Inhibitory Effects of Urtica thunbergiana Ethanol Extract on Atopic Dermatitis-Induced NC/Nga Mice

Atopic dermatitis (AD) is a chronic, inflammatory skin disease that persists or repeatedly recurs in both childhood and adulthood. Urtica thunbergiana (UT) is an aroma herb with little-known pharmacological effects and anti-inflammatory activities against AD. This study investigated the immunomodulatory efficacy of 50% ethanol-extracted UT in necrosis factor-alpha/interferon-gamma (TNF-α/IFN-γ)-stimulated HaCaT cells in vitro and AD-Biostir-induced NC/Nga mice in vivo. The results showed that UT exhibits a dose-dependent increase in scavenged free radicals, reaching 76.0% ± 1.4% of scavenged 1,1-diphenyl-2-picrylhydrazyl at a concentration of 250 µg/mL. In addition, UT significantly downregulated the mRNA expression of the following pro-inflammatory cytokines and chemokines in TNF-α/IFN-γ-stimulated HaCaT cells: interleukin (IL)-6, IL-8, thymus- and activation-regulated chemokine, macrophage-derived chemokine, and regulated on activation normal T expressed and secreted. UT-treated HaCaT cells showed inhibition of the overexpression of chemokine-regulated signaling molecules, such as nuclear factor-kappa B, inhibitor of kappa B (IκBα), signal transducer and activator of transcription 1, and mitogen-activated protein kinases (MAPKs). UT dietary administration in AD-Biostir-induced NC/Nga mice treated and improved AD-like symptoms, such as scales, epidermal thickening, the dermatitis severity score, high trans-epidermal water loss, reduced skin hydration, increased mast cells, elevated serum immunoglobulin E levels, and an enlarged spleen. UT treatment inhibited the expression of phosphorylated forms of MAPKs, nuclear factor of activated T-cells 1, and regulator IκBα. It also upregulated filaggrin (FLG) production. Therefore, UT shows high anti-AD activity both in vitro and in vivo, and can be a useful anti-AD agent.


Introduction
Atopic dermatitis (AD) is a chronic, relapsing inflammatory disease of the skin, which is characterized by inflamed skin, pruritus, redness, and blisters with oozing and crusting through dry, rough skin. Intense pruritus leads to the behavior of scratching, which is an important symptom of AD associated with filaggrin (FLG)-regulated epidermis disruption [1]. AD onset is attributed to defective immune cells, including keratinocytes, monocytes, and Langerhans cells. These defective cells are hypersensitive to environmental agents, such as dust mites, food, weeds, and bacteria, leading to an overwhelming immune and inflammatory response. CC chemokines, such as thymus-and activation-regulated chemokine/CC chemokine ligand 17 (TARC/CCL17), macrophage-derived chemokine/CC chemokine ligand 22 (MDC/CCL22), regulated on activation normal T expressed, and secreted/CC chemokine Table 1. Pharmacological activities of Urtica species.

Pharmacological
Activity Animals/Cell Lines Inducers Reference PMID * or DOI **

Sample Preparation
Dried UT leaves were purchased from Luvama Nature Co., Ltd. (Gyeonggi-do, Korea), and 5 g of dried UT was extracted using 500 mL of 50% ethanol in a digital orbital shaker (SHO-1D; Daihan, Seongbuk-gu, Korea) for 24 h at 24-25 • C three times. After incubation, the extract was filtered and evaporated under vacuum at 40 • C, providing a yield of 14.6% (w/v). The active phenolic compounds of UT detected were the same as described by Hwang et al. [33].

Cell Culture, UT Treatment, and Stimulation
HaCaT cell lines originating from human keratinocytes were purchased from the Korea cell line bank (Seoul, Korea). HaCaTs were cultured in Dulbecco's Modified Eagle Medium (DMEM) with 10% heat-inactivated fetal bovine serum (FBS; Gibco-BRL, Grand Island, NY, USA). The cells were seeded at densities of 5.0 × 10 5 cells/mL and 1.0 × 10 5 cells/mL. Inflammatory responses were stimulated in HaCaT cells by treatment with 10 ng/mL of TNF-α/ IFN-γ (Sigma-Aldrich). Next, 1-100 µg/mL of the UT extract was supplemented to the cell culture. Samples were first incubated for 30 min, and the cells were subsequently induced with 10 ng/mL of TNF-α and IFN-γ for 30 min or 24 h.

Induction of AD-Like Skin Lesions and Topical Application
Six-week-old male NC/Nga mice (body weight of 21-26 g) were purchased from Central Lab Animals, Inc. (Seoul, Korea). The mice were randomly divided into five groups of five mice per cage in standardized conditions. They were adapted for 2 weeks before the study, during which time, one mouse died. The experimental protocol KHUASP(SE)-17-014 was approved by the Institutional Animal Care and Use Committee of Kyung Hee University, Korea.

Measurement of Secretion Proteins
MDC/CCL22 and TARC/CCL17 concentrations of cell supernatants were measured using commercially available enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems, Minneapolis, Antioxidants 2020, 9,197 5 of 16 MN, USA). In in vivo experiments, blood was drawn from NC/Nga mice and centrifuged at 14,000× g for 20 min at 4 • C. The supernatant's serum IgE levels were evaluated using a mouse IgE ELISA kit (BD Bioscience, San Jose, CA, USA), based on the manufacturer's instructions.

Evaluation of AD-Like Skin Symptoms
The relative AD severity was evaluated macroscopically on the basis of the following five symptoms: erythema, edema, erosion, dryness, and lichenification [49][50][51]. The total dermatitis severity score was defined as the sum of component scores (0, no symptoms; 1, mild; 2, moderate; 3, severe), ranging from 0 to 15. Dermatitis scoring was recorded by using a blind test during the experimental period.

Evaluation of Scratching Behavior
Scratching behavior was measured once a week for 3 weeks [52]. All groups were videotaped for 15 min per mouse using a digital camera placed on the top of the cages. One scratching bout was defined as a series of scratching movements by the hind paw.

Measurement of Physiological and Histological Skin Functions
Subcutaneous hydration, trans-epidermal water loss (TEWL), and the erythema index (EI) were measured using appropriate probes (DermaLab ® ; Combo, Cortex Technology, Denmark). The mice were euthanized, and the skin of mice was fixed in 4% paraformaldehyde for 24 h. Dorsal skin specimens were embedded in paraffin, 10-µm-thick slices were cut, and the slices were stained with hematoxylin and eosin (H&E) or toluidine blue (TB).

Western Blot Analysis
Cell lysates were prepared in lysis buffer. Protein concentrations were determined using Bradford reagent (Bio-Rad, Hercules, CA, USA) with bovine serum albumin (BSA) as the standard. Equal amounts of total protein were electrophoresed using SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to a nitrocellulose membrane (Amersham Pharmacia Biotech, Buckinghamshire, UK). Transfer membranes were blocked, and a primary antibody was added (Santa Cruz Biotechnologies, Santa Cruz, CA, USA) overnight. After incubation with a secondary antibody (Cell Signaling, Danvers, MA, USA), protein levels were determined using electrochemiluminescence (ECL) detection reagents (Fujifilm, LAS-4000, Tokyo, Japan) and ImageMaster TM 17 2D Elite software version 3.1 (Amersham Pharmacia Biotech, Piscataway, NJ, USA).

Statistical Analysis
Data were expressed as the mean ± standard deviation (SD). One-way analysis of variance (ANOVA) was used for a statistical comparison of different treatments. GraphPad Prism 5.0 (GraphPad Software Inc., San Diego, CA, USA) was used. P < 0.05 was considered statistically significant. # P < 0.05 was considered statistically significant compared to either basal cells or Biostir-untreated mice, while * P < 0.05 was considered statistically significant compared to either only TNF-α/IFN-γ-induced keratinocytes or only the AD-induced group.

Antioxidative Activity of UT
The free-radical inhibition activity of UT increased dose-dependently; the DPPH inhibition ratio of UT at a concentration of 250 µg/mL was 76.0% ± 1.4% ( Figure 1A).

Effects of UT on Cell Viability and TARC and MDC Production in TNF-α/IFN-γ-Stimulated HaCaT Cells
UT treatment induced no remarkable cytotoxicity compared to unstimulated conditions ( Figure 1B). Therefore, we used UT concentrations of 10 and 100 µg/mL in subsequent experiments. TNF-α/IFN-γ stimulation increased TARC and MDC production in HaCaT cells by 579.4% and 1193.0%, respectively, compared to unstimulated cells ( Figure 1C,D). However, a UT concentration of 100 µg/mL inhibited TARC overproduction by 57.7% and MDC overproduction by 68.7% compared to only TNF-α/IFN-γ-stimulated cells.

Effect of UT on AD Symptoms on Mouse Skin
We observed the typical symptoms of AD on NC/Nga mouse skin, including increased scratching behavior, followed by the rapid development of erythematous and erosive lesions with edema, resulting in skin lichenification ( Figure 4A). These skin lesions were treated by 0.1% tacrolimus and 1% UT ( Figure 4B). Compared to group 1, no noticeable changes in body weight were observed in the other groups ( Figure 4C). Furthermore, no abnormal symptoms were recorded in all groups, indicating that the tested samples ensure safety, without toxicity or adverse effects.

Effect of UT on AD Symptoms on Mouse Skin
We observed the typical symptoms of AD on NC/Nga mouse skin, including increased scratching behavior, followed by the rapid development of erythematous and erosive lesions with edema, resulting in skin lichenification ( Figure 4A). These skin lesions were treated by 0.1% tacrolimus and 1% UT ( Figure 4B). Compared to group 1, no noticeable changes in body weight were observed in the other groups ( Figure 4C). Furthermore, no abnormal symptoms were recorded in all groups, indicating that the tested samples ensure safety, without toxicity or adverse effects.
The dermatitis severity was determined on the basis of the sum of the score for each symptom once a week for 3 weeks. In AD-induced mice, skin symptoms such as erythema, edema, erosion, dryness, and lichenification were observed [50,51]. After 3 weeks, topical Biostir-AD-administered mice exhibited all typical symptoms of AD, with a maximum dermatitis score of 7.2 ± 2.2. The topical application of tacrolimus and UT significantly improved these symptoms ( Figure 4D). The most effective sample in decreasing AD symptoms on NC/Nga mouse skin was 1% UT compared to 0.1% tacrolimus and 0.1% UT.
However, the use of tacrolimus and UT decreased the spleen weight. In groups 4 and 5, the spleen weight significantly decreased by 28.4% and 31.8%, respectively, compared to group 2.
In group 2, the number of scratches was significantly increased by 25 times by the topical application of Biostir-AD, compared to group 1. The topical application of tacrolimus and UT for 3 weeks significantly decreased the number of scratches by 87.6% in group 3, 84.7% in group 4, and 76.6% in group 5, compared to group 2 ( Figure 4G).

Effect of UT on Histological and Biophysical Characteristics of AD-Induced Skin
Epidermal thickness significantly decreased in groups 3, 4, and 5 compared to group 2 ( Figure  5A,B). In group 5, epidermal thickness improved to 16.9 ± 5.2 μm, while in group 1, this was 12.7 ± 2.7 μm. Groups 3, 4, and 5 also showed a decrease in the elevated number of mast cells on the skin ( Figure 5C). AD can cause skin dryness and increased EI [53,54]. At the end of 3 weeks, subcutaneous hydration decreased by 87.2% in group 2 compared to group 1. In contrast, TEWL and EI increased by 278.1% and 962.2%, respectively ( Figure 5D,E). The topical application of tacrolimus and UT improved these physiological changes. Subcutaneous hydration significantly increased by >600%, while TEWL and EI decreased by 40.9% and 49.7% in group 5.
In group 2, serum IgE levels increased by 3320.5% compared to group 2 ( Figure 5F) and significantly decreased in groups 3, 4, and 5. Both groups 4 and 5 exhibited a reduction of serum IgE levels by 66.8% and 62.0%, respectively, compared to group 2 ( Figure 5G). A variety of responses in AD can influence the weight of immune organs, such as the spleen. In group 2, the spleen weight significantly increased by 202.7% compared to group 1 ( Figure 4E,F). However, the use of tacrolimus and UT decreased the spleen weight. In groups 4 and 5, the spleen weight significantly decreased by 28.4% and 31.8%, respectively, compared to group 2.
In group 2, the number of scratches was significantly increased by 25 times by the topical application of Biostir-AD, compared to group 1. The topical application of tacrolimus and UT for 3 weeks significantly decreased the number of scratches by 87.6% in group 3, 84.7% in group 4, and 76.6% in group 5, compared to group 2 ( Figure 4G).

Effect of UT on Histological and Biophysical Characteristics of AD-Induced Skin
Epidermal thickness significantly decreased in groups 3, 4, and 5 compared to group 2 ( Figure 5A,B). In group 5, epidermal thickness improved to 16.9 ± 5.2 µm, while in group 1, this was 12.7 ± 2.7 µm. Groups 3, 4, and 5 also showed a decrease in the elevated number of mast cells on the skin ( Figure 5C).
AD can cause skin dryness and increased EI [53,54]. At the end of 3 weeks, subcutaneous hydration decreased by 87.2% in group 2 compared to group 1. In contrast, TEWL and EI increased by 278.1% and 962.2%, respectively ( Figure 5D,E). The topical application of tacrolimus and UT improved these physiological changes. Subcutaneous hydration significantly increased by >600%, while TEWL and EI decreased by 40.9% and 49.7% in group 5.

Effect of UT on AD-Related Proteins of NC/Nga Mice Skin
FLG, a skin barrier function regulator, decreased in group 2 compared to group 1 ( Figure 6A-E). p-p38 and p-JNK expression increased in group 2 by 1277.4% and 163.8%, respectively, while IκBα expression significantly decreased by 98.3% compared to group 1. The topical application of tacrolimus and UT downregulated p-p38 and p-JNK expression and upregulated FLG and IκBα expression. In group 5, p-p38 and p-ERK expression decreased by 84.9% and 58.9%, respectively, while FLG and IκBα expression increased by 438.8% and 3359.4%, respectively, compared to group 2.
In group 2, serum IgE levels increased by 3320.5% compared to group 2 ( Figure 5F) and significantly decreased in groups 3, 4, and 5. Both groups 4 and 5 exhibited a reduction of serum IgE levels by 66.8% and 62.0%, respectively, compared to group 2.

Effect of UT on AD-Related Proteins of NC/Nga Mice Skin
FLG, a skin barrier function regulator, decreased in group 2 compared to group 1 ( Figure 6A-E). p-p38 and p-JNK expression increased in group 2 by 1277.4% and 163.8%, respectively, while IκBα expression significantly decreased by 98.3% compared to group 1. The topical application of tacrolimus and UT downregulated p-p38 and p-JNK expression and upregulated FLG and IκBα expression. In group 5, p-p38 and p-ERK expression decreased by 84.9% and 58.9%, respectively, while FLG and IκBα expression increased by 438.8% and 3359.4%, respectively, compared to group 2.

Discussion
UT is found in Korea, China, Japan, and Taiwan. Traditionally, UT is a perennial herb used to treat inflammatory-related skin diseases. The phenolic components of Urtica spp. include hydroxycinnamic acid derivatives, flavones and flavonols, C-and O,C-glycosides, and 3-hydroxy-3methylglutaroyl derivatives, which contribute to the antioxidant potential and anti-inflammatory activity of UT. This study clarified the inhibitory effects of UT with effective anti-AD activity in both TNF-α/IFN-γ-induced HaCaT cells and Biostir-induced NC/Nga mice, opening up new approaches for its further exploitation in the pharmaceutical industry.
Keratinocytes are the most abundant cell type in the epidermis, one of the foremost protection layers of the immune system. They are also an important source of activating dendritic cells to prime naive T helper cells producing their pro-inflammatory cytokines. Following mechanical stimulation, such as scratching or exposure to TNF-α and IFN-γ, keratinocytes of AD patients secrete a variety of cytokines and chemokines. In particular, keratinocyte-derived TARC/CCL17, MDC/CCL22, and RANTES/CCL5 play a major role in AD initiation. In addition, IL-6 is mainly found in human keratinocytes regulating T-cell maturation, chemokine production, antibody secretion, and B-cell and dendritic cell development. Besides IL-6, IL-8 is an important cytokine that activates chemotaxis in neutrophils and granulocytes, which then migrate toward the infection site. In this study, UT dosedependently inhibited both mRNA and secreted protein expression of pro-inflammatory cytokines IL-6 and IL-8 and chemokines TARC, MDC, and RANTES in TNF-α/IFN-γ-induced HaCaT cells. We believe that UT has potential for application as an anti-AD agent.

Discussion
UT is found in Korea, China, Japan, and Taiwan. Traditionally, UT is a perennial herb used to treat inflammatory-related skin diseases. The phenolic components of Urtica spp. include hydroxycinnamic acid derivatives, flavones and flavonols, C-and O,C-glycosides, and 3-hydroxy-3-methylglutaroyl derivatives, which contribute to the antioxidant potential and anti-inflammatory activity of UT. This study clarified the inhibitory effects of UT with effective anti-AD activity in both TNF-α/IFN-γ-induced HaCaT cells and Biostir-induced NC/Nga mice, opening up new approaches for its further exploitation in the pharmaceutical industry.
Keratinocytes are the most abundant cell type in the epidermis, one of the foremost protection layers of the immune system. They are also an important source of activating dendritic cells to prime naive T helper cells producing their pro-inflammatory cytokines. Following mechanical stimulation, such as scratching or exposure to TNF-α and IFN-γ, keratinocytes of AD patients secrete a variety of cytokines and chemokines. In particular, keratinocyte-derived TARC/CCL17, MDC/CCL22, and RANTES/CCL5 play a major role in AD initiation. In addition, IL-6 is mainly found in human keratinocytes regulating T-cell maturation, chemokine production, antibody secretion, and B-cell and dendritic cell development. Besides IL-6, IL-8 is an important cytokine that activates chemotaxis in neutrophils and granulocytes, which then migrate toward the infection site. In this study, UT dose-dependently inhibited both mRNA and secreted protein expression of pro-inflammatory cytokines IL-6 and IL-8 and chemokines TARC, MDC, and RANTES in TNF-α/IFN-γ-induced HaCaT cells. We believe that UT has potential for application as an anti-AD agent.
In this study, we investigated the early signaling pathways of AD-induced keratinocytes, including MAPK and NF-κB/STAT1. The MAPK family includes ERK, JNK, and p38, which control a variety of physiological processes. In keratinocytes, pro-inflammatory cytokines, such as TNF-α/IFN-γ, activate the intercellular MAPK signaling pathway, which activates NF-κB/STAT1, which, in turn, regulates the expression of AD-related pro-inflammatory cytokines and chemokines.
On the one hand, NF-κB is a well-known transcription factor that regulates many immune and inflammatory responses. In the normal state, NF-κB combines with IκB in the cytoplasm. Stimulation by agents such as TNF-α and IFN-γ activates the IKK complex, which, in turn, phosphorylates IκB, leading to the proteasomal degradation of IκB. The free NF-κB is translocated to the nucleus, where it activates target genes. On the other hand, STAT1 plays an essential role in the dysregulation of immune responses in AD, including exaggeration of the Th2 cell response, the maturation of B-cells, the suppression of regulatory T-cells, and the activation of eosinophils. STAT1 resides in the cytoplasm before phosphorylation and translocates into the nucleus in order to control the transcription of target genes, such as TARC, MDC, and RANTES. As expected, UT dramatically inhibits the TNF-α/IFN-γ-induced activation of MAPKs and NF-κB/STAT1 in AD-induced HaCaT cells. Lee et al. (2018) showed a similar trend of a reversed effect of mineral-balanced deep seawater (DSW) on STAT1 and JNK phosphorylation to UT, but DSW did not control NF-κB, IκBα, and ERK phosphorylation [54].
On the basis of in vitro data, in vivo studies were performed to investigate the effect of UT on AD-like skin lesions in Biostir-induced NC/Nga mice. The clinical, immunological, biochemical, and histological features of NC/Nga mice resemble those typical of skin lesions observed in AD patients. The NC/Nga mice model is useful for not only elucidating AD pathogenesis, but also evaluating new therapeutic agents. Therefore, we used this model in pharmacological in vivo studies.
Previous studies have shown that the continuous application of Biostir induces AD-like skin lesions in NC/Nga mice, which are characterized by an increase in serum IgE levels in the blood and the number of mast cells in the skin, clinical and histological changes, and skin barrier abnormalities. Elevated serum IgE levels and mast cell numbers are important features of AD development. Although the role of mast cells is not fully understood, a large number of AD patients (80-85%) tend to have greatly increased serum IgE levels. In this study, UT treatment significantly inhibited high serum IgE levels, as well as the number of mast cells. In addition, AD is characterized by clinical and histological changes, such as epidermal thickening, erythema, and skin dryness, which are commonly observed in AD patients. Increased TEWL and reduced skin hydration are important indexes in studies, which significantly decrease after the topical application of UT. Together, histological analysis of the skin showed that UT significantly reduces the epidermal thickness.
The skin plays an important role in protecting the body from external stresses, including chemical, physical, and biological stresses, and moisture loss. AD patients present with defects in the skin barrier, resulting in inflammatory characteristics caused by facilitated pathogenic infection and allergen infiltration. Therefore, this study assessed the effect of UT treatment on the protein expression of FLG, a major component of the stratum corneum (SC), and the NF-κB-inhibitor IκBα. We found that UT treatment recovers the FLG and IκBα expression downregulated by a TNF-α/IFN-γ mixture. In addition, UT treatment significantly regresses MAPK and NFATc1 phosphorylation in AD-induced skin.
In our previous study, the anti-aging activity of an ethanolic UT extract was considered, specifically that of hydroxycinnamic acids, including caffeic acid and chlorogenic acid [55]. We found that chlorogenic acid significantly enhanced the production of type I procollagen by regulating NFATc1 dephosphorylation. These phenolic bioactive compounds are well-known to possess high antioxidant activity and potential inhibitory effects against various conditions, such as microbial infections, inflammation, skin aging, cancer, obesity, hypertension, and neuron diseases [56][57][58]. In both in vitro and in vivo experiments, caffeic acid and chlorogenic acid suppressed UVB-induced skin aging and carcinogenesis through the MAPK/AP-1/NF-κB-regulated mechanism [32]. These promising compounds are present in a variety of herbs, found in green tea and coffee, readily accessible, cheap, and easy to isolate [59][60][61]. Both demonstrated significant biological activity, even at low concentrations (1-10 µg/mL) [32]. Our data suggests that these compounds could contribute to the anti-inflammatory role of UT.

Conclusions
This study provided some important data on UT as an alternative therapeutic for treating AD-like skin lesions. UT significantly downregulates the expression levels of MDC, TARC, RANTES, IL-6, and IL-8 in TNF-α/IFN-γ-stimulated HaCaT cells via controlling MAPKs/IκBα/STAT1 signaling. In in vivo experiments, the topical application of UT protects the skin by decreasing scratching behavior, the epidermal thickness, and EI and IgE production, as well as promoting SC hydration, which, in turn, can upregulate the expression of FLG and IκBα, and downregulate MAPK and NFATc1 phosphorylation in AD-induced skin. Therefore, UT is a potential candidate for AD therapy.