Chamaejasmine Isolated from Wikstroemia dolichantha Diels Suppresses 2,4-Dinitrofluoro-benzene-Induced Atopic Dermatitis in SKH-1 Hairless Mice

Plants of the genus Wikstroemia have long been used as traditional medicines to treat diseases like pneumonia, rheumatism, and bronchitis. This study was designed to determine the effect of chamaejasmine, a biflavonoid present in W. dolichantha, on atopic dermatitis (AD)-like skin lesions in a 2,4-dinitrochlorobenzene (DNCB)-induced murine model of AD. Initially, we examined the anti-allergic activities of ten flavonoids from W. dolichantha by measuring β-hexosaminidase release from RBL-2H3 cells. Subsequently, an SKH-1 hairless mouse model of AD was developed based on the topical application of DNCB. Chamaejasmine (0.5%) or pimecrolimus (1%, positive control) were applied to dorsal skins of DNCB-sensitized AD mice for two weeks. Serum IL-4 and IgE levels were determined using enzyme-linked immunosorbent assay kits and transepidermal water loss (TEWL) and skin hydration were measured using a Tewameter TM210 and a SKIN-O-MAT, respectively. Of the ten flavonoids isolated from W. dolichantha, chamaejasmine most potently inhibited DNP-specific IgE-induced degranulation in RBL-2H3 cells. Topical administration of chamaejasmine attenuated the clinical symptoms of DNCB-induced dermatitis (i.e., itching, dryness, erythema, and edema). Histological analyses demonstrated that dermal thickness and mast cell infiltration in dermis were significantly reduced by chamaejasmine. In addition, 0.5% chamaejasmine inhibited DNCB-induced increases in total IL-4 and IgE levels in serum, improved skin barrier function, and increased epidermis moisture. Our findings suggest chamaejasmine might be an effective therapeutic agent for the treatment of atopic diseases.


Introduction
The atopic dermatitis (AD) is a chronic skin disorder provoked by immune system disturbance and is characterized by itching, redness, skin keratinization, and exudates. AD is caused by a complex

Plant Material and Extraction
The aerial parts of Wikstroemia dolichantha Diels were collected in Yunnan Province, Lijiang, China and identified by Dr. Sang Woo Lee (Korea Research Institute of Bioscience and Biotechnology). A voucher specimen (PNU-0024) was deposited at the Medicinal Herb Garden, Pusan National University. Dried aerial parts of W. dolichanta (5 kg) were extracted with 95% EtOH (12 L × 3) and evaporated under reduced pressure to yield W. dolichanta EtOH extract (WDE) (466.8 g, 9.3% extract yield).

β-Hexosaminidase Release Assay
RBL-2H3 cells were treated overnight with anti-dinitrophenyl immunoglobulin E (anti-DNP IgE), rinsed out with Siraganian buffer, and incubated in buffer for 10 min. Cells were then treated with DMSO or separately with each of the ten isolates of W. dolichantha (30 µM) for 1 h and sensitized with DNP-BSA antigen (10 µg/mL) for 20 min to provoke degranulation. Supernatants were transferred to 96-well plates and incubated with 1 mM of 4-nitrophenyl-N-acetyl-β-d-glucosaminide as substrate in 0.1 M citrate buffer for 3 h at 37 • C. Absorbances were measured using a microplate reader at 405 nm.

Animals
Six-week-old female SKH-1 hairless mice purchased from Orient Bio Inc. (Seongnam, Korea) were housed in an air-controlled animal room (25 ± 5 • C, 55 ± 5% RH). Animals were fed standard laboratory food and water ad libitum. All animal experiments were conducted in accord with the Guide for the Care and Use of Laboratory Animas issued by the National Institute of Health (NIH publication No. 85-23, revised 2011) and authorized by the Institutional Animal Care and Use Committee of KIST (Certification No. KIST-2016-011).
2.7. Induction of DNCB-Induced AD and Treatment with Chamaejamine from W. dolichantha 2,4-Dinitrochlorobenzene (DNCB; 1%) (Sigma-Aldrich, Seoul, Korea) in propylene glycol:EtOH (7:3 v/v) was used to induce AD-like symptoms. First, mice in the DNCB control group and the DNCB/isolated compound groups were sensitized by applying 1% DNCB in propylene glycol:EtOH (7:3 v/v) to dorsal skin once daily for a week (experimental day 1 (ED1) to ED7). DNCB controls were then challenged by administering 100 µL of 0.1% DNCB from ED8 every 3 days for 2 weeks. Animals were painted with 100 µL solution of 0.5% chamaejamine or 1% pimecrolimus, a positive control, in propylene glycol:EtOH (7:3 v/v) and with DNCB, as described above, twice a day from ED8 for 2 weeks. In these groups, chamaejamine or pimecrolimus were applied 4 h after DNCB application. Mice in the untreated control group were treated with propylene glycol:EtOH (7:3 v/v) daily using the method described above.

Histological Examination
To examine histologic changes, dorsal skins were fixed in 10% formalin for 24 h and embedded in paraffin. Fixed tissues were sectioned at 2-3 mm, dried for 24 h at 37 • C, and stained with hematoxylin and eosin (H&E) or toluidine blue. Histological changes were observed under an optical microscope (Olympus CX31/BX51, Olympus Optical Co., Tokyo, Japan) and photographed (TE2000U, Nikon Instruments Inc., Melville, NY, USA).

Measurement of Total Serum IgE and IL-4 Levels
Blood samples were obtained and centrifuged at 10,000 rpm for 15 min at 4 • C. Total serum IgE and IL-4 levels were measured using enzyme-linked immunosorbent assay kits (eBioscience, San Diego, CA, USA).

Statistical Analysis
All quantitative data for this study were obtained through at least two independent experiments. In vitro data are shown as the means ± SDs and in vivo data are denoted as means ± SEMs for seven animals. Statistical analyses were carried out by a one-way analysis of variance (ANOVA) and a Tukey's multiple comparisons post hoc analysis.

Chamaejasmine (7) from W. dolichantha Ameliorated AD-like Skin Symptoms in DNCB-Induced Atopic Mice
To investigate the effects of 7 on the skin lesions of DNCB-induced atopic mice, dermatitis levels were evaluated using skin lesion images. The procedure used to establish the DNCB-induced AD mouse model is shown in Figure 4A. On the last day of the experiment, DNCB application produced significant AD-like lesions, including erythema (with scratch marks), erosions, and dryness in the DNCB controls. Reduced AD-like symptom severity was observed in 0.1% DNCBtreated SKH-1 hairless mice co-treated with 0.5% 7 (DNCB-chamaejasmine group) ( Figure 4B), in which the epidermal thickness of dorsal skin was 81% thinner than in the DNCB control group (Figures 5A,C). In addition, 7 application reduced the number of mast cells by 62% as compared with the DNCB controls group (Figures 5B,D). Pimecrolimus cream is a well-known treatment for atopic skin disorders and was used as a positive control. 1% Pimecrolimus cream reduced dorsal skin dermal thickness by 54% versus DNCB controls (Figures 5A,C) and mast cell infiltration by 33% ( Figures 5B,D).

Chamaejasmine (7) from W. dolichantha Ameliorated AD-like Skin Symptoms in DNCB-Induced Atopic Mice
To investigate the effects of 7 on the skin lesions of DNCB-induced atopic mice, dermatitis levels were evaluated using skin lesion images. The procedure used to establish the DNCB-induced AD mouse model is shown in Figure 4a. On the last day of the experiment, DNCB application produced significant AD-like lesions, including erythema (with scratch marks), erosions, and dryness in the DNCB controls. Reduced AD-like symptom severity was observed in 0.1% DNCB-treated SKH-1 hairless mice co-treated with 0.5% 7 (DNCB-chamaejasmine group) (Figure 4b), in which the epidermal thickness of dorsal skin was 81% thinner than in the DNCB control group (Figure 5a,c). In addition, 7 application reduced the number of mast cells by 62% as compared with the DNCB controls group (Figure 5b,d). Pimecrolimus cream is a well-known treatment for atopic skin disorders and was used as a positive control. 1% Pimecrolimus cream reduced dorsal skin dermal thickness by 54% versus DNCB controls (Figure 5a,c) and mast cell infiltration by 33% (Figure 5b,d).
which the epidermal thickness of dorsal skin was 81% thinner than in the DNCB control group (Figures 5A,C). In addition, 7 application reduced the number of mast cells by 62% as compared with the DNCB controls group (Figures 5B,D). Pimecrolimus cream is a well-known treatment for atopic skin disorders and was used as a positive control. 1% Pimecrolimus cream reduced dorsal skin dermal thickness by 54% versus DNCB controls (Figures 5A,C) and mast cell infiltration by 33% (Figures 5B,D).

Chamaejasmine (7) Reduced Serum IgE and IL-4 Levels in DNCB-Induced Atopic Mice
The major biochemical characteristics of AD are increased serum levels of IgE and IL-4. Serum IgE levels were significantly higher in DNCB controls than in untreated controls. Treatment with 0.5% 7 resulted in a 38% decrease in serum IgE concentration versus DNCB controls ( Figure 6A). Mean total serum IL-4 level (42.5 pg/mL) was also higher in DNCB controls than in untreated controls (17.4 pg/mL). Mean total serum IL-4 levels in the DNCB-chamaejasmine group (25.2 pg/mL) were markedly lower than in DNCB controls ( Figure 6B). Pimecrolimus cream decreased serum IgE levels by 43% (Figure 6a) and serum IL-4 levels by 27% ( Figure 6B) versus those in DNCB controls.

Chamaejasmine (7) from W. dolichantha Recovered Skin Barrier Function in DNCB-Induced Atopic Mice
A significant increase in dorsal skin TEWL was observed in DNCB controls on the last day of the experiment. In contrast, treatment with 0.5% 7 (43.5 g/m 2 h) and 0.5% pimecrolimus cream (46.7 g/m 2 h) markedly reduced TEWL as compared with DNCB controls (68.9 g/m 2 h) (Figure 7a). Furthermore, a significant decrease in skin hydration was observed in DNCB controls during the 21-day experimental period, but skin hydration was significantly higher in the DNCB-chamaejasmine group (45% increase) than in DNCB controls (Figure 7b).

Chamaejasmine (7) from W. dolichantha Recovered Skin Barrier Function in DNCB-Induced Atopic Mice
A significant increase in dorsal skin TEWL was observed in DNCB controls on the last day of the experiment. In contrast, treatment with 0.5% 7 (43.5 g/m 2 h) and 0.5% pimecrolimus cream (46.7 g/m 2 h) markedly reduced TEWL as compared with DNCB controls (68.9 g/m 2 h) ( Figure 7A). Furthermore, a significant decrease in skin hydration was observed in DNCB controls during the 21-day experimental period, but skin hydration was significantly higher in the DNCBchamaejasmine group (45% increase) than in DNCB controls ( Figure 7B).

Discussion
Flavonoids are a large family of plant secondary metabolites, are found in all fruits and vegetables [21], and are primarily responsible for the characteristic red and blue colors of flowers and berries [22]. Over 4,000 flavonoids have been identified to date and are classified as flavonols, flavones, flavanones, catechins, anthocyanidins, isoflavones, dihydroflavonols, or chalcones [23]. The bioactivities of flavonoids are related to their chemical structures and functionalities [24,25]. Flavonoids are known to have health-promoting effects, which have been largely attributed to their

Discussion
Flavonoids are a large family of plant secondary metabolites, are found in all fruits and vegetables [21], and are primarily responsible for the characteristic red and blue colors of flowers and berries [22]. Over 4,000 flavonoids have been identified to date and are classified as flavonols, flavones, flavanones, catechins, anthocyanidins, isoflavones, dihydroflavonols, or chalcones [23]. The bioactivities of flavonoids are related to their chemical structures and functionalities [24,25]. Flavonoids are known to have health-promoting effects, which have been largely attributed to their antioxidant, anti-inflammatory, and chelating properties [25][26][27]. However, experimental evidence regarding their anti-allergic effectiveness is scarce and limited to a few individual flavonoids under highly specific experimental conditions.
Hapten-induced mouse models of contact dermatitis have been commonly used to investigate anti-AD properties, such as those of oxazolone, DNCB, and 1, 3, 5-trinitrochlorobenzene (TNCB). DNCB is considered to be useful for inducing sensitization due to its availability and reproducibility [30,31]. Topical administration of 0.5% chamaejasmine (7) for two weeks markedly attenuated DNCB-induced AD-like skin lesions, which included hyperkeratosis, epidermal thickening, and mast cell infiltration, in our murine model. Moisturizers increase skin hydration and improve skin barrier function, thereby serving as an important first-line therapeutic option for patients with AD increasing skin hydration [32]. Chamaejasmine (7) also protected skin barrier functions by increasing TWEL and skin hydration as a moisturizer and suppressing inflammatory cytokine IL-4 production versus DNCB controls. Reduced skin barrier function and inflammation have been reported to be associated as impaired skin barrier function facilitates allergen entry and thus inflammation and inflammation, which in turn, may comprise barrier functions [33,34]. Accordingly, our observations suggest chamaejasmine (7) treatment may inhibit T H 2-dominated inflammatory response. In addition, chamaejasmine (7) suppressed allergic inflammation by strongly inhibiting DNCB-induced increases in serum IgE and IL-4 levels. Application of chamaejasmine (7) markedly suppressed the expression level of IL-4 in vivo. It is suggested a possible mechanism that inhibition of IL-4 prevents the differentiation of Th0 to Th2 cells in lymph nodes, the IgE class switch of B cells caused by IL-4 secreted from Th2 cells, and the degranulation of mast cells by IgE, which mediate lowering allergic response. When Th2 cells migrate from lymph nodes to skin tissues, IL-4 increases the expression of proteases such as Kallikreins in keratinocytes of the epidermis, which weaken the binding between epidermal cells and skin barrier function. In addition, macrophage activates the STAT6 pathway downstream of IL-4, leading to polarization to M2. Treatment of chamaejasmine (7) is expected to prevent the above symptoms by IL-4 and reduce the worsening of AD.
A considerable body of evidence suggests that plants flavonoids are health-promoting and disease-preventing dietary compounds [35]. However, the potential therapeutic use of flavonoids and flavonoid-rich extracts in infants and children remains a matter of debate [36]. Some well-known flavones and flavonols, such as, quercetin and kaempferol, have been especially well analyzed for their immunological effects [37]. Previous structure-activity studies with flavonoids have shown that flavones or flavonols were active in inhibiting degranulation of RBL-2H3 cells but flavanones, isoflavones, and catechins were not active [38,39]. In contrast to previous results on the flavanones, our present study suggests that flavanones (aromadendrin and taxifolin) and flavanone dimers (wikstaiwanone C, neochamaejasmine B, and chamaejasmine) may be useful inhibitors of β-hexosaminidase. Further studies on whole cell systems and on other flavonoids, especially flavanones, will be reported in due course.
Six flavonoids isolated from W. dolichantha significantly reduced IgE-induced β-hexosaminidase release from RBL-2H3 cells, and chamaejasmine (7) was found to be the most active. Topical chamaejasmine (7) application impressively reduced erythema, edema, erosion, dryness, and lichenification, in our SKH-1 mouse model by suppressing serum IgE and IL-4 levels. Collectively, these results suggest 7 be viewed as a novel anti-inflammatory and anti-atopic agent with promising therapeutic potential.