenem MICs.

In this study, we observed significantly differential distribution and molecular characteristics of CREL isolates in two geological regions. In Ningxia, China, MBLs, especially NDM-1, was identified in the highest proportion, indicating a great potential risk of spread of drug-resistant strains in northwest of China. However, lower prevalence of carbapenemase producers has been described in Guangdong, China during our investigation, and that approximately half of isolates showed intermediate resistance to one or two carbapenems. This phenomenon might be mainly due to additional β-lactamases and impaired permeability. No carbapenemase producer was found in hospital B (Guangdong Provincial Hospital of Chinese Medicine, Guangdong, China). The possible interpretation was that carbapenems might be utilized less frequently in these hospitals. In most cases, patients with infections caused by CREL usually suffer from critical basic diseases, lower immune function, and all kinds of invasive operations, and they may be less inclined to TCM hospitals. Another explanation, probably different genetic events among E. cloacae isolates have occurred over time in TCM hospitals and carbapenemase genes was very rare or unique. Besides, about other mechanisms, especially high level of AmpC β-lactamases plus porin loss, and overexpression of efflux pumps, would play a more important part in carbapenem resistance among these E. cloacae isolates. Therefore, a variety of mechanisms could induce resistance in varying degrees in the process of drug resistance in E. cloacae.

The most susceptible antimicrobials to CREL isolates was amikacin (94.0%), followed by tigecycline (86.9%), and polymyxin B (82.1%). Amikacin, an aminoglycoside group antibiotic, has been restricted in recent years due to its side effects such as nephrotoxicity and ototoxicity (Wargo and Edwards, 2014). Polymyxin B and tigecycline are currently the most active therapeutic option for carbapenem-resistant Enterobacteriaceae including CREL (Rafailidis and Falagas, 2014). However, polymyxin B and tigecycline resistance has sporadically occurred in recent years (Kumar, 2016; Karaiskos et al., 2017). We found 13 isolates resistant to polymyxin B and 11 isolates non-susceptible to tigecycline, but mcr-1 and tetX were not detected. Resistance to polymyxins might be associated with alterations in the lipopolysaccharide structure and overexpression of efflux pumps (Lim et al., 2010). Besides, ribosome protection and excessive expression of efflux pumps might contribute to decreased susceptibility to tigecycline (Pournaras et al., 2016). Remarkably, polymyxin B heterogeneous subpopulations were detected in two isolates, indicating that more consideration should be taken to heterogeneously colistin resistant E. cloacae.

Our results suggested that NDM-1-harboring plasmids contained a highly conserved region around blaNDM−<sup>1</sup> (blaNDM−1-bleMBL-trpF-dsbD-cutA1-groES-groEL), which might be involved in the further spread of blaNDM−1. The highly similar genetic structure was also obtained from various NDM-1-encoding plasmids, such as C. freudii from China (Yang et al., 2015) and K. pneumoniae from Australia (Wailan et al., 2016). More importantly, the region flanking blaNDM−<sup>1</sup> was highly homologous to some Acinetobacter spp. isolated from China (Bogaerts et al., 2013). Recent studies proposed that the acquisition of blaNDM−<sup>1</sup> in Enterobacteriaceae might be derived from Acinetobacter spp. via horizontal transmission of drug-resistant plasmids (Bogaerts et al., 2013). These findings suggested that the potential dissemination of acquired blaNDM−<sup>1</sup> gene among different bacterial species had become an emerging global threat. IMP-type MBL was located within a variety of integrons. Similar to previous studies, IMP-producing E. cloacae presented a diverse genomic environment due to the transfer and rearrangement of plasmids and integrons in all probability (Aoki et al., 2018). Different genetic events might occur among E. cloacae isolates and other isolates, leading to the diversity of integrons in our study. The presence of various integrons associated with plasmids could facilitate horizontal or clonal transmission among bacteria of different genera and species through conjugation.

PFGE revealed that 84 CREL isolates possessed 54 pulsotypes. Nineteen NDM-1-producing E. cloacae isolates obtained from Ningxia had a single dominant pulsotype. These predominated clones have obtained fast transmission in hepatobiliary surgery, vascular surgery, and ICU during a peak period (from Jan. 2016 to Sep. 2016), which might be the evidence of intra-hospital clonal dissemination. The further examining revealed that all outbreak strains were ST78, belonging to clonal complex CC74, which was considered as the most versatile genetic lineages among the world (Izdebski et al., 2015). The uniform genetic background among NDM-1-producing E. cloacae with conjugative IncX3 plasmids showed horizontal transmission between epidemic isolates. These results suggested that IncX3-type plasmids might contribute significantly to the outbreak of blaNDM−<sup>1</sup> within E. cloacae in China. Previous studies have reported that the IncX3-type plasmid was a serious threat, particularly because of the global spread of NDM-1-producing Enterobacteriaceae spp (Yang et al., 2015). The outbreak strains producing NDM-1 co-harbored multiple resistance genes, including blaACT−5, blaOXA−1, sul1, dfrA15b, mphA, catB3, arr3 as well as the qnrA1 and aac(6')Ib-cr genes encoding quinolone resistance. Studies have reported the emergency and spread of E. cloacae ST120, ST74, ST418, and ST88 producing NDM-1 in China (Liu et al., 2015; Jia et al., 2018; Jin et al., 2018). This implied that the blaNDM−<sup>1</sup> gene might have been transmitted to multiple clones of E. cloacae in many parts of China. To the extent of our knowledge, this is the first identification of an outbreak of blaNDM−<sup>1</sup> harboring E. cloacae ST78 isolates in China. Our study suggested that ST78 has become a successful hospitalrelated clone with a unique ability to accept plasmids for further dissemination in China. ST78 was one of the leading clonal lineages with increased epidemic potential, which might be associated with the spread of carbapenem resistance (Miyoshi-Akiyama et al., 2013). In summary, we reported that CREL isolates were still distributed popularly in Guangdong, China during our investigation. However, we emphasized the spread of NDM-1-producing E. cloacae ST78 with contribution of IncX3 plasmids in Ningxia, China. The outbreak has provided a new model for the dissemination of the blaNDM−1-haboring E. cloacae in China. Therefore, epidemiological surveillance of resistance is vital and should be performed routinely, especially in high-risk departments.

MLST distinguished 42 STs with ST78 (32.1%), ST51 (8.3%), ST1226 (6.0%), and ST114 (4.8%), and being the predominant STs. Previous studies revealed that ST66, ST78, ST108, and ST114 were the most prevalent and widespread E. cloacae STs (Izdebski et al., 2015). Our study showed that 36.9% (31/84) of the isolates belonged to the potentially high-risk international clones, with ST78 prevailing. We found that E. cloacae ST78 producing NDM-1 with a higher epidemic potential might be more prone to cause severe drug-resistant outbreaks. The ST78 clones demonstrated various β-lactamase profiles. Early studies also indicated a lack of strict correlation between β-lactamase profiles and ST78. The ST78 E. cloacae, originally identified in Japan, was a major international clone. The ST78 clones was widely distributed and independently obtained plasmids with β-lactamase genes, making specific pulsotypes more suitable for further spread (Miyoshi-Akiyama et al., 2013). In addition, seven ST51 isolates belonged to CC234, one of dominant clonal groups with high prevalence and wide distribution. Four ST114 isolates were diversified into multiple pulsotypes and caused some sporadic isolates. As the central genotype of CC114, ST114 has a wide global distribution formerly detected in Japan, France, Italy, Spain, Greece and Israel, mainly associated with blaCTX−M−<sup>15</sup> (Izdebski et al., 2015). Moreover, we identified a novel ST type, ST1226 detected in 5 isolates with highly similar pulsotypes, and might be related to the carriage of blaIMP−<sup>26</sup> in our study. Strikingly, we described in detail one E. cloacae ST97, which co-harbored blaNDM−<sup>1</sup> and blaIMP−<sup>1</sup> as well as multiple resistance genes.

Compared with carbapenem-resistant K. pneumoniae (CRKP), our study demonstrated that NDM-1-type MBL was the most commonly identified in E. cloacae. Recent surveillance data in China indicated that NDM was predominant in E. coli and E. cloacae, whereas KPC appeared to be most common in K. pneumoniae (Wang et al., 2018). Most previous studies revealed that the substantial burden of KPC-type carbapenemase mainly lied in North America, Latin America and Europe (Munoz-Price et al., 2013). However, NDM-1 mediated carbapenem resistance was widespread in Asia, such as India, Pakistan, and China (Nordmann et al., 2012). In contrast to the close connection between ST258/ST11 and blaKPC in K. pneumoniae (Liu et al., 2018), blaNDM−<sup>1</sup> apparently has a higher association with IncX3-type plasmids regardless of ST type in E. cloacae. Diverse clones of NDM-1 carrying E. cloacae have been widely distributed in geography. Evidence showed that clonal expansion was the cause of most transmitted cases. Thus the rapid spread of NDM-1-producing E. cloacae has comprised one of the greatest challenges to global health.

### REFERENCES


This study had several limitations. Our collection may not represent the prevalence and evolution of carbapenemresistant E. cloacae in China. Our study lacks a comprehensive analysis of clinical risk factors. We were unable to determine complete genetic structures owing to the limitation of shortread sequencing. Long-read sequencing techniques would provide more detailed analysis of integron structures and plasmid backbones.

In conclusion, our study revealed that carbapenemase, ESBL and/or AmpC overexpression combined with porin loss were the primary mechanisms responsible for carbapenem resistance. Moreover, we initially reported the nosocomial outbreak caused by NDM-1-producing E. cloacae ST78 in Northwest China. Therefore, it emphases a critical concern to monitor and control the further spread of NDM-1 in China.

### AUTHOR CONTRIBUTIONS

BH, LC, and Y-WT designed the study. YC, XY, KLa, MZ, XM, and YH carried out the experiments. CC, WJ, JZ, KLi, PG, and WZ analyzed the data. YC wrote the manuscript.

# FUNDING

The study was supported in part by the National Natural Science Foundation of China (grants 81572058, 81672081, 81772249, 81871703), the US National Institutes of Health Cancer Center Support Grant (grant P30 CA008748), the Guangdong Natural Science Foundation (grant 2014A030313143), and the State Education Ministry Scientific Research Foundation for the Returned Overseas Chinese Scholars.

### ACKNOWLEDGMENTS

This thesis is supported by the China-America Carbapenem-Resistant Enterobacteriaceae Consortium (CACREC). We thank all the hospitals that provided carbapenem-resistant E. cloacae isolates, including First Affiliated Hospital of Sun Yat-sen University, Guangdong Provincial Hospital of Chinese Medicine, and Ningxia Hospital of Ningxia Medical University.

### SUPPLEMENTARY MATERIAL

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb. 2019.01610/full#supplementary-material

heavy metal tolerance genes including a chromosomal sil operon (for acquired silver resistance). Front. Microbiol. 9:539. doi: 10.3389/fmicb.2018. 00539


polymorphisms in ampD with cephalosporin resistance and phylogenetic variations in ampC, ampR, ompF, and ompC in Enterobacter cloacae isolates that are highly resistant to carbapenems. Antimicrob. Agents Chemother. 60, 2383–2390. doi: 10.1128/AAC.02835-15


hospital environment in the west of Algeria. Med. Mal. Infect. 44, 146–152. doi: 10.1016/j.medmal.2014.01.008


**Conflict of Interest Statement:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2019 Cai, Chen, Zhao, Yu, Lan, Liao, Guo, Zhang, Ma, He, Zeng, Chen, Jia, Tang and Huang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.