Methanol, acetonitrile and water (HPLC grade) were purchased from J.T. Baker Inc. (Phillipsburg, NJ, USA). Albiflorin, paeoniflorin, cinnamaldehyde, glycyrrhizin, 6-gingerol, and schizandrin were obtained from Wako Pure Chemical Industries Ltd. (Osaka, Japan). Gallic acid, benzoic acid, cinnamic acid and methyl eugenol were purchased from Sigma-Aldrich (St. Louis, MO, USA). Liquiritin and isoliquiritin were from NPC BioTech (Geumsan, Korea) and Chengdu Biopurity Phytochemicals Ltd. (Chengdu, China), respectively. All compounds had a purity of >98%. The chemical structures of the standard compounds are shown in Fig. 1. Herbal Medicines were purchased from a Herbal Medicine company, Kwangmyungdang Medicinal Herbs (Ulsan, Korea), and identified by Professor Je-Hyun Lee (Department of Herbology, Dongguk University, Korea). A voucher specimen (2008-KE13-1-8) has been deposited in the Herbal Medicine Formulation Research Group of the Korea Institute of Oriental Medicine.

A combination of chopped herbal medicines comprising SCRT (total weight, 3.5 kg)(Table I) was extracted with 35 l distilled water at 100°C for 2 h using an electric extractor (COSMOS-660; Kyungseo Machine Co., Incheon, Korea). The extracted decoction was filtered using a standard sieve (no. 270; 53 μm; Chunggyesangongsa, Seoul, Korea) and lyophilized to obtain a powder (760.6 g; yield, 21.7%), which was kept at 4°C until further use for HPLC analysis and in vitro assay.

The standard compounds were weighed accurately and dissolved in methanol to make stock solutions at concentrations of 1,000 μg/ml. The stock solution containing a standard compound was diluted to produce working solutions, which were used to construct a calibration curve. Accurately weighed SCRT powder (400 mg) was dissolved in 20 ml distilled water, the solution was then filtered through a 0.2-μm syringe filter (SmartPor^®; Woongki Science, Seoul, Korea) before it was injected into the HPLC apparatus (Shimadzu LC-20A; Shimadzu, Kyoto, Japan).

The HPLC system was equipped with a solvent delivery unit (LC-20AT), an autosampler (SIL-20AC), column oven (CTO-20A), degasser (DGU-20A3) and photodiode array detector (SPD-M20A). Separation was performed on a Gemini C18 column (4.6×250 mm, 5 μm particle size of stationary phase; Phenomenex, Torrance, CA, USA) maintained at 40°C. The mobile phase consisted of water and acetonitrile both containing 1% acetic acid. Gradient elution of the mobile phase was applied: 5–45% at 0–30 min, 45–95% at 30–45 min and held for 5 min. The flow rate was 1.0 ml/min and the injection volume was set to 10 μl. The detection wavelengths were optimized according to the maximum absorption wavelengths of standards (230, 254, 275, 290 and 360 nm).

The linear regression and correlation coefficients (r2) of the compounds were calculated on the basis of previously determined calibration curves. The values of limit of detection (LOD) and limit of quantification (LOQ) for each compound were measured using signal concentrations higher than noise concentration at three and ten fold, respectively. The intra-day and inter-day precisions were calculated by analyzing sample extracts spiked with two different concentrations of standard compounds (low and high) and their values were estimated as the relative standard deviation (RSD). Repeatability was evaluated using three replicates of sample solutions that were also represented as the RSD value. To measure the accuracy of the method, a recovery test was performed by adding two different concentration levels of reference compounds (low and high) to the samples, which were then extracted with the above methods. The recovery was calculated as follows: Recovery(%) = ((detected concentration − initial concentration)/spiked concentration) × 100.

A human bronchial epithelial cell line, BEAS-2B, was obtained from the American Type Culture Collection (Rockville, MD, USA). The cells were cultured in Dulbecco’s modified Eagle’s medium (Gibco-BRL, Grand Island, NY, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Gibco-BRL), penicillin (100 U/ml; Gibco-BRL) and streptomycin (100 μg/ml; Gibco-BRL) at 37°C in an atmosphere of 5% CO₂/95% air under saturating humidity.

Cell viability was assessed using the Cell Counting Kit-8 assay (CCK-8; Dojindo, Kumamoto, Japan) according to the manufacturer’s instructions. BEAS-2B cells (6×10³ cells/well) were incubated in 96-well plates with various concentrations of SCRT water extract and the 13 constituent compounds for 24 h. CCK-8 reagent was added to each well and incubated for 4 h. The absorbance was measured at 450 nm with a Benchmark Plus microplate reader (Bio-Rad Laboratories, Hercules, CA, USA). The percentage of cell viability was calculated using the following formula: Cell viability(%)=(mean absorbance in test wells/mean absorbance in vehicle-treated control wells)×100.

BEAS-2B cells (5×10⁵ cells/well) were cultured in six-well plates in medium containing 10% FBS. Once cells had reached confluence, they were washed and incubated with 1 ml serum-free medium containing 10 ng/ml tumor necrosis factor (TNF)-α to produce RANTES for 24 h. Other cells were incubated for 48 h with 50 ng/ml interleukin (IL)-4 + TNF-α (R&D Systems Inc., Minneapolis, MN, USA) to produce eotaxin and eotaxin-3.

Culture supernatants were used to measure RANTES, eotaxin and eotaxin-3 using an ELISA (R&D Systems) according to the manufacturer’s instructions (Catalog nos. DY278, DY320 and DY346). The absorbance was measured at 450 nm using a Benchmark plus microplate reader (Bio-Rad Laboratories).

MMP-9 activity was measured using gelatin zymography. Cell supernatant was mixed with 5X non-reducing sample buffer (Fermentas Inc., Pittsburg, PA, USA) proir to loading onto 10% SDS-PAGE (Bio-Rad Laboratories) containing 1% gelatin as the MMP substrate. The sample was subjected to electrophoresis at 80 volts for 2 h. Following electrophoresis, gels were washed twice in 2.5% Triton X-100 (Sigma-Aldrich) for 1 h to remove SDS and then incubated for 16 h at 37°C in developing buffer (1 M tris-HCl, pH 7.5; 10 mM CaCl₂). Following incubation, gels were stained with Coomassie Brilliant Blue G (Sigma-Aldrich, St. Louis, MO, USA) for 35 min, then de-stained in 25% methanol and 8% acetic acid solution for 20 min and then rinsed twice with de-staining solution to visualize the digested bands in the gelatin matrix. Gelatinase activity was manifested as white bands on a blue background, representing areas of proteolysis of the substrate protein. Images of the gels were captured and the averages of the band intensity were measured using the commercially available ChemiDocTM XRS+ imaging system (Bio-Rad Laboratories).

Data are presented as the mean ± standard error of the mean. One-way analysis of variance was used to identify significant differences between the treated and stimulated cells. Dunnett’s post-hoc test was used for multi-group comparisons. P<0.05 was considered to indicate a statistically significant difference between values. PCA and HCA were performed using a data matrix consisting of rows for the samples (the three concentrations of SCRT water extract and 13 compounds) and columns for the inhibited values of chemokines (RANTES, eotaxin, and eotaxin-3) using open source R software (version 2.15.1; R-project.org).

The linear regression and correlation coefficient r2 for each compound ranged from 0.9993 to 0.9999, demonstrating high linearity. The LODs and LOQs were calculated at the concentrations of each compound that produced signal-to-noise ratios of three and ten, respectively; their values were LOD = 0.01–0.75 μg/ml and LOQ = 0.03–2.49 μg/ml (Table II). All compounds were detected in the sample extracts and their chromatograms showed clearly separated maximum absorption wavelengths (Fig. 2). The RSD values for the intraday and interday precision were 0.01–3.21% and 0.20–4.23%, respectively. Repeatability ranged from 0.08–1.99% (Table III). Recovery was used to test the accuracy of the experimental methods. The recovery of each standard compound was 88.30–111.13%, with an RSD of <4.0% (Table IV).

The 13 constituent compounds were quantified in the SCRT water extract. Paeoniflorin was the most abundant compound in the SCRT water extract, followed by liquiritin, albiflorin and glycyrrhizin, while methyl eugenol was observed at the lowest quantity of ~100 times lower than that of paeoniflorin (Table V).

Chemokine levels, including RANTES, eotaxin and eotaxin-3 in BEAS-2B cells were determined using ELISA following treatment with SCRT water extract. Nontoxic concentrations of samples were used for the subsequent experiments by the determination of the cytotoxicity on BEAS-2B cells (data not shown). The production of RANTES was significantly increased following treatment with TNF-α; however, following treatment with SCRT water extract, production was reduced in a dose-dependent manner (Fig. 3A). Cells treated with IL-4 + TNF-α exhibited significantly increased secretion of eotaxin and eotaxin-3. However, SCRT water extract markedly decreased the production of eotaxin and significantly attenuated the production of eotaxin-3 in a dose-dependent manner (Fig. 3B and C).

The production of chemokines was measured in BEAS-2B cells treated with the constituent compounds of SCRT in order to identify the specific compounds which may contribute to its inhibitory effect. With the exception of isoliquiritin, all compounds were found to significantly suppress the increased productions of RANTES in TNF-α-treated BEAS-2B cells in a dose-dependent manner (Fig. 4A). In addition, eotaxin-3 production was significantly and dose-dependently inhibited by the treatment with eight of the compounds, including gallic acid, paeoniflorin, liquiritin, benzoic acid, couomarin, cinnamic acid, 6-gingerol and methyl eugenol, whereas only three compounds, gallic acid, 6-gingerol and methyl eugenol, reduced the release of eotaxin in BEAS-2B cells pretreated with IL4 and TNF-α (Fig. 4B and C).

To evaluate the influence of constituent compounds on the inhibition of chemokines, multivariate statistical analysis, including PCA and HCA, were performed.

PCA was represented in a plot consisting of two principal component (PC) scores. PC1 contributed 71% of the total variance, and therefore is a highly significant variable for the distribution of samples. Three concentrations (low, medium and high) of SCRT water extract, 6-gingerol and methyl eugenol, two levels (medium and high) of gallic acid, and high levels of albiflorin, liquiritin, isoliquiritin, benzoic acid, cannamic acid, cinnamaldehyde and schizandrin, were distributed at the negative PC1 score. Samples were clearly classified by PC1 score. PC2, the next most significant variable of distribution of samples orthogonal to PC1, contributed 18% of the total variance. The samples with negative PC1 scores consisted of three concentrations of SCRT water extract, two concentrations of 6-gingerol (medium and high) and methyl eugenol (low and medium), and high concentrations of coumain were assigned positive PC2 score. All arrows of chemokines (RANTES, eotaxin and eotaxin-3) were directed at negative PC1 score, with RANTES and eotaxin assigned a negative PC2 score, whereas eotaxin-3 was assigned a positive PC2 score (Fig. 5).

HCA represented the distribution of the concentrations of SCRT water extract and its 13 constituent compounds as a dendrogram. The results exhibited a similar pattern of classification to that of the PCA. Two main groups, A and B, were classified at a height of ~400. Three concentrations of SCRT water extract, two concentrations (medium and high) of gallic acid, 6-gingerol, and methyl eugenol, and a high concentration of isoliquiritin were involved in group A, which was clearly distinct from group B (Fig. 6).

Gelatin zymography was used to investigate the effect of SCRT water extract, gallic acid, 6-gingerol and methyl eugenol on MMP-9 activity in BEAS-2B cells induced with IL4 and TNF-α. The purpose of this assay was to elucidate the inhibitory mechanism of SCRT against airway-associated inflammation. As shown in Fig. 7, MMP-9 activity was markedly increased following treatment of BEAS-2A cells with IL4 and TNF-α; by contrast, MMP-9 activity was significantly reduced following treatment with SCRT water extract in a dose-dependent manner. It was also observed that gallic acid, one of the 3 contributing compounds, dose-dependently inhibited the increased MMP-9 activity (Fig. 7A and C). In addition, high levels of 6-gingerol significantly inhibited MMP-9 activity and methyl eugenol demonstrated a non-significant reduction of increased MMP-9 activity (Fig. 7B and D).