Linezolid Resistance in Enterococcus faecalis Associated With Urinary Tract Infections of Patients in a Tertiary Hospitals in China: Resistance Mechanisms, Virulence, and Risk Factors

Background: Enterococcus faecalis has been commonly considered as one of the major pathogens of the urinary tract infection (UTI) in human host worldwide, whereas the molecular characteristics of E. faecalis clinical isolates from the patients with UTI in China remains seldomly reported. This study aimed to investigate the resistance mechanism, molecular characteristics and risk factors of E. faecalis clinical isolates from patients with UTI in China. Methods: A total of 115 non-duplicated E. faecalis clinical isolates from patients with UTI were retrospectively collected in a tertiary hospital in China and their clinical data was further analyzed. The linezolid and tedizolid susceptibility were determined by agar dilution. The resistance genes, including erm(A), erm(B), erm(C), tet(M), optrA, cfr, cfr(B), poxtA, and MLST-based housekeeping genes were investigated by PCR. Results: In 115 non-duplicated E. faecalis clinical isolates from the patients with UTI in this hospital setting, the frequency of linezolid or tedizolid-resistant/intermediate isolates were 22.61 and 13.04%, respectively, and the frequency of linezolid-resistant/intermediate E. faecalis clinical isolates carrying with erm(A) were 86%. Among the five linezolid-resistant E. faecalis strains found in this study, three optrA-positive isolates and the other two linezolid-resistant strains were G2576U genetic mutations in the V domain of the 23S rRNA genes. The ST clonality analysis indicated that 31.42% (11/35) of ST16 E. faecalis UTI isolates were not susceptible to linezolid. Moreover, the univariable analysis indicated that the high risk factors of linezolid-resistant/intermediate E. faecalis infections involved the indwelling catheter, trachea cannula catheter and the carriage of erm(A) or optrA. Furthermore, the indwelling catheter and trachea cannula catheter were demonstrated as the independent predictors of linezolid-resistant/intermediate E. faecalis strains in patients with UTI by multivariable analysis. Conclusion: Linezolid-resistant/intermediate E. faecalis associated with urinary tract infections of patients in this hospital setting from China might be explained by the high carriage frequency of optrA genes and moreover, indwelling catheter and trachea cannula should be considered as the independent predictors of linezolid-resistant/intermediate E. faecalis infections. The transmission mechanism of linezolid-resistant/intermediate E. faecalis in this hospital setting should be further studied.


INTRODUCTION
Enterococcus faecalis has been widely considered as the commensal inhabitants of the intestinal tract of both humans and animals (1). E. faecalis is the most prevalent species of Enterococcus genus that is isolated from the clinical specimens among human hosts with a series of infectious diseases, such as sepsis, abdominal infections, endocarditis, cholecystitis, peritonitis, and neonatal meningitis (2). Moreover, E. faecalis has been regarded as one of the major pathogens from patients with the urinary tract infection (UTI) in clinics worldwide (3,4). Because of the inherent resistance of E. faecalis to several antibiotic agents and their natural competence for acquired resistance, the treatment difficulty of E. faecalis infections has gradually increased in recent years (5). Linezolid, the first synthetic antimicrobial agent of oxazolidinone class, inhibits the initial ribosome assembly and protein synthesis of multiple gram-positive bacteria species by targeting the 50S ribosome subunits and impacting its binding affinity with formylmethionyl-tRNA (6). Due to its broad antimicrobial spectrum, linezolid has been widely used as one of the most important options for the treatment of infectious diseases caused by multi-drug resistant gram-positive pathogens, especially including methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant enterococci (VRE), penicillinresistant streptococci and mycobacteria (7). In recent years, with the widespread application of linezolid in clinics, the gradual increasing reports of linezolid resistant gram-positive pathogens highlights the enhanced risk of linezolid resistance transmission (8,9). Our previous data indicated the possible presence of high frequency of linezolid resistance in E. faecalis clinical isolates. However, the frequency and clinical significance of linezolidresistant/intermediate E. faecalis in patients with UTI remains elusive (10).
In this study, the E. faecalis clinical isolates from the patients with UTI were collected from a tertiary hospital in China. Subsequently, the clinical data of the patients with UTI was further analyzed. The antimicrobial susceptibility of linezolid and tedizolid was determined by agar dilution. The resistance genes, including erm(A), erm(B), erm(C), tet(M), the domain V region of the 23S rRNA gene, cfr, cfr(B), poxtA, as well as optrA and several commonly detected virulence factors were investigated by PCR. The ST genotype was determined by detecting MLST-based housekeeping genes and their relationship with linezolid-resistant/intermediate E. faecalis infections was further analyzed.

Bacterial Isolates and Patients Clinical Data
A total of 115 non-duplicate clinical E. faecalis UTI isolates were collected from January 1, 2010 to September 30, 2015 in Nanshan People's Hospital (A teaching hospital) of Shenzhen, China(It is a grade A class three general hospital located in Nanshan District, Shenzhen, with more than 1,300 open beds). E. faecalis clinical strains were isolated from the urine samples and identified by the VITEK 2 system (BioMérieux, Marcyl'Etoile, France). Species-appropriate quality control strains were used to ensure laboratory standards, as directed by the Clinical and Laboratory Standards Institute (CLSI 2020) (11). E. faecalis ATCC29212 and OG1RF (ATCC47077), obtained from the American Type Culture Collection (ATCC), were used as quality control strains. Patient clinical data including age, gender, admission to intensive care unit (ICU), venous catheter, indwelling catheter, D-J tube catheter, trachea cannula catheter and antibiotics therapy, were collected from hospital information system. E. faecalis clinical isolates of inpatients in Nanshan People's Hospital were analyzed retrospectively and approved by the institutional ethical committee of Shenzhen Nanshan people's hospital. This trial followed the ethical principles of the Declaration of the Chinese Ethical Guidelines. All procedures involving human participants were performed in accordance with the ethical standards of Shenzhen University and the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Antibiotic Susceptibility Testing
The antimicrobial susceptibility of several commonly-used antibiotics, such as tetracycline, erythromycin, vancomycin, minocycline, tigecycline, vancomycin, tedizolid, linezolid and doxycycline in E. faecalis were automatically tested through VITEK 2 Compact system (BioMérieux, France). The susceptibility breakpoints of these antibiotics in E. faecalis were recommended by CLSI 2020 (11). The MIC values of linezolid, tedizolid and tigecycline were further determined by agar dilution according to related reports (10,12,13). The linezolid susceptible breakpoint recommended in E. faecalis by CLSI was adopted: ≤2 µg/mL for susceptibility, 4 µg/mL for intermediate status, and ≥8 µg/mL for resistance. The susceptible breakpoint of tedizolid to E. faecalis was defined as MIC ≤ 0.5 µg/mL (11). The MIC breakpoints for tigecycline recommended by the European Committee on Antimicrobial Susceptibility Testing (EUCAST), the strains with MIC > 0.25 µg/mL was classified as resistant to tigecycline (14).

DNA Extraction and Polymerase Chain Reaction for Detection of Resistance Genes and Virulence Factors
The genomic DNA of the bacteria was extracted by DNeasy Blood & Tissue Kit DNA extraction kit (MGI Tech Co, Ltd, Shenzhen, China) according to the performance procedure of gram-positive bacteria, and the extracted DNA was stored at −20 • C. The primers in this study listed in Supplementary Table 1 were synthesized by BGI company (13,15,16). PCR was carried out for the detection of the following resistance genes: erm(A), erm(B), erm(C), tet(M), the domain V region of the 23S rRNA gene, cfr, cfr(B), poxtA as well as the ABC transporter optrA (13,15,16). Several commonly found virulence factors in the E. faecalis, including asal, esp, gelE, cyl, hyl, efaA, and ace, were amplified by PCR based on published documents (17,18).

Multilocus Sequence Typing
On the basis of established Multilocus sequence typing (MLST) schemes (http://www.mlst.net), seven housekeeping genes of E. faecalis (gdh, gyd, pstS, gki, aroE, xpt, and yiqL) were amplified and sequenced as described previously and the primers of the housekeeping genes were listed in Supplementary Table 3 (13). Sequence types (STs) were determined by comparison with published locus types in the E. faecalis MLST.net database (http:// efaecalis.mlst.net/). A. Allelic profile or STs were assigned seven integers, corresponding to the allele numbers at the seven loci. STs were assigned to isolates in such a way that the same ST names were kept as far as possible for the same analyzed strains.

Statistical Analysis
The prevalence of antibiotic susceptibility among the isolates is presented as the number (percentage). This prevalence was compared between groups using the chi-square test or Fisher's exact test. Univariable and multivariable conditional logistic regression were performed to determine patient characteristics associated with the development of infection. The tests were performed using SPSS software (version 19.0, Chicago, IL, USA). P < 0.05 were regarded as statistically significant.

Antimicrobial Susceptibility of E. faecalis UTI Isolates
The 115 non-duplicated E. faecalis clinical isolates were obtained from urine samples in the patients with UTI and the distribution characteristics of E. facecalis clinical isolates from the hospital wards was shown in Supplementary Figure 1. Our data indicated

DISCUSSION
E. faecalis is one of the major causative pathogens of UTI among gram-positive bacteria (19). Due to acquired and intrinsic resistance, E. faecalis exhibits a high level of resistance to many commonly used antibiotics, including cephalosporin and  Multiple reports have demonstrated that ST16 might become more adaptable to the hospital environment and acquire the multi-drug resistance (10,13). Whereas, whether ST16 E. faecalis with linezolid resistance has been widely transmitted in this hospital setting or this district needs to be further studied.
The reports have indicated the outbreak of the high detection frequency of some linezolid-resistant gram-positive bacteria in the hospital settings and this may be explained by different causes, such as antibiotics exposure, environmental contamination factors, person-to-person contact transmission (23)(24)(25). The complex mechanism of linezolid resistance in E. faecalis can be explained by three mechanisms: (1) genetic mutations in linezolid target sites, including the domain V region of 23S rRNA genes; (2) mutations in rpl(D) or rpl(C) genes encoding 50S ribosomal proteins L3 and L4; and (3) acquisition and dissemination of the plasmid-borne genes cfr, cfr(B), poxtA, and ATP-binding cassette (ABC) transporter gene optrA (13,16,26,27). The plasmid-borne genes cfr and cfr(B) haven't been found in this study and moreover, two linezolid-resistant strains have the G2576U genetic mutations in the V domain of the 23S rRNA, which is consistent with previous reports in our laboratory (28). Our data indicated the high carriage of erm(A) in linezolid-resistant/intermediate E. faecalis UTI isolates and three linezolid-resistant E. faecalis isolates were positive with optrA. The mechanism of erm(A) that participate in macrolide or clindamycin resistance is mainly mediated by methylating the V domain of 23S rRNA gene. No report supported linezolid resistance could be caused by erm(A) and we hypothesized linezolid-resistant/intermediate isolates might facilitate the carriage or transmission of this resistance gene in E. faecalis. Therefore, the correlation of erm(A) with linezolid susceptibility needs to be further studied. The plasmid-borne optrA can result in cross resistance to multiple antibiotics in gram-positive bacteria, including oxazolidinones (linezolid and tedizolid) and phenicols (13). The optrA gene was firstly demonstrated for the explanation of linezolidresistant/intermediate E. faecalis and subsequently the rapid and transmission of this gene among Enterococcus spp. and other gram-positive bacteria was further reported worldwide (13,15,16,23,25). Recently, the carriage frequency of optrA in Enterococcus spp. from the animals of human food chain in China was reported to be higher than that from human host (15.9% vs. 2-2.9%, respectively) and then, this phenomenon was mainly explained by the continuous and wide application of florfenicol in the food-producing animals or the environment in China from 1999 (15,29). Considering the high frequency of linezolid resistance in this study, we presume that the transmission of optrA may exist in the environment, food products, medical device surface and so on (30). Overall, the transmission routine and mechanism of optrA in linezolid-resistant/intermediate E. faecalis in this hospital setting should be further elucidated. It is worthy of note that one linezolid-intermediate strains was found optrA gene and it's still unknown for the mechanism explanation of linezolidintermediate/resistance in majority of E. faecalis UTI isolates. We hypothesized that some unknown proteins or other resistance mechanisms, including the efflux pumps or some membrane proteins, might participate in the MIC enhancement of linezolid. Therefore, the mechanisms of linezolid-intermediate/resistance E. faecalis in this hospital setting need to be further studied. Some reports have shown that linezolid exposure is an independent risk factor for linezolid-resistant/intermediate E. faecalis infections in UTI (Case-control Studies) (31,32). The univariate and multivariable conditional logistic regression of the risk factors of E. faecalis infection with linezolid resistance in this study suggested indwelling catheter and trachea cannula catheter as the independent predictors of linezolid-resistant/intermediate E. faecalis infections. It's well-known that indwelling catheter and trachea cannula are invasive operations in clinics and they are considered as one of the important causes of nosocomial infection, indicating the hospital environment and invasive operation might prompt the occurrence or dissemination of linezolid resistance in this hospital setting. However, linezolid exposure was not considered as a risk factor in this study, which could be explained by the narrow application of this drug in this hospital. Our data indicated that linezolid resistance, even in some medical environments without the wide application of linezolid, should be alert and might exhibit the high level due to the environmental transmission of linezolid-resistant/intermediate bacteria that possibly caused by invasive operations.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

AUTHOR CONTRIBUTIONS
XM: participated in the design of the study, carried out RNA silencing test, analyzed, and interpreted the data, and drafted the manuscript. FZ and BB: performed antibiotic susceptibility testing, detected virulence genes by PCR, carried out the RNA silencing test, and participated in the data analysis. ZL and GX: conducted the MLST and CC analysis, and provided a critical revision of the manuscript. ZC, XS, and JZ: participated in the acquisition of the samples, isolated DNA, conducted MLST. QD and ZY: designed the study, participated in the data analysis, and provided critical revisions of the manuscript for important intellectual content. All authors contributed to the article and approved the submitted version.