AP was purchased from a herb shop at Gyeongdong medicinal herb market (Seoul, Korea). High-performance liquid chromatography (HPLC)-grade methanol, and ethanol, ethyl acetate and dichloromethane were purchased from Duksan Pure Chemicals Co., Ltd. (Ansan, Korea). Dimethyl sulfoxide-d₆ (DMSO-d₆), a common solvent used in nuclear magnetic resonance (NMR) spectroscopy, was purchased from Sigma-Aldrich; Merck Millipore (Darmstadt, Germany).

Dried AP (5 kg) was extracted with ethanol (95%) for 3 or 4 days at room temperature. Following filtration through a 400-mesh filter, the product was passed through filter paper (Whatman^® Grade 5) and concentrated under reduced pressure by rotary evaporation (EYELA N-1000; Tokyo Rikakikai Co., Ltd., Tokyo, Japan). The ethanol extract of AP (279.4 g) was suspended in H₂O and extracted with ethyl acetate to obtain an ethyl acetate soluble layer (97.8 g). The ethyl acetate soluble layer (97.8 g) was mixed to a silica column (97.8 g) and concentrated, followed by oven drying to obtain a coating loading powder. Then 20 g of powder was loaded into a silica open column (4.5×40 cm) filled with 100 g silica. The solvent was developed by gradient elution with methanol in dichloromethane (80:1, 50:1, 30:1, 20:1, 15:1, 5:1, 1:1). Sub-fractions (n=6; APEA-1 to APEA-6) were collected, and the concentrate (22.5 g) of APEA-3 containing the active ingredient was dissolved in 20 ml of 30% (v/v) methanol and filtered through a 0.45 µm filter. Then, in order to purify the active ingredient, a Shim-Pack Prep-ODS column (20×250 mm; 5 µm), packed with silica gel at a flow rate of 10 ml/min and an injection volume of 50 µl was dissolved in a methanol-H₂O solution (50:50). Following purification using a Prep LC column (Shimadzu Co., Kyoto Japan) and vacuum drying, pure ISTP was obtained (111 mg). This active component was identified by ¹H-NMR and ¹³C-NMR (16,21). Nuclear magnetic resonance (NMR) spectroscopy was used to analyze the structure of the compounds separated by H1-NMR for Varian (GEMINI, 400 MHz) and C13-NMR for Varian (GEMINI, 100 MHz). DMSO (Sigma Aldrich) was used as a solvent and the NMR results were consistent with the previous studies (data not shown). ISTP: syrup, ¹H-NMR (DMSO-d6, 400 MHz): δ1.70 (1H, m, H-8), 1.85 (1H, m, H-8′), 2.05 (3H, s, H-14), 2.08 (1H, d, H-2), 2.11 (3H, s, H-15), 2.45 (2H, m, H-9), 2.66 (1H, dd, J=18.2, 6.2 Hz, H-2′), 3.05 (1H, m, H-7), 4.55 (1H, s, H-6), 5.02 (1H, d, J=5.6 Hz, H-3), 5.61 (1H, brs, -OH), 5.73 (1H, d, J=2.5 Hz, H-13), 6.11 (1H, d, J=2.9 Hz, H-13′), ¹³C-NMR (DMSO-d6, 100 MHz): δ13.4 (C-15), 26.8 (C-8), 29.7(C-14), 39.1(C-9), 41.3(C-7), 44.3(C-2), 69.7(C-3), 75.4(C-6), 121.6(C-13), 135.7(C-5), 139.1(C-11), 169.6(C-12), 175.0(C-4), 203.4(C-1), 207.6(C-10).

A modular Shimadzu LC-20A System was utilized. A Capcell Pak C-18 Column (250×4.6 mm internal diameter×5 µm; Shiseido Co., Ltd., Tokyo, Japan) was used at 30°C. Isocratic elution [mobile phase, solvent mixture of methanol (15%)] was performed for 1 h at a flow rate of 1 ml min⁻¹ with an injection volume of 20 µl. The UV detector was set at a wavelength of 220 nm.

HaCaT cells (immortalized human keratinocyte cell line, obtained from American Type Culture Collection, Manassas, VA, USA) were cultured in Dulbecco's modified Eagle's medium (DMEM; Welgene Biotech, Taipei, Taiwan) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Inc., Waltham, MA, USA) and 100 U/ml penicillin/streptomycin at 37°C and 5% CO₂. Cells (5×10⁵ cells/dish in 60 mm dishes) were pretreated with 10 ng/ml TNF-α and 10 ng/ml IFN-γ at 37°C and 5% CO₂ for 30 min and subsequently incubated with 2.5, 5 or 10 µg/ml ISTP or 125, 250 or 500 µg/ml AP extract (APE) for 30 min at 37°C and 5% CO₂. For induction of IL-33, HaCaT cells were treated with 20 ng/ml TNF-α and 20 ng/ml IFN-γ at 37°C and 5% CO₂ for 30 min (22).

HaCaT cells (2.5×10⁴ cells/well) were seeded into 96-well plates and their proliferation was measured using a CCK-8 assay (Dojindo Molecular Technologies, Inc., Rockville, MD, USA). Cells were treated with 62.5, 125, 250 and 500 µg/ml APE or 1.25, 2.5, 5 and 10 µg/ml ISTP for 24 h. Cells were subsequently incubated with TNF-α (20 ng/ml) and IFN-γ (20 ng/ml) for 30 min at 37°C and 5% CO2. CCK-8 solution (10 µl) was added to the cells in 1 ml DMEM and incubated for 2 h at 37°C. Absorbance was measured at a wavelength of 450 nm using a microplate reader (SpectraMax 340; Molecular Devices, LLC, Sunnyvale, CA, USA).

Western blot analysis was performed as described previously (23). Proteins were quantified using a Bio-Rad DC Protein assay kit II (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Equal amounts of protein were separated by SDS-PAGE, transferred to polyvinylidene difluoride membranes and incubated with the following primary antibodies: Rabbit anti-phosphorylated (p)-STAT-1 phosphorylated at tyrosine 701 (pY-STAT-1; 1:1,000; cat. no. 9167; Cell Signaling Technology, Inc., Danvers, MA, USA); rabbit anti-p-STAT-1 phosphorylated at serine 727 (pS-STAT-1; 1:1,000; cat. no. 9177; Cell Signaling Technology, Inc.); rabbit anti-total STAT-1 (1:1,000; cat. no. 9172; Cell Signaling Technology, Inc.); mouse anti-ICAM-1 (1:1,000; cat. no. ab2213; Abcam, Cambridge, MA, USA); rabbit anti-β-actin (1:1,000; cat. no. 4967; Cell Signaling Technology); and anti-IL-33 (1:1,000; cat. no. sc-98659; Santa Cruz Biotechnology, Inc., Dallas, TX, USA).

The concentration of six cytokines and chemokines [IL-1β, IL-6, monocyte chemoattractant protein-1 (MCP-1)/CCL2, TARC, soluble ICAM-1 (sICAM-1) and IL-33] were measured in cell supernatants using human ELISA kits as follows: Human IL-1β (ELISA ready-SET-GO; cat. no. 88-7261, eBioscience, Inc., San Diego, CA, USA), human CCL2 (MCP-1) (ELISA ready-SET-GO; cat. no. 88-7399, eBioscience, Inc.), human sICAM-1 (Platinum ELISA; cat. no. BMS201CE; eBioscience, Inc.), human IL-33 (Platinum ELISA; cat. no. BMS2048, eBioscience, Inc.), and human CCL17/TARC (Quantikine ELISA kit; cat. no. DDN00, R&D Systems, Inc., Minneapolis, MN, USA). ELISA was performed according to the manufacturer's protocol.

RNA was isolated from HaCaT cells using an RNeasy Plus Mini kit (Qiagen, Inc., Valencia, CA, USA) according to the manufacturer's protocol. Reverse transcription was performed using the RevertAid First Strand cDNA synthesis kit (cat. no. #K1622; Thermo Fisher Scientific, Inc.). Quantitative polymerase chain reaction (qPCR) was performed using the EmeraldAmp GT PCR Master Mix (cat. no. RR310A; Takara Biotechnology Co., Ltd., Dalian, China). The primers for the qPCR were as follows: Human GAPDH, 5′ AGG GCT GCT TTT AAC TCT GGT 3′ (sense) and 5′ CCC CAC TTG ATT TTG GAG GGA 3′ (antisense); ICAM-1, 5′ CAC CCT AGA GCC AAG GTG AC 3′ (sense) and 5′ CAT TGG AGT CTG CTG GGA AT 3′ (antisense); TARC, 5′ CTT CTC TGC AGC ACA TCC 3′ (sense) and 5′ AAG ACC TCT CAA GGC TTTG 3′ (antisense); IL-33, 5′ AGC CTT GTG TTT CAA GCT GG 3′ (sense) and 5′ ATG GAG CTC CAC AGA GTG TTC 3′ (antisense). The thermocycling conditions were as follows: An initial denaturation step at 94°C for 2–10 min, followed by 30–35 cycles of denaturation at 94°C for 30 sec to 3 min, annealing at 50–58°C for 30 sec to 1 min and extension at 72°C for 30 sec to 1 min, and a final extension step at 72°C for 4–7 min. The digitized gel images were analyzed using Quantity One 1-D Analysis software version 4.6.5 (Bio-Rad Laboratories, Inc.).

HaCaT cells (1.5×10⁴ cells/well) were seeded onto a 4-well chamber slide and treated with TNF-α (20 ng/ml), IFN-γ (20 ng/ml), ISTP (10 µg/ml) or APE (500 µg/ml) for 24 h at 37°C incubation. Following fixation in 100% methanol (chilled at −20°C) at room temperature for 5 min and permeabilization, cells were incubated with 1% BSA/ 22.52 mg/ml glycine in PBS/0.1% Tween-20 (PBST) for 30 min to block unspecific binding of the antibodies. Then, cells were incubated overnight at 4°C with an anti-IL-33 primary antibody (1:100; cat. no. sc-98659, Santa Cruz Biotechnology, Inc.), followed by a fluorescein isothiocyanate-labeled goat anti-rabbit IgG (1:1,000; cat. no. NB730-F; Novus Biologicals, LLC, Littleton, CO, USA) in 1% BSA/ PBST for 1 h at room temperature in the dark. DAPI was used to counterstain the nuclei. Stained cells were visualized using a confocal microscope (Olympus FluoView FV10i; Olympus Corporation, Tokyo, Japan).

Data are presented as the mean ± standard deviation (n=3). Statistical significance was calculated by one-way analysis of variance followed by Duncan's multiple range test using PASW Statistics software version 18.0 (SPSS Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

The structure of ISTP is presented in Fig. 1A. ISTP, isolated from Artermisia rutifolia and Artermisia iwayomogi, has a sesquiterpene lactone structure and has been demonstrated to inhibit nitric oxide synthase. The present study investigated whether APE contained other primary compounds, including ISTP, eupatilin (24) and jaceosidin (25). Compounds were extracted from APE by HPLC and various other peaks were detected (Fig. 1B). To exclude the possibility that the cytotoxicity of ISTP may contribute to its suppressive effects, cell viability was determined using a CCK-8 assay. HaCaT cells were stimulated with TNF-α and IFN-γ in the absence or presence of ISTP or APE. As presented in Fig. 2, ISTP and APE exhibited no significant cytotoxic effect on HaCaT cells at the concentrations assessed.

The mechanism of action of ISTP inhibition of chemokine and cytokine release from TNF-α/IFN-γ-stimulated HaCaT cells was investigated. Previous studies reported that TNF-α/IFN-γ stimulation activates signaling molecules, including STAT-1, extracellular signal-regulated kinase, c-Jun N-terminal kinase, p38 mitogen-activated protein kinases (MAPKs) and nuclear factor (NF)-κB in HaCaT cells (26–30). Thus, the present study evaluated whether ISTP affects the STAT signaling pathway in TNF-α/IFN-γ-stimulated HaCaT cells using western blot analysis. HaCaT cells were pretreated with APE or ISTP, followed by incubation with TNF-α and IFN-γ. Treatment with ISTP or APE reduced ICAM-1 protein expression levels and decreased STAT-1 phosphorylation in a dose-dependent manner (Fig. 3A). In addition, the ability of ISTP to inhibit ICAM-1 and TARC mRNA expression levels was investigated. RT-PCR demonstrated that the expression levels of TARC and ICAM-1 mRNA were increased by TNF-α/IFN-γ (Fig. 3B and D); this increase was abrogated by ISTP or APE treatment (Fig. 3B and D). ELISA was performed to determine the inhibitory effects of ISTP and APE on TARC production. ISTP and APE significantly inhibited TNF-α/IFN-γ-induced TARC production in HaCaT cells compared with TNF-α/IFN-γ treatment only (P<0.05; Fig. 3C), in a dose-dependent manner. The results indicated that ISTP may inhibit TNF-α/IFN-γ-induced TARC expression by suppression of ICAM-1 and STAT-1 activation.

Subsequently, the present study investigated whether ISTP inhibits inflammatory cytokine and chemokine production in TNF-α/IFN-γ-stimulated HaCaT cells. The STAT family serves an important role in cytokine production. The production of the majority of cytokines and chemokines are primarily regulated at the transcriptional level through activation of specific sets of transcription factors, which are controlled by NF-κB and MAPKs. The results demonstrated that pretreatment with ISTP or APE inhibited the production of IL-1β (Fig. 4A), IL-6 (Fig. 4B), MCP-1/CCL-2 (Fig. 4C) and sICAM-1 (Fig. 4D) by HaCaT cells, in a dose-dependent manner. The results indicated that ISTP inhibits the release of pro-inflammatory cytokines.

It has previously been reported that IL-33 is upregulated when keratinocytes are exposed to pro-inflammatory stimuli such as TNF-α/IFN-γ, and may therefore be important in the pathogenesis of chronic inflammatory skin disorders, including AD and psoriasis (31). Therefore, the present study evaluated the effect of ISTP on IL-33 expression in TNF-α/IFN-γ-treated HaCaT keratinocytes and detected a high level of IL-33 at 20 ng/ml TNF-α/IFN-γ. As presented in Fig. 5A, protein and mRNA expression levels of IL-33 were increased following TNF-α/IFN-γ stimulation, compared with cells that were not treated with TNF-α/IFN-γ. Conversely, ISTP or APE treatment appeared to reduce IL-33 expression (Fig. 5A). In addition, pretreatment with ISTP or APE inhibited IL-33 production dose-dependently in supernatants from cultured HaCaT cells (P<0.001; Fig. 5B). Immunocytochemistry indicated that stimulation of HaCaT keratinocytes with TNF-α/IFN-γ led to an increase in IL-33 compared with untreated cells. However, IL-33 staining reduced following pretreatment of HaCaT cells with 10 µg/ml ISTP or 500 µg/ml APE (Fig. 5C). The results indicated that ISTP inhibits IL-33 production and may be important in the crosstalk between pro-inflammatory cytokines.