Thorns of G. sinensis were purchased in October 2008 from Omniherb (Yeongcheon, Republic of Korea). The origin of these materials was confirmed taxo-nomically by Professors Je-Hyun Lee (Dongkuk University, Gyeongju, Republic of Korea) and Young-Bae Seo (Daejeon University, Daejeon, Republic of Korea). A voucher specimen (2008-ST22) has been deposited at the K-herb Research Center, Korea Institute of Oriental Medicine (Daejeon, Republic of Korea).

The dried thorns of G. sinensis (10 kg) were extracted three times with 70% ethanol by sonication for 1 h. The extracted solution was filtered through filter paper and evaporated to dryness (450 g). The ethanol extract was suspended in water (1 liter) and then successively partitioned with n-hexane, ethyl acetate and n-butanol (1.5 liters each, three times) to give extracts of 33.8, 103.2 and 119.1 g, respectively.

(-)-Epicatechin, ethyl gallate and quercetin (all with purity ≥99.0%) were purchased from ChromaDex, Inc., (Santa Ana, CA, USA). Caffeic acid, eriodictyol and (+)-catechin (all with purity ≥99.0%) were obtained from Acros Organics (Fair Lawn, NJ, USA), Extrasynthèse S.A. (Genay, France) and Fluka AG (Buchs, Switzerland), respectively. Methanol, high-performance liquid chromatography (HPLC)-grade reagents, water and acetonitrile were purchased from JT. Baker Chemical Co. (Phillipsburg, NJ, USA). Glacial acetic acid was of analytical reagent grade and was procured from Junsei Chemical Co. (Tokyo, Japan).

This analysis was performed using a Shimadzu LC-20A HPLC system (Shimadzu Co., Kyoto, Japan), which consisted of an on-line degasser, a solvent-delivery unit, an autosampler, a photoiode array (PDA) detector and a column oven. The data processor employed the LC solution software (version 1.24; Shimadzu Co.). A Gemini^® C18 analytical column (250×4.6 mm; particle size 5 µm; Phenomenex, Torrance, CA, USA) was used. Solvent A (1.0%, v/v, aqueous acetic acid) and solvent B (acetonitrile with 1.0%, v/v, acetic acid) comprised the mobile phases. The gradient flow was as follows: 0–50 min, 5–70% solution B; 50–55 min, 70–100% solution B; 55–60 min, 100% solution B; 60–65 min, 100–105% solution B, 65–80 min, 5% solution B. The column temperature was maintained at 40°C The analysis was carried out at a flow rate of 1.0 ml/min with PDA detection from 280 nm. The volume of the injection was 10 µl.

Standard stock solutions of four flavonoids [(+)-catechin, (-)-epicatechin, eriodictyol and quercetin)] and two phenolic compounds (caffeic acid and ethyl gallate) were dissolved in methanol at a concentration of 1.0 µg/ml and maintained at <4°C. Working standard solutions were prepared by serial dilution of stock solutions with methanol. The extracts (40 mg) of each lyophilized fraction, such as 70% ethanol, n-hexane, ethyl acetate, n-butanol and water fractions, were each dissolved in 70% ethanol (20 ml, respectively). The solutions were filtered through a 0.2-µm syringe filter (Woong Ki Science Co., Ltd, Seoul, Republic of Korea).

All calibration curves were obtained by the assessment of peak areas from standard solutions in the seven different concentration ranges, from 0.78 to 50.00 µg/ml. The LOD and LOQ data were determined based on signal-to-noise ratios of 3 and 10, respectively.

The HaCaT human keratinocyte and RAW 264.7 murine macrophage cell lines were obtained from the American Type Culture Collection (Rockville, MD, USA). These cells were cultured in Dulbecco's Modified Eagle's Medium (Gibco-BRL, Grand Island, NY, USA) supplemented with 10% (HaCaT) or 5.5% (RAW 264.7) heat-inactivated fetal bovine serum (Gibco-BRL), streptomycin (100 µg/ml) and penicillin (100 U/ml) at 37°C in a 5% CO₂ i nc ub at or.

Cell viability was assessed using the Cell Counting Kit-8 (CCK-8) reagent (Dojindo Laboratories, Kumamoto, Japan), according to the manufacturer's instructions. RAW 264.7 (3×10³ cells/well) and HaCaT (1×10³ cells/wel l) cells were incubated in 96-well plates with various concentrations of the test materials for 24 h. Following the addition of CCK-8 reagent to each well the cells were further incubated for 4 h. A Benchmark Plus microplate reader (Bio-Rad, Hercules, CA, USA) was used to measure absorbance at 450 nm, and the cell viability percentage was calculated using the following formula: Cell viability (%) = [(mean absorbance in test wells - mean absorbance in blank wells)/(mean absorbance in control wells - mean absorbance in blank wells)] x 100.

RAW 264.7 cells were plated in 48-well plates at a density of 2.5×10⁵ cells/well and incubated overnight. The cells were then treated with LPS (1 µg/ml; Sigma-Aldrich, St. Louis, MO, USA) in the presence or absence of various concentrations of the test materials. Following incubation for 24 h, ELISA kits were used to analyze the supernatants for the levels of PGE₂ (Cayman Chemical Co., Ann Arbor, MI, USA) and NO (Griess Reagent System; Promega Biotech Co., Ltd., Madison, WI, USA), according to the manufacturers' instructions. N(G)-Monomethyl-L-arginine (L-NMMA; Sigma-Aldrich) and indomethacin (Sigma-Aldrich) were used as positive controls to inhibit the production of NO and PGE₂, respectively.

HaCaT cells (1×10⁶ cells/well) were cultured in six-well plates in medium containing 10% fetal bovine serum. After reaching confluence, the cells were washed and treated with the test materials in 1 ml serum-free medium containing TI (10 ng/ml TNF-α and 10 ng/ml IFN-γ; R&D Systems Inc., Minneapolis, MN, USA) for 24 h. The supernatants of the cells were harvested, and the production of TARC, MDC and RANTES was quantified using ELISA (R&D Systems, Inc.), which was performed according to the manufacturer's instructions.

All experiments of the present study were performed at least in triplicate. Significant differences between the treatment groups were identified using one-way analysis of variance and multigroup comparisons were performed using the Dunnett's test. P<0.05 was considered to indicate a statistically significant difference.

The linearity of the peak area (y) versus concentration (x, µg/ml) curve for each component was used to calculate the contents of the six constituents in each fraction of the G. sinensis extract. The correlation coefficients (r²) of the calibration curves of the six constituents were ≥0.9998. The line equations and r² of the calibration curves are shown in Table I. The ranges of LOD and LOQ were 0.037–0.425 and 0.124–1.418 µg/ml, respectively (Table I).

Good separation chromatograms were obtained using mobile phases consisting of 1.0% (v/v) aqueous acetic acid (solvent A) and acetonitrile with 1.0% (v/v) acetic acid (solvent B). Quantification was achieved using PDA detection at 280 nm, based on retention times and ultraviolet spectra. Fig. 1 shows the representative HPLC chromatogram of the standards and each fraction of the G. sinensis extract, with detection of eluents at 280 nm. Reproducibility was assessed by measuring repeatedly the retention times and peak areas of six independently prepared samples of analytes. The reproducibility of the six compounds [(+)-catechin, caffeic acid, (-)-epicatechin, eriodictyol, ethyl gallate and quercetin] was less than the relative standard deviation (RSD) 1.5% for peak responses and less than the RSD 0.2% for retention times (data not shown). The contents of the six constituents are summarized in Table II.

The cytotoxic effect of the G. sinensis extract on RAW 264.7 and HaCaT cells was measured first. The cells were exposed to various concentrations (2–200 µg/ml) of the G. sinensis extract for 24 h. The non-toxic concentrations of the test materials were used in subsequent experiments (data not shown). To determine the effects of the G. sinensis extract on NO and PGE₂ production in LPS-stimulated RAW 264.7 cells, the cells were treated with different concentrations of the G. sinensis extract (25, 50 and 100 µg/ml) and then stimulated by LPS (1 µg/ml) for 24 h. The G. sinensis extract suppressed LPS-stimulated NO production in a dose-dependent manner (Fig. 2A). LPS greatly stimulated NO production (5.846±0.220 nM) in RAW 264.7 cells, whereas the G. sinensis extract significantly decreased NO production [1.872±0.513 µM (P<0.01) at a dose of 25 µg/ml, 0.974±0.128 µM (P<0.01) at a dose of 50 µg/ml and 0.205±0.020 µM (P<0.01) at a dose of 100 µg/ml] compared with that observed in LPS-stimulated RAW 264.7 cells. Furthermore, LPS-stimulated RAW 264.7 cells exhibited increased PGE₂ production compared with the controls, whereas the G. sinensis extract decreased the production of PGE₂ relative to the production observed in the LPS-stimulated RAW 264.7 cells (Fig. 2B). The effects of the G. sinensis extract on TARC production in TI-stimulated HaCaT cells were assessed by treating the cells with different concentrations of the G. sinensis extract (50, 100 and 200 µg/ml) and then initiating TI stimulation for 24 h. The G. sinensis extract suppressed TI-stimulated TARC production in a dose-dependent manner (Fig. 2C). TI-treated cells showed significantly increased production of TARC (28.27±0.900 ng/ml) relative to that observed in the control cells (P<0.01). This increase was inhibited to 22.44±0.030 ng/ml (P<0.01) and 10.23±0.480 ng/ml (P<0.01) by the administration of the G. sinensis extract at 100 and 200 µg/ml, respectively.

Fractions of the G. sinensis extract were tested to determine whether each fraction inhibited the production of NO and PGE₂ in LPS-treated RAW 264.7 cells and the production of TARC in HaCaT cells treated with TI. NO production is the hallmark of the activation of macrophages. In the present experiment, NO production (4308±0.225 µM) was observed after 24 h of LPS incubation (1 µg/ml). As shown in Fig. 3A, treatment with the n-hexane, ethyl acetate and n-butanol fractions suppressed the NO production. In addition, the n-hexane and ethyl acetate fractions decreased PGE₂ production compared with that observed in LPS-stimulated R AW 264.7 cells (P<0.01, Fig. 3B). TARC production was increased 2.7-fold in the TI-treated cells (28.30±0.820 ng/ml) compared with that in the control cells (10.16±1.540 ng/ml) (P<0.01), whereas cells treated with the ethyl acetate fraction showed the most significant reductions in TARC production compared with the TI-treated cells (P<0.01, Fig. 3C). These results showed that the biological effects of G. sinensis were maximized by the ethyl acetate fraction.

To assess the effects of the major constituents of the G. sinensis extract on LPS-stimulated NO and PGE₂ production in RAW 264.7 macrophages, cells were treated with various concentrations of these major constituents and 1 µg/ml LPS for 24 h. LPS stimulation caused a marked accumulation of proinflammatory mediators (NO and PGE₂) in the culture medium. Of note, all major constituents of the G. sinensis extract significantly reduced the LPS-stimulated production of NO (Fig. 4A) and PGE₂ (Fig. 4B) in a dose-dependent manner. Treatment of HaCaT cells with TI for 24 h led to a 1.9-fold increase in TARC levels (6.85±0.51 ng/ml) compared with that observed in the vehicle-treated control group (3.52±0.29 ng/ml) (P<0.01); however, TARC production was significantly inhibited in a dose-dependent manner in the ethyl gallate-, eriodictyol- and quercetin-treated cells (P<0.01) (Fig. 4C). In addition, MDC production was increased in the TI-treated cells (264.80±12.76 ng/ml) compared with that in the vehicle-treated control group (P<0.01), and its levels were significantly reduced following ethyl gallate and quercetin treatment (P<0.01) (Fig. 4D). TI-treated cells exhibited a significantly increased production of RANTES (4,652.93±66.95 ng/ml), relative to that observed in the control cells (P<0.01). This increase was inhibited dose-dependently by ethyl gallate (2,530.11±8.67 ng/ml at 50 µg/ml, P<0.01; 1,068.30±95.01 ng/ml at 100 µg/ml, P<0.01), eriodictyol (3,987.98±127.92 ng/ml at 50 µg/ml, P<0.05; 2,433.12±184.17 ng/ml at 100 µg/ml, P<0.01) and quercetin (2,728.46±119.93 ng/ml at 50 µg/ml, P<0.01; 41434±2759 ng/ml at 100 µg/ml, P<0.01) (Fig. 4E). Among these compounds, ethyl gallate and quercetin, which were most abundant both in the G. sinensis extract and in each solvent fraction (particularly in the ethyl acetate fraction) significantly inhibited the RANTES, MDC and TARC expression in HaCaT cells in a dose-dependent manner, whereas the same expression was not significantly reduced by other constituents in TI-treated cells.