Impressic Acid Ameliorates Atopic Dermatitis-Like Skin Lesions by Inhibiting ERK1/2-Mediated Phosphorylation of NF-κB and STAT1

Impressic acid (IPA), a lupane-type triterpenoid from Acanthopanax koreanum, has many pharmacological activities, including the attenuation of vascular endothelium dysfunction, cartilage destruction, and inflammatory diseases, but its influence on atopic dermatitis (AD)-like skin lesions is unknown. Therefore, we investigated the suppressive effect of IPA on 2,4-dinitrochlorobenzene (DNCB)-induced AD-like skin symptoms in mice and the underlying mechanisms in cells. IPA attenuated the DNCB-induced increase in the serum concentrations of IgE and thymic stromal lymphopoietin (TSLP), and in the mRNA levels of thymus and activation regulated chemokine (TARC), macrophage derived chemokine (MDC), interleukin-4 (IL-4), interleukin-5 (IL-5), interleukin-13 (IL-13), tumor necrosis factor-alpha (TNF-α) and interferon-gamma (IFN-γ) in mice. Histopathological analysis showed that IPA reduced the epidermal/dermal thickness and inflammatory and mast cell infiltration of ear tissue. In addition, IPA attenuated the phosphorylation of NF-κB and IκBα, and the degradation of IκBα in ear lesions. Furthermore, IPA treatment suppressed TNF-α/IFN-γ-induced TARC expression by inhibiting the NF-κB activation in cells. Phosphorylation of extracellular signal-regulated protein kinase (ERK1/2) and the signal transducer and activator of transcription 1 (STAT1), the upstream signaling proteins, was reduced by IPA treatment in HaCaT cells. In conclusion, IPA ameliorated AD-like skin symptoms by regulating cytokine and chemokine production and so has therapeutic potential for AD-like skin lesions.


Introduction
Atopic dermatitis (AD)-like skin diseases are the most common chronic inflammatory skin disorders due to innate and adaptive immune responses based on genetic, seasonal, and environmental causes [1]. AD-like skin diseases are characterized by pruritus, dry skin, abnormal immune responses, and IgE-mediated allergies that respond to various antigens [2]. The prevalence of these skin diseases is in millions of people around the world.
There are two hypotheses on the etiology of skin inflammation similar to AD-like skin disease. The first is primary immune dysfunction, leading to IgE sensitization and subsequent epithelial barrier disorders [3]. The second is an inherent genetic defect in the formation of the epidermal skin barrier, or blockage of the skin barrier due to environmental metastasis, causing skin symptoms similar to AD [4]. In addition, various internal and formation of the epidermal skin barrier, or blockage of the skin barrier due to environmental metastasis, causing skin symptoms similar to AD [4]. In addition, various internal and external factors contribute to the development of AD-like skin barrier dysfunction and immunomodulatory disorders [5].
Several commercial preparations for treating AD-like skin conditions are available as creams, gels, lotions, or ointments, but their effectiveness is limited [6]. In addition, there are several treatment regimens for AD such as topical glucocorticosteroids, calcineurin inhibitors, and immunosuppressive agents [7,8]. These remedies attenuate skin inflammation but have serious side effects. Topical steroids are the most effective treatment for AD-like skin lesions, but continued repeated application causes side effects [9]. Therefore, a new treatment without side effects is needed for AD-like skin symptoms. Various natural agents have been examined for the treatment of AD-like skin disorders. Impressic acid (IPA; 3α-11α-dihydroxylup-20(29)-en-28-oic acid) is the first known lupane-type triterpenoid from Schefflera impressa [10]. In Korea, it is found in the leaves and roots of the Acanthopanax koreanum and is widely used as an herbal medicine for diseases such as hepatitis, rheumatism, type 2 diabetes, and inflammatory diseases [11,12]. IPA enhances endothelial nitric oxide synthase activation via the 5′ AMP-activated protein kinase, Ca 2+ calmodulin-dependent protein kinase II, p38 MAPK, and the JNK1/2 pathways. In addition, IPA-induced nitric oxide production suppresses vascular inflammation by reducing intercellular adhesion molecule-1 expression and NF-κB activation [13]. IPA prevents cartilage degradation disorders by suppressing matrix metalloproteinase-13 expression [14]. IPA inhibits tumor necrosis factor-alpha (TNF-α) induced NF-κB activation and upregulates the transcriptional activity of peroxisome proliferator-activated receptor γ [15]. Recently, IPA reduced the LPS-induced inflammatory response in RAW264.7 macrophages [16]. Previous studies have shown the potential of IPA to attenuate inflammatory diseases including AD via suppressing NF-κB activation, but its effect and underlying mechanism on AD-like skin lesions have not been investigated. Therefore, we examined the inhibitory effect of IPA on 2,4-dinitrochlorobenzene (DNCB)-induced AD-like skin symptoms in BALB/C mice and the underlying mechanisms in HaCaT cells.

IPA Attenuated DNCB-Induced AD-Like Skin Severity in Mice
The chemical structure of IPA is shown in Figure 1A. DNCB induces AD-like skin lesions in mice [17]. To evaluate whether IPA can attenuate AD-like skin lesions by DNCB in mice, BALB/C mice were topically treated with IPA after induction of AD-like skin lesions by DNCB ( Figure 1B). Repeated skin application of DNCB leads to AD-like skin severities, such as ear edema, scleroderma, and amassing of inflammatory and mast cells. In contrast, IPA alleviated these DNCB-induced symptoms (Figure 2A-D). Furthermore, IgE stimulates mast Repeated skin application of DNCB leads to AD-like skin severities, such as ear edema, scleroderma, and amassing of inflammatory and mast cells. In contrast, IPA alleviated these DNCB-induced symptoms (Figure 2A-D). Furthermore, IgE stimulates mast cells to release inflammatory cytokines in AD-like skin lesions [18,19]. Serum concentration of IgE was increased by DNCB but was significantly attenuated by IPA ( Figure 2E). cells to release inflammatory cytokines in AD-like skin lesions [18,19]. Serum concentration of IgE was increased by DNCB but was significantly attenuated by IPA ( Figure 2E). Results are means ± SD (n = 6). # p < 0.05, versus the control group; *p < 0.05, versus the DNCB-treated group. IPA: Impressic acid; HC: Hydrocortisone; DNCB: 2,4-dinitrochlorobenzene.

IPA Attenuated DNCB-Induced Increased Levels of Thymic Stromal Lymphopoietin (TSLP) in Serum and Ear Lesions
TSLP regulates both the onset and maintenance of AD-like skin symptoms and plays an important role in the activation and differentiation of Th2 cytokines, which trigger the secretion of IgE from AD-like skin lesions [20,21]. We examined the inhibitory effect of IPA on DNCB-induced TSLP levels in ear lesions and serum. The serum concentration and mRNA expression of TSLP were increased by DNCB but were significantly attenuated by IPA ( Figure 3).

IPA Suppressed the DNCB-Increased mRNA Levels of Th1/Th2 Cytokines and Chemokines in Ear Lesions
According to a previous study, mast cells play a critical role in the secretion of proinflammatory mediators, including cytokines and chemokines [22]. IgE activates mast cells, enhancing pro-inflammatory cytokine expression in AD-like allergic skin lesions. AD-like skin symptoms are triggered by the interaction of Th1/Th2-predominant inflammation. Chemokines also play an important role in moving lymphocytes to the skin [23]. Thus, suppression of the production of these cytokines and chemokines may ameliorate AD-like skin diseases. We evaluated the effect of IPA on the mRNA levels of thymus and activation regulated chemokine (TARC), macrophage derived chemokine (MDC), interleukin-4 (IL-4), interleukin-5 (IL-5), interleukin-13 (IL-13), TNF-α, and interferon-gamma (IFN-γ) in ear lesions. IPA significantly reduced the mRNA levels of these Th1/Th2 cytokines and chemokines increased by DNCB (Figure 4). Therefore, IPA inhibits the expression of pro-inflammatory cytokines and chemokines, thus suppressing the symptoms of

IPA Attenuated DNCB-Induced Increased Levels of Thymic Stromal Lymphopoietin (TSLP) in Serum and Ear Lesions
TSLP regulates both the onset and maintenance of AD-like skin symptoms and plays an important role in the activation and differentiation of Th2 cytokines, which trigger the secretion of IgE from AD-like skin lesions [20,21]. We examined the inhibitory effect of IPA on DNCB-induced TSLP levels in ear lesions and serum. The serum concentration and mRNA expression of TSLP were increased by DNCB but were significantly attenuated by IPA ( Figure 3).  Results are means ± SD (n = 6). # p < 0.05, versus the control group; * p < 0.05, versus the DNCB-treated group.

IPA Suppressed the DNCB-Increased mRNA Levels of Th1/Th2 Cytokines and Chemokines in Ear Lesions
According to a previous study, mast cells play a critical role in the secretion of proinflammatory mediators, including cytokines and chemokines [22]. IgE activates mast cells, enhancing pro-inflammatory cytokine expression in AD-like allergic skin lesions. AD-like skin symptoms are triggered by the interaction of Th1/Th2-predominant inflammation. Chemokines also play an important role in moving lymphocytes to the skin [23]. Thus, suppression of the production of these cytokines and chemokines may ameliorate AD-like skin diseases. We evaluated the effect of IPA on the mRNA levels of thymus and activation regulated chemokine (TARC), macrophage derived chemokine (MDC), interleukin-4 (IL-4), interleukin-5 (IL-5), interleukin-13 (IL-13), TNF-α, and interferon-gamma (IFN-γ) in ear lesions. IPA significantly reduced the mRNA levels of these Th1/Th2 cytokines and chemokines increased by DNCB ( Figure 4). Therefore, IPA inhibits the expression of proinflammatory cytokines and chemokines, thus suppressing the symptoms of AD-like skin lesions, such as thickening of the dermis/epidermis and infiltration of inflammatory and mast cells.    Results are means ± SD (n = 6). # p < 0.05, versus the control group; * p < 0.05, versus the DNCB-treated group.

IPA Inhibited the DNCB-Induced Increased Activation of NF-κB in Ear Lesions
Cytokine secretion activated by T-cells is regulated by transcription factors, including NF-κB, which activate the expression of genes encoding pro-inflammatory cytokines and chemokines [24]. NF-κB is critical to the innate/adaptive immune response and inflammatory response, especially the Th1 response, and regulates inflammation caused by Th2 cell differentiation and activation [25]. We determined NF-κB activation in DNCB-induced mice by Western blotting. IPA significantly suppressed the phosphorylation of NF-κB and IκBα, and the degradation of IκBα, in DNCB-induced ear lesions ( Figure 5). Therefore, IPA strongly inhibits the activity of NF-κB, a transcriptional regulator of pro-inflammatory cytokines and chemokines.

IPA Inhibited the DNCB-Induced Increased Activation of NF-κB in Ear Lesions
Cytokine secretion activated by T-cells is regulated by transcription factors, including NF-κB, which activate the expression of genes encoding pro-inflammatory cytokines and chemokines [24]. NF-κB is critical to the innate/adaptive immune response and inflammatory response, especially the Th1 response, and regulates inflammation caused by Th2 cell differentiation and activation [25]. We determined NF-κB activation in DNCB-induced mice by Western blotting. IPA significantly suppressed the phosphorylation of NF-κB and IκBα, and the degradation of IκBα, in DNCB-induced ear lesions ( Figure 5). Therefore, IPA strongly inhibits the activity of NF-κB, a transcriptional regulator of pro-inflammatory cytokines and chemokines.

IPA Suppressed the TNF-α/IFN-γ-Induced Increased mRNA Levels of TSLP, TARC and MDC in Keratinocytes
To evaluate the effect of IPA on cell viability and cytotoxicity, we performed the MTT reduction and LDH release assays in HaCaT cells. Cells were treated with various concentrations of IPA for 24 h. IPA at <10 μM had no significant effect on cell viability and cytotoxicity ( Figure 6A,B). Keratinocytes play an important role in AD-like skin inflammation [26]. TSLP, TARC, and MDC are important in transporting lymphocytes to skin lesions [1,3,27]. Therefore, reducing TSLP, TARC, and MDC expression in keratinocytes may have therapeutic potential for AD-like skin lesions. Next, we performed the suppressive effect of IPA on the mRNA expression of these chemokines induced by TNF-α/IFN-γ in cells. The TNF-α/IFN-γ-induced increased mRNA levels of TSLP, TARC, and MDC were suppressed by IPA in HaCaT cells ( Figure 6C-E).

IPA Suppressed the TNF-α/IFN-γ-Induced Increased mRNA Levels of TSLP, TARC and MDC in Keratinocytes
To evaluate the effect of IPA on cell viability and cytotoxicity, we performed the MTT reduction and LDH release assays in HaCaT cells. Cells were treated with various concentrations of IPA for 24 h. IPA at <10 µM had no significant effect on cell viability and cytotoxicity ( Figure 6A,B). Keratinocytes play an important role in AD-like skin inflammation [26]. TSLP, TARC, and MDC are important in transporting lymphocytes to skin lesions [1,3,27]. Therefore, reducing TSLP, TARC, and MDC expression in keratinocytes may have therapeutic potential for AD-like skin lesions. Next, we performed the suppressive effect of IPA on the mRNA expression of these chemokines induced by TNF-α/IFN-γ in cells. The TNF-α/IFN-γ-induced increased mRNA levels of TSLP, TARC, and MDC were suppressed by IPA in HaCaT cells ( Figure 6C-E).

Discussion
AD-like skin disease is a multifactorial skin disease and has a complex relationship with innate and adaptive immune responses as well as with environmental, genetic, and psychological factors [30]. AD-like skin lesions show chronic, pruritic, and recurrent progression and affect 20% of children and 3% of adults worldwide [31]. AD-like skin disease, one of the most common chronic inflammatory skin diseases, is an IgE-mediated allergic disease of the epithelial barrier manifested by traumatic lesions, itching, dry skin, abnormal immune responses and various exogenous antigens [1].
AD-like skin diseases are typically treated with topical glucocorticosteroids, and cyclosporin A as gels, creams, or ointments, but these have limited effectiveness [6]. Such medications reduce skin inflammation but have serious side effects. However, there is increasing interest in natural product derived anti-inflammatory drugs for skin diseases because of the perception that such agents are inherently safe [17]. Recently, effort has focused on developing pharmaceuticals using natural ingredients for AD-like skin symptoms because of their low toxicity and high efficacy [32].
A. koreanum is an herb native to Korea, and its root and stem bark have been used to treat rheumatism and diabetes [33]. IPA is a triterpenoid component of the root of Acanthopanax koreanum and has marked anti-inflammatory activity [13][14][15][16]34]. However, its effect on AD-like skin lesions is unknown. Therefore, we assessed the inhibitory effect of IPA on AD-like skin lesions in BALB/C mice and in HaCaT cells.
The mechanisms underlying AD-like skin diseases are unknown but involve specific inflammatory and immune system responses mediated by IgE via a series of T helper cell interactions. The etiology of AD-like skin symptoms is allergic sensitization, exposure to

Discussion
AD-like skin disease is a multifactorial skin disease and has a complex relationship with innate and adaptive immune responses as well as with environmental, genetic, and psychological factors [30]. AD-like skin lesions show chronic, pruritic, and recurrent progression and affect 20% of children and 3% of adults worldwide [31]. AD-like skin disease, one of the most common chronic inflammatory skin diseases, is an IgE-mediated allergic disease of the epithelial barrier manifested by traumatic lesions, itching, dry skin, abnormal immune responses and various exogenous antigens [1].
AD-like skin diseases are typically treated with topical glucocorticosteroids, and cyclosporin A as gels, creams, or ointments, but these have limited effectiveness [6]. Such medications reduce skin inflammation but have serious side effects. However, there is increasing interest in natural product derived anti-inflammatory drugs for skin diseases because of the perception that such agents are inherently safe [17]. Recently, effort has focused on developing pharmaceuticals using natural ingredients for AD-like skin symptoms because of their low toxicity and high efficacy [32].
A. koreanum is an herb native to Korea, and its root and stem bark have been used to treat rheumatism and diabetes [33]. IPA is a triterpenoid component of the root of Acanthopanax koreanum and has marked anti-inflammatory activity [13][14][15][16]34]. However, its effect on AD-like skin lesions is unknown. Therefore, we assessed the inhibitory effect of IPA on AD-like skin lesions in BALB/C mice and in HaCaT cells.
The mechanisms underlying AD-like skin diseases are unknown but involve specific inflammatory and immune system responses mediated by IgE via a series of T helper cell interactions. The etiology of AD-like skin symptoms is allergic sensitization, exposure to allergens, overproduction of IgE and mast cells, Th1/Th2 imbalance, inflammation and infiltration of mast cells [35]. In this study, repeated topical application of DNCB exacerbated the symptoms of AD, such as ear lesions, ear swelling, inflammation, and infiltration of mast cells in mice, but these effects were ameliorated by IPA treatment.
Th1/Th2 imbalance and a defective skin barrier contribute to the progression of ADlike skin lesions [36]. Development of acute AD-like skin lesions involves a Th2-dominant inflammatory response characterized by increased Th cells, eosinophils, allergen-specific IgE, mast cell activation, and skin infiltration of Th2 cytokine secretion. Th2 cells produce IL-4 and IL-5, promoting antibody formation by B cells, eosinophils, and mast cells. IL-4 and IL-13 play roles in B-cell differentiation and grade conversion, thus increasing the serum IgE level in AD and triggering eosinophil and mast cell penetration of the skin [4]. Chronic AD-like skin symptoms involve predominantly Th1 cells, and chronic AD-like skin lesions exhibit inflammation-induced tissue remodeling. Th1 cells release TNF-α and IFN-γ and trigger a chronic inflammatory response [37]. In addition, epidermal barrier dysfunction results in dry and itchy skin lesions, which are aggravated by mechanical injury due to scratching. Together, these behaviors allow antigen penetration of the skin and secretion of inflammatory cytokines [38]. In this study, DNCB increased the mRNA levels of Th1 and Th2 cytokines, but this effect was reversed by IPA treatment in AD-like ear lesions.
A lot of studies have shown that TSLP acts as a master switch triggering both the onset and maintenance of AD-like skin lesions [39][40][41]. TSLP acts directly on Th2 cells to promote the differentiation of Th2 cells and is activated in the epidermis after skin-scratching action, and the increased secretion of Th2 cytokines in AD-like skin lesions results in elevated serum IgE levels. TARC, MDC, IgE and TSLP are important markers of the severity of AD-like skin symptoms. In this study, the levels of TARC, MDC, TSLP and IgE were increased in DNCB-induced AD-like skin lesions, but these effects were reversed by IPA treatment.
Because Th1 and Th2 cell differentiation and function and keratinocyte activation are important for the progression of AD-like skin lesions, an appropriate therapeutic approach to AD-like skin inflammation may be to regulate T cell and keratinocyte activation [42,43]. Keratinocytes play an important role in AD-like skin inflammation development, and TARC promotes the transport of lymphocytes to the skin. Thus, suppression of the secretion of these chemokines from keratinocytes may have therapeutic potential for AD-like skin inflammation. In this study, IPA treatment suppressed TNF-α/IFN-γ-induced increased mRNA levels of TARC, MDC and TSLP in a concentration-dependent manner.
NF-κB is an important transcriptional regulator that modulates the immune and inflammatory responses involved in the transcription of several targets, including the cytokines, chemokines and growth factors involved in the initiation of immune and inflammatory responses [44]. In addition, NF-κB is critical in the inflammatory response and mediates the activity of TARC in AD-related skin inflammation. Previous studies have reported that IPA inhibits NF-κB in various models of inflammation [14][15][16]. Our data indicate that TNF-α/IFN-γ-induced NF-κB activation was inhibited by IPA in HaCaT cells. In addition, topical application of IPA attenuated DNCB-induced NF-κB activation in mice. These results imply that IPA inhibits TARC expression by suppressing DNCB-and TNFα/IFN-γ-induced NF-κB activation. Downregulation of these target genes is dependent on the activation of NF-κB transcription, which may explain the decreased mRNA levels of cytokines and chemokines in DNCB-induced mice and TNF-α/IFN-γ-induced cells.
The STAT1 signaling pathway is one of the main inflammatory signaling pathways activated by several inflammatory cytokines, such as interleukins and interferons [47,48]. Upon receptor stimulation by inflammatory cytokines, a typical cell surface receptor, receptor-associated JAK2, is phosphorylated. Subsequently, STAT1 is phosphorylated and transferred into the nucleus, activating several inflammations, whereby it activates the expression of target genes. Therefore, suppression of STAT1 phosphorylation could ameliorate AD-like skin inflammatory diseases. In this study, IPA treatment suppressed the activation of STAT1 in keratinocytes, suggesting that IPA could reduce TNF-α/IFN-γinduced STAT1 activation.

Reagents
IPA was provided by Professor Young Ho Kim (Chungnam National University, Daejeon, Korea). The structure of IPA is shown in Figure 1A

Animals and Treatment
Specific-pathogen-free male 5-week-old BALB/C mice were bought from Raon Bio (Yongin, Korea). The mice were allowed free access to Purina rodent chow (Purina, Seoul, Korea) and tap water and were maintained in a controlled environment at 22 ± 2 • C and 50 ± 5% relative humidity under a 12 h dark/light cycle and acclimatized for at least 2 weeks before use. Forty mice were randomly divided into five groups for the doseresponse model (n = 6 mice/group): Control group, DNCB group, DNCB + 0.1% (w/v) IPA group, DNCB + 0.5% (w/v) IPA group and DNCB + 1% (w/v) HC group. To induce AD-like immunologic and skin lesions, DNCB was applied to the dorsal skin and ears. After complete removal of dorsal hairs within an area of approximately 8 cm 2 , 200 µL of 1% (w/v) DNCB solution (dissolved in a 3:1 mixture of acetone and olive oil, v/v) was applied for three consecutive days for sensitization. Four days after sensitization, the dorsal skin and ears were challenged with 200 µL of 0.5% DNCB solution three times per week for 7 weeks. After inducing AD, an emulsion (a 4:3:3 mixture of PEG 400, ethanol, and water, v/v/v) containing IPA was topically applied to the dorsal skin and ears of the mice six times per week for 4 weeks. In IPA-treated mice, IPA was topically applied 1 h before DNCB application each time. Control and DNCB-treated mice underwent topical application of an emulsion without IPA on their dorsal skin and ears. The mice were euthanized 57 days after the first application of DNCB ( Figure 1B). Blood was collected from the vena cava, and the right ear was removed and subjected to histopathological analysis. The experimental protocols were approved and performed according to the regulations of the Animal Ethics Committee of Chungnam National University (CNU-01148).

Histopathological Analysis of Ear Tissue
The right ear tissue was removed, fixed in 10% buffered neutral formalin, embedded in paraffin, sectioned, deparaffinized, and rehydrated. Ear tissue fixed in 10% formalin was subjected to (H&E) or toluidine blue (TB) staining; various inflammatory cells and mast cells were detected to observe histopathological changes (KP&T, Cheongju, Korea). An arbitrary score was assigned to each microscopic field viewed at a magnification of 100×.

Measurement of Ear Thickness
Ear thickness was determined using a micrometer (Mitutoyo, Kawasaki, Japan) on the day before euthanasia [17].

ELISA
The serum concentrations of IgE and TSLP were assessed by sandwich ELISA using an OptEIA Mouse IgE Kit (BD Biosciences, San Diego, CA, USA) and a DuoSet Mouse TSLP Kit (R&D Systems, Minneapolis, MN, USA) according to the manufacturers' protocols.

Cell Culture and Treatment
Spontaneously immortalized human keratinocytes (HaCaT; provided by Fusenig, German Cancer Research, Germany) were cultured in DMEM supplemented with 10% FBS and 1% penicillin-streptomycin at 37 • C in a humidified atmosphere containing 5% CO 2 . The cells were plated on 60-mm plates and incubated until 80% confluence. The cells were pretreated with IPA for 1 h and then with TNF-α and IFN-γ (10 ng/mL each) for 6 h. IPA was dissolved in DMSO (final concentration 0.1%) and stored at −20 • C until use.

Assays of Cell Viability and Cytotoxicity
Cell viability and cytotoxicity were assessed by performing MTT reduction and LDH release assays [17].

RNA Extraction and Quantitative Real-Time Reverse Transcription PCR
RNA extraction and cDNA synthesis were performed according to the manufacturer's protocol. PCR was performed using primers for mouse TARC, TSLP, TNF-α, IFN-γ, IL-4, IL-5, IL-13, and GAPDH or human TARC, MDC, TSLP, and GAPDH. The primer sequences were used as previously described [17,49].

Transient Transfection and Luciferase Assay
Cells were transfected with the NF-κB reporter vector and Renilla luciferase reporter vector using the serum-and antibiotic-free Lipofectamine 2000 reagent. At 5 h after transfection, the transfection medium was exchanged for the complement medium. The cells were treated with IPA (1, 5, or 10 µM) for 1 h, and then stimulated with TNF-α and IFN-γ (10 ng/mL each) for 24 h. Next, the cells were lysed, and luciferase activity was determined using a luminometer (Luminoscan Ascent, Thermo Electron). The luciferase signal was normalized to that of the Renilla luciferase and expressed relative to the control value.

Western Blotting
HaCaT cells and mouse left-ear tissue were washed with ice-cold PBS, and proteins were isolated in 100 µL of Ceti Lysis buffer (TransLab, Daejeon, Korea) containing protease and phosphatase inhibitors. To evaluate the target protein for NF-κB p65, IκBα, ERK1/2, and STAT1 activation, they were resolved in 10% and 12% polyacrylamide gels, transferred to nitrocellulose membranes (Amersham Pharmacia Biotech, Piscataway, NJ, USA), blocked in 5% skim milk, and probed with the appropriate primary and secondary antibodies.

Statistical Analysis
Results are expressed as means ± standard deviation (SD). Statistical significance was assessed by analysis of variance (ANOVA) followed by the Tukey-Kramer test. In vivo data are means ± SD (n = 6). Statistical significance was defined as # p < 0.05 compared to the control group or as * p < 0.05 compared to the DNCB-treated group. In vitro data are means ± SD (n = 3). Statistical significance was defined as # p < 0.01 compared to the control group or as * p < 0.01 compared to the TNF-α/IFN-γ treatment group.

Conclusions
We evaluated the inhibitory effect of IPA on AD-like skin lesions in BALB/C mice. IPA treatment suppressed DNCB-induced AD-like skin symptoms including infiltration of inflammatory and mast cells, elevated serum levels of IgE and TSLP and increased mRNA levels of TARC, MDC, TSLP, TNF-α, IFN-γ, IL-4, IL-5 and IL-13 in mice. In addition, IPA attenuated the phosphorylation of NF-κB in ear lesions and inhibited the TNF-α/IFNγ-induced increased TARC, MDC, and TSLP level by attenuating the ERK1/2-mediated NF-κB and STAT1 activation in keratinocytes. Therefore, IPA attenuated the development of AD-like skin signs by suppressing the synthesis of cytokines and chemokines and may have therapeutic potential for AD-like skin symptoms.

Institutional Review Board Statement:
The study was approved and performed according to the regulations of the Animal Ethics Committee of Chungnam National University (CNU-01148).

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.