L007-0069 kills Staphylococcus aureus in high resistant phenotypes

Abstract

Staphylococcus aureus, a common gram-positive pathogenic bacterium, is a main cause of hospital infection. The prevalence rate of methicillin-resistant S. aureus (MRSA) has made its treatment difficult in recent decades. Moreover, S. aureus in the highly tolerant format of biofilm or persister often renders infections refractory. Thus, developing new active compounds against resistant S. aureus is urgently needed. In this study, by a high-throughput screening assay, we identified a small molecule, L007-0069, that exhibited strong and effective bactericidal activity against S. aureus and its high resistance patterns, such as biofilms and persisters, with a low probability of inducing resistance. By molecular dynamics and fluorescent probe analysis, mechanistic studies revealed that the bactericidal activity of L007-0069 was mainly mediated by membrane disruption and metabolic disorder induction. Furthermore, L007-0069 showed effective anti-MRSA effects in vivo in both a wound infection model and a peritonitis–sepsis model, with no detectable toxicity observed at the therapeutic dosage. In conclusion, L007-0069 has the potential to become an alternative for the treatment of highly resistant S. aureus-related infections.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00018-022-04588-5.

Introduction

Staphylococcus aureus is a common gram-positive pathogenic bacterium that is involved in many kinds of diseases, including skin infections and soft tissue infections, septicemia, catheter-induced endocarditis, and osteomyelitis [1]. Antibiotics, such as penicillin and methicillin, were the initial effective choices against S. aureus, but because of their inappropriate use and high selective pressure, the widespread development of antibiotic resistance has emerged [2]. With the prevalence rate of methicillin-resistant S. aureus (MRSA) increasing year by year, it has become the most frequent drug-resistant microorganism identified in many countries and one of the most urgent health crises [3]. In addition, as vancomycin- and teicoplanin-resistant isolates of S. aureus have been sporadically reported worldwide [4, 5], the treatment of S. aureus infections has become a severe clinical challenge.

Bacterial biofilm and persister cells have gained high clinical recognition due to their tolerance to antimicrobial agents and subsequent recalcitrance or failure in antibiotic therapies [6]. Bacteria often adhere to biotic or abiotic surfaces, which enables them to form biofilms, surface-attached communities with bacteria embedded in an extracellular matrix [7]. S. aureus is the most frequent cause of biofilm-associated infections on indwelling medical devices [7]. Biofilm infections are often accompanied by the formation of persister cells, which are thought to be the cause of the recurrence of refractory bacterial infections in health care contexts. Persister cells constitute transiently phenotypic antibiotic-tolerant subpopulations that are capable of evading high concentrations of antibiotic attack by turning into a physiologically dormant state [8]. Thus, killing persister cells and eradicating biofilms are among the biggest challenges in antibacterial drug discovery.

A number of conventional antibiotics target bacteria by interfering with cell wall biosynthesis, DNA replication, protein synthesis, or metabolic pathways. However, these antibiotics are known to face various types of acquired bacterial resistance mainly mediated by target mutation or modification, antibiotic inactivation by extracellular enzymes, and antibiotic efflux pump overexpression [9]. To overcome bacterial resistance, it is urgent to develop new antibacterial agents with different structures and modes of action from those of conventional antibiotics. Bacterial membranes are important potential targets of antimicrobials because they can be disrupted independently by antimicrobials regardless of the bacterial growth state, and membrane disruptors could also reduce the likelihood of developing resistance [10]. For instance, studies have identified membrane-active compounds, including NH125, nTZDpa, and bithionol, that are efficacious against MRSA and its persister cells by interfering with bacterial membranes without inducing resistance [10, 11].

In the present study, we screened 5033 small molecules and identified a small molecule, termed L007-0063, and its optimized analog, L007-0069, which was effective against MRSA. We found that L007-0069 possessed rapid bactericidal and persister killing activities without resistance development. L007-0069 in combination with gentamicin (GEN) effectively inhibited MRSA biofilm formation and eradicated preformed biofilms. Molecular dynamics (MD) simulations and other mechanistic studies demonstrated that L007-0069 could penetrate and embed into the bacterial lipid bilayers, hence disrupting the bacterial cell membranes. The antibacterial activity against MRSA was further demonstrated in vivo in murine models.

Materials and methods

The summary of all experimental procedures is described in Figure S1.

Antimicrobial susceptibility test

The minimal inhibitory concentrations (MICs) of the compounds and antibiotics used in this study were determined using the broth microdilution method according to the Clinical Laboratory Standards Institute [12]. The bacteria in the log phase were diluted to a cell density of 1 × 10⁶ colony-forming units (CFU)/mL in MH broth. Next, 50 μL of twofold diluted antimicrobials and an equal volume of bacterial suspensions were added to a 96-well microplate and mixed. After 18 h of incubation at 37 °C, the MICs were defined as the lowest concentrations of reagents that completely inhibited the visual growth of bacteria. No bacterial colonies were grown on plates after 24 h of incubation, which served as the minimum bactericidal concentration (MBC). All tests were carried out in triplicate independently. For MIC determination involving daptomycin (DAP), the medium was supplemented with 50 µg/mL CaCl₂[13].

L007-0069 synthesis

The detailed synthesis route for L007-0069 is described in Supplementary Method 1.

Resistance induction by sub-MICs

MRSA ATCC 43300 and SAJ1 were inoculated in fresh MH broth containing an extended concentration gradient of L007-0069 or (CIP, positive control), and the MIC values were determined as described above. After 24 h of incubation, 5 µL of the 1/2 × MIC cultures was diluted 1:1000 into 5 mL of fresh MH for remeasurement of the MIC. The experiment was serially passaged for 20 days with two biological replicates. The changes in MICs were calculated by comparison to the initial MIC on the first day [14].

One-step resistance-inducing assay

A single colony was cultured overnight in TSB. Then, (1–5) × 10⁹ CFU were spread on cation-adjusted MH agarose plates in the presence of 2–8 × MICs of L007-0069 or rifampicin (RFP). The viable cells were counted after 48 h of incubation at 37 °C. The frequency of resistance was defined as the number of colonies that formed after 48 h divided by the initial viable cell count [15].

Cell membrane permeability determination by SYTOX Green staining

SYTOX Green is a nucleophilic fluorescent dye used to reflect the cell membrane integrity. Bacteria in mid-log growth phase (for non-persister cells) or stationary phase (for persister cells) were centrifuged, harvested, and suspended with 5 mmol/L HEPES (pH 7.2, 5 mmol/L glucose) to obtain an OD₆₃₀ of 0.05–0.2. Next, SYTOX Green (Thermo Fisher Scientific, United States), which permeates defective membranes, was added to a final concentration of 2 µM. After incubation at 37 °C for 30 min in the dark, 100 µL of the bacteria/SYTOX Green mixture with the indicated concentrations of L007-0069 was added to a 96-well black-walled plate. Vancomycin (VAN), CIP, DAP, and GEN (32 μg/mL) served as controls. The increase in fluorescence intensity was monitored every 5 min for a total of 30 min by using a spectrophotometer at an excitation/emission of 485/525 nm (PerkinElmer EnVision, USA) [16].

Cell membrane permeability determination by DiSC3(5) staining

DiSC3(5) is a membrane potential sensitive probe. S. aureus at mid-log growth phase was washed and resuspended in 5 mM HEPES to obtain an OD₆₃₀ of 0.05 and incubated with 2 μM DiSC3(5) (AAT Bioquest, USA) in the presence of 5 mM glucose and 100 mM KCl for 1 h at room temperature in the dark. A black, clear-bottom, 96-well plate was added to 90 μL of bacterial dye mixture, and 10 μL of L007-0069 small molecule at a final concentration of 2–8 μg/mL was added. The fluorescence intensity was immediately monitored every 30 s for a total of 5 min with excitation and emission wavelengths of 622 and 670 nm, respectively. Simultaneously, 4 μg/mL melittin and 0.1% DMSO were used as positive and negative controls, respectively [17].

PI staining

PI is a DNA binding fluorescent dye, which can easily penetrate the damaged cell membrane. Mid-log-phase growth S. aureus was washed and resuspended in 1 × PBS (pH 7.4), the OD₆₃₀ was adjusted to 0.5, and the cells were incubated with 10 μM PI for 30 min. Then, the indicated concentrations of L007-0069, 0.1% DMSO (negative control), or 4 μg/mL Melittin (positive control) were added and incubated for another 1 h before measuring fluorescence intensity at an excitation/emission wavelength of 535 nm/615 nm [13].

Laurdan staining

Laurdan staining was used to assess the fluidity of bacterial membranes. Briefly, overnight cultures of MRSA ATCC 43300 were diluted 1:1000 with fresh TSB. After incubation at 37 °C and 200 rpm to an OD630 of approximately 1.0, the bacterial suspension was further incubated with 10 µM Laurdan (MedChem Express, USA) at 37 °C for 10 min in the dark. The stained bacterial suspension was washed 4 times with 1 × PBS (pH 7.4). One hundred microliters of concentrated culture was mixed with 100 µL of PBS containing the desired concentration of L007-0069 in a black 96-well plate. After incubation for 30 min at room temperature, the fluorescence intensity of Laurdan was recorded using a microplate reader with an excitation wavelength of 350 nm and emission wavelengths of 435 nm and 490 nm, respectively. Laurdan GP was calculated using the formula GP = (I440 − I490)/(I440 + I490) [10].

Giant unilamellar vesicle (GUV) assay

GUV is biomembrane-mimicking giant single-molecule vesicles. FITC-labeled GUVs consisting of DOPC/DOPG = 7:3 were purchased from SunLipo Nano Tech. The GUV suspensions were centrifuged at 3000 rpm for 2.5 min to separate the large vesicles from the small vesicles. The precipitate was then collected and diluted 1:3 in a solution containing 1 volume of 100 mM sucrose and 6 volumes of 110 mM glucose. Forty-nine microliters of the diluted GUV suspension was placed in the dark at room temperature for 15 min until all GUVs had precipitated on the bottom. After adding 1 µL of L007-0069 at a final compound concentration of 2 × MIC, the GUVs were observed and imaged using a confocal laser scanning microscope (CLSM) [10].

Persister killing assay

The formation of persister cells was induced by culturing S. aureus at 37 °C and 200 rpm overnight to stationary phase [10]. Prepared persister cells were washed 3 times with 1 × PBS (pH 7.4) and diluted to approximately (4–5) × 10⁸ CFU/mL. Serially diluted L007-0069 was added to 4 mL of the persister suspension and incubated at 37 °C and 200 rpm. Meanwhile, 10 × MICs of VAN, CIP, DAP, and GEN were used as controls. Samples were washed and harvested at 0, 1, 2, 3, and 4 h, and the viable cells were calculated by CFU counting.

Alamar blue assay

The Alamar blue assay was used to determine the metabolic activity of the persister cells based on their ability to convert the purple non-fluorescent dye resazurin to the pink reduction product resorufin. S. aureus persisters (10⁶ CFU/mL) were treated with compound L007-0069 at 37 °C and 200 rpm for 6 h. The bacterial cells were then incubated with 10 μL of 50 μg/mL resazurin solution for 1 h at 37 °C. The metabolic conversion of resazurin to pink resorufin was quantified by measuring the absorbance at 571 nm [18].

Intracellular ATP determination

Intracellular ATP levels of MRSA ATCC 43300 were determined using the Enhanced ATP Assay Kit (Beyotime, Shanghai, China) following the manufacturer’s instructions. S. aureus was grown overnight at 37 °C and 200 rpm, washed, and resuspended in 1 × PBS (pH 7.4) to obtain an OD₆₃₀ of 0.5. After treatment with various concentrations (2–8 × MIC) of the L007-0069 compound for 30 min, the bacterial cultures were centrifuged at 12,000×g for 5 min at 4 °C, and the supernatant was discarded. The bacterial sediment was lysed with lysozyme and centrifuged to harvest the supernatant for intracellular ATP level determination. The detection solution was added to a 96-well plate and incubated for 5 min at room temperature. Subsequently, the prepared supernatant was dispensed into the wells and mixed quickly, and the luminescence intensity was recorded by a microplate reader.

ROS detection

The levels of ROS in bacterial cells treated with L007-0069 were detected by a 2′,7′-dichlorofluorescein diacetate (DCFH-DA, Beyotime, Shanghai, China) probe. The procedure was performed strictly following the manufacturer’s instructions. Briefly, MRSA ATCC 43300 was grown to mid-log-phase at 37 °C and 200 rpm, washed, and resuspended in 1 × PBS (pH 7.4), and the OD₆₃₀was adjusted to 0.5. DCFH-DA was added to a final concentration of 10 μM and incubated at 37 °C for 30 min. After washing three times with 1 × PBS, 90 μL of probe-labeled bacterial cells was added to a 96-well plate, followed by the addition of 10 μL of the desired concentration of L007-0069. After 30 min of incubation, the fluorescence intensity was measured at an excitation/emission wavelength of 488 nm/525 nm.

Drug combination against persister cells

Persister cells of S. aureus were prepared according to the method described above, washed with 1 × PBS and adjusted to a concentration of 4–5 × 10⁸ CFU/mL. Subsequently, the combined therapies of conventional antibiotics (VAN or DAP or GEN) with L007-0069 were added, and the residual number of persister cells was counted at specific time points using CFU counting.

Checkerboard assay

Synergistic activity between L007-0069 and antibiotics was determined through checkerboard assays and calculated by fractional inhibitory concentration indices (FICIs). Briefly, twofold serial dilutions of L007-0069 and antibiotics were prepared in MH medium. A total volume of 100 μL of the above mixtures in the presence of bacteria (ATCC 43300, 5 × 10⁴ CFU/well) was dispensed into a 96-well plate, creating an “8 × 8” matrix. At 16–18 h, the optical density was measured at 630 nm by a microplate reader. The formula of the FICI was calculated as follows:

$$\text{FICI}=\frac{{\text{MIC}}_{\text{A}}\text{(combination)}}{{\text{MIC}}_{\text{A}}\text{(alone)}}\text{+}\frac{{\text{MIC}}_{\text{B}}\text{(combination)}}{{\text{MIC}}_{\text{B}}\text{(alone)}}.$$

A synergistic effect was defined as FICI ≤ 0.5; an additive effect was defined as 0.5 < FICI ≤ 1; an indifference effect was defined as 1 < FICI ≤ 4; an antagonism effect was defined as FICI > 4 [16].

Biofilm inhibition and eradication assays

Biofilm inhibition. Overnight cultures of MRSA ATCC 43300 in TSB (100 µL each) containing a twofold dilution series of test compound L007-0069 (initial inoculum of 10⁶ CFU per well) were seeded into a 96-well plate. DMSO (1%) served as a solvent control. After 24 h of incubation at 37 °C under static conditions, the planktonic bacteria were removed by washing with PBS, and the biofilm was stained with 0.25% crystal violet (CV) for 15 min and then dissolved in 95% ethanol. The absorbance at 570 nm (A570) of the solution was measured with a microtiter plate reader to quantify the biofilm biomass [19].

Biofilm eradication

The overnight cultured bacterial suspension was inoculated into a sterile flat-bottom 96-well plate to form S. aureus biofilms. After 24 h of static incubation, the unbound cells were removed, and each well was washed twice with PBS. Fresh TSB containing serial dilutions of compound L007-0069 (100 μL) was added and incubated for 24 h at 37 °C. After treatment, the remaining biofilm biomass was measured by CV staining as described above [20].

Biofilm persister killing assay

To obtain persisters from biofilm cultures, MRSA ATCC 43300 at mid-log growth was adjusted to (1–2) × 10⁸ CFU/mL in brain heart infusion (BHI) broth. One hundred microliters of this bacterial suspension was transferred to a 96-well plate, which was then sealed and incubated at 37 °C in a humidified atmosphere for 24 h. After washing twice with 1 × PBS to remove planktonic bacteria, the biofilms were exposed to 100 μL of BHI broth containing RFP (100 × MIC) to induce the formation of biofilm persister cells. After 24 h of incubation at 37 °C, the bacteria were washed twice, and the residual adherent bacteria were added to 100 μL of 1 × PBS and sonicated for 5 min. The obtained persisters were then exposed to 200 μL PBS containing various concentrations of L007-0069 or conventional antibiotics (GEN, VAN, CIP, or DAP) alone or in combination. The bacterial suspension was incubated at 37 °C, and the number of surviving persisters was determined by CFU counting at intervals of 1 h for a total of 4 h [21].

Murine wound infection model

All animal experiments were executed in accordance with Ethics Committee approval by the Third Xiangya Hospital of Central South University (2021sydw0245). Female ICR mice aged 6–8 weeks were obtained from SAJ Lab Animal Co., Ltd. (Changsha, China). Mice were anesthetized by intraperitoneal (i.p.) injection of sodium pentobarbital (50 mg/kg), and their backs were shaved with an electric razor as previously reported. A 5-mm-diameter wound was punched surgically on their back and inoculated with 50 μL bacterial suspension containing 1 × 10⁷ CFU of S. aureus in saline. L007-0069 was dissolved in moisturizing cream (Glaxal Base, Canada). At 1 h post-infection, saline (vehicle), 1–2% (w/w) L007-0069 or 2% (w/w) fusidic acid, was topically administered to the wound. At 24 h post-infection, the wound skin was excised and homogenized in saline for CFU counting. Meanwhile, the infected skin tissue was preserved in 4% paraformaldehyde for hematoxylin and eosin (H&E) staining [22].

Time–survival curve determination

The bacterial suspension of S. aureus was prepared in saline to a lethal concentration of 2 McF turbidity standard (~ 6 × 10⁸ CFU/mL). Female ICR mice (n = 10 per group) were infected (i.p.) with 500 μL of the bacterial suspension in the presence of 5% mucin. After incubation for 1 h, mice were treated with a single dose of 1% DMSO (Vehicle), L007-0069 (15 mg/kg), GEN (20 mg/kg) or L007-0069 + GEN (15 + 20 mg/kg). The survival conditions were recorded at intervals of 8–12 h for 7 days, and the survival rate was calculated [16]. Meanwhile, the mice treated with 20 mg/kg L007-0069 for 24 h were sacrificed by CO₂ inhalation, and the organs of liver, spleen, and kidney were removed for CFU counting.

Postantibiotic effect (PAE) and postantibiotic sub-MIC effect (PA-SME)

For PAE, glass tubes containing 5 mL of MH broth in the presence of 1–4 × MICs of tested compounds were inoculated with 5 × 10⁶ CFU/mL S. aureus. MH broth with DMSO was used as a control. After incubation at 37 °C and 180 rpm for 1 h, the cultures were diluted 1:1000 to remove the effects of the antimicrobials. Viable cell counts were determined by CFU counting at 0, 2, 4, 8, 12, and 24 h. For PA-SME, after treatment with 4 × MICs of the tested compounds as described above, the suspension was diluted 1:1000 in the presence of sub-MIC (1/2–1/4 × MIC) of L007-0069 or VAN. Tubes with DMSO were also used as a control. The viable cells were determined as described above [23].

Statistical analysis

Experiments were independently performed in triplicate except when specifically noted. All data were analyzed by GraphPad Prism 8.0 software and are presented as the mean ± standard deviation (SD). The data were compared using Student’s t test, whereas data comparisons across more than two groups were performed using one-way ANOVA followed by Dunn’s multiple comparison test. p < 0.05 was considered statistically significant. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Results

Bactericidal activity against MRSA by L007-0069

To identify the potential antibacterial molecules, we purchased 5033 small-molecule chemical compounds from the TopScience MINI Scaffold Library (https://www.tsbiochem.com/). By using the MRSA-type strain ATCC43300, we primarily identified 43 hits that had potential antibacterial activity in the first round of screening (Fig. 1A, B). Through microbroth dilution, 6 hits were identified in the second round of screening (Fig. 1C). The structures and molecular characteristics of the 6 hits are shown in Fig. 1D, E, respectively. Furthermore, the antimicrobial effects of the 6 molecules were assessed by repeated assays with the MRSA strains USA300 and ATCC 43300 (Fig. 1F). Although none of the molecules exhibited antibacterial effects against gram-negative pathogens with MICs > 32 μg/mL, C218-0546 showed bacteriostatic effects against S. aureus strains with the lowest MICs (2–4 μg/mL), and L007-0063 showed bactericidal activity with detectable MBCs (8–16 μg/mL) (Table 1). Thus, we selected both C218-0546 and L007-0069 for structure optimization. By searching structure-similar molecules in ZINC15 database, we selected top 100 hits. Then, by searching the substitutions at the side chain with halogen group, hydrocarbon group or cyanogens group, which are reported to be capable of change molecules’ antimicrobial activity [24, 25], we selected the final 21 and 9 hits for antimicrobial susceptibility test determination for C218-0546 and L007-0069, respectively. Although no lower MICs or bactericidal analogs of C218-0546 were found among these analogs (Table S1), we found that L007-0069, an analog of L007-0063, showed lower MIC (4 μg/mL) and MBC (8 μg/mL) values against ATCC 43300 (Table S2). In addition, L007-0069 had a similar cytotoxicity profile as L007-0063 (SI Appendix, Fig. S2 and Fig. S9). Thus, we selected L007-0069 for further study. The synthesis route and MRI spectrum of L007-0069 are shown in SI Appendix, Fig. S3. By detecting the antimicrobial spectrum, L007-0069 was effective against a series of S. aureus strains (including MSSA, MRSA and VISA) with MICs and MBCs of 2–4 and 4–16 μg/mL, respectively. L007-0069 also exhibited a growth inhibitory effect against Enterococcus, with MICs of 4–8 μg/mL. However, no antimicrobial activity was found against gram-negative pathogens by L007-0069 (Table 2).

Fig. 1.

Bioactive compound screening. A Flow process diagram for potential antimicrobial screening. MRSA ATCC 43300 was incubated with 100 μM molecules from the MINI Scaffold Library, and after 16 h of incubation at 37 °C, the OD630 nm was determined. Small molecules of 43 hits with bacteriostatic activity were selected for the second-round screening microdilution test, and 6 hits with MIC < 50 μM were included for further study. B Bacterial growth inhibition rate by 5033 molecules from MINI Scaffold Library. Gray band indicates the molecules with the inhibition rate ≥ 75%. C Concentration-dependent bacterial growth inhibition effects of the 43 hits by first round screening. D Structural formulas of the 6 selected molecules. E Characteristics of selected molecules. H_acceptor: hydrogen bond acceptor count. H_donor: hydrogen bond donor count. B_rotN: rotatable bond count. LogP: oil–water partition coefficient. F Confirmation of the bacterial growth inhibition activity of the 6 selected molecules against the MRSA strains USA300 and ATCC 43300

Table 1.

Antimicrobial susceptibility of selected molecules against pathogens (μg/mL)

Strains	L489-1761		8006-4679		C218-0546		L007-0063		8002-7631		6791-0043
	MIC	MBC	MIC	MBC	MIC	MBC	MIC	MBC	MIC	MBC	MIC	MBC
S. aureus
USA300a	16	> 32	8	> 32	4	> 32	8	16	8	> 32	8	> 32
ATCC 43300a	4	> 32	4	> 32	4	> 32	16	16	16	> 32	16	> 32
LZB1	> 32	> 32	8	> 32	4	> 32	8	8	8	> 32	8	> 32
Newman	16	> 32	4	> 32	2	> 32	8	16	8	> 32	8	> 32
ATCC 25923	32	> 32	4	> 32	4	> 32	8	16	8	> 32	8	> 32
S. epidermidis
ATCC 12,228	> 32	> 32	4	> 32	4	> 32	8	32	4	> 32	2	> 32
RP62A	2	> 32	> 32	> 32	4	> 32	8	> 32	16	> 32	16	> 32
E. faecalis
ATCC 29212	> 32	> 32	> 32	> 32	16	> 32	16	> 32	> 32	> 32	32	> 32
E. coli
ATCC 25922	> 32	> 32	> 32	> 32	> 32	> 32	> 32	> 32	> 32	> 32	> 32	> 32

^amethicillin-resistant Staphylococcus aureus (MRSA)

Table 2.

Antimicrobial activity of L007-0069 against pathogens

Strains	Resistant pattern	MIC (μg/mL)	MBC (μg/mL)
S. aureus
USA300	MRSA	4	4
Newman	MSSA	4	4
ATCC 25923	MSSA	4	8
ATCC 29213	MSSA	2	4
LZB1	MSSA	4	4
SA2145	MRSA	2	4
RJ-2	MSSA	4	8
SAJ1	MRSA, VISA	4	8
MW2	MSSA	4	8
SA2231	MRSA	4	16
SA2174	MRSA	4	8
S. epidermidis
ATCC 12228	MSSE	4	32
RP62A	MSSE	4	>32
E. faecalis
EF02	Non-MDR	4	> 32
EF05	Non-MDR	4	> 32
EF11	Non-MDR	8	> 32
VRE1	VRE	8	> 32
VRE2	VRE	8	> 32
E. faecium
EFM04	Non-MDR	8	> 32
EFM08	Non-MDR	8	> 32
EFM09	Non-MDR	8	16
E. coli
ATCC 25922	Non-MDR	> 32	> 32
Y9592	XDR	> 32	> 32
Y9395	XDR	> 32	> 32
P. aeruginosa
PAO1	Non-MDR	> 32	> 32
PA1059	XDR	> 32	> 32
PA9113	MDR	> 32	> 32
K. pneumoniae
ATCC 700603	Non-MDR	> 32	> 32
KPLUO	XDR	> 32	> 32
KPWANG	XDR	> 32	> 32
A. baumannii
ATCC 19060	Non-MDR	> 32	> 32
AB0756	Non-MDR	> 32	> 32
AB0130	XDR	> 32	> 32

MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin sensitive Staphylococcus aureus; VISA, vancomycin intermediate Staphylococcus aureus; MSSE, methicillin sensitive Staphylococcus epidermidis; MDR, multidrug drug resistance; XDR, extensive drug resistance

Similar to a newly identified membrane disruptor, bithionol [10], we found that L007-0069 also showed cell lysis activity against S. aureus (Fig. 2A). Furthermore, the bactericidal kinetics revealed that L007-0069 showed rapid bactericidal activities against S. aureus in a dose- and time-dependent manner (Fig. 2B). The viable cells were promptly decreased to the limit of detection at 4 × MICs of L007-0069 for only 4 h of treatment. By a resistance induction assay in the presence of sub-MICs of L007-0069 and CIP (positive antibiotic) against S. aureus, we found that CIP revealed a rapid increase in MIC values after a few passages; however, the MIC values of L007-0069 remained unchanged over the entire the study (Fig. 2C). Moreover, the CIP-induced S. aureus of the last highly resistant generation remained sensitive to L007-0069, with MICs and MBCs of 4–8 μg/mL (Fig. 2D). Similarly, there were no colonies grown by any of the tested strains (USA300, ATCC 43300, Newman, and RJ-2) in the presence of 2–8 × MICs of L007-0069 by the one-step resistance-inducing assay. However, RFP showed a relatively high frequency of resistance (SI Appendix, Table S3).

Fig. 2.

Bactericidal activity of L007-0069 with low resistance. A L007-0069 causes cell lysis. Exponential-phase MRSA ATCC 43300 cells (OD630 ~ 0.55) were treated with 32 µg/mL L007-0069, 32 µg/mL benzyldimethylhexadecylammonium chloride (16-BAC), or 40 µg/mL bithionol (BT) for 3 h. B Time-dependent killing of L007-0069 against MRSA ATCC 43300. LOD., limit of detection. C Serial passage resistance development in the presence of 1/2 × MIC concentrations of L007-0069 or ciprofloxacin (CIP). D MIC (above panel, MIC1 and MIC2 indicate the MICs of ATCC 43300 and SAJ1, respectively) and MBC (below panel) determinations of L007-0069 against CIP-induced highly resistant MRSA ATCC 43300 and SAJ1

The fast killing efficacy of L007-0069 against S. aureus was also verified in both the USA300 and Newman strains and the clinical isolate RJ-2 (SI Appendix, Fig. S4A). Moreover, L007-0069 showed better PAE results than VAN in these strains (SI Appendix, Fig. S4B). Although L007-0069 showed a relatively worse PA-SME than VAN by USA300, Neman and RJ-2 still exhibited better PA-SME values, indicating that the PA-SME of L007-0069 may be strain dependent (SI Appendix, Fig. S4C).

The antimicrobial activity of L007-0069 was mediated by cell membrane disruption

We performed scanning electron microscope (SEM) and transmission electron microscope (TEM) to visually observe the ultrastructural changes in S. aureus after treatment with L007-0069 to initially speculate on its potential mechanism of action. SEM showed that bacterial cells in the presence of L007-0069 aggregated into clumps with filamentous attachment and showed a rough cell surface with a depressed and distorted appearance (Fig. 3A). TEM images showed that cells treated with L007-0069 formed mesosome-like structures, the cell membrane dissolved and intracellular contents leaked, while the surface of untreated S. aureus cells was smooth, without obvious ultrastructural changes (Fig. 3B). Given the morphological changes in the cell membrane after L007-0069 treatment, we hypothesized that L007-0069 might target the cell wall components, like peptidoglycan (PGN) or cell membrane. However, no antimicrobial activity alteration was observed in the presence of PGN (SI Appendix, Fig. S5A). Thus, we further conducted a SYTOX Green permeability assay and found that the L007-0069 small molecule induced dose-dependent membrane permeability within 30 min (Fig. 3C). Correspondingly, double-stained images by DAPI and SYTOX Green also showed that L007-0069 could disrupt the membrane integrity of almost all the tested bacterial cells (Fig. 3D). Furthermore, increased membrane permeability was proven based on a dose-dependent uptake of PI (Fig. 3E) and was also confirmed by images captured by CLSM (Fig. 3F). We found that L007-0069 could also disrupt the bacterial membrane potential as measured by DiSC(3)5 fluorescence staining (Fig. 3G), which caused a decrease in membrane potential and resulted in a polarization effect.

Fig. 3.

L007-0069 disrupted the bacterial cell membrane and interacted with the lipid bilayers. Representative images of SEM (A) and TEM (B) of MRSA ATCC 43300 after treatment with 5 × MICs of L007-0069 for 1 h. Red asterisk indicates cell shrinkage and deformation. White and blue asterisks indicate mesosome and late-stage cell, respectively. Scale bar: 5 μm for SEM, 500 nm for TEM. C Increased permeability of the bacterial cell membrane of MRSA ATCC 43300 treated with L007-0069 for 30 min was detected by a SYTOX Green probe. Melittin (4 μg/mL) was used as a positive control. D MRSA ATCC 43300 cells stained with DAPI/SYTOX Green after treatment with 1 × MIC L007-0069 for 30 min. E Increased permeability of the bacterial cell membrane by the PI probe after 30 min of treatment with L007-0069. F PI staining of MRSA ATCC 43300 in the presence of 1 × MIC L007-0069 for 30 min and observed by CLSM. G Uptake of DiSC3(5) by logarithmic-phase MRSA ATCC 43300 cells treated with the indicated concentrations of L007-0069. H Bacterial cell membrane disruption by Laurdan probe staining. The fluorescence intensity of Laurdan was recorded using a microplate reader with excitation at 350 nm and emission at 435 nm and 490 nm. Laurdan GP was calculated as follows: Laurdan GP = (I440 − I490)/(I440 + I490). I Cell membrane disruption by GUVs detection. The GUVs consisted of DOPC/DOPG (7:3) labeled with FITC. GUVs were treated with 2 × MICs of L007-0069 and were monitored by CLSM with excitation at 485 nm and emission at 525 nm. Scale, 20 μm. J Representative configurations of MD simulations of the bacterial (upper panel) and mammal (down panel) cell membrane in the presence of L007-0069. From left to right: the onset of simulation, membrane attachment, membrane penetration and equilibrium state of the L007-0069 interaction with the cell membrane. K Change in lipid bilayer thickness after L007-0069 was embedded into the outer leaflet of the bilayer. L Detailed configuration of the nearest adjacent lipids around the embedded L007-069 compound from the top and side views. M Heatmap of differentially expressed genes of MRSA ATCC 43300 after L007-0069 treatment. Blue indicates downregulated genes and red indicates upregulated genes. The SCALE method by Pheatmap software was used to standardize the quantification of DEGs. Scale: “− 1 to 1” indicates the lowest expression value to the highest expression value

It is generally known that the insertion of specific membrane-disrupting compounds into lipid bilayers causes dramatic changes in membrane fluidity and disrupts the normal liquid crystal phase of the membrane, which further leads to leakage of cellular components and bacterial death [10]. Therefore, we proceeded to determine whether L007-0069 altered membrane fluidity in S. aureus by using Laurdan, a dye sensitive to membrane fluidity, and quantified it using the Laurdan generalized polarization (GP) index. As shown in Fig. 3H, similar to DAP, the Laurdan GP of S. aureus decreased significantly after treatment with L007-0069 at concentrations greater than 16 µg/mL. These data indicated that L007-0069 may permeate the cell membrane into the lipid bilayer and cause changes in membrane fluidity. Furthermore, we investigated the effect of L007-0069 on the lipid bilayer utilizing biomembrane-mimicking giant single-molecule vesicles. GUVs treated with 8 µg/mL L007-0069 formed lipid aggregates on their surface, and the lipid body area shrank significantly, whereas the control group presented an intact lipid structure (Fig. 3I).

By using all-atom MD simulations, we found that L007-0069 was initially recruited to the membrane surface. After a few hundred nanoseconds of sustained attachment, L007-0069 became closer to the centroid of the membrane (SI Appendix, Fig. S6A), and the binding energy was significantly decreased within 20 ns (SI Appendix, Fig. S6B), indicating that L007-0069 could rapidly attach to the phospholipid membrane surface and stabilize. L007-0069 then penetrated into the interior of the membrane and then embedded itself in the outer leaflet of the lipid bilayer. And as we expected, no interaction was found between L007-0069 and the mammal cell membrane (SI Appendix, Fig. 3J). In addition, the number of hydrogen bonds also increased with time (SI Appendix, Fig. S6C). The insertion of L007-0069 small molecules led to substantial increases in local disorder and changes in the thickness of the lipid bilayer (Fig. 3K). The 3-dimensional binding mode showed that L007-0069 lies horizontally in the phospholipid membrane and that the oxygen atom in L007-0069 can form a hydrogen bond with the DOPC molecule at 4 Å resolution (Fig. 3L). In addition, as we expected, the outer membrane disruptor SPR741 restored its antimicrobial activity against Gram-negative strains, like E. coli and K. pneumonia (Fig. S5C), which further confirmed the cytoplasm-disrupting effects of L007-0069. However, as shown in Fig. S5C, no antimicrobial effect was observed against P. aeruginosa, which was probably due to its different outer membrane proteins and efflux pumps.

To gain a deeper understanding of the molecular mechanisms underlying L007-0069, we performed transcriptional analysis of MRSA ATCC 43300 after treatment with L007-0069. Clustering analysis revealed a drastic change in gene expression, and a total of 1287 differentially expressed genes (DEGs) were observed (Fig. 3M), of which 664 DEGs were upregulated and 623 DEGs were downregulated (SI Appendix, Fig. S7A). GO annotation analysis demonstrated a rapid gene upregulation involved in biological adhesion and detoxification probably indicating the bacterial defense process against membrane disrupting. Similarly, the cellular anatomical entities-related genes were also influenced by the membrane disruptor L007-0069 (SI Appendix, Fig. S7B). In consistence, KEGG enrichment analysis showed that the DEGs were enriched in cellular component organization/assembly and structural molecule activity indicating the bacterial cellular repairment after the disrupting of cell membrane components by L007-0069 treatment (SI Appendix, Fig. S7C).

Persister killing activity by L007-0069

L007-0069 killed stationary-phase MRSA ATCC 43300 planktonic persisters in a dose-dependent manner, and the CFU of persister cultures were reduced by ~ 4 Log10 CFU/mL at 8 × MICs of L007-0069 after 4 h of treatment. In contrast, MRSA ATCC 43300 persisters were highly tolerant to high concentrations (10 × MICs, the MICs of antibiotics were shown in Table S5.) of several conventional antibiotics, including VAN, CIP, DAP, and GEN, which are commonly selected for S. aureus infection treatment (Fig. 4A). We hypothesized that L007-0069 retained fast killing potency against S. aureus persister cells via cell membrane damage. We subsequently detected the ability of L007-0069 to permeabilize the MRSA persister membrane by using SYTOX Green staining. As we expected, in contrast to VAN (Fig. 4B), CIP (Fig. 4C), DAP (Fig. 4D) and GEN (Fig. 4E), L007-0069 resulted in a dose-dependent increase in fluorescence intensity (Fig. 4F). A similar rapid persister killing effect and membrane permeability ability were also observed in S. aureus USA300, Newman and RJ-2 strains (SI Appendix, Fig. S8). Interestingly, L007-0069 did not cause the accumulation of intracellular ROS (Fig. 4G), but an increase in intracellular ATP levels was observed in L007-0069-treated S. aureus persisters for 30 min incubation [even though no obvious growth inhibition by 2–8 × MIC of L007-0069 was found after 30 min incubation (SI Appendix, Fig. S5B)], although VAN and DAP caused a decrease in ATP levels (Fig. 4H). In addition, by using the metabolism indicator Alamar blue, we found that the metabolic activities of S. aureus persisters were enhanced (Fig. 4I). Hence, we hypothesize that L007-0069 may kill persisters by turning their persister status into proliferating status. As expected, L007-0069 significantly enhanced the bactericidal activities of DAP (Fig. 4J), VAN (Fig. 4K), and GEN (Fig. 4L) against MRSA ATCC 43300 persisters. These results led to the conclusion that L007-009 increased the metabolic activity of S. aureus persisters, thus sensitizing them to antimicrobials.

Fig. 4.

Bactericidal activity of L007-0069 against stationary-phase S. aureus persisters. A Antimicrobial effects of L007-0069 and 10 × MICs of conventional antibiotics against stationary-phased persisters of MRSA ATCC 43300 within 4 h. The data represent average values ± SD. B–F Uptake of SYTOX Green by MRSA persister cells treated with a series of concentrations of VAN (B), CIP (C), DAP (D), GEN (E), and L007-0069 (F). G ROS accumulation in S. aureus cells after melittin (positive control), VAN, CIP, and L007-0069 treatment. H Levels of intracellular ATP in MRSA ATCC 43300 after treatment with VAN, CIP, DAP, and L007-0069. I Metabolic activity of MRSA ATCC 43300 persisters by Alamar blue staining after treatment with L007-0069. J–L Viability of MRSA ATCC 43300 persisters treated with L007-0069 (1–4 × MICs) alone or in combination with 10 × MICs of DAP (J), VAN (K), or GEN (L)

L007-0069 sensitizes the antimicrobial activity of GEN against S. aureus

Among different conventional antibiotics, GEN was observed to be synergistic with the small molecule L007-0069 by chequerboard microdilution assays with a FICI of 0.375 (Fig. 5A and SI Appendix, Table. S4). By time-inhibition analysis, sub-MIC levels of L007-0069 (1 μg/mL) or GEN (16 μg/mL) showed no growth inhibition, but significant synergistic effects were observed when the drugs were combined (Fig. 5B). Similarly, according to the time-killing assay, sub-MIC concentrations of L007-0069 or GEN exhibited no antimicrobial effect against S. aureus, whereas combined treatment significantly inhibit the cell growth (Fig. 5C).

Fig. 5.

Synergistic antimicrobial effects of L007-0069 and GEN against S. aureus and its resistant phenotypes. A Synergistic antimicrobial activity between L007-0069 and GEN determined by checkerboard assay. FIC, fractional inhibitory concentration; time–growth inhibition (B) and time-killing activity (C) of L007-0069 alone or in combination with GEN at sub-MICs. D Biofilm inhibition effects of L007-0069 alone or in combination with GEN against MRSA. E Biofilm eradication effects of L007-0069 alone or in combination with GEN against MRSA. F Bactericidal effects of L007-0069 alone or in combination with GEN against MRSA biofilm persisters. 10 × MICs of the conventional antibiotics, including VAN, CIP, DAP, were used as controls. G Representative CLSM images of biofilm inhibition effects of L007-0069 alone or in combination with GEN. Scale: 20 μm. H Representative CLSM images of the biofilm-eradicating effects of L007-0069 alone or in combination with GEN. Scale bar: 20 μm. Green fluorescence indicates the total cells (both of the live and dead cells). Red fluorescence indicates the dead cells

Furthermore, we found that 4 μg/mL L007-0069 inhibited biofilm formation (Fig. 5D) and killed bacterial cells within biofilms (Fig. 5E) in a dose-dependent manner. The antibiofilm efficacy was significantly enhanced in the presence of GEN. Moreover, biofilm persister killing assays showed that although MRSA persisters presented high levels of tolerance to several conventional antibiotics (including VAN, DAP, CIP, and GEN) at 10 × MICs, (2–8) × MICs of L007-0069 exhibited dose/time-dependent anti-persister activity, and the effects were significantly enhanced in the presence of GEN (Fig. 5F). In accordance, by CLSM observation, 2 × MICs (8 μg/mL) of L007-0069 effectively inhibited the biofilm formation of ATCC 43300 (Fig. 5G). In addition, 1 μg/mL L007-0069 in combination with 32 μg/mL GEN exerted an obvious synergistic biofilm inhibition effect compared with its use alone, accompanied by significantly reduced biofilm thickness and density and markedly weakened green fluorescence of live bacteria (Fig. 5G). Although eradication of the preformed biofilm was far more difficult than biofilm inhibition, 16 μg/mL L007-0069 in combination with 32 μg/mL GEN still exhibited a better biofilm eradication effect than when used alone, with the biofilm biomass markedly decreasing and the proportion of red fluorescence of dead bacteria visibly increasing (Fig. 5H). These results demonstrate that GEN could enhance the antimicrobial ability of L007-0069 against MRSA planktonic cells, persisters, and biofilms.

In vivo antimicrobial efficacy of L007-0069

Due to the encouraging bactericidal activity against S. aureus in vitro, we evaluated the potential in vivo therapeutic efficacy of L007-0069 in a wound infection model by MRSA ATCC 43300. Compared with the vehicle groups, the 2% L007-0069 treatment groups showed significant CFU reduction (1.85-log) in mice, which was much more effective than the positive control by 2% fusidic acid treatment (1.2-log CFU reduction was shown) (Fig. 6A). H&E analysis showed that mice treated with 2% L007-0069 and 2% fusidic acid had no detectable inflammatory infiltration compared to that in the vehicle group (Fig. 6B). Furthermore, in the peritonitis–sepsis murine model, we found that the mice in the vehicle group all died at 24 h post-infection, and the groups treated with L007-0069 or GEN used alone reached an approximately 50% survival rate. As we expected, the survival rate of the mice treated with the combination of L007-0069 (15 mg/kg) and GEN (20 mg/kg) achieved a 100% survival rate during the entire 7-day course of infection (Fig. 6C). Similar to DAP and VAN, a single dose (20 mg/kg) of L007-0069 significantly reduced the bacterial loads in infected organs of the liver (Fig. 6D), spleen (Fig. 6E) and kidney (Fig. 6F) by CFU counting.

Fig. 6.

In vivo antibacterial efficacy of L007-0069. A The viable bacterial cell counts after treatment with L007-0069 (1% or 2%) or 2% fusidic acid in a wound infection model by ATCC 43300 (n = 10). B Representative images of skin from the wound by H&E staining after treatment with 2% L007-0069 or 2% fusidic acid. Scale, 100 μm. C Percent survival in a peritonitis infection model after treatment with L007-0069 (15 mg/kg) alone or in combination with GEN (20 mg/kg); bacterial loads in the liver (D), spleen (E), and kidney (F) in a peritonitis-sepsis model. The mice were infected with MRSA ATCC 43300 i.p., 1 h post-infection, and a dose of 20 mg/kg L007-0069, GEN, DAP, or VAN was administered i.p. The organs were separated and counted by plate dilution assay after 24 h of treatment

In vitro and in vivo toxicity of L007-0069

We used human RBC hemolysis and cytotoxicity assays to evaluate the in vitro toxicity of L007-0069. We found that L007-0069 showed no hemolytic activity on human RBCs at concentrations of up to 64 μg/mL (SI Appendix, Fig. S9A). Although L007-0069 moderately inhibited the viability of the LO2, HepG2, HK-2, 786-O, HMC3, and U251 cell lines (SI Appendix, Fig. S9B), the cell viability inhibitory concentrations were still much higher than the concentration where L007-0069 showed effective antibacterial activity.

We further assessed in vivo toxicity by administration (i.p.) of 20 mg/kg L007-0069 alone or in combination with 20 mg/kg GEN. Compared to the vehicle group, routine blood analysis showed that there was no significant difference among the parameters of WBC, RBC, PLT, HGB, and N%, with p values > 0.05 (SI Appendix, Fig. S10A). In addition, there was no significant difference between the vehicle and combination treatment groups in the liver functional biomarkers ALT and AST (SI Appendix, Fig. S10B), the myocardial biomarker CR or the renal biomarker BUN, with p-values > 0.05 (SI Appendix, Fig. S10C and S10D). In accordance, the H&E analysis showed that L007-0069 alone or in combination with GEN did not cause significant tissue damage compared with the control group. Hemorrhage, edema, granulocyte infiltration, hyperplasia, or other morphological changes were not observed in the main organs (SI Appendix, Fig. S10E).

Discussion

Several studies have identified bioactive drugs against bacterial cells by scientific large-scale screening; although the whole screening process requires a large amount of effort, it has yielded satisfactory results [26, 27]. However, to the best of our knowledge, small molecules of the MINI Scaffold Library have not been utilized for antibacterial purposes. In this study, we identified the small molecule L007-0069 by screening a MINI backbone library. L007-0069 demonstrated potent bactericidal efficacy against S. aureus and its persistent cells without inducing resistance. However, L007-0069 exhibited resistance to gram-negative bacteria. We hypothesize that the envelope of gram-negative bacteria includes a negatively charged outer membrane, which forms a physical barrier to most antimicrobial compounds [28], thus making it challenging for L007-0069 to transmit through the cell wall to exert its antibacterial activity. A mechanistic study indicated that L007-0069 showed bactericidal activity, probably by increasing cell membrane permeability and disrupting the lipid bilayers. In addition, its in vivo efficacy was also validated in a MRSA-infected skin wound model and peritonitis model.

Persisters exhibit a metabolically low-energy state that prevents attack by antibiotics. Recent studies have shown that S. aureus can form persisters when cultured to stationary phase, accompanied by a decline in intracellular ATP. Meanwhile, cells with reduced ATP levels indicate a decrease in the energy level, which could lead to antibiotic tolerance [29]. The SYTOX Green membrane permeability assay suggested that L007-0069 killed MRSA persisters by rapid membrane disruption, and this property is applicable to other S. aureus persisters with different resistance patterns (SI Appendix, Fig. S7). We subsequently tested the intracellular ATP levels. Unexpectedly, we found that L007-0069 increased the ATP levels of persisters; hence, we speculated that the bactericidal compound L007-0069 may convert persisters from a resistant non-replicating state to a sensitive metabolically active state. This property is different from other well-studied membrane disruptors; for example, daptomycin was reported to cause rapid cell death by targeting gram-positive bacterial membranes. Although DAP has strong antibacterial activity against growing bacterial cells, no studies have reported that it is effective against persisters. In accordance, even high concentrations of VAN, GEN, CIP, and DAP all failed to kill persisters in our study. However, this antibiotic showed bactericidal activity against persisters in the presence of L007-0069, probably because L007-0069 could increase metabolic activity and prompt persisters to be sensitive to conventional antibiotics, thus facilitating the removal of persisters. Antibiotics targeting the bacterial cell membrane and activating the metabolic state may have many attractive properties, such as fast killing, synergism with other antimicrobial agents, and a low probability of inducing antibiotic tolerance, thereby providing effective therapeutic strategies for the clinical treatment of chronic recurrent infections [30].

MD simulations rendered an advanced process for cell membrane fluidity determination by using computer calculations. MD provides a detailed view of dynamics and helps us understand the interaction between compounds and membrane lipid bilayers on the nano- and microsecond timescales. It reflects how and when small molecules move through the cell membranes and reflects the binding model between molecules and membrane components [31]. MD simulations have been widely used to observe the interaction patterns of various membrane-active antibiotics with bacterial membranes [10, 32, 33]. For example, Kim et al. [32] described the whole process of membrane–molecule interactions by CD437 and bithionol. These molecules were found to be recruited to the membrane surface after attachment for a few nanoseconds, penetrated into the membrane, and finally resulted in lipid bilayer disorders. Similarly, our results demonstrated that the small molecule L007-0069 penetrated bacterial membranes even more quickly than CD437 or bithionol (100 ns vs. 310/300 ns), which illustrated the feasibility of this method to provide insight into the mechanism of rapid molecule–membrane interactions. Although MD simulations provide a certain degree of detail at the atomic level, biological experiments are still needed to prove the membrane-disrupting activities of these molecules.

As reported elsewhere, membrane-active antibacterial agents have some potential advantages, including fast killing, anti-persister activity and low probability of inducing resistance [30]. Our results are consistent with these properties. Although, in our study, the current frontline anti-MRSA antibiotic VAN and the well-studied membrane disruptor DAP did not induce rapid membrane permeabilization or kill S. aureus persisters or eliminate preformed biofilms and persister cells (Fig. 3), we discovered that L007-0069 exhibits effective antibiofilm and anti-persister potency, which are notable for the treatment of chronic infections caused by these refractory bacteria.

It is well known that antibiotic combination is a great approach to reduce such toxicity, further increase antibacterial activity, and avoid resistance evolution [34]. As we expected, L007-0069 showed significant synergism with GEN against both MRSA planktonic cells, biofilms, and persister cells in the present study, most likely a consequence of the increased passive diffusion of GEN through the bacterial cell membranes damaged by L007-0069. Although GEN has been reported to have dose-related side effects of most commonly ototoxicity and nephrotoxicity [35], L007-0069 in combination with GEN could significantly reduce the dose of both antimicrobials to reach the same antimicrobial effect as single use, thereby reducing its toxicity. However, pharmacological optimization with respect to better antimicrobial and lower toxicity profiles still needs to be carried out in further studies, such as structure–activity-related analog synthesis, simulated docking techniques [36], and new drug delivery systems (such as siderophores, liposomes, antimicrobial peptides, and nanoparticles).

In conclusion, we have identified a novel compound, L007-0069, that rapidly kills S. aureus bacterial cells and persisters by disrupting lipid bilayers. Our study suggests that L007-0069 is a promising therapeutic for hard-to-treat infections caused by MRSA.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

PS and YW designed research; PS, YL, and LX performed the study; PS, ZL, YL, and SL analyzed data; and all authors wrote and editing the paper.

Declarations

Conflict of interest

The authors declare that they have no competing interests.

Ethical approval

All animal experiments were executed in accordance with Ethics Committee approval by the Third Xiangya Hospital of Central South University (2021sydw0245).

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Data availability statement

All materials are available on request, and the manuscript includes supplemental information submitted electronically.