Abstract: Wang Junxia, Yang Ziwei, Che Yamin, Shan Shijun, Chen Hongduo Department of Dermatology and Venereology, Tianjin Medical University General Hospital, Tianjin 300052, China （Wang JX, Yang ZW, Che YM, Shan SJ）; Department of Dermatology and Venereology, China Medical University, The First Hospital of China Medical University, Shenyang 110001, China （Chen HD） Corresponding author: Shan Shijun, Email: 15822183620@163.com 【Abstract】 Objective To evaluate the effects of extracts of Semen Coicis （ESC） on a BALB/c mouse model of atopic dermatitis （AD）, and to explore its potential mechanism. Methods Forty specific pathogen-free （SPF） female BABL/c mice were randomly divided into blank group （8 mice, receiving no treatment） and AD model group （32 mice）. The mice in the model group were topically treated with 2,4-dinitrochlorobenzene （DNCB） in acetone/olive oil to establish the mouse model of AD. After modeling, 8 mice in the blank group and 8 in the model group were sacrificed immediately. The other 24 mice in the model group were randomly and equally divided into 3 groups: model control group receiving no treatment, ESC group and ESC vehicle group topically treated with ESC and ESC vehicle respectively once every day on the back and aural region of the mice for 28 consecutive days. Changes in skin lesions were observed by naked eyes every day. A thickness tester was used to measure the thickness of skin lesions on the left ear before modeling, at completion of modeling and 12 hours after the final treatment. At 12 hours after the final treatment, the mice in the above 3 groups were sacrificed, and the eyeballs were removed for collecting blood. Then, the sera were isolated, and skin tissue specimens were obtained from the skin lesions on the back. These tissue sections were subjected to hematoxylin and eosin （HE） staining and toluidine blue staining for observing the infiltration of inflammatory cells in skin lesions. An immunohistochemical study was performed to determine the of aquaporin 3 （AQP3）, Toll-like receptor 2 （TLR2） and TLR4, and enzyme-linked immunosorbent assay （ELISA） to detect the serum levels of IgE, interleukin-4 （IL-4） and interferon-γ （IFN-γ）. Results After 28-day treatment, skin lesions were improved in the ESC group. Compared with the model control group, the ESC group showed a significantly lower clinical symptom score （1.50 ± 0.58 vs. 2.50 ± 0.58, P < 0.05）, decreased lesional thickness on the left ear （［0.31 ± 0.01］ mm vs. ［0.33 ± 0.01］ mm, P < 0.05）, and lower number of infiltrating mast cells per high-power field （15.18 ± 1.64 vs. 28.94 ± 1.28, P < 0.05）. Immunohistochemical findings indicated that the ESC group showed significantly lower of AQP3, TLR2 and TLR4 compared with the model control group, and decreased AQP3 in the spinous layer. Compared with the model control group, the ESC group showed significantly lower total serum IgE and IL-4 levels, but higher IFN-γ levels （all P < 0.05）. Conclusion Topical ESC is effective for the treatment of skin lesions in mouse models of AD, likely by regulating serum levels of IgE, IL-4 and IFN-γ and affecting the of AQP3, TLR2 and TLR4.

Key words: Dermatitis, atopic, Semen Coicis, Disease models, animal, Interleukin-4, Interferon-gamma, Toll-like receptors, Aquaporin 3