In Vitro and In Vivo Activities of a Bi-Aryl Oxazolidinone, RBx 11760, against Gram-Positive Bacteria

Abstract

RBx 11760, a bi-aryl oxazolidinone, was investigated for antibacterial activity against Gram-positive bacteria. The MIC₉₀s of RBx 11760 and linezolid against Staphylococcus aureus were 2 and 4 mg/liter, against Staphylococcus epidermidis were 0.5 and 2 mg/liter, and against Enterococcus were 1 and 4 mg/liter, respectively. Similarly, against Streptococcus pneumoniae the MIC₉₀s of RBx 11760 and linezolid were 0.5 and 2 mg/liter, respectively. In time-kill studies, RBx 11760, tedizolid, and linezolid exhibited bacteriostatic effect against all tested strains except S. pneumoniae. RBx 11760 showed 2-log₁₀ kill at 4× MIC while tedizolid and linezolid showed 2-log₁₀ and 1.4-log₁₀ kill at 16× MIC, respectively, against methicillin-resistant S. aureus (MRSA) H-29. Against S. pneumoniae 5051, RBx 11760 showed bactericidal activity, with 4.6-log₁₀ kill at 4× MIC compared to 2.42-log₁₀ and 1.95-log₁₀ kill for tedizolid and linezolid, respectively, at 16× MIC. RBx 11760 showed postantibiotic effects (PAE) at 3 h at 4 mg/liter against MRSA H-29, and linezolid showed the same effect at 16 mg/liter. RBx 11760 inhibited biofilm production against methicillin-resistant S. epidermidis (MRSE) ATCC 35984 in a concentration-dependent manner. In a foreign-body model, linezolid and rifampin resulted in no advantage over stasis, while the same dose of RBx 11760 demonstrated a significant killing compared to the initial control against S. aureus (P < 0.05) and MRSE (P < 0.01). The difference in killing was statistically significant for the lower dose of RBx 11760 (P < 0.05) versus the higher dose of linezolid (P > 0.05 [not significant]) in a groin abscess model. In neutropenic mouse thigh infection, RBx 11760 showed stasis at 20 mg/kg of body weight, whereas tedizolid showed the same effect at 40 mg/kg. These data support RBx 11760 as a promising investigational candidate.

INTRODUCTION

Bacterial infections due to methicillin-resistant Staphylococcus aureus (MRSA), methicillin-resistant coagulase-negative Staphylococcus (MRCoNS), and vancomycin-resistant enterococci (VRE) have emerged as a major public health problem throughout the world. Over the past few years, there has been a dramatic increase in the number of cases of MRSA and VRE infections in hospitals and communities. In 2014, the Agency for Healthcare Research & Quality reported 23,000 deaths due to MRSA in the United States alone. The prevalence of Streptococcus pneumoniae strains resistant to β-lactam and macrolide antibiotics has increased worldwide (1–6). Gram-positive bacterial pathogens, particularly hospital-acquired MRSA, have become increasingly resistant to more than one antibiotic, such as vancomycin, trimethoprim-sulfamethoxazole, β-lactams, tetracycline, clindamycin, quinolones, and aminoglycosides (7, 8). In addition to drug resistance, the presence of high-virulence determinants in the pathogens makes the situation in the hospital and community worse. According to the Centers for Disease Control and Prevention (CDC) and others, the bacteria causing nosocomial infections are often associated with biofilms that cause treatment failure and result in high mortality (9, 10). Although vancomycin is considered the drug of choice for MRSA treatment, MIC creep of vancomycin contributed to treatment failure and highlights the need for alterative therapy against MRSA (11, 12). Moreover, coinfection with VRE and MRSA in hospitalized patients is a concern (13, 14).

Linezolid (Fig. 1A) is the first FDA-approved oxazolidinone introduced for the treatment of complicated skin and skin structure infections (cSSSI) due to MRSA and infections associated with VRE, including bloodstream infection and hospital-acquired pneumonia. However, linezolid resistance in S. aureus and VRE has already been observed (15, 16). In addition, linezolid is not approved for the treatment of patients with catheter site or catheter-related bloodstream infections due to the potential coculture of Gram-negative infection or infections due to coagulase-negative Staphylococcus (CoNS) (17). Tedizolid phosphate (Fig. 1B), an expanded-spectrum oxazolidinone, has 4- to 16-fold more activity than linezolid against MRSA and was approved for clinical use by the FDA in June 2014 (18). However, there were safety and efficacy concerns for tedizolid in patients with neutropenia (neutrophil counts of <1,000 cells/mm3) (18). In an animal model of infection, the antibacterial activity of tedizolid was reduced in the absence of granulocytes (19). It has been suggested that alternative therapies should be considered when treating patients with neutropenia and acute bacterial skin and skin structure infections (ABSSSI) (18, 19).

FIG 1.

Structures of linezolid (A), tedizolid (B), and a novel bi-aryl oxazolidinone, RBx 11760 (C).

A series of aryl-oxazolidinone compounds was synthesized and checked for antibacterial activity by our group (20, 21). RBx 11760 (Fig. 1C) is a bi-aryl oxazolidinone identified in our laboratory. The objective of the present study was to evaluate RBx 11760 for in vitro activity against Gram-positive isolates and in vivo efficacy in a skin and soft-tissue infection model and in foreign-body biofilm models. The pharmacokinetics (PK) of RBx 11760 was investigated to substantiate the pharmacodynamics (PD) in animal models. In addition, in vitro transcription/translation assays and resistance development against linezolid and RBx 11760 by serial passaging and molecular modeling were performed to confirm the mechanism of action (MOA).

MATERIALS AND METHODS

Bacterial strains and growth conditions.

A total of 378 clinical isolates, including Staphylococcus aureus (n = 126), Staphylococcus epidermidis (n = 40), Enterococcus faecalis (n = 63), Enterococcus faecium (n = 55), and S. pneumoniae (n = 94), were used in this study. These isolates include American Type Culture Collection (ATCC; Manassas, VA, USA) strains, and clinical isolates obtained from Indian hospitals were also used in in vitro screening. For the identification of clinical strains, we subcultured the strains on growth and selective media. Based on Gram-positive and -negative identification, we performed biochemical characterization using the catalase, coagulase test (Gram positive) and IMViC test (Gram negative) for presumptive identification. Following the biochemical tests, we confirmed strain identifications using bioMérieux's API system and stored them at −80°C in 20% glycerol in Trypticase soy broth (TSB) (Becton, Dickinson and Company, Cockeysville, MD). MRSA Xen-29 was purchased from Xenogen Corporation, USA (now PerkinElmer). MRSA Xen-29 showed very good reproducible infection in mouse, so we included it in our evaluation system.

Media, reagents, and antimicrobial agents.

All media and reagents were purchased from Becton Dickinson and Company and Sigma-Aldrich Co. LLC, USA, respectively, unless otherwise mentioned. Mueller-Hinton broth (MHB) was used for MIC testing and growing bacteria for the mouse thigh infection model. TSB was used for biofilm experiments, and brain heart infusion (BHI) broth was used for growing bacteria for groin abscess. RBx 11760 and tedizolid were synthesized in-house (18) in Gurgaon, India. Linezolid was obtained from the National Chemical Laboratory, Pune, India. Vancomycin, rifampin, penicillin, and erythromycin were purchased from commercial sources. For in vitro experiments, RBx 11760 and tedizolid were prepared in dimethyl sulfoxide (DMSO) with the final concentration of DMSO being below 1%; rifampin and erythromycin were prepared in methanol (maximum concentration, 640 μg/ml) and 95% ethanol, respectively. Vancomycin, penicillin, and linezolid were dissolved in sterile water and then further diluted in medium to the desired concentrations.

In vitro antimicrobial susceptibility testing.

MICs of RBx 11760, tedizolid, linezolid, and vancomycin were determined in MHB against 378 Gram-positive bacterial isolates using the broth microdilution method recommended by the Clinical and Laboratory Standards Institute (CLSI) (22). The MIC_50/90 was defined as the MIC for 50% and 90% of the bacterial isolates. The inoculum and pH effect on the activity of RBx 11760 and standard drugs were also studied using the same broth microdilution method. In order to determine the effect of plasma proteins on the activity of RBx 11760, the MIC in the presence of 50% human plasma in MHB was determined. All experiments were performed in triplicate, and higher MIC values were considered the final results. The in vitro postantibiotic effect (PAE) and frequency of resistance were also determined (see the supplemental material for methods) (22, 23).

Time-kill kinetics.

The time-kill kinetics of RBx 11760, tedizolid, and linezolid against MRSA H-29, MRSE 35984, E. faecalis 427, and S. pneumoniae 5051 were determined as described earlier (24). Briefly, 10 ml cation-adjusted Mueller-Hinton broth (CAMHB) was prewarmed at 37°C in reaction tubes, and the drug was added at multiple concentrations ranging from 1× to 8× MIC. Exponential-phase cultures were diluted to 1 McFarland standard (Densimat; bioMérieux, SA, France) and added to the reaction tubes to give final inocula of ∼5 × 10⁶ CFU/ml. For S. pneumoniae 5051, tubes containing 10 ml of CAMHB with 5% lysed horse blood and doubling drug concentrations were inoculated with ∼5 × 10⁶ CFU/ml, and the tubes were incubated at 35°C. The viable cell count was determined at predefined time points by serial dilution plating onto Trypticase soy agar (TSA) or blood agar plates for S. pneumoniae 5051. The time-kill kinetics assays were performed three independent times. Antimicrobials that reduced the original inoculum by ≥3 log₁₀ CFU/ml (99.9% killing) were considered bactericidal, and they were considered bacteriostatic if the inoculum was reduced by 0 to <3 log₁₀ CFU/ml. The reduction of 1 log₁₀ and 2 log₁₀ CFU/ml from original inocula was defined as 90% and 99% killing.

Effect of RBx 11760 on biofilm-producing and glass-adherent bacteria.

The effects of RBx 11760, linezolid, and vancomycin on biofilm production were assessed as described by previously with slight modifications (25). Briefly, the overnight growth of MRSE ATCC 35984 in TSB was diluted, adjusted to ∼10⁶ CFU/ml in TSB supplemented with 2% glucose, and incubated in polystyrene microtiter flat-bottom 96-well plates. After 4 h of incubation, the bacterial culture was exposed to various concentrations of RBx 11760, linezolid, and vancomycin, ranging from 0.03 to 16 mg/liter. The effects of RBx 11760 and comparators on inhibition of biofilm formation were assayed after 24 h of exposure at 37°C. Wells were emptied and washed three times with phosphate-buffered saline (PBS) (pH 7.3). Adherent bacteria were stained with 1% crystal violet for 20 min, excess stain was rinsed with distilled water, and adherent materials were solubilized with 200 μl of 0.2 M sodium hydroxide for 1 h at 85°C. The optical density at 544 nm (OD₅₄₄) was measured. Each experiment was performed thrice, and the mean percent inhibition and standard deviations were calculated. The relative inhibition of biofilms expressed as mean percentage was determined by the following formula: percent inhibition = 100 − [(OD₅₄₄ of drug well/OD₅₄₄ of positive-control well) × 100].

The killing effects of RBx 11760, linezolid, and vancomycin were also assessed on glass-adherent bacteria as described earlier (21). The adherent inocula were prepared using sintered glass beads with pore sizes of 60 to 300 μm. The adherent inocula (relative biomass coated on the bead) were 5 × 10⁶ to 5 × 10⁷ CFU/bead. RBx 11760, linezolid, and vancomycin were prepared from 0.125 to 32 mg/liter in 5 ml TSB in triplicate. Three beads were tested per concentration. These tested concentrations (0.125 to 32 mg/liter) for RBx 11760, linezolid, and vancomycin were 0.25× to 64×, 0.125× to 32×, and 0.0625× to 16× MIC, respectively, against MRSE ATCC 35984. Each tube was incubated with one sintered glass bead coated with adherent bacteria at 37°C for 24 h. After incubation, nonadherent bacteria were washed from the bead and the number of CFU per bead was determined. The final concentration of each drug required to kill adherent bacteria was calculated after performing three independent experiments. The mean log₁₀ killing and standard deviations were calculated for each concentration.

The mean log₁₀ killing of adherent bacteria was calculated as the geometric mean log₁₀ CFU of untreated beads minus the geometric mean log₁₀ CFU of drug-treated beads. Three independent experiments were done to determine the killing concentration of adherent bacteria. The mean log killing and standard deviations were calculated for each concentration, and the data were presented for 1× to 16× MIC for head-to-head comparison.

Generation of linezolid-resistant strains.

In vitro development of resistance for linezolid was carried out against clinical and ATCC strains of S. aureus (n = 11) and Enterococcus (n = 22). Staphylococcus and Enterococcus strains were cultivated in Mueller-Hinton II broth, containing sub-MICs and MICs of linezolid on day 1, and were subsequently observed daily for visual turbidity. Each day, the cultures growing in the highest drug concentration were repeatedly passaged on the same and next higher concentration of drug along with a drug-free control for up to 30 passages. The selected resistant isolates were checked for stability by subculturing on drug-free medium plates. MICs for linezolid, tedizolid, and RBx 11760 were checked by the broth microdilution method against these resistant cultures to confirm the development of resistance.

In vitro transcription/translation assay.

Cell-free in vitro translation of the luciferase gene was performed with a bacterial or mammalian transcription/translation system as described earlier (26). The assay was performed in the presence of various concentrations of RBx 11760 and linezolid using commercially available bacterial (Escherichia coli S30) and a mammalian (rabbit TNT) kit (Promega, USA) per the manufacturer's protocol. The percent inhibition of luciferase activity by RBx 11760 or linezolid was used to calculate the 50% inhibitory concentrations (IC₅₀).

Experimental animals.

Normal immunocompetent Swiss albino mice were used for all in vivo efficacy experiments except for the mouse thigh infection model, wherein neutropenic Swiss albino mice were used. For pharmacokinetics (PK) studies, normal immunocompetent Swiss albino mice and Wistar rats were utilized. Ethical review of this study was carried out by the Institutional Animal Ethics Committee (IAEC) of Daiichi Sankyo India Pharma Private Limited and Ranbaxy Research Laboratories, Gurgaon, India, and the IAEC approved all experimental protocols for the use of animals. The study was conducted under the strict guidelines set out by the Committee for the Purpose of Control and Supervision on Experiments on Animals (CPCSEA), India, for the correct implementation of animal care and experimentation. Specific-pathogen-free animals were used for the in vivo studies, and animals were allowed 7 days of acclimatization before commencing experiments.

Mouse groin abscess model.

The in vivo efficacy of RBx 11760 against MRSA H-29 was determined in the murine groin abscess infection model. Each experimental group was allocated with 6 animals per group. Animals were allocated to each group by random selection based on body weight. The hair was removed from the groin area of the animals using an electric hair remover (TSE Systems, Germany) a day before infection. During the development and optimization of mouse groin abscess, we tried inocula starting from 10⁴, 10⁵, 10⁶, 10⁷, and 10⁸ per mouse. Finally, the optimum inocula for MRSA H-29 in mouse groin infection was found to be <10⁵ to 10⁷ CFU/mouse. We used 0.6% semisolid nutrient agar, an abscess-promoting agent which helps in development of abscess by activating the host defense and immune infiltration. The overnight-grown culture of MRSA H-29 was diluted 1:20 in normal saline and mixed with 0.6% semisolid nutrient agar at a 1:10 ratio. The bacterial suspension was subcutaneously injected into the groin area at a volume of 0.5 ml/mouse for a total of 4.5 × 10⁶ CFU per mouse. The formulations of RBx 11760 and linezolid were prepared in 0.25% (wt/vol) methylcellulose (Sigma-Aldrich Co. LLC) for in vivo efficacy studies. Treatment with RBx 11760 (10, 20, or 40 mg/kg of body weight) and linezolid (20 or 40 mg/kg) was started 2 h postinfection and given twice a day orally for 4 consecutive days. Twelve hours after the last dose, animals were euthanized; the abscess content along with skin was excised aseptically in 1 ml normal saline and homogenized. The resultant homogenate was used for the determination of viable bacterial count.

Foreign-body mouse biofilm infection model.

The foreign-body biofilm infection model in mice was performed as described by others (27, 28). Briefly, overnight-grown culture of S. aureus Xen-29 and MRSE ATCC 35984 was reinoculated in TSB supplemented with 2%, wt/vol, glucose (TSBG) and grown for 2 h at 200 rpm at 37°C. The turbidity of cultures was adjusted to an OD₆₀₀ of 0.25 and further diluted 1:10 in fresh TSB supplemented with 2% glucose. Fourteen-gauge Teflon intravenous catheter pieces (Abbocath-T; Vet Supply, Vancouver, Canada) were incubated with culture for 3 h at 37°C to form biofilm as described previously (23). In order to remove the nonadherent bacteria, the catheter pieces were washed thrice with fresh TSB. The number of bacteria coated per catheter piece was 1 × 10⁵ to 5 × 10⁵ CFU/catheter. Preanesthetized mice (0.05-ml mixture of 5 mg/kg xylazine and 100 mg/kg ketamine) were infected in the flank region (n = 6 mice/group) with two catheters per animal through a trocar and cannula. The skin opening was closed with skin adhesive (J & J, USA). Treatment started 3 days postimplant, and mice were treated with RBx 11760, linezolid, or rifampin at a dose of 40 mg/kg twice a day orally for three consecutive days. At the end of treatment, mice were euthanized and the catheter was surgically removed for enumeration of bacteria by a conventional viable count method. To remove the adherent bacteria, the catheter piece was placed in a tube containing 3 ml PBS, placed in an ultrasonic bath, and sonicated for 5 min, followed by vortexing for 1 min (27).

Murine pyelonephritis model.

In vivo efficacy of RBx 11760 against E. faecalis 427 was evaluated in the murine pyelonephritis model as described earlier (29). Overnight-grown culture of E. faecalis 427 on a TSA plate was suspended in PBS and adjusted to 0.5 McFarland. This inoculum was further diluted to get 2 × 10⁶ to 5 × 10⁶ CFU per 0.2 ml. Seven days prior to infection, mice were given a single 200-μl intravenous dose of 2% carrageenan (Sigma, USA). Mice (n = 6/group) were infected intravenously with 2 × 10⁶ to 5 × 10⁶ CFU/mouse (0.2 ml/mouse) of E. faecalis 427, and oral treatment was started 2 h postinfection either twice or four times a day for 2 consecutive days with RBx 11760 (20 mg/kg) and linezolid (20 mg/kg). Approximately 12 h after the last dose, mice were sacrificed and their kidneys were excised and homogenized in 1 ml PBS. The homogenate was diluted and plated on bile esculin agar (BEA; Difco, BD) to determine the bacterial load. Data from two independent experiments were pooled and analyzed.

Efficacy of RBx 11760, tedizolid, and linezolid in neutropenic mouse thigh infection.

The method described earlier for mouse thigh infection was followed for this model (30). Mice were rendered neutropenic (neutrophil count of <100 cells/mm³) with two intraperitoneal (i.p.) injections of cyclophosphamide 4 days and 1 day prior to infection at dose rates of 150 mg and 100 mg per kg. Overnight-grown culture of MRSA 562 on MHB was adjusted to 0.5 McFarland. The suspension was then diluted 1:100 in fresh MHB, and 100 μl (5 × 10⁵ CFU/thigh) was injected intramuscularly in the thigh muscles of the mice (n = 5/group). The bacterial inocula were confirmed by quantitative culture analyses. Treatment started 2 h postinfection, the pretreatment control group was immediately sacrificed, and tissue homogenates of thigh muscles were quantitatively cultured to determine the bacterial density in the thigh muscle. RBx 11760 and tedizolid were administered at 10, 20, and 40 mg/kg, and linezolid was administered at 80 mg/kg twice a day. RBx 11760 was administered by the oral route, and tedizolid was administered by the i.p. route. RBx 11760 and linezolid were suspended in 0.25% methyl cellulose, and tedizolid was prepared in phosphate-buffered saline.

PK study.

Male Swiss albino mice and Wistar rats were fasted overnight and then administered a single dose of RBx 11760 at 10 mg/kg by oral (per os [p.o.] and intravenous (i.v.) routes in mouse and at 5 mg/kg i.v. and 25 mg/kg p.o. in Wistar rat. Noninfected animals were used in PK studies. Sparse blood sampling in mouse (use of multiple animals to generate a complete PK) and serial blood sampling in rat were carried out at different time points up to 24 h postdose. Plasma was harvested by centrifugation of blood and kept at −80°C until analysis. RBx 11760 concentrations in plasma samples were determined using liquid chromatography-tandem mass spectrometry (LC-MS/MS). RBx 11760 was extracted from plasma samples by acetonitrile precipitation. Briefly, 50 μl Milli-Q-grade water was added to 20 μl of plasma samples. To this, 50 μl acetonitrile and 350 μl internal standard solution (1 μM niflumic acid in acetonitrile) were added. The contents were mixed with 96-well shakers and filtered through a Captiva 96-well polypropylene filter plate. The supernatant (8 μl) was injected into LC-MS/MS (AB Sciex API 4000 with electrospray ionization-MS probe kept at 450°C) coupled with an Acquity ultraperformance LC (UPLC; Waters) system in gradient mobile flow. A Sunniest HT C₁₈, 50- by 2.1-mm, 2-μm column was used at a flow rate of 0.8 ml/min. The mobile phase consisted of (i) 100 mM ammonium acetate-acetonitrile-formic acid (50:950:2) and (ii) 100 mM ammonium acetate-Milli-Q-grade water-acetonitrile-formic acid (50:900:50:2). The selected precursor and product ion for RBx 11760 were m/z 398.2 and 268. Analyst software, version 1.4, was used to process the sample in quantitation mode using a (1/x) weighted linear regression analysis.

Structural modeling studies for interaction with ribosome.

The crystal structure of the E. coli 50S ribosome (PDB entry 2AW4) was used for molecular docking of linezolid and RBx 11760 (31). The proposed structural effects of the ribosomal mutations in this study were inferred by using the coordinates of the Deinococcus radiodurans and Haloarcula marismortui linezolid-bound 50S ribosomal subunit crystal structures (PDB entries 3DLL and 3CPW) (26). We first validated the docking simulation by reproducing the reported binding mode of linezolid in the ribosome. MGLTools 1.5.4 was used to prepare the receptor and ligands (http://mgltools.scripps.edu). Prior to the docking, three-dimensional (3D) structures of linezolid and RBx 11760 were generated with the help of Marvin Sketch and were subjected to optimization for their geometry using the energy minimization in UCSF Chimera software. All ligands were treated as being flexible. Gasteiger Huckel charges were assigned to ligands as well as rRNA. The Autogrid 4.2 program of the Autodock suite (http://autodock.scripps.edu/) was used to generate the grid maps. Grid dimensions of 90 by 82 by 86, enclosing the reported important residues involved in inhibitor binding, were constructed in x, y, and z directions, respectively. The Lamarkian genetic algorithm (LGA) was used to generate 50 conformations of each ligand with the help of Autodock 4.2. Docking results were analyzed using MGLTools and UCSF Chimera 1.6.

Statistical analysis.

All data were analyzed using GraphPad Prism (version 5.03; GraphPad Software, San Diego, CA). The statistical significance of treated (RBx 11760 and comparator antimicrobial agents) and untreated control (preinfection control or postinfection, untreated control) were calculated by nonparametric Mann-Whitney analysis. The statistical significance of killing of adherent bacteria and percent inhibition were determined and data were compared using one-way analysis of variance (ANOVA) with Dunnett's multiple-comparison posttest. A P value of <0.05 was considered statistically significant. The bar chart and scatter line graph (for kill kinetics) data, presented as means ± SD (standard deviations), were calculated using Microsoft Office Excel. The pharmacokinetics data were analyzed using the NCA module of WinNonlin professional software (version 4.1). The pharmacokinetic parameters, i.e., maximum concentration of drug in serum (C_max), time to maximum concentration of drug in serum (T_max), plasma clearance (CL_p), volume of distribution at steady state (V_ss), half-life (t_1/2), and area under the concentration-time curve (AUC), were determined. Percent bioavailability (F) was calculated as [(AUC_p.o. × dose_i.v.) × 100]/[(AUC_i.v. × dose_p.o.)].

RESULTS

In vitro antimicrobial susceptibility testing.

RBx 11760 showed in vitro antibacterial activity against S. aureus, S. epidermidis, E. faecalis, E. faecium, and S. pneumoniae. The MIC₅₀s, MIC₉₀s, and MIC ranges against these pathogens are presented in Table 1. Linezolid-resistant (LZD^r) mutants of S. aureus and enterococci were generated from wild-type strains after 30 passages in linezolid-containing growth medium. MIC₅₀s and MIC₉₀s of RBx 11760 were 4- to 8-fold lower than those of linezolid against LZD^r strains of S. aureus and enterococci. The MIC₉₀s of RBx 11760 and linezolid against laboratory-generated LZD^r S. aureus (n = 11) as well as Enterococcus (n = 22) strains were 8 and 32 mg/liter, respectively. RBx 11760 also showed in vitro activity against VISA strains. MIC ranges of RBx 11760 and vancomycin against VISA (n = 15) strains were 0.5 to 1 and 4 to 8 mg/liter, respectively. RBx 11760 did not show inoculum effects at 10⁵ (low) and 10⁷ (high) inocula; however, linezolid and vancomycin showed moderate increases in MIC values at high inocula of 10⁷ CFU/ml (see Table S1 in the supplemental material). The pH effect study of RBx 11760 and linezolid showed minimal variation in MIC at an inoculum of 10⁵ CFU/ml across a medium pH range of 5.2 to 8.7 against S. aureus or Enterococcus strains. However, vancomycin showed 2- to 4-fold variation in MIC under the same experimental conditions, and erythromycin showed high MIC at low pH even at a low inoculum of 10⁵ CFU/ml (see Table S2 in the supplemental material). RBx 11760 and linezolid displayed equipotent in vitro activity in the presence of 50% human plasma (see Table S3). RBx 11760 showed 3-h PAE at 4× MIC (4 mg/liter), while linezolid showed the same effect at 8× MIC (16 mg/liter) against MRSA H-29 (see the supplemental material for methods). RBx 11760 (4 mg/liter) exhibited PAE at 3.2 h against MRSE, whereas linezolid (8 mg/liter) showed PAE at 2.5 h at 8× MIC (see Table S4).

TABLE 1.

Antimicrobial susceptibility of Gram-positive pathogens to RBx 11760, tedizolid, linezolid, and vancomycin

Bacterial strain	MIC (mg/liter) ofg:
	RBx 11760			Tedizolid			Linezolid			Vancomycin
	50%	90%	Range	50%	90%	Range	50%	90%	Range	50%	90%	Range
S. aureusa (n = 100)	1	2	0.25–4	0.5	0.5	0.25–1	2	4	1–4	1	2	0.5–4
S. epidermidisb (CoNS) (n = 40)	0.25	0.5	0.25–0.5	0.25	0.5	0.25–1	2	2	0.5–2	1	2	0.5–4
VISA (n = 15)	1	1	0.5–1	0.5	0.5	0.25–1	1	2	0.5–4	4	8	4–8
Linezolid-resistant S. aureus (n = 11)	2	8	2–8	2	8	2–8	16	32	8–64	ND	ND	ND
Enterococcusc (n = 65)	0.5	1	0.25–2	0.5	1	0.25–2	2	4	1–8	2	4	1–4
Vancomycin-resistant Enterococcusd (n = 31)	0.5	1	0.25–2	0.5	1	0.25–2	2	2	1–4	32	32	16–64
Linezolid-resistant Enterococcuse (n = 22)	4	8	2–8	ND	ND	ND	16	32	8–64	ND	ND	ND
S. pneumoniae (n = 94), including resistant strainsf	0.25	0.5	0.06–0.5	0.25	1	0.25–2	2	2	0.5–2	0.5	1	0.5–1

^aMSSA (n = 54) and MRSA (n = 46).

^bMethicillin-sensitive Staphylococcus epidermidis (MSSE) (n = 11) and MRSE (n = 29).

^cEnterococcus faecalis (n = 39) and Enterococcus faecium (n = 26).

^dE. faecalis (n = 10) and E. faecium (n = 21).

^eE. faecalis (n = 14) and E. faecium (n = 8).

^fPenicillin (n = 27) and macrolide resistance (n = 25).

^gND, not done.

RBx 11760, tedizolid, and linezolid showed very low frequencies of resistance, i.e., <3 × 10⁻¹⁰ against MRSA 562 at 4× MIC, and the positive control, rifampin, showed high frequencies of resistance of 1.1 × 10⁻⁷ and 2.5 × 10⁻⁷ at 4× and 8× MIC, respectively (see Table S5 and methods in the supplemental material).

Time-kill kinetics.

The time-kill kinetics studies of RBx 11760, tedizolid, and linezolid were performed against MRSA H-29, MRSE ATCC 35984, E. faecalis 427, and S. pneumoniae 5051. The results from three independent experiments are summarized in Fig. 2A to D. All three oxazolidinones exhibited bacteriostatic effect against MRSA, MRSE, and E. faecalis but not S. pneumoniae. RBx 11760 showed 2-log₁₀ kill at 4× MIC, while tedizolid and linezolid showed 2-log₁₀ and 1.4-log₁₀ kills, respectively, at 16× MIC (P < 0.05) against MRSA H-29. Against S. pneumoniae 5051, RBx 11760 showed cidal potential with 4.6-log₁₀ kill at 4× MIC, compared to a 2.42-log₁₀ and 1.95-log₁₀ reduction for tedizolid and linezolid at 16× MIC as early as 6 h (P < 0.05). Similarly, RBx 11760, linezolid, and tedizolid showed 2.4-log₁₀, 2.1-log₁₀, and 2.0-log₁₀ kills at 16× MIC against MRSE ATCC 35984. However, they did not show significant bacterial killing beginning at 0 h at any concentration against E. faecalis 427 (P > 0.05), indicating stasis.

FIG 2.

Time-kill kinetics study of RBx 11760, tedizolid, and linezolid against MRSA H-29 (A), MRSE ATCC 35984 (B), E. faecalis 427 (C), and S. pneumoniae 5051 (D). Graphs represent means from three independent experiments ± standard deviations (SD).

Effect of RBx 11760 on biofilm-producing and glass-adherent bacteria.

The effect of RBx 11760 on the biofilms producing MRSE ATCC 35984 is presented in Fig. 3A. RBx 11760 inhibited biofilm production in a concentration-dependent manner and exhibited a better effect than linezolid and vancomycin at 4× (P < 0.01) and 8× MIC (P < 0.001). RBx 11760 displayed activity similar to that of vancomycin and linezolid at 1× MIC and 2× MIC against glass-adherent MRSE ATCC 35984; however, at 4×, 8×, and 16× MIC, it showed significantly better inhibition of glass-adherent bacteria than linezolid and vancomycin (Fig. 3B).

FIG 3.

Percent inhibitory effect of RBx 11760, linezolid, and vancomycin on biofilm production (A) and killing effect of RBx 11760, linezolid, and vancomycin on glass-adherent bacteria (mean log₁₀ kill of glass-adherent bacteria ± SD) (B) when drug exposure was initiated at 4 h postinoculation of MRSE ATCC 35984. The statistical significance of the indicated P values was determined as <0.05 (*), <0.01 (**), <0.001 (***), and >0.05 (not significant [ns]).

In vitro transcription/translation assay.

The sub-MIC level of RBx 11760 resulted in selective inhibition of protein synthesis (see Fig. S1A and methods in the supplemental material). RBx 11760 and linezolid specifically inhibited bacterial protein synthesis, and we determined their IC₅₀s against bacterial (E. coli S30) and mammalian (rabbit TNT) protein synthesis. The IC₅₀s of RBx 11760 and linezolid against bacterial and mammalian protein synthesis and the safety window are presented in Table 2.

TABLE 2.

IC₅₀s of RBx 11760 and linezolid against bacterial and mammalian ribosomes in a cell-free in vitro transcription/translation system

Compound	IC50a (μg/ml)		Safety window (mammalian IC50/bacterial IC50)
	Bacterial	Mammalian
RBx 11760	7.94 ± 3.97	6,476.96 ± 2,483.5	815.7
Linezolid	47.68 ± 10.33	5,125.94 ± 437.09	107.5

^aValues are means ± SD from three independent experiments.

Mouse groin abscess model.

In the mouse groin abscess model, the mean (±SD) log₁₀ bacterial count of the untreated control (96 h) mice was 7.60 ± 0.37, which was an increase from the value for the untreated control group infected for 2 h (termed the 2-h control), 6.39 ± 0.39 (Fig. 4A). The treatment with linezolid at 20 and 40 mg/kg achieved a static effect, while RBx 11760 treatment showed 1-log₁₀ killing for the 2-h control even at 10 mg/kg. Moreover, treatment with RBx 11760 at 20 and 40 mg/kg resulted in significant decreases in bacterial loads of 1.36 and 1.76 log₁₀ CFU/groin abscess tissue, respectively (Fig. 4A). The simulated concentration-time profile of RBx 11760 at 10 mg/kg, twice a day by oral dose for 4 days, yielded an AUC_0–24 of 66.9 μg · h/ml and an AUC/MIC ratio (MIC of MRSA H-29, 1 mg/liter) of 66.9 (see Fig. S2 in the supplemental material).

FIG 4.

(A) In vivo efficacy of RBx 11760 and standard drugs against MRSA H-29 in murine groin abscess model. Treatment was started 2 h postinfection and administered twice a day (BID) for 4 consecutive days. All treated groups were compared to the 2-h control. (B and C) In vivo efficacy of RBx 11760 and standard drugs against MRSA Xen-29 (B) and MRSE ATCC 35984 (C) in a murine foreign-body biofilm infection model. Treatment was started 3 days postinfection and administered at 40 mg/kg twice a day for 3 consecutive days. All treated groups were compared with precontrol (3 days) and postcontrol (6 days) values. (D) In vivo efficacy of RBx 11760 and standard drugs against E. faecalis 427 in a murine pyelonephritis model. Treatment was started 2 h postinfection and administered at 20 mg/kg BID or four times a day (QID) for 2 consecutive days. Each bar represents the means ± standard deviations for 6 mice. For statistical analysis, all treated groups were compared to the untreated controls infected for 2 h or 48 h. The statistical significance of the indicated P values was determined as <0.05 (*), <0.01 (**), <0.001 (***), and >0.05 (ns). BID, twice daily; QID, four times a day.

We have also investigated the efficacy of RBx 11760 against a Panton valentine leukocidin (PVL)-positive community-acquired MRSA (CA-MRSA) strain. The bacterial load recovered at the beginning of treatment at 2 h after infection was 6.22 log₁₀, which increased to 8.28 log₁₀. RBx 11760 showed significant reduction of 1.63 log₁₀ compared to the 2-h control (P < 0.01), and linezolid showed a reduction of 0.503 compared to the 2-h control (P > 0.05). However, compared to the 96-h late control, RBx 11760 and linezolid showed reductions of 3.67 log₁₀ and 2.56 log₁₀, respectively. RBx 11760 also showed less abscess formation even at half the dose of linezolid (see Fig. S3A to D in the supplemental material).

Foreign-body mouse biofilm infection model.

The mean bacterial log₁₀ count of S. aureus Xen-29 and MRSE ATCC 35984 on each implanted catheter was 5.57 ± 0.38 and 6.61 ± 0.20 when the treatment was started on day 3. After 6 days postinfection, average log₁₀ CFU counts of 7.48 ± 0.37 and 9.08 ± 0.46 were recovered from S. aureus Xen-29- and MRSE ATCC 35984-coated catheters, respectively, from untreated control animals. Treatment with linezolid and rifampin at 40 mg/kg, twice a day for 3 days, resulted in no advantage over stasis, while the same dose of RBx 11760 demonstrated a significant killing (*, P < 0.05) of approximately 1-log₁₀ CFU/catheter from the initial bacterial load (precontrol day 3 group) against both organisms (Fig. 4B and C).

Murine pyelonephritis model.

In the pyelonephritis model, RBx 11760 and linezolid showed comparable efficacy (Fig. 4D). At 20 mg/kg, twice-daily treatment with both RBx 11760 and linezolid did not show any significant reduction in bacterial count compared to the initial 2-h control group. In contrast, treatment with RBx 11760 at 20 mg/kg four times a day yielded a 0.5-log₁₀ reduction from value for the initial control while linezolid resulted in stasis (Fig. 4D). Compared to 48-h postcontrol RBx 11760 at 20 mg/kg four times a day, a 3.82-log₁₀ reduction (P < 0.05) was seen compared to a 3.27-log₁₀ reduction for linezolid (P < 0.05). When results were statistically compared using Mann-Whitney U test, the groups did not differ significantly (P > 0.05), indicating comparable efficacy.

Efficacy of RBx 11760, tedizolid, and linezolid in neutropenic mouse thigh infection.

In a neutropenic mouse thigh infection model, the mean (±SD) log₁₀ bacterial count for mice was 6.03 ± 0.26, which increased to 8.98 ± 0.38 at 26 h postinfection. RBx 11760 showed static effect at half the dose, i.e., 20 mg/kg, compared to tedizolid, which showed stasis at 40 mg/kg in this experiment. On the contrary, linezolid displayed a 0.53-log₁₀ increase in bacterial count compared to the 2-h controls even at 80 mg/kg (Fig. 5).

FIG 5.

Comparative efficacy of RBx 11760, tedizolid, and linezolid against MRSA-562 in neutropenic mouse thigh infection model. Treatment was started 2 h postinfection and administered twice a day for 1 day. Each bar represents the means ± standard deviations for 6 mice. For statistical analysis, all treated groups were compared to the untreated control group that was infected for 2 h. The statistical significance of the indicated P values was determined as <0.05 (*) and >0.05 (ns).

Pharmacokinetic studies.

RBx 11760 showed low plasma clearance of 11.5 and 3.6 ml/min/kg with V_ss of 0.74 and 0.4 liters/kg in mouse and rat, respectively (Table 3). Long terminal half-lives of 9.3 h and 3.2 h in mouse and rat, respectively, suggest long duration of action. RBx 11760 showed 60% and 72% oral bioavailability in mouse and rat, respectively, following administration of suspension formulation.

TABLE 3.

Pharmacokinetic study of RBx 11760 in Swiss albino mouse and Wistar rat by oral and intravenous routes

Parameter	Value fora:
	Mouse		Rat
	10 mg/kg i.v.	10 mg/kg p.o.	5 mg/kg i.v.	25 mg/kg p.o.
Cmax (μg/ml)	22.4	4.50	15.5 ± 2.8	6.6 ± 1.3
Tmax (h)		0.5		2.7 ± 1.2
AUC0–24 (μg · h/ml)	14.4	8.6	23.8 ± 4.8	86.1 ± 34.7
t1/2 (h)	9.3		3.2 ± 1.1
CLP (ml/min/kg)	11.5		3.6 ± 0.8
Vss (liter/kg)	0.74		0.4 ± 0.1
MRTb (h)	1.1		2.0 ± 0.1
F (%)		60		72

^aValues are presented either as means or means ± SD for n = 3 animals.

^bMRT, mean residence time.

Structural modeling studies for interaction with ribosome.

Initial docking of linezolid produced conformation similar to that of the previously reported binding mode (25). As reported for the crystal structure, nucleotide residues G2061, A2451, A2503, U2504, G2505, and U2506 are important residues which are involved in interaction with linezolid (Fig. 6A). Further, docking of RBx 11760 in the active site of the 50S ribosomal subunit revealed a binding conformation similar to that of linezolid (Fig. 6C and 7). We found that a few selected top conformations of RBx 11760 make an H bond with ribose moiety of U2584 and OP2 oxygen of G2505 (Fig. 6B); however, a few other conformations (with slightly less binding energy) from the same cluster showed H bonds with 5′ orientation oxygen of U2504 (Fig. 6D). The binding energy of RBx 11760 (−8.80 kcal/mol) was found to be more than that of linezolid (−6.62 kcal/mol). The preferred and top-ranked conformation indicates the H-bonding interaction with ribose sugar moiety U2584.

FIG 6.

Interaction of linezolid and RBx 11760 with active-site residues of 50S ribosomes of E. coli. Overlay of docked conformation (yellow) and cocrystal conformation (magenta) of linezolid (A) and RBx 11760 (green) (B) into active sites of 50S ribosomes of E. coli. Merged conformation of linezolid with RBx 11760 (C) and other low-scoring conformations of RBx 11760 in different orientations (D) within the active site of 50S ribosomes of E. coli. H bonds are shown by dotted green lines.

FIG 7.

Surface image of docked conformation of linezolid and RBx 11760 in the same active site of 50S ribosomes of E. coli.

DISCUSSION

The emergence of multidrug resistance in Gram-positive bacteria has dramatically limited the therapeutic options for hospital- as well as community-acquired infections. Despite extensive antimicrobial research in the last 50 years, the drugs used for hospital-acquired infections are still limited (32, 33). In the present study, antibacterial activity of RBx 11760, a bi-aryl oxazolidinone, was evaluated against a panel of Gram-positive bacteria. The MIC₅₀s and MIC₉₀s of RBx 11760 were significantly lower than those of linezolid against S. aureus (2-fold), S. epidermidis (4- to 8-fold), Enterococcus (4-fold), S. pneumoniae (4- to 8-fold), and LZD^r strains of S. aureus and Enterococcus (4- to 8-fold), indicating improved activity of RBx 11760 against methicillin-susceptible S. aureus (MSSA), MRSA, MRSE, Enterococcus, S. pneumoniae, VISA, and VRE. RBx 11760 displayed comparatively potent activity against MSSA, MRSA, E. faecalis, and E. faecium even at higher inoculum levels. When checked at different pH levels, RBx 11760 and linezolid did not show pH effect on the activity; however, vancomycin and erythromycin did show pH effect on activity. RBx 11760 showed more potent activity against S. pneumoniae than that of tedizolid and linezolid. RBx 11760 and linezolid displayed no shift in MIC in the presence of human plasma against staphylococci. RBx 11760 maintained a balanced and potent antibacterial profile and demonstrated significant advantages over linezolid against a variety of antimicrobial-resistant strains.

In order to assess the killing potential of RBx 11760, the time-kill kinetics studies were performed against MRSA, MRSE, E. faecalis, and S. pneumoniae, owing to their importance in hospital-acquired infections. RBx 11760, tedizolid, and linezolid exhibited bacteriostatic activity, except that 4× MIC of RBx 11760 and 16× MIC of tedizolid and linezolid against MRSA H-29 at 8 h exhibited an ∼2-log₁₀ kill. RBx 11760 showed better killing potential at 4× MIC than tedizolid and linezolid at the same concentration against S. pneumoniae 5051. PAE is an important pharmacodynamic parameter that can influence the dosing regimens of an antibiotic (20). In PAE experiments, RBx 11760 showed prolonged PAE against MRSA and MRSE compared to linezolid at the same concentration. However, RBx 11760 and linezolid showed comparable PAE against E. faecalis. The long PAE of aminoglycosides are thought to contribute to clinical efficacy with once-daily dosing (34). Linezolid exhibits moderate concentration-dependent in vitro PAE against Gram-positive pathogens, supporting its twice-daily dosing in humans. The long PAE of RBx 11760 could be a valuable parameter in human clinical settings. The concentrations selected for in vitro PAE determination are achievable under in vivo conditions; therefore, it is presumed that the results obtained are clinically relevant. However, in vivo PAE, which is longer than in vitro PAE, needs to be generated in animal models before making a conclusive determination of its effect on dosing regimens (34, 35).

RBx 11760, tedizolid, and linezolid displayed a very low frequency of resistance against MRSA 562, while the positive control, rifampin, showed a high frequency of resistance at 4× and 8× MIC. These data suggest that there is little chance for spontaneous resistance development to the oxazolidinone class of compounds, including RBx 11760 (36). However, in our serial passage study for the potential for resistance development, linezolid in an S. aureus mutant generated MICs of 16 to 32 mg/liter, whereas the same analyses performed with RBx 11760 yielded mutants with MICs of 2 to 8 mg/liter. This indicated that the lower cross-resistance potentials of RB11760 than those of linezolid biofilm-mediated adherence and bacterial biofilms are critical factors in the pathogenesis of S. epidermidis infections of medical devices (37, 38). RBx 11760 showed significant reduction in biofilm production by MRSE ATCC 35984 compared to linezolid and vancomycin (P < 0.05). RBx 11760 not only displayed potent biofilm inhibitory activity but also exhibited superior activity against glass-adherent bacteria (P < 0.01). This is one of the unique characteristics of RBx 11760 compared to linezolid.

The effect of RBx 11760 on macromolecular synthesis inhibition of MRSA H-29 exhibited that the sub-MIC level of RBx 11760 resulted in selective inhibition of protein synthesis, while all other macromolecular syntheses, such as DNA, RNA, cell wall, and lipid syntheses, were almost unaffected (see Fig. S1 and methods in the supplemental material). Therefore, the inhibitory target of RBx 11760 is bacterial protein biosynthesis. In addition, RBx 11760 and linezolid act specifically on bacterial protein synthesis and RBx 11760 showed significantly better IC₅₀s than linezolid against bacterial protein synthesis, while RBx 11760 displayed very poor activity against mammalian protein synthesis. Therefore, RBx 11760 may show comparatively better safety profiles against mammalian protein synthesis.

Three types of resistance mechanisms, including mutations in the domain V region of the 23S rRNA gene, acquisition of the ribosomal methyltransferase gene cfr, and mutations in the genes encoding 50S ribosomal proteins, have been reported against oxazolidinones (39). In this study, we have shown through serial passage experiments of linezolid selection that the MIC of linezolid increased 32-fold (2 to 64 mg/liter) compared to 4-fold (1 to 4 mg/liter) for RBx 11760 (see Table S6 and methods in the supplemental material). The concentration of 4 mg/liter is not very high, and this concentration of RBx 11760 can easily be maintained in human plasma to reach clinical efficacy against linezolid-resistant strains by optimization of pharmacodynamic parameters. However, in RBx 11760 selection, the MIC of both RBx 11760 and linezolid increased to 64 mg/liter. This indicated that the frequency of a linezolid selection by a mutant was greater than that of RBx 11760 selection, which is in agreement with another report (40). In mutant sequence analysis (see Table S6), domain V 23S rRNA gene mutations revealed a novel mutant, T2374C, in rrn₆ in addition to the already reported mutations. The mutation G2576T in linezolid selection mutants observed in this study was also reported to be the common mechanism for linezolid resistance through 23S rRNA, with 63.5% frequency (15). This mutant study is a preliminary one, and detailed studies in the future will help to understand the fitness of RBx 11760 resistance reported in this study.

We evaluated the in vivo efficacy of RBx 11760 in a reproducible murine groin abscess model caused by MRSA, which resembles human localized skin and soft-tissue infections. The infection was maintained throughout the 4 days of the experiment in untreated control groups and also showed groin abscess formation similar to clinical abscess formation. In the groin abscess model, the bacterial count of the untreated control increased from that of precontrol animals and was maintained until day 4. This increase of bacterial load was also substantiated by the gross pathology of the abscess. RBx 11760 showed killing potential at half the dose of linezolid. This indicated that RBx 11760 was more efficacious than linezolid against skin and soft-tissue infection, and it showed excellent activity in a dose range study. The in vivo efficacy of RBx 11760 was AUC_0–24 driven in the mouse model (data not shown). Linezolid has also been evaluated in patients who have bacteremia, complicated skin and skin structure infections, and pneumonia that is associated with MRSA or VRE. The time the concentration of an antibiotic remains above the MIC (t_>MIC) and AUC_0–24/MIC ratios are reported to be highly correlated with cure. As reported previously, the static effect of linezolid against S. aureus was achieved at a mean 24-h AUC/MIC ratio of 83, which was almost equivalent to the finding of human clinical trials of a mean 24-h AUC/MIC ratio of 110 (30, 41–43). For the recently FDA-approved tedizolid phosphate, it was reported that the pharmacodynamic index most closely linked with efficacy was AUC over 24 h divided by the MIC in the mouse thigh infection model of MRSA (40). PVL is a cytotoxin produced by some strains of Staphylococcus which is associated with severe virulence and necrosis of skin and soft tissues and necrotic pneumonia. They are commonly associated with community-acquired MRSA (CA-MRSA). In order to determine the spectrum of in vivo efficacy of RBx 11760, we investigated the efficacy against an PVL-positive MRSA strain which causes severe necrosis. We observed that even at half the dose, treatment with RBx 11760 showed significant bacterial count reduction from the initial 2-h control (P < 0.001) compared to linezolid treatment. In addition, no abscess formation was observed in RBx 11760-treated animals, whereas significant abscess formation and gross tissue damage were observed in the untreated control mice followed by mild abscess formation in the linezolid-treated group (see Fig. S3 in the supplemental material).

In the foreign-body mouse biofilm infection model, the level of bacteria recovered from the implant catheters of postcontrol animals after 6 days was significantly higher than that recovered from the catheters of precontrol animals (3 days postinfection at the time of first dose). In the implanted catheter experiments, we observed that there was a significant advantage for RBx 11760. The treatment of linezolid and rifampin at 40 mg/kg twice a day for 3 days resulted in no improvement over stasis, while the same dose of RBx 11760 demonstrated a significant killing of approximately 1 log₁₀ CFU/catheter from the initial bacterial load against both organisms. Moreover, the recovered bacteria from treated animals were also checked for drug susceptibility. There was no change in MIC in the case of RBx 11760. In contrast, MRSE 35984, recovered from the linezolid- and rifampin-treated group, showed 2- and 8-fold higher MICs, respectively. These results demonstrated that while only two organisms were studied and more studies are needed, the relative in vitro performance of RBx 11760 was recapitulated in vivo. Linezolid resistance was reported in Gram-positive organisms mainly due to biofilm formation in blood and catheters (44). However, it will be interesting to observe the antibiofilm activities of RBx 11760 in the clinic.

The increasing incidence of complicated urinary tract infections (cUTIs) in hospitals due to drug resistance of Staphylococcus and Enterococcus at present is a matter of concern. Oxazolidinones active against Gram-positive bacteria may be considered for use in the treatment of cUTIs because of the presence of Gram-positive uropathogens (45, 46). In a mouse pyelonephritis infection model, RBx 11760 and linezolid showed comparable efficacy, resulting in stasis when administered at 20 mg/kg four times a day.

The low plasma clearance suggests good plasma stability of RBx 11760. Low to moderate volumes of distribution suggest that RBx 11760 is less distributed in organs and therefore is less likely to cause organ toxicity. Long terminal half-lives in mouse and rat, respectively, suggests long duration of action. Both long PAE and long terminal half-lives of RBx 11760 contributed to the prolonged activity of the drug. The long half-lives delayed the initiation of the PAE and prolonged the activity of the drug because it was cleared more slowly. High oral bioavailability of 60% and 72% was observed in mouse and rat, respectively, following administration of suspension formulation, supporting further development of RBx 11760. Pharmacokinetics in higher species would help to understand the molecule better.

In a molecular modeling study, RBx 11760 and linezolid occupied almost identical orientations in the active site of the 50S ribosome of E. coli. However, the formation of additional van der Waals interactions and H bonding by RBx 11760 contributed to better stability and strong interactions in the active site compared to those of linezolid, which may provide an explanation for its better antibacterial activity than that of linezolid.

In conclusion, RBx 11760 demonstrated better in vitro activity against the clinically relevant bacterial isolates by exhibiting 2- to 4-fold lower MICs than linezolid and MICs comparable to those of tedizolid. Pharmacokinetics of RBx 11760 exhibited good oral bioavailability with low plasma clearance and low to moderate volumes of distribution in mouse and rat. RBx 11760 showed superior activity against biofilm-producing bacteria in vitro and also translated the activity in a foreign-body-associated mouse biofilm model. This is one of the unique characteristics of RBx 11760 compared to linezolid (44). RBx 11760 showed a 1-log₁₀ kill from initial control at a 4-fold lower dose than that of linezolid in a mouse groin abscess model. In neutropenic mouse thigh infection, RBx 11760 showed static effect at half the dose of tedizolid. These data indicate that RBx 11760 can be explored as a new therapeutic agent against Gram-positive bacterial infections.

Funding Statement

This study was supported by Daiichi Sankyo India Pharma Private Limited (DSIN), Gurgaon, India. The funder had no role in study design, data collection and interpretation, or the decision to submit the work for publication.