Deterioration of dry skin in arthritis model mice via stress-induced changes in immune cells in the thymus and spleen

Skin dryness is a characteristic of rheumatoid arthritis model mice. We previously reported that the stress hormone glucocorticoid (i.e., corticosterone) is related to the induction of dry skin in arthritic mice. However, the mechanism through which stress induces dry skin in these mice is still unclear. Therefore, in this study, we examined the relationship between stress and induction of dry skin in arthritic mice. Physical stress load in mice with DBA/1JJmsSlc collagen-induced arthritis was treated with water immersion stress, and transepidermal water loss and the expression of markers associated with allergic reactions and inflammation was evaluated. Deterioration of skin dryness was observed in stressed arthritic mice compared with that in unstressed arthritic mice. Moreover, plasma levels of interleukin-6 and corticosterone were increased in stressed arthritic mice compared with those in unstressed arthritic mice. We also observed decreased regulatory T cell numbers and increased T helper type 2 cell numbers in the thymus of stressed arthritic mice compared with those in unstressed arthritic mice. These results suggested that abnormalities in the immune system were related to deterioration of dry skin in stressed arthritic mice. Thus, reduction of stress may prevent deterioration of dry skin in mice with arthritis.

these mice is still unclear. Therefore, in this study, we examined the relationship between 23 stress and induction of dry skin in arthritic mice. Physical stress load in mice with 24 DBA/1JJmsSlc collagen-induced arthritis was treated with water immersion stress, and 25 transepidermal water loss and the expression of markers associated with allergic reactions 26 and inflammation was evaluated. Deterioration of skin dryness was observed in stressed 27 arthritic mice compared with that in unstressed arthritic mice. Moreover, plasma levels of 28 interleukin-6 and corticosterone were increased in stressed arthritic mice compared with 29 those in unstressed arthritic mice. We also observed decreased regulatory T cell numbers and 30 increased T helper type 2 cell numbers in the thymus of stressed arthritic mice compared with 31 those in unstressed arthritic mice. These results suggested that abnormalities in the immune 32 system were related to deterioration of dry skin in stressed arthritic mice. Thus, reduction of 33 stress may prevent deterioration of dry skin in mice with arthritis. barrier function (i.e., dry skin) causes itching [7], and injury to the stratum corneum caused 44 by scratching can then induce skin infections [8,9]. Therefore, elucidation of the mechanisms 45 regulating the induction of dry skin and approaches for preventing dry skin is necessary to 46 avoid such adverse reactions. 47 We previously reported that dry skin occurs in an arthritic mouse model via a 48 mechanism involving mast cells and histamine secreted from mast cells [10]. In addition, 49 reactive oxygen species (ROS), neutrophils, and thymic stromal lymphopoietin were found to 50 induce dry skin via mast cells and expression of cytokines (e.g., interleukin [IL]-6 and tumor 51 necrosis factor-α) [11]. Thus, ROS released from neutrophils stimulates thymic stromal 52 lymphopoietin, and activated thymic stromal lymphopoietin induces mast cells, which 53 degranulate histamine and release IL-6, resulting in dry skin.

54
Patients with RA also suffer from stress owing to severe joint pain. Psychological 55 stress causes increased glucocorticoid (GC) production and stimulates hypothalamic 56 corticotropin-releasing hormone production, leading to increased secretion of 57 adrenocorticotropic hormone by the pituitary gland [12]. These changes caused increased 58 synthesis and secretion of GC from the adrenal glands. GC has anti-inflammatory effects, but 59 induces abnormalities in the structure and function of the skin by upregulating Toll-like 60 receptor 2 [13]. Furthermore, treatment with the GC receptor (GCR) inhibitor RU486 61 ameliorates dry skin in arthritic mice [11]. Accordingly, stress is related to induction of dry 62 skin in arthritic mice. Additionally, stress is involved in the onset of RA itself [14]. However, 63 despite these studies, the mechanisms through which stress induces dry skin in arthritic mice 64 are still unclear.

65
In this study, we evaluated the effects of water immersion stress on dry skin in 66 arthritic mice.

69
Specific pathogen-free DBA/1JJmsSlc mice (10 weeks old) were obtained from Japan SLC 70 (Hamamatsu, Shizuoka, Japan). We used these mice as a collagen-induced arthritic mouse 71 model. The method for inducing arthritis in these mice was described previously [ After allowing the mice to acclimate for 1 week, stress treatment was applied for 4 days, and 80 mice were sacrificed on day 5 (Fig. 1). On day 4, the hair on the dorsal skin was clipped, and  Immobilized-water immersion stress 89 Water immersion stress was performed as described by Izumi et al. (1983) [15], with some     Data are presented as means ± standard deviations. Tukey's post-hoc tests were used to 149 evaluate differences between groups, and results with P values of less than 0.05 were 150 considered statistically significant.

152
Changes in skin condition 153 A schematic of the experimental model is given in Fig. 1. First, we measured transepidermal 154 water loss (TEWL) in the dorsal skin of mice to evaluate skin dryness ( Fig. 2A). The results 155 showed that arthritic mice had higher TEWL than control mice and that stressed arthritic 156 mice had higher TEWL than arthritic mice. In addition, we examined histological changes in 157 the skin of these mice using hematoxylin and eosin (H&E) and Trichrome staining (Fig. 2B   158 and 2C). The skin of arthritic mice was thicker than that of control mice; however, the skin of 159 stressed arthritic mice was thinner than that of arthritic mice and similar to that of control  164 Our previous report suggested that mast cell activation induced skin dryness [10]. Therefore, 165 we assessed the number of mast cells in each group using histological staining with toluidine 166 blue. Notably, the number of mast cells was increased in arthritic mice compared with that in 167 control mice (Fig. 3), but that in stressed arthritic mice was not different compared with that 168 in arthritic mice.

169
ROS, IL-6, corticosterone, and hyaluronan levels 170 Next, we examined the concentrations of ROS, IL-6, corticosterone (a stress hormone), and 171 hyaluronan (a natural moisture factor) in the plasma (Fig. 4). The levels of ROS and 172 corticosterone were increased in arthritic mice compared with those in control mice.

173
Although ROS levels did not change in stressed arthritic mice, the levels of IL-6 and 174 corticosterone were increased compared with those in arthritic mice. In contrast, hyaluronan 175 levels were decreased in stressed arthritic mice compared with those in control mice.

176
However, there were no differences in hyaluronan levels between arthritic mice and stressed 177 arthritic mice. 179 Thymus weights in arthritic mice were decreased compared with those in control mice and 180 significantly decreased in stressed arthritic mice (Fig. 5). In contrast, spleen weights were 181 increased in arthritic mice compared with those in control mice and tended to increase in 182 stressed arthritic mice compared with those in arthritic mice. Annexin V detects changes in the cell membrane induced by apoptosis. Thymic annexin V 188 expression was increased in arthritic mice compared with that in control mice and further 189 increased in stressed arthritic mice. However, splenic annexin V expression did not differ 190 among groups (Fig. 6B).

191
Thymic GCR expression did not differ between arthritic and control mice but 192 increased in stressed arthritic mice. In addition, splenic GCR expression was increased in 193 arthritic mice compared with that in control mice and tended to increase in stressed arthritic 194 mice (Fig. 6B).  197 we found that T-bet expression did not differ in any of the groups (Fig. 6C). In contrast, 198 although thymic GATA3 expression did not differ between arthritic and control mice, 199 GATA3 was upregulated in stressed arthritic mice compared with that in the other two groups.

200
Splenic GATA3 expression was increased in arthritic mice compared with that in control 201 mice and further increased in stressed arthritic mice (Fig. 6C). Additionally, thymic and 202 splenic RORγt expression was increased in arthritic mice compared with that in control mice, 203 but unchanged in stressed arthritic mice compared with that in arthritic mice (Fig. 6C). 204 Finally, thymic Foxp3 expression did not differ between arthritic and control mice, but was 205 decreased in stressed arthritic mice compared with that in the other groups. No differences in 206 splenic Foxp3 expression were observed among groups (Fig. 6C). In this study, we evaluated the relationship between physical stress and induction of dry skin 210 in arthritis using RA model mice. Physical stress load was applied by water immersion stress.

211
In this model, we observed increased TEWL and decreased collagen compared with that in 212 unstressed arthritic mice, indicating increased skin dryness. Additionally, plasma levels of 213 ROS and corticosterone were increased in arthritic mice compared with those in control mice.

214
Plasma levels of IL-6 and corticosterone were increased in stressed arthritic mice compared 215 with those in unstressed arthritic mice. We also observed decreased Treg numbers and 216 increased Th2 cell numbers in the thymus of stressed arthritic mice compared with that in 217 unstressed arthritic mice. 218 We previously reported that various factors, including mast cells, ROS, collagen, 219 and hyaluronan, were involved in skin dryness [10,11,21,22]. Therefore, we evaluated the 220 relationships of skin dryness with these factors. Importantly, we found that there were no between IL-6 and corticosterone is unclear. In the current study, corticosterone and IL-6 239 levels were increased following application of stress in arthritic mice. Accordingly, these 240 findings suggested that corticosterone and IL-6 were closely related and that IL-6 production 241 may have been induced by corticosterone.  In summary, our results suggested that abnormalities in the immune system resulted 296 in dry skin in stressed arthritic mice. Stresses vary from being acute to chronic, and can be 297 physical or mental. In this study, we applied physical stress in mice, which is difficult to 298 quantify, and hence was a limitation of this experiment. However, because patients with RA 299 tend to experience both mental and physical stresses, it is necessary to reduce them in order