Introduction

Coagulase-negative Staphylococcus epidermidis (S epidermidis) is one of the most common opportunistic pathogens involved in implanted medical device-associated nosocomial infections1, 2, 3. With the increasing use of implanted medical devices, such as intravascular catheters, artificial pacemakers, cerebrospinal fluid shunts, and artificial organs, S epidermidis biofilm-associated infections have become a common problem. The ability to form biofilms on the surfaces of the implanted devices is the primary pathogenic trait of S epidermidis, and the bacteria in the biofilms are resistant to antimicrobial treatments and host immune defenses4, 5, 6. Because of the increasing emergence of multidrug-resistant strains, the discovery of novel antibiotics to combat staphylococcal biofilm infections is imperative7. Completion of the genome sequencing of S epidermidis8, 9 makes it possible to discover potential antimicrobial agents using genomics-based drug discovery strategies.

The majority of bacteria use a phosphotransfer mechanism termed a two-component system (TCS), comprised of a histidine kinase (HK) and a response regulator (RR), to sense environmental conditions and bring about appropriate changes in cellular behavior10, 11, 12. TCSs are important in regulating the virulence and propagation of pathogenic bacteria13, 14, 15, and they are considered attractive targets for the development of novel anti-bacterial drugs. The two-component YycG/YycF system, originally identified in Bacillus subtilis16, 17, is highly conserved and specific to low G+C gram-positive bacteria, including S epidermidis. YycG/YycF is essential for cell viability, cell wall metabolism, autolysin synthesis, and biofilm formation in staphylococcal species. Upon sensing a signal from the external milieu, YycG autophosphorylates by transferring a phosphate group from ATP to a histidine residue within the kinase and then transferring it to an aspartate residue within the conserved receiver domain of the response regulator YycF. Phosphorylation of YycF results in changes in its ability to bind gene promoters and regulate transcription18, 19, 20.

Several inhibitors that target YycG in B subtilis, Staphylococcus aureus (S aureus), and S epidermidis have been documented to inhibit bacterial growth21, 22, 23. We previously described two inhibitors (Compounds 2 and 5) that target the HK domain of S epidermidis YycG and have bactericidal and biofilm-killing activities24. To obtain more effective and less toxic inhibitors, the structure of Compound 2 was optimized, and a series of derivatives were designed and synthesized, some of which inhibit the growth of planktonic S epidermidis cells25. In the present study, we explored the characteristics of these derivatives, including their YycG phosphorylation-inhibiting activity, bactericidal activity, biofilm-killing activity, and cytotoxicity. The inhibitory and bactericidal activities of these derivatives against S aureus were also assessed.

Our study of structure-based modification of a leading anti-TCS compound may contribute to the discovery of new antibiotics to treat staphylococcal biofilm infections.

Materials and methods

Bacterial strains, media, and derivatives of Compound 2

The bacterial strains used in this study were S epidermidis ATCC 12228 (biofilm negative), S epidermidis ATCC 35984 (biofilm positive), S aureus ATCC 49230, S aureus ATCC 25923, and E coli ATCC 25922. E coli strains were grown in Luria-Bertani (LB) broth, and staphylococcal strains were cultivated in tryptic soy broth medium (TSB; Oxoid Ltd, Basingstoke, UK). The six derivatives of Compound 2 mentioned in this report (including five reported previously26 and one newly synthesized) are H2-10, H2-12, H2-20, H2-27, H2-28, and H2–29. These compounds were synthesized by modifying the functional groups while keeping the thiazolidione core structure intact. The structural formulas of the derivatives are listed in Figure 1. To generate the derivative compounds, the terminal (2-methoxy-4-vinyl-phenoxy)-acetic acid functional group of Compound 2 was replaced by a (5-vinyl-furan-2-yl)-benzoic acid fragment with similar structure and by a shortened side alkyl chain, according to the principle of multi-target-directed ligands. The derivatives were dissolved in dimethyl sulfoxide (DMSO; AMRESCO, USA) to create 200 mmol/L stock solutions.

Figure 1
figure 1

The structures of the thiazolidione derivatives. Based on the thiazolidione core structure of Compound 2 {2-{4-{3-(2-ethylphenyl)-2-[(2-ethylphenyl)imino]-4-oxothiazolidin-5-ylidene}methyl}-2-methoxyphenoxy}acetic acid, six derivatives were designed and synthesized by modifying the functional groups through cyclization, aldol condensation, substitution, and hydrolyzation reactions. H2-10, 2-{4-{[3-(2-ethylphenyl)-2-[(2-ethylphenyl)imino]-4-oxothiazolidin-5-ylidene}methyl}phenoxy}acetic acid; H2-12, 3-{5-{{3-(2-ethylphenyl)-2-[(2-ethylphenyl)imino]-4-oxothiazolidin-5-ylidene)methyl}furan-2-yl}benzoic acid; H2-20, 3-{5-{{4-oxo-3-(o-tolyl)-2-(o-tolylimino)thiazolidin-5-ylidene]methyl}furan-2-yl}benzoic acid; H2-27, 3-{5-{[4-oxo-3-(p-tolyl)-2-(p-tolylimino)thiazolidin-5-ylidene]methyl}furan-2-yl}benzoic acid; H2-29, 3-{5-{{3-(4-methoxyphenyl)-2-[(4-methoxyphenyl)imino]-4-oxothiazolidin-5-ylidene}methyl}furan-2-yl}benzoic acid; and H2-28, 4-{5-{[4-oxo-3-(o-tolyl)-2-(o-tolylimino)thiazolidin-5-ylidene]methyl}furan-2-yl}benzoic acid.

Full size image

Expression and purification of the YycG HATPase_c and HisKA domains

The HisKA and HATPase_c domains (amino acids 370–610, NCBI accession number YP_190074) of YycG (termed YycG′) from S epidermidis ATCC 35984 were expressed as a fusion protein with an N-terminal maltose binding protein six-histidine (MBP-His) tag. The 720-bp YycG ORF was PCR-amplified from S epidermidis ATCC 35984 chromosomal DNA with the primers YycG For-BamH I (5′-CGCGGATCCGAACAACAACAAGTCGAACGT-3′) and YycG Rev-Xho I (5′-CCGCTCGAGTTATTCATCCCAATCACCGTCT-3′). After digestion with BamH I and Xho I, the PCR product was cloned into pET28a (Novagen). The resultant plasmid was transformed into E coli BL21 (DE3). The BL21 (DE3) strain carrying the plasmid was grown in LB to an optical density of 0.6 at 600 nm (OD600), and then induced with 0.8 mmol/L isopropyl β-D-1-thiogalactopyranoside (IPTG). After 8 h induction at 25 °C, the cells were harvested, and recombinant YycG′ protein was purified by Ni2+ affinity chromatography using a Ni-NTA column (Qiagen, Germany). The MBP-His-tagged YycG′ was then cleaved with TEV protease, and the resultant un-tagged YycG′ was further purified by Ni-NTA and Superdex 75 gel filtration columns (GE Healthcare, USA).

Inhibition assay for YycG′ autophosphorylation activity

The Compound 2 derivatives were tested for inhibition of YycG′ autophosphorylation using the Kinase-Glo™ Luminescent Kinase Assay (Promega, Madison, WI, USA) in a solid black flat-bottomed 96-well plate24. Briefly, 0.13 μmol/L recombinant YycG′ was pre-incubated with serial dilutions of the derivatives in reaction buffer (40 mmol/L Tris pH 8.0, 20 mmol/L MgCl2, and 0.1 mg/mL BSA) at 25 °C for 30 min. Then, 4 μmol/L ATP was added and the plates were incubated for 30 min at 25 °C, and Kinase-Glo™ Reagent was added to detect the remaining ATP, recorded by luminescence measurement (RLU). Recombinant YycG′ incubated with Compound 2 was used as a positive control, and recombinant protein with no derivative added was used as a negative control in these experiments. The rate of protein phosphorylation (Rp) inhibition by the derivatives was calculated using the equation (Eq 1).

The half maximal inhibitory concentration (IC50, the concentration of the derivatives required to inhibit half of the autophosphorylation of the recombinant YycG′) was determined by Origin 8.0 software (OriginLab, Northampton, USA)24.

Minimal inhibitory concentration (MIC) and minimal bactericidal concentration (MBC) assays

MIC assays for the antibacterial activities of the derivatives were performed according to the macrodilution (tube) broth method of the Clinical and Laboratory Standards Institute (CLSI) of America. Briefly, the derivatives were serially diluted twofold into eight tubes containing 4 mL Mueller-Hinton Broth (OXOID, UK), yielding final concentrations of 100 μg/mL to 0.78 μg/mL. The turbidity of cultures incubated for 6 h was adjusted to match that of a 0.5 McFarland standard (approximately 108 CFU/mL), and 0.02 mL of the bacterial inoculum was added to each tube. Inoculated broth containing no antibiotic was included as a bacterial growth control and a tube of un-inoculated broth was used as a sterility control. The bacteria were incubated at 37 °C for 12 h. The lowest concentration that completely inhibits visible growth of the organism as detected by the unaided eye was recorded as the MIC27.

The MBC was obtained by sub-culturing 100 μL from each negative (no visible bacterial growth) tube from the MIC assay onto drug-free Mueller-Hinton agar plates. The plates were incubated at 37 °C for 24 h, and the MBC was defined as the lowest concentration of drug which produced subcultures growing no more than five colonies on each plate27.

Biofilm-killing assays

Biofilm formation was detected using a semiquantitative plate assay26. An overnight culture of S epidermidis ATCC 35984 was diluted 1:200 into TSB medium, and incubated statically for 6 h in a polystyrene 96-well plate at 37 °C. Subsequent to the removal of planktonic cells, fresh TSB containing the serially diluted derivatives was added to each biofilm, and the plates were incubated for another 18 h at 37 °C. After incubation, the wells were washed gently three times with phosphate buffered saline (PBS), fixed with methanol, and stained with 2% (w/v) crystal violet28. The absorbance of the wells was determined at 570 nm using a spectrophotometer (DTX880, Beckman Coulter, USA).

To determine the effect of the derivatives on cell viability in mature biofilms, the Live-Dead Bacterial Viability method (Live-Dead BacLight, Molecular Probes, USA) was used. Overnight cultures of S epidermidis ATCC 35984 grown in TSB medium were subcultured for another 6 h. The turbidity of the culture was adjusted to match that of a 0.5 McFarland standard (approximately 108 CFU/mL) and inoculated (1:200) into cover-glass cell-culture dishes (WPI, USA), which were then incubated at 37 °C for 24 h. After removal of the planktonic cells, fresh TSB containing the derivatives at concentrations corresponding to 4×MIC was added, and the dishes were incubated at 37 °C for another 16 h. After incubation, the dishes were carefully washed three times with normal saline (NS), stained with Live-Dead reagents (containing SYTO 9 and PI) at room temperature for 15 min, and observed under a Leica TCS SP5 confocal laser scanning microscope (CLSM).

Cytotoxicity and erythrocyte hemolysis

The cytotoxicity of the derivatives to Vero cells (African green monkey kidney cells) was detected using the Cell Proliferation Kit I (Roche, Indianapolis, USA) in 96-well cell culture plates. After exposure of the cells (∼104 cells per well) to the serially diluted derivatives for 48 h, relative cell numbers were assayed by co-incubation with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) for 4 h at 37 °C in 5% CO2. After incubation, the purple formazan salts were dissolved with DMSO, and the absorbance of each well was measured at 595 and 655 nm then converted to percentages of the control (cells treated with the solvent, 0.1% DMSO). The concentration of the derivatives with 50% cytotoxicity for Vero cells (CC50) was calculated using the Origin 8.0 software (Origin Lab, Northampton, USA).

To determine the hemolytic activities of the derivatives on erythrocytes, 5% (v/v) healthy human erythrocytes re-suspended in NS were co-incubated with the derivatives at MIC, 4×MIC, or 100 μg/mL for 1 h at 37 °C in 96-well microtiter plates. After the incubation, the suspensions were centrifuged at 350×g for 10 min, and the level of hemolysis was determined by measuring the absorbance of the supernatant at 570 nm25. Cells treated with DMSO (0.1%) and Triton-X100 (1%) were used as negative and positive controls, respectively.

Results

In our previous study, five derivatives were designed and synthesized based on the chemical structure of Compound 224, 25. The biological characteristics of the Compound 2 derivatives, including these five derivatives that were previously reported (H2-10, H2-12, H2-20, H2-29, and H2-27) and the newly synthesized H2-28, were further explored in the present investigation.

Inhibition of YycG′ enzyme activity by the derivatives

To detect the enzyme inhibiting activities of the derivatives on the YycG protein, recombinant YycG′ (∼29 kDa) was expressed as a fusion protein with an N-terminal maltose binding protein tag and purified using the ProBond™ Purification System (protein purity >90%). The enzymatic activity of the recombinant YycG′ was measured using the Kinase-Glo™ Luminescent Kinase assay, and all 6 derivatives displayed dose-dependent inhibition of its autophosphorylation activity. At 100 μmol/L (50 μg/mL), H2-10, H2-12, H2-20, H2-29, H2-27, and H2-28 inhibited YycG′ (0.13 μmol/L) enzyme activity by 55.22%, 63.70%, 83.65%, 50.75%, 78.31%, and 58.59%, respectively. The IC50 values of the six derivatives were 88.35 μmol/L (42.9 μg/mL), 61.15 μmol/L (31.9 μg/mL), 34.83 μmol/L (17.2 μg/mL), 66.68 μmol/L (35.1 μg/mL), 22.15 μmol/L (10.9 μg/mL), and 82.51 μmol/L (40.7 μg/mL), respectively, and the IC50 of Compound 2 was 47.9 μmol/L (24.9 μg/mL) (Table 1).

Table 1 Biological effects of the six derivatives.
Full size table

Antimicrobial activity of the derivatives

The MIC values of the six derivatives for S epidermidis ATCC 35984 and 12228 were determined. All 6 derivatives were found to inhibit bacterial growth more effectively than Compound 2. The MIC values of H2-10, H2-12, H2-20, H2-29, H2-27, and H2-28 on S epidermidis ATCC 35984 were 24.3, 6.5, 6.2, 3.3, 3.1, and 1.5 μg/mL, respectively; and the MBC values were 48.6, 52.2, 12.4, 52.6, 12.4, and 6.2 μg/mL, respectively (Table 1). Additionally, all derivatives inhibited the growth of Sepidermidis ATCC 12228 and S aureus (ATCC 49230 and ATCC 25923) (Table 2), whereas they had no effect on the growth of E coli strain ATCC 25922 at the highest tested concentration (100 μg/mL). The MIC/MBC values of these 6 derivatives ranged from 1/2 to 1/4, except for those of H2-12 and H2-29 (which had MIC/MBC values of 1/8 and 1/16, respectively).

Table 2 Anti-Staphylococcus activities of the derivatives.
Full size table

Effects of the derivatives on S epidermidis biofilm proliferation

Effects of the derivatives on the proliferation of 6-h-old S epidermidis ATCC 35984 biofilms were detected by a microtiter plate assay. All derivatives showed biofilm-killing effects on immature (6-h-old) S epidermidis ATCC 35984 biofilms (Figure 2), especially H2-28 and H2-29 [minimal biofilm-killing concentration (MBKC)=3.1 μg/mL and 3.3 μg/mL, respectively], which were stronger than Compound 2 (MBKC=51.9 μg/mL) (Table 1).

Figure 2
figure 2

Macroscopic profiles of the biofilms co-cultured with the derivatives. S epidermidis strain ATCC 35984 was cultured in polystyrene microtiter plates at 37 °C for 6 h. After removal of planktonic cells, fresh medium containing serial dilutions of the derivatives was added to the 6-h-old biofilm, and then incubated at 37 °C for another 18 h. After incubation, the biofilms were visualized by crystal violet staining. The biofilm treated with 0.1% DMSO (NC) as well as the 6-h-old biofilm without further incubation served as controls. The images are representative of results from three independent experiments.

Full size image

Bactericidal effects of the derivatives on mature S epidermidis biofilms

The bactericidal activities of the derivatives on mature (24-h-old) S epidermidis biofilms were detected by a confocal microscope; treatment with 0.1% DMSO or vancomycin (128 μg/mL) served as the controls. Each derivative was added individually to a mature biofilm at a concentration of 4×MIC, and incubated for 16 h. Significant reductions of viable cells in the mature biofilms were observed by CLSM with Live-Dead staining. All of the derivatives (24.3 μg/mL H2-10, 26.1 μg/mL H2-12, 24.7 μg/mL H2-20, 13.1 μg/mL H2-29, 12.4 μg/mL H2-27, and 12.4 μg/mL H2-28) showed bactericidal activity against bacteria in mature biofilms, especially H2-20, H2-27, and H2-29 (Figures 3). In contrast, DMSO (0.1%) and vancomycin (128 μg/mL, Bio Basic Inc, Canada) showed little effect on bacterial viability within a biofilm (Figure 3).

Figure 3
figure 3

Bactericidal effects of the derivatives on mature S epidermidis biofilms. S epidermidis ATCC 35984 was grown in cover-glass cell-culture dishes at 37 °C for 24 h. Subsequent to the removal of planktonic cells, the 24-h-old biofilms were further incubated at 37 °C for another 16 h with fresh TSB containing the following substances: 0.1% DMSO, 128 μg/mL vancomycin, 24.3 μg/mL H2-10, 26.1 μg/mL H2-12, 24.7 μg/mL H2-20, 13.1 μg/mL H2-29, 12.4 μg/mL H2-27, or 12.4 μg/mL H2–28. After incubation, the biofilms on the dishes were washed with normal saline and stained with Live-Dead reagents (containing SYTO9 and PI), and observed under CLSM using a 63×objective lens. Images representative of the results from three independent experiments were three-dimensionally reconstructed using Imaris software based on CLSM data at approximately 0.5 μm increments. The green fluorescent cells are viable, while red fluorescent cells indicate dead bacteria.

Full size image

Cytotoxicity and hemolysis of the derivatives in vitro

The cytotoxicity of the derivatives to mammal cells was investigated using Vero cells and an MTT assay. Cells treated with 0.1% DMSO and untreated cells were used as controls. Compared with the control group, after treatment with the derivatives, no obvious cytotoxicity to Vero cells was detected. The CC50 values of all the derivatives were higher than 100 μg/mL, which was the highest concentration used in the present study (Table 1).

Hemolysis of healthy human erythrocytes induced by the derivatives was examined, with vancomycin treatment as a control. Erythrocytes treated with 1% Triton-X100 and untreated erythrocytes served as complete hemolysis (100%) and no hemolysis (0%) controls, respectively. At the MIC concentrations, none of the derivatives displayed obvious induction of hemolysis of healthy human erythrocytes (<1% compared to the control), even at the highest concentrations (100 μg/mL). Derivatives H2-20, H2-27, and H2-28 had the lowest hemolytic activities (Table 1).

Discussion

S epidermidis and S aureus biofilm formation have become two of the most prevalent causes of nosocomial infections, especially in patients with prosthetic medical devices such as indwelling catheters and implanted foreign polymer bodies28, 29. As biofilms render bacteria increasingly resistant to multiple antibiotics30, 31, 32 and host defenses33, 34, chronic biofilm infections persist in patients, and it often becomes necessary to remove the implanted devices29. New antimicrobial agents are thus urgently needed to combat biofilm-associated infections.

TCSs are composed of a sensor histidine kinase capable of autophosphorylation in response to an environmental signal and a response regulator that interacts with the phosphorylated HK and regulates the expression of specific genes35. The YycG/YycF TCS has been extensively studied in recent years because of its essential role in pathogenic bacteria16, 18, 36. YycG/YycF regulates bacterial murein biosynthesis19, 37, 38, 39, 40, 41, 42, cell division16, 17, 39, lipid integrity18, 37, 42, 43, virulence factor expression19, 40, 42, 44, 45, 46, exopolysaccharide biosynthesis, and biofilm formation19, 44, 46, 47, 48. The YycG/YycF TCS is highly conserved in low G+C gram-positive bacteria but is absent in mammals and is thus considered a potential drug target in pathogenic bacteria24, 27, 49, 50, 51, 52. We previously reported two newly discovered compounds targeting YycG24 that clearly possess biofilm-killing activities against S epidermidis31. To improve the antibacterial activities of these compounds, a series of derivatives were designed and synthesized by modifying the functional groups of Compound 2 while keeping the thiazolidione core structure intact25. When the (5-vinyl-furan-2-yl)-benzoic acid fragment was incorporated into the thiazolidine-4-ones scaffold, the rigidity of the determined compound was enhanced. That change might stabilize the thiazolidiones in the bonding pocket. Six out of thirty-five derivatives were found to possess higher antibacterial activities than the compound they were derived from, and they inhibited autophosphorylation of YycG, suggesting that the bactericidal activity of these derivatives is based on inhibiting the enzyme activity of the YycG HK domain. However, the antibacterial activities of the derivatives did not always correlate with their IC50 values, as was the case with derivative H2-28 in the present study. This may be because the antibacterial activities of the derivatives may be affected by absorption by the bacterial cell, cell membrane permeability, or cellular metabolism, whereas the IC50 of the YycG inhibitors was determined by an autophosphorylation assay analyzing the direct interaction between the inhibitor and purified YycG′ protein in vitro. The YycG-targeting property of these derivatives accounts for their effectiveness against S epidermidis and S aureus, which are both low G+C gram-positive bacteria with highly conserved YycG/YycF TCSs; this is also consistent with the low cytotoxicity of these derivatives to mammalian cells and with the absence of inhibitory activity on the growth of E coli strain ATCC 25922, as mammalian and E coli cells do not have genes homologous to yycG/yycF.

Bacteria in biofilms are 100–1000 times more resistant to antibiotics than planktonic cells30, 31, 32, and vancomycin has little effect on S epidermidis biofilms even at high concentrations28, 53. The anti-biofilm activities of the six derivatives described here toward S epidermidis ATCC 35984 were improved compared to Compound 2. Lower concentrations of the derivatives compared with Compound 2 killed all bacteria in immature (6-h-old) biofilms and eliminated biofilm proliferation. More diluted derivatives did not eliminate biofilm formation but disrupted its structure so that it lost mechanical stability and could be washed away easily (Figure 2). The derivatives also displayed strong bactericidal activities toward cells in mature (24-h-old) biofilms, especially H2-20, H2-27, and H2-29, whereas Compound 2 mainly killed cells located at the bottom of the biofilm28.

In summary, six additional effective YycG inhibitors were designed and synthesized by modifying the chemical structure of the YycG inhibitor Compound 2. Their bactericidal and biofilm-killing activities were significantly better than those of Compound 2. The modification of anti-YycG leading compounds will help to discover new agents to combat biofilm infections and multidrug-resistant bacterial infections.

Author contribution

Ren-zheng HUANG, Hua-yong LIU, Shi-qing HAN, and Di QU designed the research; Ren-zheng HUANG, Li-kang ZHENG, Hua-yong LIU, and Bin PAN performed most of the experiments; Tao ZHU, Dan-bin JIANG, and Wei WANG provided critical technical support; Jian HU and Yang WU analyzed the data; Ren-zheng HUANG, Hua-yong LIU, You-cong WU, Shi-qing HAN, and Di QU wrote the paper.