Salt-processed CPE and SOE were purchased from Kyung-Dong Herb market (Seoul, Korea) in 2010, were confirmed by Dr Byung Seob Ko (Korean Herbal Medicine Institute, Daejeon, Korea), and voucher specimens (nos. 2010-04 and 2010-05) were deposited at the herbarium at the Department of Food and Nutrition, Hoseo University (Asan, Korea). Salt-processed CPE is commercially produced by boiling Phellodendron amurense bark and spraying the bark with 2% salt water prior to drying. Since 1,3-butylene glycol is an effective solvent for producing skin lotion (16), the salt-processed CPE and SOE (1 kg) were extracted at room temperature for 12 h using 3.3 liters of 1,3-butylene glycol (Sigma-Aldrich, St. Louis, MO, USA), prior to being filtered with filter paper (Watman; GE Healthcare, Little Chalfont, UK) and centrifuged at 450 × g at room temperature for 30 min to produce 30% extracts. Over 30% of the salt-processed extracts of CPE and SOE in 1,3-butylene glycol formed a precipitate. The supernatants were used for topical application in the subsequent experiments.

The levels of total phenolic compounds of each 30% extract of salt-processed CPE and SOE were measured using Folin-Ciocalteu reagent (97.5% purity; Sigma-Aldrich), and were expressed as mg gallic acid equivalents·g⁻¹. The extracts were dissolved in ethanol and the total flavonoid contents were measured using a previously described method (17) with minor modifications. The extract was added to 2 N HCl prior to being filtered. The solution was then mixed with bromocresol green solution (Sigma-Aldrich) and phosphate buffer (1:5:5) and the mixture was transferred to a separating funnel. Chloroform (Sigma-Aldrich) was subsequently added and mixed by vigorous agitation. The chloroform fraction was separated and its absorbance was measured at 470 nm using a UV/Visible spectrophotometer (Lambda 850; Perkin Elmer, Waltham, MA, USA). Berberine chloride (>90% purity; Sigma-Aldrich) was used as a standard. The total alkaloid content was expressed as mg berberine/g extract (18).

A total of 20 male six-week-old NC/Nga mice were purchased from Charles River Japan (Yokohama, Japan), and maintained in conventional conditions of a 12 h light/12 h dark cycle, room temperature of 22–23°C and humidity of 55±15%. The mice had free access to food and water. All surgical and experimental procedures were performed according to the guidelines of the Animal Care and Use Review Committee of Hoseo University (Asan, Korea).

The mice were anesthetized with a mixture of ketamine and xylazine (100 and 10 mg/kg body weight, respectively; Bayer AG, Leverkusen, Germany), following which the hair on the back and right ear were shaved 1 day prior to sensitization. On the first day, 1% DNCB in acetone/olive oil (3:1; 150 µl per mouse) was applied to the dorsal skin and right ear, following which 0.2% DNCB (150 µl per mouse) was applied every other day for five weeks, as previously described (10). The same volume of acetone/olive oil vehicle was applied, instead of the DNCB solution, to the controls. Repeated application of DNCB onto the dorsal region caused apparent dermatitis in the NC/Nga mice (19).

Topical application of the 1,3-butylene glycol extracts of the salt-processed CPE and SOE were used to determine the effect of CPE and SOE on atopic dermatitis. Based on the preliminary cell-based investigations and the maximum dosage of the extract, two doses were assigned. The preliminary investigation demonstrated that 20 and 50 µg/ml of salt-processed CPE and SOE extracts, respectively, were effective against house mites (Arthropods of Medical Importance Resource Bank, Yonsei University, Seoul, Korea) in the HaCaT human keratinocyte cell line (American Type Culture Collection; Manassas, VA, USA). Therefore, the topical application of 200 µl 30% 1,3-butylene glycol was considered to be an effective dosage for use the animal model, when compared with a previous study (10). Following the induction of the atopic dermatitis-like skin lesions, the animals were divided into four groups, each containing 10 mice. The mice in these groups were then treated topically on the dorsal skin with a 200 µl dose of one of the following four agents for five weeks: 1,3-butylene glycol (BG; control); 30% CPE; 30% SOE; 15% CPE+15% SOE; or 0.1% hydrocortisone butyrate (HC; Sigma-Aldrich; positive control) twice a day. Mice without induction of atopic dermatitis-like skin lesions were treated with 1,3-butylene glycol as a normal control. At the end of the study, rats were anesthetized with ketamine and xylazine (100 and 10 mg/kg body weight, respectively). Rats were sacrificed and tissues were collected for further experiments.

The relative dermatitis severity was assessed macroscopically using a previously described scoring procedure (20). The total skin severity scores were assessed weekly and defined as the sum of the individual scores for each of the following four symptoms: i) Erythema and hemorrhage, ii) edema, iii) erosion (excoriation) and iv) scaling (dryness). For each symptom, 0 was defined as exhibiting no symptoms, 1 as mild symptoms, 2 as moderate symptoms, and 3 as severe symptoms. To minimize technique variations, a single investigator performed the measurements throughout each experiment in a blinded-manner.

Total serum levels of IgE and IgG1 were quantified using an ELISA Quantification kit (BD Biosciences, San Jose, CA, USA), according to the manufacturer's instructions. In addition, the serum concentrations of the TNF-α, IL-4 and IFN-γ cytokines were quantified using a Mouse Cytokine Enzyme Immunoassay kit (R&D Systems, Minneapolis, MN, USA).

The dorsal skin tissue samples from five rats of each group were collected at the end of the experiment, and each skin tissue sample was individually powdered with a cold steel mortar and pestle, prior to being mixed with a monophasic solution of phenol and guanidine isothiocyanate (TRIzol reagent; Invitrogen Life Technologies, Carlsbad, CA, USA) for total RNA extraction, according to the manufacturer's instructions. The quantity and purity of RNA was measured at 260 and 280 nm using a Lambda 850 spectrophotometer (Perkin Elmer, Inc.) and cDNA was reverse transcribed from 1 µg RNA extracted from the individual rats using a Superscript III Reverse Transcriptase kit (Invitrogen Life Technologies). A total of five cDNAs were produced from each group, and each cDNA was used for RT-qPCR. Equal quantities (1 µg) of cDNA and primers for specific genes were mixed with SYBR^® Green (Bio-Rad Laboratories, Inc., Hercules, CA, USA) in duplicate, and amplified using a real-time PCR instrument (CFX Connect™ Real-Time PCR Detection System; Bio-Rad Laboratories, Inc.). The following thermocycling conditions were used to perform the PCR: 55°C for 2 min, 95°C for 10 min followed by 40 cycles of 94°C for 20 sec, 65°C for 30 sec and 72°C for 20 sec. To assess which genes were associated with inflammation and degradation of articular cartilage, primers were used to detect the expression levels of rat-inducible nitric oxide synthase, TNF-α, IL-1β, IL-6, matrix metalloprotinase (MMP)-3, and MMP-13 genes, as previously described (21,22). The cycle threshold (CT) for each sample was subsequently determined. The mRNA expression levels in the unknown samples were quantified using the comparative CT method (ΔΔCT method), as previously described by Livak and Schmittgen (23). ΔCT was calculated using the following formula: ΔCT = CT (target gene) − CT (endogenous reference gene, β-actin). The relative fold-change in expression was calculated using the following equation: ΔΔCt = ΔCt_treatment −ΔCt_control. The results were presented as 2^−ΔΔCT.

The following primers were used for the PCR reactions: Mouse IFN-γ, sense 5′-CGGCACAGTCATTGAAAGCCTA-3′ and antisense 5′-GTTGCTGATGGCCTGATTGTC-3′; IL-4, sense 5′-TCTCGAATGTACCAGGAGCCATATC-3′ and antisense 5′-AGCACCTTGGAAGCCCTACAGA-3′; IL-13, sense 5′-CAGCTCCCTGGTTCTCTCAC-3′ and antisense 5′-CCACACTCCATACCATGCTG-3′; TNF-α, sense 5′-CCCTCACACTCAGATCATCTTCT-3′ and anti-sense 5′-GCTACGACGTGGGCTACAG-3′; and β-actin, sense 5′-CATCCGTAAAGACCTCTATGCCAAC-3′ and anti-sense 5′-ATGGAGCCACCGATCCACA-3′. The primers were designed to surround at least one intron in order to distinguish between the products derived from mRNA and genomic DNA.

Dorsal skin tissue samples were harvested 24 h following final DNCB administration on day 35, and fixed in 10% buffered-neutral formaldehyde (Sigma-Aldrich) and embedded in paraffin wax (Leica Microsystems, Wetzlar, Germany). Histological skin tissue sections (6 µm) were stained with hematoxylin and eosin (Sigma-Aldrich), in order to count the number of eosinophils. The sections were also stained with 0.5% toluidine blue (Sigma-Aldrich) in order to determine the number of mast cells. The cell counts were performed using microscope (Axio Imager 2; Carl Zeiss AG, Oberkochen, Germany) in six consecutive microscopic fields at ×400 magnification.

Statistical analysis was performed using SAS software (version 9.3; SAS Institute Inc., Cary, NC, USA) and all results are expressed as the mean ± standard deviation. The biological and metabolic effects of CPE, SOE, CPE + SOE, HC (positive control), and vehicle (negative control) were compared using one-way analysis of variance. Significant differences in the treatment effects among the groups were identified using Tukey's test. The significance of differences between the mice with and without atopic dermatitis-like skin lesion were determined using a two-sample Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

CPE was predominantly composed of alkaloids, particularly berberine, and low levels of phenols and flavonoids. However, SOE was predominantly composed of phenols and flavonoids (Table I).

Following repeated topical applications of DNCB on the backs of the NC/Nga mice atopic dermatitis lesions, the dorsal skin of the mice in the control group that developed DNCB-induced atopic dermatitis exhibited hypertrophy, hyperkeratosis, intercellular edema and liquefaction degeneration of the basal layer (data not shown). Clinical severity was calculated from the sums of the scores for erythema and hemorrhage, edema, erosion (excoriation) and scaling (dryness). The clinical severity of the skin lesions in the control group increased markedly until three weeks post-topical application of DNCB, however, no further exacerbated was observed at four weeks (Fig. 1). Hypertrophy, hyperkeratosis, intercellular edema and liquefaction degeneration of the basal layer in the dorsal skin tissue samples were alleviated by treatments with CP and SOE to a similar extent, and were alleviated more by SOE+CPE, compared with the control group (data not shown). Treatment with CPE and SOE decreased the severity of atopic dermatitis symptoms after four weeks. In addition, treatment with CPE and SOE synergistically slowed symptom progression and exhibited similar activity levels to treatment with HC over time (Fig. 1).

Following topical application of DNCB onto the dorsal skin of the mice, the serum expression levels of IgE, an activator of mast cells, increased by ~7 and 15-fold at 2 and 5 weeks, compared with the control. Compared with the control, the increase in serum levels of IgE were suppressed by treatment with SOE CPE, and suppressed further by treatment with SOE and CPE combined, compared with SOE and CPE alone, and this suppression level was similar to that induced by treatment with HC (Fig. 2A). Serum levels of IgG1 increased in the control group by 3.3-fold after two weeks, and these expression levels decreased to 2.3-fold after five weeks (Fig. 2B). Treatment with SOE, CPE, and CPE and SOE combined decreased the serum levels of IgG1, and the decrease in expression levels induced by treatment with CPE and SOE combined were similar to those observed following treatment with HC after 2 and 5 weeks (Fig. 2B). In addition, treatment with CPE and SOE combined, and treatment with HC decreased the serum expression levels of IgG1 to similar levels as the non-induction group.

Mast cells secrete TNF-α and activate Th2-derived cytokines, including IL-4, IL-6 and IL-13. The serum levels of TNF-α in the control group increased significantly by 2.1-fold, compared with those of the non-induction group (Fig. 2C). The levels of TNF-α expression levels were suppressed by treatment with CPE, however treatment with SOE and CPE combined inhibited them to similar levels as those induced following treatment with HC. In addition, the serum levels of IL-4 and IFN-γ were higher in the control group, compared with the non-induction group; IL-4 and IFN-γ expression was suppressed by CPE, and this suppression was markedly increased by treatment with SOE and CPE combined, which reached similar expression suppression as that induced by treatment with HC (Fig. 2C). SOE also inhibited the serum expression levels of INF-γ, as compared with the control group, but did not inhibit the expression of IL-4. These results suggest that both Th2 and Th1 were activated during the experiment.

As determined using histological staining, skin thickness was markedly higher in the control group, compared with the non-induction group. Treatment with CPE or SOE alone decreased skin thickness, however, treatment with SOE and CPE combined markedly decreased skin thickness to similar levels as treatment with HC, as compared with treatment with SOE or CPE alone (Fig. 3A).

As seen in Fig. 3B, the infiltration of inflammatory cells, including mast cells and eosinophils was greater in the control group than in the normal control group that did not exhibit atopic dermatitis. The number of mast cells and eosinophils markedly increased in the skin lesions of the control group, compared with the non-induction group (Fig. 3C). SOE and CPE alone decreased the number of eosinophils in the DNCB-treated mice, and treatment with SOE+CPE further decreased the number of eosinophils, which was similar to that observed following treatment with HC. However, the number of mast cells was decreased by CPE, but not SOE, compared with the control group, whereas treatment with CPE and SOE synergistically inhibited the number of mast cells, and this decrease was marginally greater compared with treatment with HC (Fig. 3C).

To investigate cytokine production in the dorsal skin lesions, the mRNA expression levels of the TNF-α, IL-4, IL-6, IL-13 and INF-γ cytokines were quantified. The dorsal skin of the DNCB-treated control mice exhibited significantly higher expression levels of TNF-α, IL-4, IL-6, IL-13, and IFN-γ, compared with those of the non-induction group (Fig. 4A and 4B). CPE, but not SOE, suppressed the mRNA expression of TNF-α, IL-4, IL-6, IL-13 and IFN-γ, and treatment with CPE+SOE further decreased their expression levels (Fig. 4A and 4B). The suppression of TNF-α, IL-4, IL-6, IL-13 and IFN-γ following treatment with CPE and SOE combined was marginally more effective than treatment with the HC a positive control (Fig. 4A and 4B).