Manifestation of atopic dermatitis‐like skin in TNCB‐induced NC/Nga mice is ameliorated by topical treatment of substance P, possibly through blockade of allergic inflammation

Atopic dermatitis (AD) is a chronic inflammatory skin disorder characterized by intense pruritus and eczematous lesion. In this study, topically applied substance P (SP) significantly alleviated AD‐like clinical symptoms in 2, 4, 6‐trinitrochlorobenzene (TNCB)‐induced dermatitis in NC/Nga mice. This effect was nullified by pretreatment of the neurokinin‐1 receptor (NK‐1R) antagonist CP99994. SP treatment significantly reduced the infiltration of mast cells and CD3‐positive T cells as well as inflammatory cytokines, such as tumor necrosis factor‐α (TNF‐α) and thymic stromal lymphopoietin (TSLP), in AD‐like skin lesions and decreased the levels of IgE and thymus and activation‐regulated chemokine in serum. This SP‐induced alleviation of allergic inflammatory responses was also confirmed as reduced activation in the axillary lymph nodes (aLN) and spleen, suggesting the systemic effect of SP on immune responses in TNCB‐induced NC/Nga mice. Furthermore, SP‐mediated TSLP reduction was confirmed in human keratinocyte culture under pro‐inflammatory TNF‐α stimulation. Taken together, these results suggest that topically administered SP may have potential as a medication for atopic dermatitis.

extrinsic allergens and itching result in a vicious cycle of the worsening of the skin lesion, causing the allergic inflammatory disease to become chronic.
Thymic stromal lymphopoietin (TSLP) plays a critical role in the initiation of the inflammatory cascade of AD at the interface between keratinocytes and dendritic cells (DCs) and also mediates itching. [5] Accordingly, elevated TSLP in atopic skin activates immature DCs to produce chemokines, such as eotaxin-2, thymus and activationregulated chemokine (TARC), and macrophage-derived chemokine (MDC), which attract Th2 cells. These Th2 cells then migrate back to the affected skin and initiate allergic inflammation by activating B cells and triggering IgE production. [6] TSLP is mainly produced by epithelial cells and pro-inflammatory cytokines, such as TNFα and IL-1, upregulate its production. Mast cells are one of the essential sources of TSLP, as well as several pro-inflammatory cytokines including TNFα. These cytokines are released by IgE-activated mast cells and also upregulate TSLP expression. [7] Consistently, the expression of TSLP is highly increased in the skin of patients with AD, but absent in normal skin. [7][8][9][10][11][12][13] SP is an endogenous neuropeptide belonging to the tachykinin family and is known to mediate intracellular signalling, mainly through G-protein coupled receptor (GPCR)-linked NK-1R, in a variety of non-neuronal as well as neuronal tissues, including the peripheral and central nervous systems. [14][15][16][17] The role of SP in a variety of cell types and diseases is rather controversial. For example, at very high concentrations (10 −5 M), SP induces histamine release, which is actively implicated in neurogenic inflammation, but at low local concentrations (10 −11 -10 −6 M), SP alleviates AD progression by trophic and immune modulatory functions. [18,19] Recently, several reports have supported the anti-inflammatory and tissue repairing roles of SP in spinal cord injury, demonstrating increases in IL-10 and in the M2 type of macrophages and decreases in TNFα. [20] Furthermore, decreases in TNFα and IL-17 and increased regulatory T cells (Treg) have been demonstrated in a collagen II-induced arthritis animal model. [21] SP also improves MSC's inhibitory effect on T-cell activation by stimulating tumor growth factorβ (TGFβ) secretion in long-term ex vivo expanded MSC. [22] Furthermore, SP has been newly identified as an injuryinducible messenger mobilizing MSC from the bone marrow for recruitment to the injury site. [23,24] These newly identified physiological roles of SP led us to suggest that SP contributes to wound healing and to the regulation of chronic inflammation, rather than its classically characterized role as a pain-sensing neurotransmitter or mast cell activator.
NC/Nga mice provide a useful animal model for evaluating potential therapeutic drugs for human AD. [25] Moreover, repeated sensitization with TNCB can cause an apparent chronic dermatitis in NC/Nga mice which is similar to human AD. [26][27][28] In this study, using a NC/Nga AD-like dermatitis murine model induced by cutaneous sensitization with TNCB, we evaluated the effect of topical SP treatment in AD-like skin progression. We found that topical SP treatment reduces infiltration of mast cells and T cells in AD-like skin lesions, as well as reducing local inflammation in lymph nodes. These results suggest the efficacy of SP in the treatment of human diseases such as AD. These finding should be further evaluated in future clinical studies.

| Animals
Male 5-week-old NC/Nga mice were purchased from SLC (Shizuoka, Japan) and maintained under conventional circumstances. Animals were maintained under a 12-h light/12-h dark illumination cycle in an animal holding room and allowed to adapt to the new environment for a period of 1 week before the experiments were started.
All animals received standard chow and water ad libitum. All animal studies were performed in accordance with the Animal Experiment Guidelines of Kyung Hee University and were approved by the Animal Use Committee at Kyung Hee University (approval number; KHUASP[SU]-10-010).

| Induction of dermatitis
AD-like dermatitis was induced in NC/Nga mice as described by Takahashi et al [29] ( Figure 1A Evaluation of clinical skin severity scores (SCORAD). SCORAD was expressed as the summation of each score based on four symptoms: erythema, haemorrhage, scaling and dryness (no sign = 0, mild = 1, moderate = 2, severe = 3). Each value represents the mean ± SD. The significance of these differences was determined using a Dunnett's post hoc multiple comparison test following ANOVA and p< 0.05 was considered statistically significant. **P < .01 which correspond to clinically used doses. [21,24,30] During the drug treatment period, 0.1% TNCB was applied once a week to maintain chronic dermatitis. The mice were kept in the same environment without sensitization and were considered to be the normal (Nor) group (n = 5 for Nor).

| Evaluation of skin dermatitis severity
The severity of dermatitis in the dorsal skin was evaluated by the scoring atopic dermatitis (SCORAD) index at 0, 1, 2 and 4 weeks after drug application. [31] The clinical skin severity of dermatitis was scored as 0 (none), 1 (mild), 2 (moderate) and 3 (severe), according to the criteria of the four major clinical symptoms of AD: erythema/haemorrhage, oedema, excoriation/erosion and scaling/dryness. SCORAD was evaluated by three individuals in single-blind testing.

| Enzyme-linked immunosorbent assay (ELISA)
The levels of TARC in serum were measured using TARC ELISA kits (R&D Systems, Wiesbaden, Germany). All experiments were performed in triplicate with at least three independent experimental replicates.

| Analysis of lymph nodes
To evaluate immunological alterations in the entire body, axillary lymph nodes (aLNs) were collected when the mice were sacrificed at 2 weeks. Briefly, aLNs were isolated after incision of the arm skin.
Isolated tissues were measured by wet weight and normalized to body weight of each animal, and then, the specimens were fixed in 4% PFA for histological analysis.

| Statistical analysis
Results were expressed as mean ± standard deviation. Statistical analysis was performed by one-way analysis of variance (ANOVA) followed by a Dunnett's post hoc multiple comparison test using Graph Pad Prism 5.0. (GraphPad Software, Inc. La Jolla, CA, USA). P < .05 was accepted as significant. The thickness of the epidermis in the AD controls was approximately 88 μm, which was about fivefold thicker than the normal control (18.3 ± 4.3 μm), while SP treatment (42.8 ± 8.4 μm, P < .05) reduced epidermal thickness by greater than 50%, compared to the AD controls ( Fig. S3B-C). However, the measured thickness in the AD-VE and AD-SP-A groups was not significantly altered relative to the AD controls.

| SP treatment inhibits mast cell recruitment and degranulation in TNCB-induced AD-like lesions
As shown in Figure 1, topical treatment with SP inhibited the clinical features of AD in NC/Nga mice. To investigate the mechanisms of action of SP in the amelioration of AD-like symptoms, infiltration of mast cells and their degranulation in the skin lesions were examined by toluidine blue staining at 1, 2 and 4 weeks (Figure 2 and Fig. S4).
In normal skin, infiltrated mast cells were barely detected and were found to be predominantly in an inactive state (indicated by an arrow head), although a few active state degranulated cells (indicated by an arrow) were observed. Highly infiltrated mast cells, regardless of their degranulation, were observed in AD skin, which was significantly prevented in AD-SP skin from 1 week (Fig. S4). This SP-mediated pre-

| SP reduces chronic dermatitis in TNCB-induced NC/Nga mice
TSLP has a critical role in the initiation of the inflammatory cascade of AD, and TNFα stimulates the release of TSLP expression. [8] Therefore, we examined TNFα and TSLP expression by immunohisto-  (Fig. S7B). In particular, the number of CD3 cells infiltrating to the epidermis of the AD-SP group (4.5 ± 2.5 cells/field) was significantly reduced to the levels seen in normal skin (2.3 ± 1.1 cells/field), while AD, AD-VE and AD-SP-A showed levels approximately fivefold higher than the normal group (Fig. S7C). Topical treatment with SP reduced IgE levels, infiltration of mast cells and secretion of TNFα in vivo. Therefore, we carried out in vitro assays to examine the direct effects of SP in immune cells, such as RBL-2H3 cells (a rat-derived mast cell line) and U266 cells (a humanderived B-cell line). We performed mast cell migration assay using a transwell system and a β-hexosaminidase assay to examine degranulation of mast cell by SP treatment (Fig. S10). Moreover, we examined the cytotoxicity assay and determined the amount of IgE in response to various concentration of SP (Fig. S11). Collectively, SP did not stimulate RBL-2H3 migration and degranulation as well as U266 cell proliferation and production of IgE according to concentration of SP in vitro.

| DISCUSSION
In the present study, topical treatment with SP on the skin of TNCBinduced chronic dermatitis in NC/Nga mice, which is similar to human AD skin, markedly reduced the typical symptoms (Figure 1 and Fig.   S2). This alleviation of AD symptoms was accompanied by reduced allergic and cutaneous inflammation (Figures 2 and 3 and Fig. S3-S5). As one possible mechanism for the beneficial role of SP in the alleviation of AD symptoms in TNCB-sensitized NC/Nga mice, marked reductions in TNFα, TSLP and TARC were detected (Figure 3 and Fig. S5) and SP-mediated blockade of TNFα-induced TSLP production was shown in cultured human keratinocytes (Fig. S6). These immune modulatory effects of SP were also found in a secondary lymphoid organ, the aLNs adjacent to the atopic dorsal skin (Figure 4). In addition, the concentration of SP used for topical treatment in this study reduced the symptoms of AD, specifically the thickness of ear skin and the infiltration of immune cells (Fig. S3). SP may be considered as a potential medication for AD-like dermatitis.
The manifestations of AD in TNCB-sensitized NC/Nga mice are caused by recurrent allergen exposure and resulting chronic inflammation of the skin. [33] The allergic response itself may be difficult to completely eradicate unless allergen exposure is completely eliminated, because memory plasma B cells are present. [34,35] [36] In the case of intravenous treatment with SP, an SP-induced TNFα reduction has been consistently observed in various experimental models, including spinal cord injury and collagen II-induced rheumatoid arthritis. [20,21] SP treatment also decreased TNFα levels in the dermal compartment of AD skin (Fig. S5), possibly due to reduced infiltration and degranulation of mast cells, as mast cells are the major source of TNFα (Figure 2 and Fig. S4). Despite the ability of SP to induce acute mast cell activation and mediate neurogenic inflammation, several studies also have suggested that the modulation of neuropeptide activities is associated with the pathogenesis of AD. [37][38][39] At the concentration of SP used for AD-like dermatitis treatment in this study, SP did not affect mast cell degranulation and migration (Fig. S10), contrary to the findings of a previous report that was evaluated at the 100-fold higher SP concentrations. [18] SP also did not activate B-cell activity (Fig. S11). Therefore, the reduced infiltration of mast cells and their degranulation may not be due to the direct effect of SP on mast cells, but may rather reflect a less vigorous immune response in the early stages. All the responses to SP in ADlike chronic dermatitis of TNCB-sensitized NC/Nga mice were mediated by NK-1R activation, which was blocked by pretreatment of NK-1R antagonist. However, the IgE in the serum was not reduced to the level of normal unchallenged animal by SP treatment, possibly due to continuous exposure of low-dose allergen during the SP treatment period. In other word, SP may reduce allergen-mediated cutaneous and local cascades of immune response at the early stage but not be effective in terminating systemically pre-established allergic response itself.
Recently, TSLP has been regarded as a master switch for allergic inflammation. [6] TSLP can perform its biological functions in various types of cells. TSLP activates T cells in mice as well as induces B-cell proliferation and activation in humans. [8,[40][41][42] Moreover, TSLP can stimulate the production of Th2 cytokines in human mast cells in combination with IL-1 and TNFα. [8] In patients with AD, elevated TSLP production has been reported. [43] This may be caused by TNFα-mediated chronic dermatitis and due to elevated Th2 cytokines. In this study, we clearly showed the regulation of TSLP expression by SP treatment in TNFα-treated keratinocytes, as well as in AD-like skin in NC/Nga mice (Fig. S5-S6). These effects in the down-regulation of TNFα, mediated by reduced mast cell infiltration and TSLP expression in the epidermis, could synergistically ameliorate cutaneous inflammation in AD-like skin.
In conclusion, SP treatment alleviated typical AD symptoms in a TNCB-sensitized chronic dermatitis model in NC/Nga mice, as demonstrated by reduced allergic and cutaneous inflammation.
Therefore, we propose that SP may be a potential medication for AD.

SUPPORTING INFORMATION
Additional Supporting Information may be found online in the supporting information tab for this article.