Antibacterial Effects of Glycyrrhetinic Acid and Its Derivatives on Staphylococcus aureus

Staphylococcus aureus is a major pathogen in humans and causes serious problems due to antibiotic resistance. We investigated the antimicrobial effect of glycyrrhetinic acid (GRA) and its derivatives against 50 clinical S. aureus strains, including 18 methicillin-resistant strains. The minimum inhibitory concentrations (MICs) of GRA, dipotassium glycyrrhizate, disodium succinoyl glycyrrhetinate (GR-SU), stearyl glycyrrhetinate and glycyrrhetinyl stearate were evaluated against various S. aureus strains. Additionally, we investigated the bactericidal effects of GRA and GR-SU against two specific S. aureus strains. DNA microarray analysis was also performed to clarify the mechanism underlying the antibacterial activity of GR-SU. We detected the antimicrobial activities of five agents against S. aureus strains. GRA and GR-SU showed strong antibacterial activities compared to the other three agents tested. At a higher concentration (above 2x MIC), GRA and GR-SU showed bactericidal activity, whereas at a concentration of 1x MIC, they showed a bacteriostatic effect. Additionally, GRA and GR-SU exhibited a synergistic effect with gentamicin. The expression of a large number of genes (including transporters) and metabolic factors (carbohydrates and amino acids) was altered by the addition of GR-SU, suggesting that the inhibition of these metabolic processes may influence the degree of the requirement for carbohydrates or amino acids. In fact, the requirement for carbohydrates or amino acids was increased in the presence of either GRA or GR-SU. GRA and GR-SU exhibited strong antibacterial activity against several S. aureus strains, including MRSA. This activity may be partly due to the inhibition of several pathways involved in carbohydrate and amino acid metabolism.


Introduction
Staphylococcus aureus is a commensal bacterium in humans that can also be pathogenic primarily as an opportunistic infectious agent. S. aureus causes various suppurative diseases, food

S. aureus growth curve
Overnight cultures of S. aureus MW2 were adjusted to an OD 660 of 1.0. Then, 100 μl of bacterial culture was inoculated into 5 ml of TSB and incubated at 37°C with shaking. When the OD 660 reached 0.3, various concentrations of either GRA or GR-SU were added to the medium, and the growth was monitored. To investigate cell viability, appropriate dilutions of TSB after the addition of either GRA or GR-SU were inoculated onto a tryptic soy agar (TSA) plate. After overnight incubation at 37°C, the colony forming units (CFUs) were counted. Three independent experiments were performed, and the mean ± SD was calculated. The data were analyzed for statistically significant differences compared to untreated control at each timepoint by a two-way ANOVA followed by Dunnett's post hoc test.

Effects of GRA and GR-SU on cell viability
Overnight cultures of S. aureus MW2 cells were washed with 10 mM sodium phosphate buffer (PB; pH 6.8) and suspended in PB. The bacterial suspension was diluted to 10 7 cells/ml with PB, and 10 μl of the diluted suspension was inoculated into 500 μl of PB containing various concentrations of GR-SU (1x MIC: 128 mg/L) or GRA (1x MIC: 512 mg/L) and incubated for 1 or 2 hours at 37°C. Dilutions of the reaction mixture (100 μl) were plated onto TSA plates and incubated at 37°C overnight. The antibacterial effect was calculated as the ratio of the number of surviving cells (survival rate %) to the total number of bacteria incubated in PB. To verify the cell number of each strain inoculated in PB, the dilutions of the bacterial suspension prior to inoculation were plated onto TSA and incubated at 37°C overnight. Three independent experiments were performed, and the mean ± SD was calculated. The data were analyzed for significant differences compared to untreated control using two-way ANOVA followed by Dunnett's post hoc test.

Effects of GRA and GR-SU in combination with antibiotics
The effects of combination treatments of gentamicin (Wako Pure Chemical Industries Ltd, Osaka, Japan), chloramphenicol (Sigma-Aldrich Corporation, Tokyo, Japan), tetracycline (Wako Pure Chemical Industries Ltd), oxacillin (Sigma-Aldrich Corporation) or ofloxacin (Daiichi Sankyo Company, Tokyo, Japan) with either GRA or GR-SU were investigated. Overnight cultures of S. aureus MW2 were adjusted to 10 7 cells/ml, and 10 μL of the bacterial suspension was applied to each well in a 96-well plate. Antibiotics were dissolved in TSB alone or TSB containing either GRA or GR-SU (1/2, 1/4 or 1/8 MICs for each agent). Stock solutions (1 g/L) of gentamicin, oxacillin or ofloxacin were prepared in TSB or TSB containing either GRA or GR-SU. Stock solutions (10 g/L) of chloramphenicol and tetracycline were prepared in ethanol. The final concentrations of gentamicin, chloramphenicol, tetracycline, oxacillin and ofloxacin were 32 mg/L, 32 mg/L, 64 mg/L, 128 mg/L, and 4 mg/L, respectively, and these solutions were then serially diluted 2-fold with the same media. After a 24 h incubation at 37°C, the MICs were determined as described above. The fractional inhibitory concentration (FIC) was calculated with a method previously described [34]. FIC indexes of 0.5, >0.5 -4.0 and >4.0 were defined as synergy, no interaction and antagonism, respectively.

Microarray analysis
Overnight cultures of S. aureus (10 8 cells) were inoculated into 30 ml of fresh TSB and cultured at 37°C with shaking. When the OD 660 reached 0.3, GR-SU (1x MIC: 128 mg/L) was added to the medium. After 2 h of incubation, the bacterial cells were collected by centrifugation at 5,000 x g for 5 min at 4°C and stored at -80°C until further use. Additionally, bacterial cells without GR-SU treatment were prepared as the control. Total RNA was extracted from the bacterial cells using the FastRNA Pro Blue Kit (MP Biomedicals, Cleveland, OH, USA) according to the manufacturer's protocol. Then, cDNA was synthesized from 10 μg of total RNA using the FairPlay III Microarray Labeling Kit (Agilent Technologies, Santa Clara, CA, USA) according to the manufacturer's instructions. The Agilent eArray platform (Agilent Technologies) was used to design the microarray; 13,939 probes (60-mers) were designed for the 2,628 protein-coding genes of S. aureus MW2 (up to five probes per gene). For microarray analyses, the test (GR-SU treatment) and control cDNAs were labeled with Cy3 mono-Reactive Dye and Cy5 mono-Reactive Dye (GE Healthcare, UK), respectively. The fluorescently labeled cDNAs were purified using the FairPlay III Microarray Labeling Kit (Agilent Technologies) The Cy3-labeled and Cy5-labeled cDNAs were mixed and hybridized onto an array using the Hi-RPM Gene Expression Hybridization Kit (Agilent Technologies). The arrays were scanned with an Agilent scanner, and data extraction, filtering and normalization were conducted using the Agilent Feature Extraction software according to the manufacturer's instructions. The experiments were performed using two biological replicates (two technical replicates for each condition), and the expression data were deposited into the Gene Expression Omnibus (http:// www.ncbi.nlm.nih.gov/geo/) under accession GSE80500. We compared difference between control and test sample for each gene by student t-test using Cyber-T (http://cybert. microarray.ics.uci.edu/). Factors that exhibited a greater than 2-fold change (increase/decrease) and P<0.05 were defined as being altered by GR-SU.

Gene expression analysis by quantitative PCR
Overnight cultures of S. aureus MW2 (10 8 cells) were inoculated into 5 ml of TSB and aerobically incubated at 37°C with shaking. When the OD 660 reached 0.3, either 1/4 or 1x MIC of GR-SU or GRA was added to the culture. The bacterial cells were collected after 1, 2, 4 and 8 h of incubation. RNA extraction and cDNA synthesis were performed as described above. Quantitative PCR was performed with the LightCycler Nano system (Roche, Tokyo, Japan) using cDNA as the template DNA. The primers used are shown in Table 1. The results were normalized to the housekeeping gene gyrB. Three independent experiments were performed, and the mean ± SE was calculated. doi:10.1371/journal.pone.0165831.t001

Effect of GRA and GR-SU on carbohydrate and amino acids requirements
Chemically defined medium (CDM) and CDM containing 1/2 MIC or 1/4 MIC of GRA-K, GR-SU or GRA were used in this experiment. Overnight cultures of S. aureus MW2 were adjusted to 1x10 7 cells/ml. Glucose, trehalose, sucrose or fructose (each 50 mM) was serially diluted 2-fold with either TSB or TSB containing GRA, K-GRA or GR-SU (1/2 MIC or 1/4 MIC at the final concentration for each agent). Then, 10 μl of the bacterial culture (1x10 5 cells) was applied to each well. After incubation for 24 h at 37°C, the minimum concentration of each sugar that showed visible bacterial growth was determined.
To investigate the effect on amino acid requirement, we used CDM containing 18 amino acids necessary for S. aureus growth. The amino acids used in this experiment and their concentrations have been previously described [33]. The CDM was prepared using 2-fold serial dilutions with CDM depleted of all amino acids. Then, 10 μl of the bacterial culture (1x10 5 cells) was applied to each well. After incubation for 24 h at 37°C, the minimum dilution rate that showed visible bacterial growth was determined.

Antimicrobial activities of GRA and its derivatives against S. aureus strains
The MICs of GRA and its 4 derivatives are shown in Table 2. The MICs were evaluated using 50 S. aureus strains (19 MRSA strains and 31 MSSA strains). The MICs of GR-K and GR-S were higher than GA, GR-SU and GR-SA. The MICs of GR-K and GR-S against all strains were more than 256 mg/L. While GR-SA exhibited an intermediate level of MICs on the tested strains (ranged from 64 to 4,096 mg/L except 2 strains [more than 4,096 mg/L]), the MICs of GRA and GR-SU ranged from 16 to 512 mg/L and from 16 to 256 mg/L, respectively. The distribution of MICs in the MRSA and MSSA strains is shown in Table 2. We found no differences in susceptibility between the MRSA and MSSA strains.
Effects of GRA and GR-SU on S. aureus growth As shown in Fig 2, S. aureus MW2 growth was inhibited by the addition of either GRA or GR-SU in a dose-dependent manner. The addition of more than 1x MIC of GRA or GR-SU fully inhibited the growth of S. aureus MW2 as expected, but growth was already inhibited by both agents at sub-MIC concentrations (1/16 and 1/4 MIC). We also investigated the effects of GRA and GR-SU on cell viability in a time course experiment (Fig 2). The result was similar to that obtained with the OD. The viable cell number was suppressed by the addition of either GRA or GR-SU in a dose-dependent manner. The bacterial cell numbers were decreased after 4 h by the addition of 2x MIC of GRA or GR-SU. Additionally, we analyzed a methicillin-sensitive strain (MS23513) and obtained similar findings (S1 Fig).

Effects of GRA and GR-SU on cell viability
Because GRA and GR-SU exhibited antimicrobial activity against S. aureus strains during active growth, we investigated whether this activity was observed under non-growing conditions. We treated S. aureus MW2 cells with either GRA or GR-SU in 10 mM PB. As shown in Fig 3, the viable cell number of the MW2 strain treated with 1x MIC and 2x MIC of GRA or GR-SU showed no significant difference compared to the control (no treatment). Synergistic effects of GRA and GR-SU against 8 S. aureus strains Next, we evaluated the synergistic effects of GRA and GR-SU in combination with other antibiotics. As shown in Table 3, all strains tested showed synergy with the combination of gentamicin with either GRA or GR-SU, while one strain showed synergy with the combination of tetracycline or chloramphenicol with GR-SU, and 2 strains showed synergy with the combination of oxacillin with GRA. Oxacillin, ofloxacin and erythromycin had no effect (no interaction) when combined with GR-SU, and chloramphenicol, ofloxacin and erythromycin had no effect when combined with GRA. Furthermore, the MICs of antibiotics in the absence or presence of GRA or GR-SU are shown in S2 Table. A synergistic effect of GRA and GR-SU with gentamycin against both gentamicin-resistant (4 to 128 mg/L) and susceptible strains (3 strains with MIC of 1 mg/L) was found (S2 Table).  Table).

Microarray analysis
We performed a microarray analysis to identify genes with altered expression patterns in response to GR-SU. As summarized in Fig 4, more than 200 genes exhibited altered expression levels in response to GR-SU. Among these, 39 phage-related genes were increased by the addition of GR-SU. In the S. aureus MW2 genome, two phage-related operons (MW1380-MW1451 and MW1882-MW1952) were found. Specifically, in the operon of MW1380-MW1451 (72 genes), the expression of 31 genes was increased. Many genes responsible for metabolic and biosynthetic pathways also exhibited altered expression patterns (S3-S6 Tables). The expression levels of 6 genes involved in protein synthesis, 10 genes in RNA synthesis and 2 genes in amino acid metabolism were increased, whereas the levels of 8 genes for RNA synthesis, 12 genes for carbohydrate metabolism and 8 genes for amino acid metabolism were decreased. In addition, the expression levels of 2 transporter and substrate-binding protein genes were increased, and 16 genes were decreased after GR-SU treatment. Regarding pathogenic factors, the expression of 3 factors was increased while that of 8 factors was decreased. Effect of GRA and GR-SU on virulence factor expression As we found in the microarray analysis, the expression levels of several virulence-related genes were altered in response to GR-SU. We analyzed the gene expression of RNAIII, which is a key factor involved in the expression of many virulence genes, in more detail by quantitative PCR. As shown in Fig 5A and 5B, RNAIII expression was significantly increased during the exponential phase in the absence of GRA or GR-SU, whereas GRA and GR-SU both fully suppressed RNAIII expression at 1/4 MIC. The expression of seh, which is a RNAIII-dependent virulence factor, exhibited a pattern similar to that of RNAIII in response to GRA and GR-SU added to the medium. The growth and colony counts were monitored. Three independent experiments were performed, and the mean ± SD was calculated. The data were analyzed for statistically significant differences compared to untreated control by a two-way ANOVA followed by Dunnett's post hoc tests. *P<0.05. doi:10.1371/journal.pone.0165831.g002 (Fig 5C and 5D). These results suggest that the repression of RNAIII by GR-SU may explain the alterations in gene expression of several virulence-related genes, which were observed in the microarray analysis (Fig 4). Three independent experiments were performed, and the mean ± SD was calculated. The data were analyzed for statistically significant differences compared to the untreated control by a two-way ANOVA followed by Dunnett's post hoc test. Significant differences were not observed.  Therefore, we investigated the expression levels of 12 virulence factors and found that several toxins such as sea, clfAB and hlgC were downregulated by GRA, whereas the cell surface factors fnbA and ica were upregulated (Fig 6). The effect of GR-SU on virulence factor expression was similar to that of GRA.

Effects of GRA and GR-SU on sugar and amino acid requirements
Because amino acid and carbohydrate metabolism tended to decrease following a 2-hour exposure to GRA and GR-SU based on the microarray analysis, we hypothesized that GRA and GR-SU may inhibit the growth of S. aureus by influencing amino acid and carbohydrate metabolism. Therefore, we investigated sugar and amino acid requirements in the presence of GRA, GR-SU or GRA-K (1/2 MIC or 1/4 MIC at the final concentration for each agent). As shown in Table 4, the minimum concentrations of each sugar for the growth of S. aureus MW2 in CDM containing glucose, sucrose, lactose or trehalose was increased by 2-or 4-fold in the presence of 1/2 MIC of GRA or GR-SU, whereas GRA-K did not increase the sugar requirement. Additionally, we investigated the amino acid requirements. GRA and GR-SU treatment caused a 2-fold higher requirement compared to untreated bacteria. When 1/4 MIC of GRA or GR-SU was used, the minimal concentration of each sugar or amino acids was not altered.

Discussion
In this study, we demonstrated the antibacterial activities of GRA and GR-SU against S. aureus strains. These two compounds exhibited strong antibacterial activities, but the antibacterial activities of the other GRA derivatives (GR-S, GR-K and GR-SA) were much weaker than those of GRA and GR-SU. GR-SU is synthesized from GRA by adding a succinic acid moiety to increase water solubility. GR-K (dipotassium glycyrrhizate) is a potassium salt of glycyrrhizin (GRA with two glucuronic acids) and also has high water solubility. Both GR-SU and GR-K have high water solubility, but the antibacterial effect of GR-K is much lower than that of GR-SU. Thus, difference in water solubility is not related to the difference in antibacterial activity among these compounds. Instead, some other structural differences between the two agents (Fig 1) may affect their antibacterial activities against S. aureus strains. GRA and GR-SU have been clinically used as anti-inflammatory agents [21,32]. Taken together with our findings, GRA and GR-SU are expected to exert two biological activities of clinical importance: antibacterial and anti-inflammatory activities.  Although several investigations regarding the antibacterial effects of GRA have been reported [28,29], this is the first report of the use of GR-SU as an antibacterial agent. Glycyrrhetinic acid derivatives are known anti-inflammatory drugs [30,31,32]. Stronger Neo-Minophagen C, which is the most well-known glycyrrhetinic acid derivative drug, contains 2 mg/ml of monoammonium glycyrrhizinate hydrate at the dose used to clinically treat chronic hepatitis; this dose is approximately 4-to 20-fold higher than the concentration used in this study [26,27]. Therefore, GRA and GR-SU may have a potential use against S. aureus infectious diseases caused by antibacterial and anti-inflammatory effects.
Long, D.R. et al. previously reported the bacteriostatic activity of GRA and the bactericidal effect of GRA at a high concentration (above 62.5 mg/L) against S. aureus [28]. In this study, we also observed bactericidal activity of GRA and GR-SU at a 2x MIC dose (GR-SU: 256 mg/L, GRA: 1,024 mg/L). As shown in the microarray analysis (Fig 4), the expression levels of many phage-related genes were increased by the addition of 1x MIC of GR-SU, and we speculated that the bactericidal effect of a high dose of GR-SU (2xMIC) was due to phage induction-mediated cell lysis.
Among the genes that exhibited altered expression following the addition of GR-SU, many genes involved in sugar (S3 Table), amino acid (S4 Table) and nucleic acid (S5 Table) metabolism were significantly decreased. Moreover, the expression levels of many genes that encoded transporters were also decreased (S6 Table). These results suggest that GRA and GR-SU-treatments inhibit the uptake of nutrients, which then causes alterations in the expression of genes related to nutrient metabolism. Supporting this hypothesis, S. aureus MW2 required a higher concentration of sugars and amino acids in the presence of GRA and GR-SU than in the absence of these compounds (Table 4). Based on these findings, we propose that GRA and GR-SU have no specific mechanism for antibacterial activity but induce antimicrobial effects by inhibiting nutrient acquisition and affecting bacterial metabolism. Additionally, the expression of several transporters responsible for the efflux and/or influx of compounds was increased in the microarray analysis. The transporters with altered expression levels in the presence of either GRA or GR-SU included some transporters that might be involved in the high susceptibility to gentamicin in the presence of either GRA or GR-SU. GRA and GR-SU had weak synergistic activities with other protein synthesis inhibitors, speculating that transporters specific for gentamicin influx may be influenced by GRA and GR-SU.
In this study, we found that the expression of RNAIII, which is responsible for the regulation of many S. aureus virulence factors, was fully repressed by GRA and GR-SU [35,36]. Because RNAIII is related to the agr quorum sensing system, RNAIII expression is growth dependent [35]. We found that GRA and GR-SU suppressed growth in a dose-dependent manner, suggesting that RNAIII expression was inhibited. Then, we investigated the expression of several virulence factors in the presence or absence of GRA or GR-SU. As shown in Fig 6, the expression levels of toxins such as enterotoxins (seh and sea), staphylokinase (sak) and hlgC were decreased, whereas the expression levels of cell surface factors such as fnbA and ica were increased. From these results, we hypothesize that GRA and GR-SU can modulate virulence in addition to their antimicrobial effects on S. aureus.
In conclusion, we have found indications that the antibacterial activity of GRA and GR-SU are mostly occurring via effects on nutrient metabolism. Additionally, these agents suppressed the expression of several virulence genes. Taken together with previous reports of inflammatory effects, GRA and GR-SU may have potential for clinical use against S. aureus infectious diseases.