Effect of Topically Applied Wikstroemia dolichantha Diels on the Development of Atopic Dermatitis-Like Skin Symptoms in Mice

Plants of the genus Wikstroemia are traditionally used to treat inflammatory diseases like bronchitis and rheumatoid arthritis. In the present study, the anti-atopic effects of an EtOH extract of Wikstroemia dolichantha (WDE) on oxazolone- and DNCB (2,4-dinitrochlorobenzene)-induced dermatitis in mice were investigated. Both ears of BALB/c mice were exposed to oxazolone, and dorsal skins of SKH-1 hairless mice were sensitized with DNCB to induce acute eczematous atopic skin lesions. 1% WDE was applied daily to oxazolone- and DNCB-induced AD mice for two or three weeks, respectively. Total IL-4 and IgE concentrations in serum, transepidermal water loss (TEWL) and skin hydration were assessed. High-performance liquid chromatography/mass spectrometry (HPLC/MS) was used to determine the composition of WDE. Dermal application of 1% WDE grossly and histopathologically improved oxazolone- and DNCB-induced AD skin symptoms. Epidermal thickness and mast cell infiltration were significantly lower in animals treated with WDE than in vehicle controls. Furthermore, in addition to reducing DNCB-induced increases in serum IL-4 (interleukin 4) and IgE (immunoglobulin E) levels, WDE also decreased TEWL and increased skin hydration (indicative of improved skin barrier function). The four flavonoids taxifolin, aromadendrin, padmatin and chamaejasmine were tentatively identified in WDE by HPLC-DAD/QTOF-MS. The above results show WDE protected against oxazolone- and DNCB-induced AD in mice by down-regulating the TH2-associated cytokine IL-4 and improving skin barrier function and suggest WDE might be useful for the management of atopic dermatitis.


Introduction
Atopic dermatitis (AD) is a chronic inflammatory skin disease with a multifactorial pathogenesis, and both genetic and environmental factors determine the expression of the disease [1]. Typical symptoms of AD, like pruritus, scaly skin, lichenification, and serous exudate, usually appear on the

Measurements of Skin Inflammation in The Oxazolone-Induced Model and WDE Treatment
AD was induced on the ears of BALB/c mice by applying 20 µL of oxazolone (4-ethoxymethylene-2-phenyl-2-oxazolin-5-one) in vehicle (propylene glycol: EtOH = 7:3) as previously described [22]. One week after first challenge, 20 µL of 0.1% oxazolone was applied every other day to both ears for 3 weeks (the OX group). On the other hand, the ears of BALB/c mice were exposed to 20 µL of 1% WDE twice daily over the same 3-week period (the OX-WDE group). WDE was applied 4 h before and after oxazolone application. The normal control group (CON group) was treated with distilled water only. On the last experimental day, no materials were applied and inflammatory indices, e.g., ear swelling and erythema, were obtained.

Measurement of Skin Severity in The DNCB-Induced Model and Treatment with WDE
DNCB (100 µL) (Sigma-Aldrich, Seoul) was used to provoke AD in SKH-1 hairless mice. During the first 7 experimental days, 1% DNCB was rubbed into the dorsal skin of hairless mice every day. Subsequently, mice were challenged with 0.1% DNCB (100 µL) every 3 days for an additional 2 weeks (DNCB group). In addition, 100 µL of 1% WDE (DNCB-WDE group) or 1% Elidel cream (DNCB-Elidel group) as positive control was applied to the dorsal skins of hairless twice daily over this 2-week period. When DNCB and WDE were applied on same days, WDE was applied 4 h before and after applying DNCB. Normal controls (the CON group) were administered distilled water.

Histological Examinations
To assess histopathological variations, ear skins from BALB/c mice or dorsal skins from SKH-1 hairless mice were fixed in 10% formalin for 24 h and paraffin infiltrated. Tissue sections were sliced to 2-3 mm, placed on slides, dried overnight at 37°C, and stained with H&E (hematoxylin and eosin) or toluidine blue. Histopathological examinations were performed and photographs obtained using an optical microscope (Olympus CX31/BX51, Olympus Optical Co., Tokyo) fitted with a TE-2000U camera (Nikon Instruments Inc. Melville, USA).

Total Serum IgE and IL-4 Levels
Blood specimens obtained from the abdominal aorta of SKH-1 hairless mice were centrifuged at 10,000 rpm for 15 min at 4°C. Serum samples obtained on experimental day 21 were stored at −70 • C until required for IgE and IL-4 determinations. Total serum IgE and IL-4 levels were measured using ELISA (enzyme-linked immunosorbent assay) kits (eBioscience, San Diego, USA).

HPLC/MS Analysis of WDE
This analysis was performed using an Agilent 6530 Accurate-Mass Q-TOF LC/MS system (Agilent Technologies). The analysis was performed using a Poroshell 120 EC-C18 column (3.0 × 100 mm, 2.7 µm, Agilent) at a flow rate of 0.3 mL/min, and detection was performed at 254 nm. The mobile phase consisted of acetonitrile (A) and ultrapure water (B) and the flow rate was 0.3 mL/min. The gradient profile was as follows: 0-5 min 10% A; 5-30 min, linear gradient from 10 to 70% A; and 30-35 min 70% A. All acquisitions were performed in positive ionization mode. Mass spectra were recorded over the range m/z = 100-1500 and accurate mass measurements were obtained for all peaks.

Statistical Analysis
The analysis was performed using ANOVA (one-way analysis of variance) and a statistical software program. Results are presented as means ± SEMs and significance was accepted for p values < 0.05.

Effects of WDE on Oxazolone-Induced AD Mice
The experimental procedure is summarized in Figure 1A. When BAKB/c mice were exposed to oxazolone (OX) for 3 weeks, ears exhibited severe AD-like skin symptoms ( Figure 1B). However, when mice were cotreated with oxazolone and 1% WDE, these symptoms were reduced. On experimental day 28, 1% WDE co-treatment was found to have markedly prevented lesion formation as compared with OX-treated mice. In fact, erythema, erosion, and dryness, were significant lower in the OX-WDE group than in the OX group on experimental day 28. The analysis was performed using ANOVA (one-way analysis of variance) and a statistical software program. Results are presented as means ± SEMs and significance was accepted for p values < 0.05.

Effects of WDE on Oxazolone-Induced AD Mice
The experimental procedure is summarized in Figure 1A. When BAKB/c mice were exposed to oxazolone (OX) for 3 weeks, ears exhibited severe AD-like skin symptoms ( Figure 1B). However, when mice were cotreated with oxazolone and 1% WDE, these symptoms were reduced. On experimental day 28, 1% WDE co-treatment was found to have markedly prevented lesion formation as compared with OX-treated mice. In fact, erythema, erosion, and dryness, were significant lower in the OX-WDE group than in the OX group on experimental day 28.

Histopathological Analysis of Oxazolone-Treated Skins
Histological changes in the dorsal skins of oxazolone-sensitized BALB/c mice were observed by H&E and toluidine blue staining. As shown in Figure 2A, H&E staining showed dorsal skin sections were thicker in the OX group than in the CON group. Toluidine blue staining also confirmed an increased infiltration of inflammatory cells in the oxazolone-induced CON group ( Figure 2B). Ear and epidermal thicknesses and mast cell infiltration counts are shown in Figs 2C, 2D, and 2E, respectively. Epidermal thicknesses and numbers of mast cells were significantly lower in the OX-WDE group (40.7 ± 5.8% decrease and 48.9 ± 4.6% decrease, respectively) than in the OX group.

Histopathological Analysis of Oxazolone-Treated Skins
Histological changes in the dorsal skins of oxazolone-sensitized BALB/c mice were observed by H&E and toluidine blue staining. As shown in Figure 2A, H&E staining showed dorsal skin sections were thicker in the OX group than in the CON group. Toluidine blue staining also confirmed an increased infiltration of inflammatory cells in the oxazolone-induced CON group ( Figure 2B). Ear and epidermal thicknesses and mast cell infiltration counts are shown in Figure 2C-E, respectively. Epidermal thicknesses and numbers of mast cells were significantly lower in the OX-WDE group (40.7 ± 5.8% decrease and 48.9 ± 4.6% decrease, respectively) than in the OX group.

Effects of WDE on Skin Severity in The DNCB-Induced Model and Histopathological Analysis
The experimental procedure is summarized in Figure 3A. Topical application of 1% WDE for two weeks to SKH-1 hairless mice exposed to DNCB effectively reduced AD symptoms including erythema, erosion and hemorrhage ( Figure 3B). Dorsal skins collected on experimental day 21 were subjected to H&E staining to determine epidermal thickness or toluidine blue staining to detect cells infiltrating tissues. Epidermal thickening and inflammatory cell infiltration were observed in the DNCB group, although they were not as high as the DNCB-WDE group ( Figure 4A,B). WDE reduced DNCB-induced epidermal thickness and mast cell numbers by 57.1 ± 6.2% and 16.7 ± 2.3%, respectively, as compared with the DNCB group ( Figure 4C and D). Application of 1%

Effects of WDE on Skin Severity in The DNCB-Induced Model and Histopathological Analysis
The experimental procedure is summarized in Figure 3A. Topical application of 1% WDE for two weeks to SKH-1 hairless mice exposed to DNCB effectively reduced AD symptoms including erythema, erosion and hemorrhage ( Figure 3B). Dorsal skins collected on experimental day 21 were subjected to H&E staining to determine epidermal thickness or toluidine blue staining to detect cells infiltrating tissues. Epidermal thickening and inflammatory cell infiltration were observed in the DNCB group, although they were not as high as the DNCB-WDE group ( Figure 4A,B). WDE reduced DNCB-induced epidermal thickness and mast cell numbers by 57.1 ± 6.2% and 16.7 ± 2.3%, respectively, as compared

Effects of WDE on Skin Barrier Function in DNCB-Treated SKH-1 mice
Skin barrier functions were evaluated by measuring TEWL and skin hydration. DNCB treatment decreased stratum corneum water content, but two weeks of 1% WDE enhanced skin hydration as compared with the DNCB group ( Figure 5). Mean TEWL was 12.5 ± 7.5% lower in the DNCB-WDE group than in the CON group ( Figure 5C) and skin hydration was significantly greater in the DNCB-WDE group than in the CON group ( Figure 5D). On day 21, co-treatment with 1% Elidel, a positive control, reduced the decrease in TEWL observed in the DNCB control by 28.5 ± 5.2% ( Figure 5C) and increased skin hydration by 12.0 ± 8.2% ( Figure 5D).

HPLC/MS of WDE
Optimization of the mobile phase to acetonitrile/water (1:9→70% acetonitrile, gradient system) resulted in a satisfactory S/N at a wavelength of 254 nm for all detected peaks. The presence of four compounds, that is, 1: taxifolin (m/z 303.05 at tR 14.4 min), 2: aromadendrin (m/z 287.06 at tR 18.3 min), 3: padmatin (m/z 317.07 at tR 21.2 min) and 4: chamaejasmine (m/z 539.24 at tR 23.3 min), was confirmed by comparing retention times and UV spectra with standard compounds (Figure 6).

Effects of WDE on Skin Barrier Function in DNCB-Treated SKH-1 mice
Skin barrier functions were evaluated by measuring TEWL and skin hydration. DNCB treatment decreased stratum corneum water content, but two weeks of 1% WDE enhanced skin hydration as compared with the DNCB group ( Figure 5). Mean TEWL was 12.5 ± 7.5% lower in the DNCB-WDE group than in the CON group ( Figure 5C) and skin hydration was significantly greater in the DNCB-WDE group than in the CON group ( Figure 5D). On day 21, co-treatment with 1% Elidel, a positive control, reduced the decrease in TEWL observed in the DNCB control by 28.5 ± 5.2% ( Figure 5C) and increased skin hydration by 12.0 ± 8.2% ( Figure 5D).

HPLC/MS of WDE
Optimization of the mobile phase to acetonitrile/water (1:9→70% acetonitrile, gradient system) resulted in a satisfactory S/N at a wavelength of 254 nm for all detected peaks. The presence of four compounds, that is, 1: taxifolin (m/z 303.05 at t R 14.4 min), 2: aromadendrin (m/z 287.06 at t R 18.3 min), 3: padmatin (m/z 317.07 at t R 21.2 min) and 4: chamaejasmine (m/z 539.24 at t R 23.3 min), was confirmed by comparing retention times and UV spectra with standard compounds (Figure 6).

Discussion
Plant extracts have been adopted as complementary and alternative medicines for the treatment and/or prevention of mild-to-severe AD [23]. Somewhat surprisingly, the major causes of skin diseases like atopic dermatitis (AD), psoriasis and pruritus have not yet been clearly identified and currently, the pathogenesis of AD is believed to involve complex interrelationships between genetic, environmental, skin barrier, pharmacologic, psychologic and immunologic factors [23][24][25]. Recent studies have indicated that skin barrier disruption contributes to the onset of AD [25,26]. Skin barrier disruption is the hallmark of AD and represents the main cause of the subsequent release of pro-inflammatory mediators [27]. Certain emollients or moisturizers, such as ceramide and hyaluronic acid, enhance skin barrier function, and thus, it has been suggested that barrierstrengthening moisturizers might delay the relapse of AD [28]. Topical corticosteroids are used in the management of AD as a second-line therapy and long-term treatment is problematic due to steroid-induced side effects, such as, skin atrophy and telangiectasia [8]. Accordingly, efforts are being made to identify naturally derived agents because they are likely to provide safer, more effective therapies.
In our preliminary in vitro experiment, we found that WDE had inhibited IL-4 production in RBL-2H3 cells (data not shown). Based on this result, we conducted a series of in vivo experiments to investigate its potential as a therapeutic agent for AD. The ears of female BALB/c mice were sensitized with 1% oxazolone for 1 week and then treated topically with 0.1% oxazolone for 3 weeks to induce AD-like skin symptoms. After oxazolone treatment mice showed typical skin lesions of AD (dry skin, cutaneous thickening, erythema, and edema). However, when 1% WDE was applied during the same 3-week period, these symptoms were significantly alleviated, as were oxazoloneinduced epidermal thickening and mast cell infiltration. Topical treatment with WDE was found to have an anti-AD effect in our oxazolone-induced murine AD model.
To determine whether the topic application of WDE relieves atopy, additional in vivo testing was conducted using a DNCB-induced SKH-1 hairless mouse model of AD. The dorsal skins of hairless mice were treated with 1% DNCB daily for 1 week and then with 0.1% DNCB every 3 days for 2 weeks. The dermal irritation induced by this treatment provoked the expression of IL-4 and massive release of IgE to serum. AD, associated with TH1 type cytokine IFN-γ and TH2 response,

Discussion
Plant extracts have been adopted as complementary and alternative medicines for the treatment and/or prevention of mild-to-severe AD [23]. Somewhat surprisingly, the major causes of skin diseases like atopic dermatitis (AD), psoriasis and pruritus have not yet been clearly identified and currently, the pathogenesis of AD is believed to involve complex interrelationships between genetic, environmental, skin barrier, pharmacologic, psychologic and immunologic factors [23][24][25]. Recent studies have indicated that skin barrier disruption contributes to the onset of AD [25,26]. Skin barrier disruption is the hallmark of AD and represents the main cause of the subsequent release of pro-inflammatory mediators [27]. Certain emollients or moisturizers, such as ceramide and hyaluronic acid, enhance skin barrier function, and thus, it has been suggested that barrier-strengthening moisturizers might delay the relapse of AD [28]. Topical corticosteroids are used in the management of AD as a second-line therapy and long-term treatment is problematic due to steroid-induced side effects, such as, skin atrophy and telangiectasia [8]. Accordingly, efforts are being made to identify naturally derived agents because they are likely to provide safer, more effective therapies.
In our preliminary in vitro experiment, we found that WDE had inhibited IL-4 production in RBL-2H3 cells (data not shown). Based on this result, we conducted a series of in vivo experiments to investigate its potential as a therapeutic agent for AD. The ears of female BALB/c mice were sensitized with 1% oxazolone for 1 week and then treated topically with 0.1% oxazolone for 3 weeks to induce AD-like skin symptoms. After oxazolone treatment mice showed typical skin lesions of AD (dry skin, cutaneous thickening, erythema, and edema). However, when 1% WDE was applied during the same 3-week period, these symptoms were significantly alleviated, as were oxazolone-induced epidermal thickening and mast cell infiltration. Topical treatment with WDE was found to have an anti-AD effect in our oxazolone-induced murine AD model.
To determine whether the topic application of WDE relieves atopy, additional in vivo testing was conducted using a DNCB-induced SKH-1 hairless mouse model of AD. The dorsal skins of hairless mice were treated with 1% DNCB daily for 1 week and then with 0.1% DNCB every 3 days for 2 weeks. The dermal irritation induced by this treatment provoked the expression of IL-4 and massive release of IgE to serum. AD, associated with T H 1 type cytokine IFN-γ and T H 2 response, increased levels of T H 2-associated cytokines (IL-4, IL-5 and IL-13) [29]. Herbal anti-inflammatory agents are often used to prevent and treat skin and allergic diseases [30]. Plant-derived phytochemicals with anti-AD properties are known to exert their effects by disrupting the activities of inflammatory cytokines and receptors [30,31]. In the present study, treatment with 1% WDE during the final 2 weeks of DNCB application significantly reduced DNCB-induced increases in serum IL-4 (the critical T H 2 cytokine) and IgE production. Furthermore, the lesioned skins of DNCB-treated mice had greater TEWL and lower skin hydration values than treatment naïve controls, indicating impaired barrier function [32,33]. In particular, the 1% WDE treatment produced the similar inhibitory effect when compared to the positive control, 1% Elidel, regarding reducing serum IL-4 level. Application of WDE during the last 2 weeks of DNCB application markedly improved skin biophysical properties of dry atopic skin versus DNCB application alone. These results indicate WDE acts as a moisturizer and improves skin barrier function in atopic skin and reduces atopic response to irritants.
Nine flavonoids, that is, apigenin, afzelechin, afzelechin 3-O-glucoside, aromadendrin, catechin, kaempferol, taxifolin, padmatin, and chamaejasmine, were identified in the aerial parts of W. dolichantha during the present study (unpublished data). Furthermore, phytochemical screening of WDE revealed the presence of the flavonoids taxifolin, aromadendrin, padmatin, and chamaejasmine. Flavonoids are phenolics commonly found in fruits, vegetables, grain, flowers, tea, and wine [34] that possess diverse bioactivities [11]. However, the role of flavonoids in human health have been inconclusive and their potential use, especially in infants and children, is still a matter of debate [35]. Taken together, these four flavonoids would appear to be responsible for the anti-atopic effects of WDE.

Conclusions
In summary, an EtOH extract of the aerial parts of W. dolichantha (WDE) was found to effectively reduce the AD-like symptoms induced by oxazolone or DNCB via regulating IgE synthesis and T H 2-mediated cytokine IL-4 expression. Topical treatment with WDE strengthens the skin barrier function, thereby exhibiting an excellent effect of improving skin moisturization in our BALB/c and SKH-1 mouse models with AD. Phytochemical analysis supports the hypothesis that the anti-inflammatory and anti-allergic effects of WDE could, in part, be due to the presence of flavonoids like taxifolin, aromadendrin, padmatin, and chamaejasmine as major constituents. Our results suggest that W. dolichantha may offer a potential means of preventing/treating atopy.