Eczematous plaques and intense pruritus are the main characteristics of atopic dermatitis (AD), a chronic inflammatory skin disease (Davidson et al., 2019). Up to 20% of children and infants experience AD (Leung and Guttman-Yassky, 2017). Immunoglobulin E (IgE) levels are elevated in serum and T_H2 cytokines (IL-4 and IL-13), contributing to the pathogenesis of AD (Vestergaard et al., 1999; Leung and Soter, 2001; Davidson et al., 2019). Although T_H2 helper cells are crucial in the acute phase of AD, T_H1/T_H17/T_H22 helper cells play fundamental roles in the chronic phase of AD pathogenesis (Spergel et al., 1999; Chen et al., 2004; Koga et al., 2008; Nograles et al., 2009; Souwer et al., 2010). Although anti-inflammatory steroidal corticosteroids are helpful and widely used in children, several symptoms of atrophy, perioral dermatitis, striae, rosacea, and purpura have been reported as the common adverse effects of topical corticosteroids (Hengge et al., 2006; Blume-Peytavi and Wahn, 2011). Therefore, strategies to develop novel drugs for the treatment of AD are required.

Many studies have reported the effectiveness of traditional Chinese medicines and their active components in treating AD (Chen et al., 2020; Min et al., 2022; Zhang et al., 2022). Interestingly, suppression of AD by traditional Chinese medicines has been reported to be mediated through the inhibition of mitogen-activated protein kinases (MAPKs), such as, ERK, JNK, and p38 MAPK, and NF-κB components, such as p65 (Chen et al., 2020; Min et al., 2022; Zhang et al., 2022). In fact, overexpression of MAPK mRNA in the skin tissue was observed in the AD group (Chen et al., 2020; Min et al., 2022; Zhang et al., 2022). However, the role of each MAPK in AD pathogenesis has not yet been studied. Moreover, the p38 MAPK signaling pathway plays a key role in inflammation and other physiological processes (Saklatvala, 2004). Therefore, we hypothesized that p38 MAPK inhibitors might have a promising therapeutic efficacy in AD, which has still not been explored.

There are four p38 MAPKs: α, β, γ, and δ (Ono and Han, 2000). Both α and β have 75% homology, and γ and δ possess more distant isoforms (Ono and Han, 2000). p38α MAPK is highly expressed in macrophages, monocytes, neutrophils, CD4⁺ T cells, and endothelial cells, while p38β expression is restricted in endothelial cells (Hale et al., 1999). Among the four isoforms of p38 MAPK, p38α, which has been extensively studied, plays an important role in inflammatory responses (Kumar et al., 2003). Therefore, targeting p38α MAPK is a reasonable approach. Previously, we developed a highly specific inhibitor of p38α/β MAPKs, namely NJK14047 (Heo et al., 2015; Choi et al., 2016; Kim et al., 2017; Gee et al., 2018, 2020).

Here, we investigated whether NJK14047 administration ameliorates the development of AD-like skin lesions in mice epicutaneously exposed to 1-chloro-2,4-dinitrobenzene (CDNB). CDNB is an electrophilic benzene derivative that induces AD-like skin lesions in BALB/c mice (Ku et al., 2018; Song et al., 2019; Bai et al., 2020). A parallel study using dexamethasone (DEX), a commonly used anti-inflammatory steroid for skin diseases was conducted to compare its efficacy with that of NJK14047.

NJK14047 was synthesized using a previously reported procedure (>97%, HPLC) (Heo et al., 2015; Gee et al., 2020). 1-Chloro-2,4-dinitrobenzene was purchased from Sigma-Aldrich (St. Louis, MO, USA).

Animals and treatment

Seven-week-old male BALB/c mice were purchased from DBL (Seoul, Korea) and housed in the laboratory animal facility at Kyung Hee University under conditions of 20-24°C with a 12-h light-dark cycle. Mice were provided with standard laboratory chow and water ad libitum (Huang et al., 2018). The Institutional Animal Care and Use Committee of Kyung Hee University reviewed and approved the experimental protocol for mice (KHSASP–21-548). After 2 weeks of acclimation, the mice were divided into four groups (n =5 per group): (1) vehicle control, (2) CDNB, (3) CDNB+2.5 mg/kg NJK14047, and (4) CDNB+DEX (Kang and Im, 2020). The mice were sensitized by applying CDNB (1%, 150 µL) to the dorsal skin (day 0). Mice were challenged by applying CDNB (0.3%, 200 μL) to both ears (10 µL each) every alternate day (days 7-48). NJK14047 (2.5 mg/kg body weight) or DEX (DEX, 10 mg/kg) was administrated 30 min prior to challenge on days 19-47 via intraperitoneal injection (Kang and Im, 2020). On day 49 all mice were sacrificed (Fig. 1A). DEX was used as a positive control.

Figure 1. Experimental protocol and macroscopic views. (A) Mice were divided into four groups: (1) vehicle control, (2) CDNB+vehicle, (3) CDNB+2.5 mg/kg NJK14047, (4) CDNB+DEX. To induce AD-like symptoms, CDNB was topically applied on the dorsal skin of BALB/c mice. CDNB (1%) was applied on day 0. After seven days, 0.3% CDNB was challenged on the ears during alternative days for the experiment (day 7-48). NJK14047 or DEX was administered on alternate days, starting from day 19 for 3 weeks (day 19-47). Mice were sacrificed on day 49. (B) Photographed images of ears from each group were taken on day 49.

Histological examination

To evaluate epidermal thickening, the ear skin of each mouse was prepared on day 49. The ears were fixed with neutral-buffered formalin (10%), dehydrated in sucrose solution (30%), and embedded in the O.C.T. compound. The sections were stained with toluidine blue O (TBO), hematoxylin, and eosin (H&E). Then, the thickness of the ears were measured based on the images of the H&E-stained sections.

Measurement of IgE and cytokines levels

Serum was collected on day 49 and stored at –80°C until use. We measured the serum IgE levels by using an ELISA kit (eBioscience, San Diego, CA, USA). The levels of T_H2 cytokines (IL-13, IL-4, and TARC), T_H1 cytokine (IL-12A, IFN-γ, and TNF-α), T_H17 cytokine (IL-17A), T_H22 cytokine (IL-22), IL-33, and TSLP in the ears or lymph nodes were also measured using reverse transcriptase polymerase chain reaction (RT-PCR). The synthesis of first-strand cDNA, specific primers, and PCR conditions have been described previously (Park and Im, 2019). Aliquots (7 μL) were electrophoresed on 1.2% agarose gels and stained with StaySafe™ Nucleic Acid Gel Stain (Real Biotech Corporation, Taipei, Taiwan). The intensity of each PCR product was quantified using ImageJ software (NIH, Bethesda, MD, USA) and normalized to GAPDH levels (Lee et al., 2017).

Flow cytometry

To determine the cell populations of CD4⁺Foxp3⁺ T cells and CD4⁺T-bet⁺ T cells in lymph nodes, we isolated single cells from lymph nodes. The cells were stained at 4°C for 15 min with a FITC-labeled rat anti-CD4 (cat. 11-0041-82, eBioscience), which is a surface marker. The cells were fixed at room temperature for 1 h using an intracellular fixation buffer (cat. 00-8222-49, eBioscience). After fixation, the cells were permeabilized using a permeabilization buffer (cat. 88-8824-00, eBioscience) and stained at room temperature for 1 h with APC-labeled rat anti-Foxp3 (cat. 17-5773-82, eBioscience), or with labeled rat anti-T-bet (cat. 50-5825-82, eBioscience).

Statistics

Statistical analysis was performed using the GraphPad Prism software (GraphPad Software, Inc., La Jolla, CA, USA). Data are expressed as means ± standard error of the mean (SEM). One-way analysis of variance (ANOVA) followed by Tukey’s multiple comparison test was used to compare the differences among multiple groups. Differences were considered statistically significant at p<0.05.

NJK14047 ameliorated CDNB-induced AD-like symptoms in BALB/c mice

To investigate the roles of p38 MAPKs in the pathogenesis of AD and determine the therapeutic potential of p38 MAPK inhibitor in AD, we administrated NJK14047 to BALB/c mice, which were sensitized and challenged with CDNB (Fig. 1A). Seven days after the sensitization of CDNB on the dorsal skin, we treated the mice with CDNB on both ears on every alternative day. NJK14047 (2.5 mg/kg) or DEX (10 mg/kg) was administered by intraperitoneal injection 30 min prior to CDNB challenge from day 19 to evaluate its therapeutic potential for AD-like symptoms (Fig. 1A). On day 49, the ears of CDNB-treated BALB/c mice showed severe cracking, dryness, erosion, erythema, and skin lesions (Fig. 1B). Administration of NJK14047 or DEX, decreased severity of the AD-like symptoms (Fig. 1B). H&E staining revealed CDNB-induced increased infiltration of immune cells compared to that in the control group (Fig. 2A). In addition, the epidermis was apparently thickened in the CDNB-treated group upon visual inspection owing to hyperkeratosis, where the ears showed epidermal hypertrophy (Fig. 2A). In contrast, the ears from NJK14047- or DEX-treated mice exhibited slightly severe epidermal hypertrophy than those from CDNB-treated mice (Fig. 2A). Thus, it was concluded that treatment with NJK14047 or DEX markedly suppressed the CDNB-induced epidermal hypertrophy (Fig. 2B). Moreover, the efficacy of NJK14047 (2.5 mg/kg) was similar to that of DEX (10 mg/kg), a well-known anti-inflammatory steroid for skin diseases (Fig. 2).

Figure 2. Effect of NJK14047 on CDNB-induced AD-like lesions in mice. (A) H&E-stained ear section on day 49 (400×). (B) Ear thickness was measured from the sections (n=5). Data represent the mean ± SEM (n = 5). ***p<0.001 vs. the control group, ^###p<0.001 vs. the CDNB-treated group.

NJK14047 decreased CDNB-induced infiltration of mast cells into the skin lesions

Mast cells infiltrate the epidermis and regulate the recruitment and function of many cell types involved in the pathogenesis of AD (Liu et al., 2011). Therefore, we investigated the effects of NJK14047 on mast cell infiltration in skin lesions. Through toluidine blue staining of collected tissue sections, we found a significant increase in mast cell numbers in skin lesions in samples obtained from CDNB-treated mice compared to those in the control group (Fig. 3A). Administration of NJK14047 or DEX considerably reduced the number of infiltrated mast cells in the ears compared to that in CDNB-treated mice (Fig. 3). The efficacy of NJK14047 (2.5 mg/kg) was similar to that of DEX (10 mg/kg) (Fig. 3).

Figure 3. Effect of NJK14047 on CDNB-induced mast cell accumulation in skin lesions of mice. (A) Representative images depicting the histological features of skin lesions collected on day 49 are shown. TBO staining was used to identify mast cells. Red arrows indicate mast cells. Cells were counted under a microscope at 400× magnification. (B) Numbers of mast cells. The numbers of mast cells in the skin lesions was calculated. Data represent the mean ± SEM (n = 5). ***p<0.001 vs. the control group, ^###p<0.001 vs. the CDNB-treated group.

Effects of NJK14047 on the CDNB-induced increased serum IgE level and T_H2 cytokine levels

Serum IgE levels are usually increased in AD-like animal models. Therefore, we measured the serum IgE levels in the experimental animal of BALB/c mice. CDNB treatments significantly increased serum IgE levels compared to those in control mice (Fig. 4A), while administration of DEX significantly decreased the serum IgE levels (Fig. 4A). However, administration of NJK14047 did not reduce the level of serum IgE considerably (Fig. 4A), thus validating the presence of different modes of action of NJK14047 and DEX.

Figure 4. Effect of NJK14047 on CDNB-induced increased serum IgE and TH2 cytokine levels in the ears. (A) Serum was collected on day 49. Serum IgE level was determined using ELISA. Data are presented as means ± SEM (n = 5). ***p<0.001 vs. the control group, ^##p<0.01 vs. the CDNB-treated group. (B-F) RT-PCR analyses for TH2 cytokines IL-13 (B), IL-4 (C), and a chemokine TARC(CCL17) (D). RT-PCR analyses for IL-33 (E) and TSLP (F). Results are presented as means ± SEM (n=5). **p<0.01, ***p<0.001 vs. the control group, ^#p<0.05, ^##p<0.01, ^###p<0.001 vs. the CDNB-treated group.

T_H2 cells play a major roles in AD. Thus, the mRNA levels of the T_H2 cytokines IL-13 and IL-4 and chemokine TARC (CCL17) in the ears were determined. The mRNA levels of the T_H2 cytokine IL-13 were significantly increased in the skin lesions of CDNB-treated mice compared to those in the control mice (Fig. 4B). Levels of IL-4 and TARC in the skin lesions also increased, but the difference was not statistically significant (Fig. 4C, 4D). Thus, administration of NJK14047 or DEX significantly decreased IL-13 levels in the ear skin of CDNB-treated BALB/c mice (Fig. 4B).

IL-33, an inflammatory cytokine, is up-regulated in the keratinocytes of AD patients and its expression in the skin causes AD-like eczema in mice (Imai, 2019). Thymic stromal lymphopoietin (TSLP) expression is induced by IL-13 in the skin and its expression is increased in inflammatory AD-like skin lesions (Oh et al., 2011). Therefore, the mRNA levels of the IL-33 and TSLP in the ears were determined. The mRNA levels of the IL-33 and TSLP were significantly increased in the skin lesions of CDNB-treated mice compared to those in the control mice (Fig. 4E, 4F). CDNB-induced increases of the levels of IL-33 and TSLP were suppressed by administration of NJK14047 or DEX significantly (Fig. 4E, 4F).

NJK14047 suppressed T_H1 cytokines in BALB/c mice

As T_H1/T_H17/T_H22 cells are anticipated to be involved in the late stages of AD in addition to T_H2 cells (Koga et al., 2008; Nograles et al., 2008; Souwer et al., 2010; Kim et al., 2014; Muraro et al., 2016), we measured the mRNA levels of the T_H1 cytokine IFN-γ, TNF-α, and IL-12A, the T_H17 cytokine IL-17A, and the T_H22 cytokine IL-22 in the ears. Here, we found that the mRNA levels of T_H1 cytokines TNF-α and IL-12A were significantly increased in the skin lesions of CDNB-treated mice compared to those in control mice (Fig. 5A, 5B). However, administration of NJK14047 or DEX considerably decreased IL-12A and IFN-γ levels in the ear skin of CDNB-treated BALB/c mice (Fig. 5A, 5B). Moreover, administration of NJK14047 significantly decreased the level of TNF-α, while that of DEX did not (Fig. 5C), again confirming the different modes of action of NJK14047 and DEX. The mRNA levels of T_H17 cytokine IL-17A and T_H22 cytokine IL-22 were not significantly increased in skin lesions from CDNB-treated mice compared to those in control mice (Fig. 5D, 5E).

Figure 5. Effect of NJK14047 on CDNB-induced increased TH1, TH17, and TH22 cytokine levels in the ears. RT-PCR analyses for TH1 cytokines IL-12A (A), IFN-γ (B), and TNF-α (C), TH17 cytokine IL-17A (D), and TH22 cytokine IL-22 (E). Results are presented as means ± SEM (n=5). *p<0.05, **p<0.01 vs. the control group, ^#p<0.05, ^##p<0.01 vs. the CDNB-treated group.

NJK14047 decreased CDNB-induced AD-like responses in lymph nodes

We found enlarged cervical lymph nodes in CDNB-treated mice compared to those in control mice (Fig. 6). The growth in lymph node weight after CDNB was 1370% (Fig. 6B). However, treatment with NJK14047 or DEX significantly decreased the size of lymph nodes in CDNB-treated mice (Fig. 6). The reduction in the NJK14047-induced lymph node weight was 46% and that for the DEX-induced was 94% (Fig. 6B).

Figure 6. Effect of NJK14047 on the size of lymph nodes in CDNB-induced AD-like responses in mice. (A) Images of lymph nodes. (B) Weights of lymph nodes. ***p<0.001 vs. the control group, ^##p<0.01, ^###p<0.001 vs. the CDNB-treated group.

Effects of NJK14047 on the CDNB-induced increased T_H2/T_H1/T_H17 cytokine levels in lymph nodes

Because T_H2 cells as well as T_H1/T_H17 cells can affect pathogenesis of AD, we measured the mRNA levels of the T_H2 cytokines IL-13 and IL-4 and the T_H1 cytokines IFN-γ, TNF-α, and T_H17 cytokine IL-17A in lymph nodes. The mRNA levels of the T_H2 cytokine IL-4 and IL-13 were significantly increased in the lymph nodes of CDNB-treated mice compared to those in the control mice (Fig. 7A, 7B). The CDNB-induced increases of levels of IL-4 and IL-13 were suppressed by NJK14047, but the difference was not statistically significant (Fig. 7A, 7B). The mRNA levels of T_H1 cytokines IFN-γ and TNF-α were also significantly increased in lymph nodes of CDNB-treated mice compared to those in control mice (Fig. 7C, 7D). Administration of NJK14047 or DEX decreased IFN-γ and TNF-α levels in lymph nodes of CDNB-treated BALB/c mice (Fig. 7C, 7D). Moreover, The mRNA levels of T_H17 cytokine IL-17A were also significantly increased in lymph nodes of CDNB-treated mice compared to those in control mice and administration of NJK14047 significantly decreased the level of IL-17A (Fig. 7E). However, DEX administration could not significantly suppress the increased IL-17A levels (Fig. 7E), again confirming the different modes of action of NJK14047 and DEX.

Figure 7. Effect of NJK14047 on CDNB-induced increased TH2/TH1/TH17 cytokine levels in lymph nodes. RT-PCR analyses for TH2 cytokines IL-4 (A) and IL-13 (B), TH1 cytokines TNF-α (C) and IFN-γ (D), and TH17 cytokine IL-17A (E). Results are presented as means ± SEM (n=5). *p<0.05, **p<0.01 vs. the control group, ^##p<0.01 vs. the CDNB-treated group.

Many studies have reported novel natural products from traditional Chinese medicine for AD treatment due to the incidence of severe adverse effects caused by long-term use of corticosteroids (Buyanravjikh et al., 2018). As AD pathogenesis shares common features with allergic diseases, such as pro-inflammatory immune responses (Akdis et al., 2006; Bantz et al., 2014), various anti-inflammatory immune regulators from the natural extracts of medicinal plants have shown therapeutic potentials in AD (Ku et al., 2018; Song et al., 2019; Bai et al., 2020). Notably, from the previous studies, we understand that a common mode of action of natural products exists, i.e., by suppression of MAPKs and NF-κB (Chen et al., 2020; Min et al., 2022; Zhang et al., 2022). Thus, by using NJK14047, a potent and specific p38 MAPK inhibitor, for the first time, we recognized the importance of p38 MAPK inhibitors to develop novel anti-AD drugs because p38 MAPK inhibition distinctly leads to the suppression of the immune responses of AD.

Compared to DEX, there are several common features, 1) improved macroscopic appearance of the skin, including thickness, 2) reduced mast cells accumulation in the skin, 3) reduced T_H1 and T_H2 cytokine levels, and 4) reduced sizes of lymph nodes. Although the dose of NJK14047 (2.5 mg/kg) was one-quarter of the DEX dose (10 mg/kg), the efficacy of NJK14047 was similar to that of DEX, implying that NJK14047 is more potent than DEX. There were also differences observed between the NJK14047 and DEX groups. DEX suppressed the increased levels of serum IgE but not NJK14047. The efficacy of NJK14047 in suppressing lymph node enlargement was lower than that of DEX, suggesting a unique mode of action.

There have been many studies on the effects of specific p38 MAPK inhibitors on the production of major inflammatory cytokines in inflammatory disease models (Saklatvala, 2004). In particular, attempts to apply p38 MAPK inhibitors to allergic asthma, a chronic airway disorder, needs consideration as it shares common characteristics with AD, such as T_H2-mediated inflammatory responses. For examples, SB239063 (12 mg/kg), a p38 MAPK inhibitor, markedly reduced inflammatory cytokine production and eosinophil infiltration in bronchoalveolar lavage fluid of ovalbumin-sensitized mice (Underwood et al., 2000). Another p38 MAPK inhibitor, SB203580 (10-100 mg/kg), inhibited TNF-α production in bronchoalveolar lavage fluids by approximately 95%, but did not inhibit eosinophil recruitment in ovalbumin-sensitized rats (Escott et al., 2000). Moreover, SD282 (30 and 90 mg/kg), which is also a p38 MAPK inhibitor, inhibited bronchial hyperresponsiveness but had no effect on serum IgE levels and on the recruitment of eosinophils in a chronic model exposed to 15 ovalbumin (Nath et al., 2006). Local airway delivery of antisense oligonucleotides to p38α MAPK inhibits ovalbumin-induced airway hyperresponsiveness, mucus hypersecretion, and pulmonary eosinophilia in a mouse model of asthma (Duan et al., 2005). Therefore, common results in the asthma models using these p38 MAPK inhibitors are seen in the present AD model with NJK14047; 1) suppression of T_H2 and T_H1 cytokines production, 2) suppression of immune cell infiltration, eosinophils in asthma, and mast cells in AD, and 3) no suppression of serum IgE levels.

As T_H2 and T_H1 cytokines orchestrate immune responses in disease lesions, the suppression of cytokine productions might be the key mode of action of p38 MAPK inhibitors. This suppression by p38 MAPK inhibitors has been studied in various cell types. Two p38 MAPK inhibitors, SB 203580 and SB 220025, have been shown to inhibit the in vitro production of TNF-α and IL-1β (Lee et al., 1994). Furthermore, p38 MAPK signaling pathway has been shown to be involved in cytokine production in a variety of cell types, such as T, B, and mast cells (Kawakami and Galli, 2002; Nel, 2002; Niiro and Clark, 2002). In particular, activation of p38 kinase in T_H2 cells contributes to an increase in the expression of T_H2 cytokine genes and differentiation of T_H2 cell (Schafer et al., 1999; Chen et al., 2000; Skapenko et al., 2001; Lu et al., 2004). Furthermore, p38 MAPK-mediated phosphorylation of GATA-3 and subsequent nuclear translocation of GATA-3 were found to be the optimum mechanisms of action in human T cells (Maneechotesuwan et al., 2007). p38 MAPK inhibition by SB203580 inhibits the synthesis of T_H2 cytokines (IL-4 and IL-13) from activated human T cells (Koprak et al., 1999; Mori et al., 1999). In addition, the p38 MAPK signaling pathway has been implicated in cytokine production by mast cells (Kalesnikoff et al., 2001).

In summary, previous studies have suggested the important roles of p38 MAPK in cytokine production in T and mast cells, where the anti-inflammatory effects of p38 MAPK inhibitors have been proven in allergic asthma models. In the present study, for the first time, we experimentally demonstrated the efficacy of NJK14047, a p38 MAPK inhibitor, on a CDNB-induced AD model; the suppression of cytokine productions in immune cells might be an optimal mechanism of action of NJK14047, suggesting that using p38 MAPK inhibitors could be a promising strategy for developing anti-AD drugs.

This research was supported by the Basic Research Laboratory Program (BRL) and the Basic Science Research Program of the Korean National Research Foundation funded by the Korean Ministry of Science, ICT, and Future Planning (NRF-2020R1A4A1016142 and NRF-2019R1A2C1005523).

The authors declare that there is no conflict of interest.