The Antibacterial and Antibiofilm Activity of Telithromycin Against Enterococcus spp. Isolated From Patients in China

Telithromycin has been reported to possess robust in vitro antibacterial activity against many species of gram-positive bacteria, and telithromycin is also effective against Staphylococcus aureus biofilms. However, the in vitro antimicrobial susceptibility of telithromycin against clinical enterococci isolates in China is rarely reported and the impacts of telithromycin on the biofilm formation and eradication of enterococci remain elusive. Therefore, this study aimed to explore the inhibitory effects of telithromycin on planktonic cells and biofilms of Enterococcus strains. A total of 280 Enterococcus faecalis and 122 Enterococcus faecium isolates were collected from individual inpatients in China. The 50% minimum inhibitory concentration (MIC50) values of telithromycin against the E. faecalis and E. faecium strains carrying erythromycin-resistant methylase (erm) genes such as the ermA, ermB, or ermC, were 2 and 4 μg/mL, respectively. In addition, these isolates were typed using multilocus sequence typing (MLST) based on housekeeping genes. The predominant sequence types (STs) of E. faecalis were ST16, ST30, and ST179, and the main STs of E. faecium isolates were ST18, ST78, and ST80. Among these major STs, 87.1% (135/158) of E. faecalis and 80.4% (41/51) of E. faecium carried erm genes. Furthermore, at the subinhibitory concentrations (1/4 and 1/8 × MIC) of telithromycin, the biofilm formation of 16 E. faecalis isolates were inhibited by approximately 35%. Moreover, treatment with 8 × MIC of telithromycin or ampicillin led to an almost 40% reduction in the established biofilms of E. faecalis isolates, whereas vancomycin or linezolid with 8 × MIC had minimal effects. The combination of telithromycin and ampicillin resulted in an almost 70% reduction in the established biofilms of E. faecalis. In conclusion, these results revealed that telithromycin significantly decreased the planktonic cells of both E. faecalis and E. faecium. In addition, the data further demonstrated that telithromycin has the robust ability to inhibit E. faecalis biofilms and the combination of telithromycin and ampicillin improved antibiofilm activity. These in vitro antibacterial and antibiofilm activities suggest that telithromycin could be a potential candidate for the treatment of enterococcal infections.

Telithromycin has been reported to possess robust in vitro antibacterial activity against many species of gram-positive bacteria, and telithromycin is also effective against Staphylococcus aureus biofilms. However, the in vitro antimicrobial susceptibility of telithromycin against clinical enterococci isolates in China is rarely reported and the impacts of telithromycin on the biofilm formation and eradication of enterococci remain elusive. Therefore, this study aimed to explore the inhibitory effects of telithromycin on planktonic cells and biofilms of Enterococcus strains. A total of 280 Enterococcus faecalis and 122 Enterococcus faecium isolates were collected from individual inpatients in China. The 50% minimum inhibitory concentration (MIC 50 ) values of telithromycin against the E. faecalis and E. faecium strains carrying erythromycin-resistant methylase (erm) genes such as the ermA, ermB, or ermC, were 2 and 4 µg/mL, respectively. In addition, these isolates were typed using multilocus sequence typing (MLST) based on housekeeping genes. The predominant sequence types (STs) of E. faecalis were ST16, ST30, and ST179, and the main STs of E. faecium isolates were ST18, ST78, and ST80. Among these major STs, 87.1% (135/158) of E. faecalis and 80.4% (41/51) of E. faecium carried erm genes. Furthermore, at the subinhibitory concentrations (1/4 and 1/8 × MIC) of telithromycin, the biofilm formation of 16 E. faecalis isolates were inhibited by approximately 35%. Moreover, treatment with 8 × MIC of telithromycin or ampicillin led to an almost 40% reduction in the established biofilms of E. faecalis isolates, whereas vancomycin or linezolid with 8 × MIC had minimal effects. The combination of telithromycin and ampicillin resulted in an almost 70% reduction in the established biofilms of E. faecalis. In conclusion, these results revealed that telithromycin significantly decreased the planktonic cells of both E. faecalis and E. faecium. In addition, the data further demonstrated that telithromycin has the robust ability to inhibit E. faecalis biofilms and the combination of telithromycin and ampicillin improved antibiofilm activity. These in vitro antibacterial and antibiofilm activities suggest that telithromycin could be a potential candidate for the treatment of enterococcal infections.

INTRODUCTION
Enterococci are gram-positive cocci which are commonly found in the gastrointestinal tracts of nearly all land animals, including humans (Fiore et al., 2019). Although a core member of the microbiome, enterococci are capable of resulting in various infectious diseases, such as urinary tract infections, wound infections, intra-abdominal and pelvic regions infections, and bloodstream infections (Murray and Weinstock, 1999;Richards et al., 2000;Mohamed and Huang, 2007). Enterococci are now the third most common nosocomial pathogen. Statistics showed that enterococci caused almost 15% of hospital-acquired infections in the United States between 2011 and 2014 (Weiner et al., 2016), an increase of 12% from 2006 to 2007 (Hidron et al., 2008). In addition, enterococci are also responsible for 5-20% of cases of infective endocarditis (Megran, 1992). Enterococcus faecalis, the most common species of Enterococcus in the clinical setting, causes 85 to 90% of human enterococcal infections, while Enterococcus faecium is responsible for 5 to 10% of the remainder (Jett et al., 1994;Jones et al., 2004). Because E. faecalis and E. faecium usually carry a range of intrinsic and acquired resistance genes, these Enterococcus strains are frequently resistant to many commonly used antibiotics, such as glycopeptides (vancomycin and teicoplanin), betalactams (ampicillin, penicillin), aminoglycoside (gentamicin or streptomycin), and macrolides (van Harten et al., 2017;Ch'ng et al., 2019). Of note, the widespread emergence of vancomycinresistant enterococci (VRE) has caused further concern due to the high mortality rate (Eliopoulos, 1997). In 2013, almost 70% of clinical E. faecium isolates in the United States displayed vancomycin resistance, while this was up to 20% in Europe (Mendes et al., 2016). In contrast, E. faecalis isolates are less frequently resistant to vancomycin (<10%) (Sievert et al., 2013).
Besides antibiotic resistance, enterococci are also known for their ability to form biofilms which is a population of cells growing on a surface and surrounded by a matrix of macromolecules like polysaccharides, proteins, lipids, and extracellular DNA (Jakubovics and Burgess, 2015). The term biofilm was introduced into medicine in 1982 by Costerton as Staphylococcus aureus biofilms were observed on a cardiac pacemaker lead (Marrie et al., 1982). Bacteria that are not innately resistant to antibiotics can also become resistant by forming persistent biofilms that lead to chronic infections (Lebeaux et al., 2014). In today's healthcare environment, the diversity of biofilm-associated infections has risen with time, it is suggested that biofilms are present in more than 65% of all bacterial infections (Costerton et al., 1999;Lewis, 2001). Especially, due to a sharp increase in the number of patients receiving implanted medical devices in recent years, the rates of infection are 40% for ventricular assist devices, 10% for ventricular shunts, and 4% for mechanical heart valves, pacemakers, and defibrillators (Darouiche, 2004).
Abbreviations: AMP, ampicillin; CIP, ciprofloxacin; DOX, doxycycline; ERY, erythromycin; LZD, linezolid; MICs, minimum inhibitory concentrations; MIN, minocycline; MLST, multilocus sequence type; NIT, nitrofurantoin; PCR, polymerase chain reaction; STs, sequence types; TEC, tetracycline; TED, tedizolid; TEL, telithromycin; TET, tetracycline; VAN, vancomycin. Biofilms serve as a new nidus for bacterial dissemination and as a reservoir for antimicrobials resistant genes. Additionally, biofilms protect bacteria from detergent solutions, antimicrobial agents, environmental stress, and effectively make bacteria 10 to 1000-fold more resistant to antibiotic treatment, making their eradication extremely difficult (Lewis, 2001). Therefore, biofilm infections are becoming increasingly difficult to effectively treat due to the decreasing efficacy of antibiotics (Paganelli et al., 2012;van Harten et al., 2017). Enterococci frequently cause biofilm-associated infections such as catheter-related bloodstream infections, urinary tract infections, and infective endocarditis (Arias and Murray, 2012). Among Enterococcus species, E. faecalis isolates usually have a higher capacity of producing biofilms than E. faecium isolates and the prevalence of E. faecalis biofilms varies in different regions (Mohamed and Huang, 2007). For example, in Sardinia, biofilm production was identified among 87% of E. faecalis clinical isolates and 16% of E. faecium clinical isolates (Duprè et al., 2003). In Rome, 80% of E. faecalis and 48% of E. faecium isolates from infected patients were able to form biofilms (Baldassarri et al., 2001). Other study showed similar results and indicated that E. faecalis (95%) isolates produced biofilms more often than E. faecium (29%) (Di Rosa et al., 2006). In contrast with planktonic cells, biofilm-embedded enterococci are involved in multidrug resistance and may even be untreatable with conventional antibiotics (Lewis, 2001). Therefore, there is an urgent need to identify novel treatments for enterococcal infections.
Telithromycin (HMR 3647), a semi-synthetic derivative of erythromycin, belongs to a new chemical class of antibiotics called ketolides that have been added to the macrolidelincosamide-streptogramin class of antibiotics (Douthwaite and Champney, 2001). Telithromycin has been approved by the U.S. Food and Drug Administration (FDA) to treat several infectious diseases, such as community-acquired pneumonia, acute exacerbations of chronic bronchitis, and acute maxillary sinusitis (Nguyen and Chung, 2005). The antimicrobial action of telithromycin, and probably that of the other 14-memberring macrolides, is compounded by binding to 23S rRNA and blocking protein synthesis at the early stages (Menninger, 1995). Macrolides resistance is dominantly explained by the prevalence of erm genes in a wide spectrum of gram-positive bacteria (Westh et al., 1995). erm gene classes can encode a series of methyltransferase that specifically methylate the N-6 position of adenosine 2058 (A2058) or neighboring nucleotides in domain V of 23S rRNA within the large ribosomal subunit (Weisblum, 1995;Vester and Douthwaite, 2001), this prevents interaction with macrolides. In addition, erythromycin-resistant enterococci often exhibit macrolide-lincosamide-streptogramin B (MLSB) antibiotic resistance phenotypes due to the presence of erm genes. However, compared with MLSB antibiotics like erythromycin, telithromycin has an additional target site at position A752 in domain II of 23S rRNA. U747 methylation promotes G748 methylation, resulting in the increased binding of telithromycin to ribosomes and enhanced telithromycin susceptibility (Shoji et al., 2015). Telithromycin has been found to conquer erm-mediated resistance in S. pneumonia and S. aureus (Van Laethem and Sternon, 2003;Sun et al., 2018), and it also showed less efficacy against erythromycinresistant Enterococcus isolates from different regions (TEL, MIC ≥ 4 µg/mL) (Singh et al., 2000;Min et al., 2011). However, the antimicrobial impacts of telithromycin on Enterococcus carrying erm genes remain less clear in China. Moreover, a previous study reported that telithromycin exhibited effective antibiofilm activity against S. aureus in vitro . Therefore, the effect of telithromycin on Enterococcus biofilms should be further investigated.
In this study, the in vitro antibacterial activity of telithromycin was tested against enterococcal clinical isolates from inpatients in Shenzhen Nanshan People's Hospital, and then the antibacterial activity of telithromycin was further compared with that of other antimicrobials. In addition, the multilocus sequence types (MLSTs) and erm genes expression in enterococci were detected by PCR assay. Moreover, the effect of telithromycin on the biofilms of E. faecalis isolates was further explored.

Antimicrobials Susceptibility Testing
Antimicrobial susceptibilities of Enterococcus to several clinical antibiotics, including TEL, ERY, AMP, VAN, TET, DOX, MIN, CIP, NIT, LZD, TED, and TEC were tested with the VITEK 2 system. The MICs of AMP, VAN, TEL, and ERY were determined by the broth macrodilution method in cationadjusted Mueller-Hinton broth (CAMHB) according to the 2019 Clinical and Laboratory Standards Institute (CLSI) guidelines 1 . The isolates were cultured overnight in TSB at 37 • C with a shaker at 220 rpm, then the strains were diluted at 1:200 [2.0-3.0 × 10 7 colony-forming units (CFU) ml −1 ], and inoculated into 96 polystyrene microtiter plates with 200 µL of CAMHB containing the indicated concentrations of antibiotics. As the MIC breakpoint of telithromycin against enterococci has not been established, the MIC value of telithromycin against S. aureus (≤1, 2, ≥4 µg/mL) was based on the 2019 CLSI guidelines. Four MIC levels were thus employed for telithromycin in the 1 https://eucast.org/clinical_breakpoints/ antimicrobial susceptibility analysis (≤1, 2, 4, ≥8 µg/mL). The used concentrations of indicated antibiotics are given in the figure legends and Supplementary Table 4. All experiments were performed at least three times.

Multilocus Sequence Type
The genotypes of the enterococcal isolates were analyzed by MLST. Seven pairs of housekeeping genes: gdh, gyd, pstS, gki, aroE, xpt, and yqiL for E. faecalis, and atpA, ddl, gdh, purK, gyd, pstS, and adk for E. faecium were amplified by PCR. As previously reported (Zheng et al., 2017), the purified PCR products were sequenced, and the results were submitted to the MLST database 2 for comparison, and the sequence types (STs) of enterococci were determined. The primers used for PCR amplification are listed in Supplementary Tables 5, 6.

Inhibition and Eradication of E. faecalis Biofilms
A detailed protocol has been previously reported . As for the inhibition experiments, E. faecalis isolates were cultured overnight in TSB at 37 • C with a shaker at 220 rpm, then the strains were diluted at 1:200 [2.0-3.0 × 10 7 colony-forming units (CFU) ml −1 ], and inoculated into 96 polystyrene microtiter plates with 200 µL of TSBG (TSB with 0.5% glucose) containing the indicated concentration of antibiotics. TSBG without antimicrobials was used as an untreated control. After incubation for 24 h, the biofilm biomasses were washed three times with ddH 2 O before and after crystal violet staining, the stained biofilms were detected by optical density (OD 570 ). Solithromycin was used as a positive control (Wang et al., 2020). As for the eradication assays of the established biofilms, E. faecalis isolates were cultivated in tryptic TSB medium at 37 • C for 24 h to form matured biofilms, then they were treated with antimicrobials (8 × MIC) for 48 h with the medium replaced daily. TSBG without antimicrobials was used as an untreated control. Biofilm biomasses were stained and then detected. The used concentrations of indicated antibiotics are given in the figure legends and Supplementary Table 4. All data are representative of three independent experiments.

Detecting the Adherent Cells in the Established Biofilms
The adherent cells in the established biofilms of E. faecalis were identified by the CFU numbers as described previously (Zheng et al., 2019). Briefly, the E. faecalis isolates were inoculated into 24 polystyrene microtiter plates with TSBG and formed mature biofilms after 24 h of static incubation. The supernatant was discarded and plates were washed, fresh TSBG containing antimicrobials was added, and TSBG without antimicrobials was used as an untreated control. After 48 h of static incubation with the medium being replaced daily, the supernatant was discarded and the remaining adherent cells in the established biofilms were collected by scratching the wall of the wells with a flat end toothpick. Finally, the bacteria were centrifuged and the numbers of CFU were determined. The used concentrations of indicated antibiotics were given in the figure legends and Supplementary Table 4. All data are representative of three independent experiments.

Time-Kill Curve Assay
Two E. faecalis 16C3 and 16C6 isolates were cultured in TSB at 37 • C for 16 h, then diluted 200 times with TSB and antibiotics were added to make the final concentrations at 4 × MIC. A colony count was performed after 0, 1, 3, and 24 h. Data are representative of three independent experiments.

Statistical Analysis
The SPSS software (version 19.0) and GraphPad Prism software (version 5.0) were used for statistical analysis. For multiple comparisons, one-way analysis of variance (ANOVA) and a post hoc-Dunnett test were applied to analyze the data. All experiments were repeated at least three times. Sample size, n, for each experiment is given in the figure legends. Results are shown as mean ± SEM. Value differences were considered significant when * p < 0.05 (not significant p > 0.05, * * p < 0.01, * * * p < 0.001).
In addition, as shown in Tables 1, 2

Telithromycin Against the Enterococcus Clinical Isolates Harboring erm Genes
Next, this study further examined the effects of E. faecalis and E. faecium isolates carrying erm genes on the sensitivity of telithromycin. At first, the presence of ermA, ermB, or ermC in Enterococcus isolates was detected by PCR assays. As shown in Table 3, the rates of E. faecalis strains harboring ermA and ermB genes were 3.9 and 67.1%, respectively. However, there were no ermC-positive E. faecalis isolates. Additionally, among 122 E. faecium isolates, the rates of E. faecium isolates carrying ermA, ermB, and ermC gene were 9.8, 32.8, and 45.9%, respectively (Table 4). Furthermore, telithromycin MIC 50 /MIC 90 values of ermA and ermB-positive E. faecalis strains or E. faecium strains both were 4/8 µg/mL, and ermC-positive E. faecium strains had a telithromycin MIC 50 /MIC 90 of 8/8 µg/mL (Tables 3, 4). As mentioned, the MIC 50 /MIC 90 values of telithromycin against E. faecalis and E. faecium strains were 2/8 and 4/8 µg/mL, respectively (Tables 1, 2), suggesting that the presence of erm genes slight impacted telithromycin susceptibility in the Enterococcus isolates.

Relationship Between Telithromycin MICs Distribution and ST Clonality
Subsequently, MLST was performed to determine the ST distribution of E. faecalis and E. faecium isolates. As shown in Supplementary Tables 1, 2, ST16, ST30, and ST179 were the predominant STs in 44 STs detected from E. faecalis isolates. In total, 25 STs were identified in E. faecium isolates, the main STs were ST18, ST78, and ST80. In addition, the relationships between telithromycin and ERY MICs distributions in the predominant ST isolates are shown in Tables 5, 6. The data indicated that E. faecalis with telithromycin MIC ≤ 1, 2, 4,

The Effects of Telithromycin Against Biofilm Formation and Eradication of E. faecalis Clinical Isolates
Many studies have demonstrated that E. faecalis isolates have a higher capacity of producing biofilms than E. faecium (Duprè et al., 2003;Sandoe et al., 2003;Zheng et al., 2017), thus E. faecalis isolates are usually chosen for biofilm analysis. Among our 280 E. faecalis isolates, 16 E. faecalis isolates showed a higher biofilm-forming ability, these specific 16 strains were thus tested for biofilm formation. The isolation sites of the 16 E. faecalis strains are listed in Supplementary Table 3. The inhibitory effect of telithromycin on the biofilm formation of these 16 E. faecalis isolates was determined by crystal violet staining.
The MIC values of AMP, VAN, LZD, and telithromycin against these isolates are listed in Supplementary Table 4. As shown in Figure 1, 1/2 × MIC, 1/4 × MIC, or 1/8 × MIC of telithromycin could inhibit the biofilm formation of the 16 E. faecalis isolates. Finally, to dissect the effect of telithromycin on the established biofilms of E. faecalis isolates. Here, eight specific E. faecalis isolates were chosen for analysis by crystal violet staining. As shown in Figures 2A,B, the established biofilms of E. faecalis reduced by almost 40% after treatment with 8 × MIC of telithromycin or ampicillin, whereas 8 × MIC of vancomycin or linezolid only showed slight effects. Due to the high susceptibility rate of the E. faecalis isolates toward ampicillin (Table 1), the effects of telithromycin combined with ampicillin on the established biofilms of these eight E. faecalis isolates were further evaluated. The data showed that the combination of telithromycin and ampicillin resulted in an approximate 70% reduction in the established biofilms than with telithromycin or ampicillin alone (Figures 2C,D). Additionally, a colonyforming unit (CFU) assay was performed to quantify the viable cells of the established biofilms. Consistently, the data further confirmed that telithromycin combined with ampicillin could kill more than 70% of adherent cells in the established biofilms than telithromycin or ampicillin alone (Figures 2E,F). Therefore, these results suggest that the combination of telithromycin and ampicillin is an effective way to reduce the established biofilms in the E. faecalis isolates.

DISCUSSION
Enterococci species are commensal bacteria in the human gastrointestinal tract with the ability to cause various nosocomial infections. They not only have multiple inherent antimicrobial resistances, but are also able to acquire mutations and/or new resistance genes (Cetinkaya et al., 2000;Raza et al., 2018). Telithromycin, a novel ketolide antimicrobial agent, can be utilized for the treatment of respiratory infections. It retains its activity against most macrolide-resistant strains of Streptococcus pneumoniae and Streptococcus pyogenes (Spiers and Zervos, 2004;Wolter et al., 2008;Togami et al., 2009;Uzun et al., 2014). In addition, the efficacy of telithromycin against Enterococcus has been studied worldwide (Baltch et al., 2001;Bonnefoy et al., 2001;Mensa et al., 2003;Min et al., 2011), and it has been proven to be more effective against enterococci   than some of first-and second-generation macrolides. For instance, telithromycin was found to be more potent against enterococci than erythromycin in a mouse peritonitis model (Singh et al., 2000). The MIC 50 /MIC 90 of telithromycin were 8/32 µg/mL against vancomycin-resistance enterococci strains, whereas erythromycin and clarithromycin were completely inactive (Hamilton-Miller and Shah, 1998). In this study, the MIC 50 /MIC 90 of ERY were >256/>256 µg/mL, and the resistance rate of ERY reached more than 75% in both E. faecalis and E. faecium isolates (Tables 1, 2). However, the MIC 50 /MIC 90 of telithromycin in E. faecalis and E. faecium isolates were found to be as low as 2/8 and 4/8 µg/mL, respectively (Tables 1, 2). This study thus delineated additional evidence to support the fact that telithromycin could be a promising antimicrobial drug for use against enterococci. Of note, the majority of clinical E. faecalis isolates remain susceptible to β-lactams which are inhibitors of cell wall synthesis, and time-kill curve studies demonstrated that telithromycin (4 × MIC) combined with ampicillin (4 × MIC) could kill more than 100 times more planktonic cells in comparison to telithromycin or ampicillin alone in two E. faecalis isolates (Supplementary Figures 2A,B), suggesting that the combination of telithromycin and ampicillin could improve bactericidal activities against E. faecalis. Ever since erythromycin has been clinically applied, the MLSB resistance phenotype has been widely found in erythromycinresistant isolates of many bacteria species. For instance, Weisblum B et al. demonstrated that erythromycin-resistant S. aureus was related to the MLSB resistance phenotype after the clinical application of erythromycin (Weisblum and Demohn, 1969). This MLSB phenotype was determined by the erm genes producing erythromycin-resistant methylases, which induced methylation at specific adenosine residues on the 23S rRNA, thus leading to the resistance of the newly synthesized ribosome to MLSB antibiotics (Westh et al., 1995;Ackermann and Rodloff, 2003). Previous studies have described that the emergence of high-level erythromycin resistance narrowed the clinical application of macrolides and their application might result in a poor prognosis, more severe recurrence, and higher mortality for the treatment of bacterial infections (Min et al., 2011;Celik et al., 2014). While the chemical structure of telithromycin is derived from erythromycin, telithromycin has a low potential for drug-drug interaction, and is less likely to induce MLSB FIGURE 1 | Telithromycin inhibiting biofilm formation of 16 E. faecalis isolates. (A,B) The four isolates per group were treated with TEL at 1/2 ×, 1/4 ×, and 1/8 × MICs, respectively. All of the eight isolates with the TEL MIC at 8 µg/mL. (C,D) The eight isolates of the two groups were treated with TEL at 1/2 ×, 1/4 ×, and 1/8 × MICs. The MICs were 0. 25, 0.5, 0.125, 0.125, 0.5, 0.125, 0.25, and 0.25 µg/mL, respectively. TSBG without antimicrobials was used as an untreated control. The data were presented as the mean ± SEM (n = 3 experiments), one-way analysis of variance (ANOVA), and a post hoc-Dunnett test, *p < 0.05, **p < 0.01, ***p < 0.001, compared with the CTRL group.
resistance than macrolides with a 14-or 15-member ring (Benes, 2004). In this study, the ermA and ermB genes were found in ERY-resistant isolates of E. faecalis, and ermA, ermB, and ermC genes were detected in E. faecium isolates (Tables 3, 4). However, telithromycin remained active against enterococci with erm genes. Recombination is involved in the genetic variation of resistance and virulence determinants, which might promote the hospital adaptation of Enterococcus bacteria such as E. faecalis and E. faecium (Homan et al., 2002;Ruiz-Garbajosa et al., 2006). In this study, E. faecalis isolates were grouped into 44 distinct STs, with the predominant STs being ST16 and ST179, which belonged to the clonal complex (CC16) . In addition, 25 STs were determined in E. faecium isolates, among which ST18 and ST78 were the main positive STs (Tables 5, 6). These results further confirmed that the presence of erm genes had minimal effects on the sensitivity of the predominant ST strains to telithromycin, suggesting its potential application for the treatment of some multi-resistant enterococci infections.
Evidence continues to accrue documenting the crucial role of biofilm formation in enterococcal infection (Ch'ng et al., 2019). The majority of clinical E. faecalis isolates are capable of forming biofilms on inanimate and living surfaces, which may promote antibiotic tolerance and reduce susceptibility to environmental influences or antimicrobial pressures (Paganelli et al., 2012;Holmberg and Rasmussen, 2016). Thus, targeting the biofilm formation of E. faecalis may potentially contribute to the treatment of enterococcal infections. As previously described, telithromycin could be considered as a novel inhibitor of FIGURE 2 | Telithromycin alone or combined with ampicillin eradicating the established biofilms of eight E. faecalis isolates. (A,B) A total of eight isolates formed mature biofilms for 24 h, then treated with AMP, VAN, LZD, or TEL at 8 × MICs for 48 h. The MICs of AMP, VAN, and TEL for these isolates were 2, 1, and 8 µg/mL, respectively. The MICs of LZD for 16C3, 16C6, 16C25, and 16C28 isolates were 4 µg/mL, the MICs for 16C109, 16C137, 16C139, and 16C170 were 2 µg/mL. (C,D) A total of eight isolates formed mature biofilms for 24 h, then were treated with AMP and TEL alone or TEL combined with AMP at 8 × MICs for 48 h. The MICs of AMP and TEL for these isolates were 2 and 8 µg/mL, respectively. (E,F) A total of eight isolates formed mature biofilms for 24 h, then were treated with AMP and TEL alone or TEL combined with AMP at 8 × MICs for 48 h. Then the adherent cells in these established biofilms were detected by the CFU numbers. The MICs of AMP and TEL for these isolates were 2 and 8 µg/mL, respectively. TSBG without antimicrobials was used as an untreated control throughout. The data were presented as the mean ± SEM (n = 3 experiments), one-way analysis of variance (ANOVA), and post hoc-Dunnett test, *p < 0.05, **p < 0.01, ***p < 0.001, compared with the CTRL group.
S. aureus biofilms, which may be a result of the decreased expression of biofilm formation-related genes (Woo et al., 2017;Zheng et al., 2020), but there are no available data on the contribution of telithromycin susceptibility to biofilm phenotype in E. faecalis. The present study firstly showed that subinhibitory concentrations of telithromycin could inhibit the biofilm formation of 16 E. faecalis isolates (Figure 1). However, the data regarding biofilm formation inhibition does not show a dose-dependent effect. Therefore, further investigation is needed to characterize the inhibitory effect of telithromycin on the biofilm formation of E. faecalis.
Because biofilm prevention is not always possible, the removal of pre-existing enterococcal biofilms remains a necessity. Although the antibiofilm activity of antibiotics is likely hampered by poor penetration or slowed cell division in metabolically dormant biofilms, the use of antibiotics as first-line treatment for biofilm-associated infections is commonplace (Stewart, 2002). For example, the most recommended antibiotic treatment of endocarditis caused by E. faecalis involves ampicillin combined with gentamicin for 4-6 weeks (Habib et al., 2015). In this study, both telithromycin and ampicillin exhibited promising biofilminhibiting activity (40% reduction) in E. faecalis (Figures 2A,B), these results are consistent with those previously reported in the in vitro activities of telithromycin or ampicillin against S. aureus or E. faecalis biofilms (Di Domenico et al., 2019;Zheng et al., 2020). Notably, treatment with an 8 × MIC telithromycin/ampicillin combination exhibited a visible reduction (∼70%) in mature E. faecalis biofilms (Figures 2C,D) and killed 70% of adherent cells in the established biofilms (Figures 2E,F). Therefore, the combination of telithromycin with ampicillin as an anti-adherence or antibiofilm strategy appears to be much more promising for the treatment of biofilm-associated enterococcal infections.
Accumulating evidence has revealed that there is a lower likelihood of resistance developing through the clinical use of telithromycin. For example, spontaneous resistance to telithromycin was at a low frequency in vitro. Telithromycin does not induce MLSB resistance and it shows low potential in selecting resistance or cross-resistance (Felmingham, 2001). In addition, pharmacodynamic studies suggested that telithromycin was generally well tolerated and a once-daily 800 mg oral dose of telithromycin maintains an effective concentration in plasma for the treatment of respiratory tract infections involving the key respiratory pathogens (Drusano, 2001;Namour et al., 2001). However, information regarding kill kinetics and post-antibiotic effects for telithromycin against enterococci is limited. Thereafter, there is inadequate clinical evidence to suggest an optimal dosage regimen for telithromycin against enterococci. The present study demonstrated the antimicrobial susceptibility and antibiofilm activity of telithromycin against Enterococcus isolates in vitro, suggesting that this compound might be useful, alone or in combination, in some difficult to treat enterococcal infections. The further results of in vivo studies should provide some evidence to support such a possibility.

CONCLUSION
In summary, this study presents the effective antimicrobial activity of telithromycin against clinical enterococci isolates from China in comparison to that of ERY. Importantly, this study further demonstrated that telithromycin could inhibit the biofilm formation of E. faecalis and that telithromycin combined with ampicillin resulted in enhanced antimicrobial and antibiofilm activity. To our knowledge, this is the first study to present new insight into the antibiofilm activity of telithromycin against enterococci and provide further evidence for the potential clinical application of telithromycin/ampicillin combination for the treatment of Enterococcus infections.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.

AUTHOR CONTRIBUTIONS
YX conducted the PCR analyses, MLST analysis, biofilms inhibition and eradication of E. faecalis, and drafted the manuscript. JC collected the bacterial isolates, performed the antimicrobials susceptibility tests, and gene manipulation. XS participated in gene manipulation, MLST analysis, and the inhibition of E. faecalis biofilms. GX and PL participated in the collection of bacterial isolates, MLST analysis, and the inhibition and eradication of E. faecalis biofilms. ZY and QD participated in the collection of bacterial isolates, and the inhibition and eradication of E. faecalis biofilms. ZC and JZ designed the study, analyzed the experimental data, and revised the manuscript. All authors have read and approved the manuscript.