The leaves of A. henryi (Oliv.) Harms were collected in October 2012 in Xinhua, Changsha, China. The plant species was confirmed by Professor Xiang-Qian Liu (Hunan Key Laboratory of Traditional Chinese Medicine Modernization, Hunan University of Chinese Medicine, Changsha, China) and the voucher specimen (no. 20121125) was deposited at the School of Pharmacy, Hunan University of Chinese Medicine.

The dried leaves of A. henryi (Oliv.) Harms (10 kg) were cut into small pieces, extracted three times with MeOH (3×100 ml) at room temperature, and concentrated under reduced vacuum to obtain a dark-green residue (0.8 kg). The residue was then suspended in H₂O and partitioned with petroleum ether. The water fraction was fractionated using column chromatography (CC) on macroporous resin and eluted with a gradient of EtOH/H₂O (0, 30, 50, 75 and 95%) into five fractions (1–5). Fraction 4 (75% EtOH, 14.0 g) was subjected to silica gel CC and eluted with CHCl₃/MeOH/H₂O (25:1:0/1:1:0.2) to give fifteen fractions (A-O). Fraction C (119 mg) was re-fractionated on silica gel H CC and eluted with CHCl₃/MeOH/H₂O (15:1:0/6:1:0.1) to give six sub-fractions (C1-C6). Sub-fraction C3 (106.0 mg) was subjected to ODS CC and eluted with a gradient of MeOH/H₂O (70, 80, 90 and 100%) to yield 12.0 mg URS (16).

The compound structures were identified using mass spectros copy, 1D-nuclear magnetic resonance (NMR) and 2D-NMR and the spectral data were compared with those reported previ ously (16). ¹H NMR and ¹³C NMR spectra were measured on a Varian INOVA 400 M spectrometer (Agilent Technologies, Inc., Santa Clara, CA, USA) with chemical shifts reported as ppm (tetramethylsilane as internal standard). Electrospray ionization mass spectra were then measured using an Agilent 6530 Accurate-Mass Q-TOF (Agilent Technologies, Inc.).

The purity of URS was >98%, as determined via HPLC as previously described (17). Briefly, URS was dissolved in MeOH to a concentration of 0.1 mg/ml for HPLC analysis by using a Kinetex XB-C18 analytical column (100×4.6 mm ×2.6 µm; Phenomenex, Inc., Torrance, CA, USA) at 30°C. Elution was conducted using mobile phase A (water) and mobile phase B (acetonitrile) with a gradient as follows: 0–2 min, 29–31% B; 2–13 min, 31–35% B; 13–15 min, 35–40% B; 15–23 min, 40–44% B; 23–25 min, 44–46% B; 25–31 min, 46–49% B; and 31–38 min, 49–55% B. The flow rate was constant at 1.0 ml/min and the effluents were monitored at 210 nm using an Agilent 1200 HPLC system with variable wavelength detector (Agilent Technologies, Inc.). The purity value was found to be >98% using a peak area normalization method. The purity value was obtained by calculating the percentage of the URS peak area to that of the total peaks in the HPLC chromatogram.

Among the eight strains of S. aureus used in the present study, two clinical MRSA isolates, DPS-1 and DPS-2, as the references (18,19) mentioned, were collected from two different patients at Wonkwang University Hospital (Iksan, Korea); two strains were MRSA ATCC 33591 and methicillin-susceptible S. aureus (MSSA) ATCC 25923, purchased from the American Type Culture Collection (Manassas, VA, USA); and the remaining four MRSA strains, CCARM 3090, CCARM 3091, CCARM 3095, CCARM 3102, were provided by the Culture Collection of Antimicrobial Resistant Microbes (National Research Resource Bank, Seoul, Korea). All bacteria were cultured on either Mueller-Hinton agar (MHA) or Brain Heart Infusion agar at 37°C for 24 h. The bacterial strains were suspended in either Mueller-Hinton broth (MHB) or brain heart infusion broth (BHIB) and grown at 37°C for 24 h in order to perform the experiments. The bacteria were stored in 30% glycerol and frozen at −80°C.

Difco™ Mueller-Hinton agar, Difco™ Mueller-Hinton broth, Difco™ brain heart infusion agar, Bacto™ Brain Heart Infusion broth and Difco™ skim milk were obtained from Difco Laboratories (Baltimore, MD, USA). Glycerol was obtained from Sigma-Aldrich; Merck KGaA (Darmstadt, Germany). MTT, Triton X-100 (TX-100), N,N′-dicyclohexylcarbodiimide (DCCD), paraformaldehyde, glutaraldehyde, sodium cacodylate buffer, osmium tetroxide, uranyl acetate, EtOH, MeOH, propylene oxide, Spurr's resin, OXA and solvents were purchased from Sigma-Aldrich; Merck KGaA. SMART™ bacterial protein extraction solution was purchased from Intron Biotechnology, Inc. (Seongnam, Korea). The chemiluminescent ECL assay kit was purchased from ATTO Corp. (Tokyo, Japan).

A total of 8 bacterial strains were subjected to antimicrobial susceptibility and MIC assays. The MIC values of URS and the antibiotic, OXA, against MRSA and MSSA were determined via broth microdilution assay using a 96-well microplate, according to a previous study (19). A series of 2-fold dilutions of URS in MHB and BHIB were prepared and the bacteria colonies were picked with a 1 µl white sterile loop and needle to be suspended in either MHB or BHIB. The inocula were adjusted to the 0.5 of the McFarland standard scale [~1.5×10⁸ colony-forming units (CFU)/ml] and the final inocula were adjusted to 1.5×10⁵ CFU/spot. The inoculated broth was incubated at 37°C for 24 h and the MIC was determined using MTT reagent. Following the 24 h incubation, MTT (1 mg/ml) was added to the broth suspension in every well and the plate was incubated for 30 min in a 37°C incubator. Blue color indicated the presence of bacteria (18,20). The MIC was defined as the lowest concentration that inhibited bacterial growth.

The synergistic effect between URS and OXA was determined using a checkerboard dilution test (21). Serial dilutions of URS with different concentrations OXA were mixed in cation-supplemented MHB and BHIB. Each test strain of the final inocula concentration was adjusted to 1.5×10⁵ CFU/ml and incubated at 37°C for 24 h. For the synergy studies, the range of concentrations used was determined according to the previously determined MIC of OXA for each specific isolate. The concentration of URS ranged from 6.25 to 0.19 µg/ml. Following a 24 h incubation, the MICs were interpreted. The in vitro interaction between the drugs was quantified by determining the fractional inhibitory concentration (FIC). The FIC of each agent was calculated as the MIC of the agents in combination, divided by the MIC of the agent alone. The FIC index (FICI) was calculated using the following formula: FICI = FIC_A + FIC_B = [A]/MIC_A + [B]/MIC_B, where [A] and [B] were the concentrations of drug A and B, respectively, and MIC_A/FIC_A and MIC_B/FIC_B were the MIC/FIC of drug A and B, respectively. The FICI was interpreted as follows: <0.5, synergy; 0.5–0.75, partial synergy; 0.75–1, additive effect; 1–4, no effect; and >4, antagonism. Finally, the different values of synergy between the two agents were calculated (22).

The synergistic antimicrobial effect was determined using a time-kill assay, as described previously (23). URS combined with OXA, OXA alone and URS alone were compared to control (drug-free) regarding the synergistic effect on the bacterial growth curve (24). At five different time phases (0, 4, 8, 16 and 24 h), bacterial growth curves were observed. Bacterial cultures were diluted with fresh MHB to ~1.5×10⁵ CFU/ml and the bacteria suspensions were incubated at 37°C for 24 h. Aliquots (0.1 ml) of the suspension were taken at 0, 4, 8, 16 and 24 h of incubation and serial 10-fold dilutions were prepared in saline as needed. Following incubation for 24 h, the number of viable cells was determined on a drug-free MHA plate. Colony counts were performed on plates and 30–300 colonies were calculated. The lower limit of sensitivity of the colony counts was 100 CFU/ml. The antimicrobial agents used were considered bactericidal at the lowest concentration that reduced the original inoculum by 3 log₁₀ CFU/ml (99.9%) for each of the indicated times. However, they were designated bacteriostatic if the inoculum was reduced by 0–3 log₁₀ CFU/ml.

To explore whether the antibacterial activity of URS was associated with the altered membrane permeability or the action of ATPase, the authors evaluated the antibacterial activity of URS in the presence of a detergent and an ATPase-inhibiting agent (19). To determine the detergent-induced permeabilization, an appropriate concentration of URS was determined using the detergent TX-100 (25), which significantly increases bacterial sensitivity to antibiotics (26). DCCD, a metabolic inhibitor that can decrease ATP levels by disrupting electrochemical proton gradients in a bacterial environment, was used as an inhibitor of ATPase (19,27). The bacteria culture were adjusted to 1.5×10⁵ CFU/ml and 100 µl/well was added to 96-well microplate. A total density of 10 µl/well of URS, at a concentration of 1/64 MIC, with the presence of 0.00001% TX-100 and 250 µg/ml DCCD were individually added to the 96-well microplate and incubated at 37°C for 24 h. The results were read at optical density 600 nm.

On the basis of biological activity, morphological changes in MRSA ATCC 33591 following the addition of URS were evaluated using TEM, according to a previously described protocol with some modifications (28,29). MRSA exponential phase cultures were prepared by diluting cultures in MHB overnight and cell growth was continued at 37°C until the cultures reached the mid-logarithmic phase of growth. The MHB-grown exponential-phase MRSA was treated with 31.25 µg/ml OXA alone, 1/2 MIC of URS alone, and 31.25 µg/ml OXA + 1/2 MIC of URS for 4 h. Following treatment, 2 ml of the culture was collected by centrifugation at 10,000 × g for 10 min. Following the removal of the supernatant, pellets were washed with MHB and primary fixed with 2% paraformaldehyde and 2% glutaraldehyde at 4°C for 2 h. Samples were washed and resuspended thrice in 0.05 M sodium cacodylate buffer (pH 7.2) at 4°C for 10 min, then post-fixed with 1% OsO₄ at 4°C for 2 h. The samples were then washed twice with sterile-distilled water at room temperature. Thereafter, the samples were en bloc stained with 0.5% uranyl acetate at 4°C for 30 min and dehydrated using a graded EtOH series. Finally, 100% propylene oxide was used for transit and infiltration with propylene oxide and Spurr's resin reagents in a specific ratio. The specimens were examined using an energy-filtering transmission electron microscope (LIBRA 120; Carl Zeiss GmbH, Oberkochen, Germany) operated at an accelerating voltage of 120 kV. Transmitted electron signals were recorded using a 4×4 k slow-scan charge-coupled device camera (Ultrascan 4000 SP; Gatan, Pleasanton, CA, USA) attached to an electron microscope.

The western blot assay was performed according to the standard procedures to measure the translated protein level (30,31). The MRSA culture (ATCC 33591) was grown at an OD600 of 0.4 in MHB and treated with various concentrations and combinations of OXA and URS for 4 h for western blot analysis (32). Briefly, cells were harvested and suspended in SMART™ bacterial protein extraction solution containing Tris-HCl (pH 7.5). The extraction was performed according to the manufacturer's protocol. Protein concentrations were measured using the Bio-Rad protein assay reagent (Bio-Rad Laboratories, Inc., Hercules, CA, USA) and cell lysates were separated using 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The electrophoresed gels were transferred to Amersham™ Hybond™-P-membranes (GE Healthcare Life Sciences, Chalfont, UK). The membranes were blocked with 5% skim milk for 1 h and hybridized with monoclonal mouse anti-PBP2a primary antibody (1:500, cat. no. 70PB001; DiNonA Inc., Seoul, Korea) overnight at 4°C. Loading differences were normalized with monoclonal anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) antibody (1:500, cat. no. sc-166574, Santa Cruz Biotechnology, Inc., Dallas, TX, USA). Following incubation with anti-mouse IgG secondary antibody (1:1,000, cat. no. G-21040; Enzo Life Sciences, Ann Arbor, MI, USA) at room temperature for 1 h, immunoreactive proteins were detected using a chemiluminescent ECL assay kit (ATTO Corp.) according to the manufacturer's instructions. Western blot bands were visualized using ImageQuant LAS 4000 Mini Luminescent image analyzer (GE Healthcare Life Sciences) and the quantitative measurement of band intensity was performed using ImageJ software (version 1.45S; National Institutes of Health, Bethesda, MA, USA)

Analyses were performed in triplicate and data were presented as the mean ± standard deviation. Results were statistically analyzed using an independent Scheffe's t-test (SPSS software version 22.0; IBM SPSS, Armonk, NY, USA). P<0.05 was considered to indicate a statistically significant difference.

Antimicrobial susceptibility studies were performed using a broth microdilution method. The MIC values of URS and OXA against eight strains of S. aureus are presented in Table I. The data indicated that URS, which was isolated from the leaves of A. henryi (Oliv.) Harms, had high antimicrobial activity against MRSA and MSSA. The MIC values of URS against MSSA and MRSA were 3.125 and 6.25 µg/ml, respectively. The MIC of OXA against MSSA was <0.97 µg/ml, whereas the MIC against MRSA ranged from 3.9 to 2,000 µg/ml. The high MIC values of OXA against MRSA confirmed that the studied strains were resistant to OXA, whereas the MIC value of OXA against MSSA indicated the susceptibility of this strain to the antibiotic (11).

Evaluation of the synergistic effect of URS and OXA in combination against MRSA was performed using a checkerboard dilution method. The results are presented in Table II and suggested that in the presence of URS, the susceptibility of MRSA to OXA increased. Treatment with 1/2 MIC URS in combination with OXA reduced the MIC of OXA by 2–32-fold.

On the basis of FIC indices, the synergism of URS and OXA against MRSA was confirmed using a time-kill assay. In the present study, two strains, CCARM 3090 and DPS-1, were used to perform the analysis. The results are reported in Fig. 2. The time-kill curves were constructed with control, 1/2 MIC OXA alone, 1/2 MIC URS alone, 1/2 MIC OXA and 1/2 MIC URS in combination, and 1/2 MIC OXA and 3/4 MIC URS in combination at time-points 0, 4, 8, 16 and 24 h. The results are presented as the log value of the number of surviving bacteria in the antimicrobial test at the different time intervals. As the figures indicated, at 24 h, treatment with 1/2 MIC OXA and 3/4 MIC URS in combination resulted in combined group bacteria counts that decreased to 3 log₁₀. However, the original antibacterial-free control count was 1.5×10⁵ CFU/ml, which increased to almost 10¹⁰ CFU/ml after 24 h (33). In addition, the time-kill curves indicated a concentration-dependent bactericidal effect against MRSA strains.

MRSA CCARM 3090 was used to investigate the effects of enhanced membrane permeability by using detergents and the diversification of susceptibility by using ATPase inhibitors on the activity of URS. The results are presented in Fig. 3. The membrane-permeabilizing agent, TX-100, can increase the permeability of the outer membrane in gram-negative bacteria (34). Compared to the OD600 value of URS alone (1/64 MIC), the OD600 value of the suspension in the presence of 1/64 MIC URS and 0.00001% TX-100 was reduced 56.6%. From the results of the OD600 values, in the presence of either URS alone (1/64 MIC) or 250 µg/ml DCCD alone, MRSA maintained its viability. However, the OD600 value of the suspension significantly decreased by URS in combination with DCCD. Bacterial viability in the presence of 1/64 MIC URS and 250 µg/ml DCCD reduced to 11.7%.

To determine morphological changes, MRSA ATCC 33591 treated with OXA (31.25 µg/ml) alone, URS (3.125 µg/ml) alone, and OXA and URS in combination was examined by TEM analysis. The results are indicated in Fig. 4. The images indicated characteristic morphological changes in the cells of MRSA ATCC 33591 after treatment with OXA and URS. The untreated bacterial cells had normal morphology with distinct septa (Fig. 4A). In the presence of OXA and URS individually, the cytoplasmic membranes of the bacterial cells were damaged and had rougher surfaces (Fig. 4B and C) compared to those of control cells. Following exposing MRSA to the combination of OXA and URS, deformation of bacterial cells was observed compared to groups treated with OXA alone and URS alone (Fig. 4D). This caused cell membrane disintegration, cell lysis and release of cytoplasmic contents. In addition, the cells appeared almost absent. This suggested a strong bactericidal activity against MRSA. The notable changes in bacterial cell morphology indicated that bacterial cell membrane viscosity and permeability were compromised by treatment with the combination of URS and OXA (24).

To detect the protein level of PBP2a in MRSA, western blotting was performed. PBP2a expression levels following the tested treatments are summarized in Fig. 5. GAPDH, which served as an internal control, was detected after all treatments (results not shown). The experimental samples consisted of control, OXA (31.25 µg/ml), URS (6.25 µg/ml) and the combination of OXA (31.25 µg/ml) and URS (6.25 µg/ml). As the figure demonstrated, PBP2a was not completely inhibited; however, the PBP2a protein level decreased non-significantly on the addition of URS and OXA alone. Compared to control, the expression of PBP2a of the combination group was reduced 36.98%. The decrease in the PBP2a level may indicate that URS interrupted the process of protein synthesis by damaging RNA.