Inhibition of Dermatitis Development by Sopungsan in Nc/Nga Mice

Sopungsan (SS) is a traditional Korean decoction used for the treatment of dermatitis. The aim of this study is to confirm whether or not SS has a preventive effect on the development of atopic dermatitis in dinitrochlorobenzene-applied Nc/Nga mice. SS was administered orally to Nc/Nga mice, which led to the remarkable suppression of the development of dermatitis, as determined by a histological examination and the serum IgE levels. Moreover, SS inhibited the production of thymus- and activation-regulated chemokine (TARC) and its mRNA expression in a keratinocyte cell line, HaCaT, which had been stimulated with tumor necrosis factor-α (TNF-α) and interferon-γ (IFN-γ). Activation of the nuclear factor-κB (NF-κB) or activator protein-1 (AP-1) is one of key steps in the signaling pathways mediating induction of TARC. In this study, SS selectively suppressed NF-κB activation which may be essential for TARC expression in TNF-α/IFN-γ treated keratinocytes. The inhibitory effect of SS on NF-κB activation and TARC production might be associated with the anti-dermatitic effects of SS.


INTRODUCTION
Atopic dermatitis is a common disease with an increasing incidence in industrialized countries. The clinical symptom of atopic dermatitis is characterized by pruritic skin lesions that are distinguished by infiltrating Iymphocytes, macrophage and granulated mast ceils (Hanfin and Rajka, 1980;Soter, 1989). Although topical steroids or anti-histamines are widely prescribed to manage this disease (leung, 1997), long-term use of these agents causes serious side effects. Hence, a great deal of effort has been directed toward identifying safer compounds or herbai remedies that can inhibit the development of atopic dermatitis. Some traditional Korean herbai prescriptions have been clinicaily applied as curative agents against atopic dermatitis. Sopungsan (SS) is one of herbal prescriptions used in Korea for the treatment of dermatitis. SS is a decoction consisting of 15 medicinal plants, Spirode/ae Herba, G/ycyrrhizae Radix, Sophorae Radix, Ange/icae Gigantis Radix, Cannabis Fructus, Akebiae Cau/is, Ledebouriellae Radix, Rehmanniae Radix, Gypsum Fibrosum, Cicadae Periostracum, Forsythiae Fructus, Arcth Fructus, Atracty/odis Rhizoma, Schizonepetae Herba and Sesami Semen Nigrum. In Dongeuibogam, a classical oriental medicine literature SS is introduced as an effective prescription for skin diseases. Moreover, it has been recently reported that SS has inhibitory effect in contact dermatitis animal model (Park et al., 2006).
Thymus-and activation-regulated chemokine (TARCI CCl17) is known as an endogenous ligand to CC chemokine receptor 4 (CCR4) and TARC binding to CCR4 results in attraction of Th2 ceils and regulatory T cells (Imai et a/., 1997). TARC produced from basal keratinocytes plays a significant role in recruiting Th2type Iymphocytes to dermatitic skin lesions and a subsequent deterioration of dermatitis (Vestergaard et a/., 2000;Furukawa et a/., 2004). In the present study, we first investigated wh ether SS administration protects the development of dermatitis in NdNga mice topicaily applied with 1-chloro-2,4-dinitrobenzene (DNCB). We also determined the effect of SS on TARC production in a keratinocyte ceil line, HaCaT to examine a possible mechanism for the antidermatitic effects of SS.

MATERIALS AND METHODS
Animals. The Institutional Animal Care and Utilization Committee of Chosun University approved all the animal procedures used in this study. Male Nc/Nga mice (25 g) were purchased from Joong-Ang Experimental Animals Co. (Seoul, Korea), placed in cages at a temperature between 20"C and 23°C with a 12 h light and dark cycle and a relative humidity of 50%. The animals were given commercial mouse chow (Purina, Korea) and water ad libitum. Controlled dermatitis was induced by topically applying 0.2% DNCB dissolved in acetone/olive oil (1 : 3) to the hair-removed back of the mice 3 times per week (Monday, Wednesday and Friday) for 10 weeks. The mice were then housed for 3 days without any further treatment. In case of the SStreated group, SS was orally administered at a dose of 10 or 30 mg/mouse (6 times per week for 12 weeks) in their drinking water. Control animals received the vehicle. The mice were sacrificed on the day of the experiment (on 73 days after first applying the DNCB) and blood was collected from the vena cava. The skin tissues from the backs of the mice were excised and subjected to a histological examination.
Other materials. The recombinant TNF-a and IFN-y were obtained from Peprotech Inc. (Rocky Hili, NJ). The human TARC and mouse IgE enzyme linked-immunosorbent assay kits were purchased from R&D Systems (Minneapolis, MN) and Shibayagi Co. (Gunma, Japan), respectively. 1-chloro-2,4-dinitrobenzene (DNCB) was obtained from Aldrich (Milwaukee, WI). Most of the reagents used for the molecular studies were purchased from Sigma (St. Louis, MO).
Cell eulture. The HaCaT cells, a human keratinocyte cell line were cultured in Dulbecco's modified Eagle's media (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 mg/mi penicillin and 100 mg/ ml streptomycin, in 5% CO2 at 37°C. The production of TARC was stimulated by incubating HaCaT cells with 20 ng/ml TNF-a and 20 ng/ml IFN-y. The SS extract was dissolved in sterile PBS and added to the incubation medium 10 min before adding TNF-a/IFN-y.

Enzyme linked immunosorbent assay (ELlSA).
Both the serum levels of IgE in the Nc/Nga mice and the supernatant levels of TARC in the HaCaT cells were measured according to the manufacturer's instructions. The absorbance was measured at a wavelength of 450 nm, and the concentration was calculated from a standard curve using serially diluted IgE or recombinant TARC.
Histopathologie examination. The skin tissues were isolated from each mouse and fixed in 10% formalin in 50 mM of a phosphate buffer (pH 7.0) for 24 h at 4°C. The skin tissues were subsequently embedded in paraffin, sectioned (4 flm), stained with hematoxylin and eosin, and examined by optical microscopy (Olympus, Tokyo, Japan). A certified pathologist analyzed and scored the sampies in a blinded manner. A minimum of 2 sections per experimental animal were examined for the presence and degree of the following: incrustation, thickening of the epidermis, epidermal necrosis, bleeding, hyperkeratosis, and inflammation of the epidermis and dermis.
Transient transfection and NF-KB and AP-1 reporter gene assay. The cells were plated on a 12weil dish and transfected the following day. The promoter activity was determined using a dual-Iuciferase reporter assay system (Promega, Madison, WI). Briefly, the cells were transiently transfected with 1 /-1g of pNF-KB-Luciferase or pAP-1-Luciferase plasmid and 20 ng of the pRL-SV plasmid (Renilla luciferase expression for normalization) (Promega, Madison, WI) using the Genejuice® reagent (Novagen, Madison, WI), and then exposed to TNF-a (20 ng/ml)/IFN-y (20 ng/ml) for 18 h. The firefly and Renilla luciferase activities in the cell Iysates were measured using a luminometer (Turner Biosystems; TD-20, Sunnyvale, CA). The relative luciferase activity was calculated by normalizing the NF-KB or AP-1 promoter-driven firefly luciferase activity to that of Renilla luciferase.
5tatistics. The differences between the treatment groups were examined using unpaired Student's t-test. The criterion for statistical significance was set at either p < 0.05 or p < 0.01.

RESULTS
55 protects progression of dermatitis in Nc/Nga mice topically applied with DNCB. Nc/Nga mice develop dermatitis with depilation and hemorrhage when housed in conventional conditions but not in specific pathogen-free conditions (Matsuda et a/., 1997). However, the NclNga mice bred in conventional conditions for 10 weeks showed only mild dermatitis and even higher variations in our experimental condition. Hence, 250 /-11 of 0.1 % DNCB was topically applied to the back of the mice for 10 weeks (3 times a week) in order to induce strong and controlled dermatitis. The DNCB applied Nc/Nga mice showed progressive diffuse erythematous changes, scaling, lichenified areas, oozing and crusting on the back (Fig. 1A) as weil as histological findings of hyperkeratosis, parakeratosis, acanthosis with varying degrees of spongiosis, exocytosis of mononuclear cells in the epidermis and the infiltration of inflammatory cells into the upper dermis (Fig.  1A), wh ich suggests that the chronic application of DNCB can accelerate the development of dermatitis in Nc/Nga mice.
Next, we examined whether SS administration prevents DNCB-induced dermatitis in Nc/Nga mice. SS (10 B c D 500 400 E g.  (Matsuda et al., 1997;Kotani et al., 2000). We found that chronic application of DNCB for 10 weeks significantly enhanced the serum IgE levels compa red with the vehicle-treated group and the increase in serum IgE levels were reversed in the 88-treated groups (Fig. 10). In the mice treated with 10 and 30 mg 88, the inhibition percentage of serum IgE was 71 and 60%, respectively (Fig. 10). These results demonstrated that the oral administration of 88 prevents the development of dermatitis in Nc/Nga mice.

SS inhibits TARC production in HaCaT cells treated
with TNF-a/IFN-y. TARC is a chemokine secreted from dermal dendritic ceils and is involved in the development of Th2-mediated inflammation such as atopic dermatitis (Imai et al., 1997;8andoval-Lopez and Teran, 2001). Vestergaard et al. (2000) also showed that TARC actively participate in the pathogenesis of atopic dermatitis in NclNga mice. In cell culture studies using primary human keratinocytes or HaCaT ceils, TNF-a and IFN-y synergisticaily induce TARC production, which might playa role in the development of atopic dermatitis (Vestergaard et al., 2000). Because 88 showed suppressive effect on the development of der-matitis in the DNCB-applied NclNga mice, we hypothesized that 88 could affect the production of TARC in HaCaT ceils by TNF-a./IFN-y. In ELl8A assay using a human TARC specific antibody, we found that the level of TARC production was The concentration of TARC in the medium was determined using TARC-specific ELiSA assays. The results shown represent the mean ± SEM of 4 different sampies (significant compared with control, **p < 0.01; significant compared with TNF-a/IFN-y-treated group, ##p< 0.01).
(B) The TARC mRNA expression levels were determined by RT-PCR analysis. The mRNA expression of the S16 ribosomal protein was comparable among the sam pies.
increased in cells treated with TNF-a (20 ng/ml)/IFN-y (20 ng/ml) for 24 hand the increase in TARC production was significantly inhibited by 1000 fl9/ml SS pretreatment ( Fig. 2A). To further determine whether SS inhibits the transcription of the TARC gene, the expres- NF-KB reporter gene analysis. The induction of luciferase activity by TNF-a/IFN-y in HaCaT ceils transiently transfected with the pNF-KB-Luc construct, which contained the three-times repeated NF-KB binding sequences, was confirmed using a luminometer. A dual luciferase reporter gene assay was performed on the Iysed ceils that had been cotransfected with the pNF-KB-Luc (firefly luciferase) and pRL-SV (Renilla luciferase) (a ratio of 100: 1) after exposing them to TNF-a/IFN-y (20 ng/ml, each) and SS (300 or 1000 Ilg/ml) for 18 h. The activation of the reporter gene was calculated as the relative change in the Renilla luciferase activity. The data represent the mean ± SO of 4 separate experiments (significant compared with control, **p < 0.Q1; significant compared with the TNF-a/IFN-y-treated group, ##p < 0.01).
(B) AP-1 reporter gene analysis. The ceils were transfected with the pAP-1-Luc plasmid, and reporter gene analysis was performed as reported in panel (A). The data are represented as a mean ± SEM of 4 separate experiments.
sion levels of TARC mRNA in HaCaT ceils exposed to TNF-a (20 ng/ml)/IFN-y (20 ng/ml) for 3 h were measured. The expression of TARC mRNA was increased markedly by TNF-a/IFN-y, and this increase was diminished in HaCaT ceils pretreated with 1000 fl9/ml SS (Fig. 28). In contrast, mRNA levels of S16 ribosomal protein were similar in ail sampies (Fig. 2B). Hence, SS may target transcription process of TARC gene .

inhibits NF-KB activation in HaCaT cells stimulated with TNF-a/IFN-y.
The promoter region of the TARC gene contains putative NF-KB and AP-1 binding sites (Nakayama et a/., 2004) and recent studies also showed that the transcriptional regulation of TARC gene is also controiled by the NF-KB or AP-1 activity (Berin et a/., 2001;Komine et a/., 2005;Nakayama et a/., 2004). Thus, we performed reporter gene analyses using luciferase reporter plasmids containing the NF-KB or AP-1 binding sequences to determine if the suppressive effect of SS on the induction of TARC gene occurs alongside the inhibition of NF-KB or AP-1. TNF-a/IFN-y (20 ng/ml each, 18 h) caused a 3.2-fold increase in NF-KB reporter activity (Fig. 3A), which was inhibited by pretreating the ceils with 1000 fl9/ml of SS for 10 min (Fig. 3A). However, TNF-a/IFN-y did not increase the AP-1 reporter activity in HaCaT cells. Moreover, the basal AP-1 reporter activity was not affected by 300 or 1000 fl9/ml SS (Fig. 3B). These results show that SS selectively inhibits the NF-KB activation process, and is associated with its suppressive effect on TARC induction caused by inflammatory insults.

DISCUSSION
Here, we found that topical application of DNCB on the back of NclNga mice was an efficient tool to induce severe and controiled dermatitis within 10 weeks. In contrast, DNCB itself did not cause any pathological changes in ICR mice (data not shown), which suggests that the dermatitis observed in our system is mainly the result of the DNCB-induced potentiation of spontaneous dermatitis in Nc/Nga mice. Chronic administration of SS effectively prevented the development of dermatitis, as evidenced by histological examination and serum IgE determination. SS is often prescribed for patients with atopic dermatitis in Korean oriental medicine hospitals, and our results provide a scientific evidence to support its clinical efficacy.
TARC functions as a selective chemoattractant and assists in the recruitment and migration of Th2 cells expressing CCR4 (Sandoval-Lopez and Teran, 2001). In fact, TARC antibody therapy reduces the develop-ment of allergie airway inflammation and hyper-responsiveness (Kawasaki et al., 2001). Hence, TARC has been suggested to be an important mediator that exaggerates atopic dermatitis. The clinical and pharmacological efficacy of SS against dermatitis might partly be associated with its inhibitory actions on TARC production in keratinocytes because both the release of TARC and its mRNA expression were diminished in keratinocyte cell line exposed to SS. The promoter of the human TARC gene contains several homologous consensus sequences for the binding of transcription factors including STAT, NF-KB and AP-1 (Nakayama et al., 2004). Among these transcription factors, it was reported that either NF-KB or AP-1 is important for the transcription of the human TARC gene (Komine et al., 2005;Nakayama et al., 2004). In the present study, we showed that SS selectively inhibited TNF-a./IFN-y-inducible NF-KB reporter activity. Therefore, the inhibition of NF-KB by SS may be related with the suppression of TARC induction in TNF-a/lFN-y-treated keratinocytes.
In summary, the chronic administration of SS, at doses of 10 or 30 mg/mice/day prevents the developme nt of dermatitis in NclNga mice. These beneficial effects may at least in part be associated with the inhibitory effect of SS on NF-KB-mediated TARC production in keratinocyte in the dermatitic lesions.