The Protective Effects of Ultraviolet A1 Irradiation on Spontaneous Lupus Erythematosus-Like Skin Lesions in MRL/lpr Mice

We investigated the effects of ultraviolet A1 (UVA1) irradiation on spontaneous lupus erythematosus- (LE-) like skin lesions of MRL/lpr mice, using a disease prevention model. UVA1 irradiation significantly inhibited the development of LE-like skin lesions, without obvious changes of the disease including renal disease and serum antinuclear antibody levels. Besides the massive infiltration of mast cells in the LE-like skin lesions, in the nonlesional skins, more mast cells infiltrated in the UVA1-irradiated group compared with the nonirradiated group. Although apoptotic cells were remarkably seen in the dermis of UVA1-irradiated mice, those cells were hardly detectable in the dermis of the nonirradiated mice without skin lesions. Further analysis showed that some of those apoptotic cells were mast cells. Thus, UVA1 might exert its effects, at least in part, through the induction of the apoptosis of pathogenic mast cells. Our results supported the clinical efficacy of UVA1 irradiation for skin lesions of lupus patients.


Introduction
Collagen disease, one of the representative autoimmune diseases, is often accompanied by photosensitivity. In particular, photosensitivity is included in its diagnostic criteria of systemic lupus erythematosus (SLE) [1,2], and it has been reported that, in Japan, about 60% of SLE patients showed photosensitivity [3,4]. In addition, exposure to sunlight is a well-known risk factor that induces or exacerbates not only skin lesions but also systemic symptoms for SLE patients. To investigate the effects of sunlight on SLE, some studies using MRL/lpr mice, one of the major lupus-prone mice, have been performed. MRL/lpr mice are characterized by dysregulated apoptosis due to defective signaling through Fas antigen [5] and by autoimmune symptoms including fetal nephritis, autoantibody production, and lymphoadenopathy. Furthermore, some of MRL/lpr mice spontaneously develop LElike skin lesions with alopecia and scab formation on their upper back region [6][7][8]. In cultured cells from MRL/lpr mice, the cytotoxicity of ultraviolet (UV) irradiation was significantly elevated [9]. In line with this, UVB exposure significantly exaggerated the development of LE-like skin lesions in MRL/lpr mice [10]. Moreover, UVA could also exacerbate the skin lesions in SLE patients [11]. On the contrary, several recent studies demonstrated that UVA1 irradiation had therapeutic effects for SLE patients [12][13][14][15].
Ultraviolet can be divided into three different parts based on the wavelength as follows: UVA (320-400 nm), UVB (290-320 nm), and UVC (200-290 nm). Moreover, UVA is made up of two parts: UVA1 (340-400 nm) and UVA2 (320-340 nm). UVC is absorbed by the ozone layer and does not reach the ground, whereas UVA and UVB have different various biological effects. For example, UVA and UVB induce suntan without an erythema response and suntan with sunburn, respectively [16,17]. Although the biological effects of UVA2 are similar to the effects of UVB, only UVA1 possesses the particular characteristic biological functions. In the previous studies with MRL/lpr mice [9,10] [18]. However, to the best of our knowledge, there have been no studies examining the UVA1 effects on the LE-like skin lesions of SLE model mice, and the biological effects of UVA1 on SLE still remain investigated. In the present study, we investigated the effects of UVA1 irradiation on MRL/lpr mice, particularly on their LE-like skin lesions, using a disease prevention model. Simultaneously, we also studied the dermal infiltration of mast cells, histamine receptor (HR) expression [19] and the mRNA expression of cytokines [20], which were previously examined in the spontaneous LE-like skin lesions in MRL/lpr mice [19,20].

2.2.
Mice. MRL/lpr mice, aged 4-6 weeks at the start of each experiment, were purchased from Japan SLC Inc. (Hamamatsu, Japan) and housed individually in cages under the specific pathogen free conditions. All animal experiments were approved by the Committee on Animal Care and Use at Wakayama Medical University.
The MRL/lpr mice were divided into the following three groups: (1) nonirradiation group (n = 22), (2) UVA1 irradiation with 5 J/cm 2 (n = 18), and (3) UVA1 irradiation with 10 J/cm 2 (n = 18). The doses of 5 and 10 J/cm 2 have been chosen by the following reasons: first, effective doses of UVA1 treatment for human SLE patients were reportedly low, such as 6 and 12 J/cm 2 [12][13][14]; second, it took about 60 minutes to irradiate 10 J/cm 2 of UVA1 using our equipment, and therefore the irradiation over 10 J/cm 2 seemed quite stressful for mice. Under the deep anesthesia, all of mice were shaved at the upper dorsal skin twice a week, because the spontaneous LE-like skin lesions would appear on this site at older ages. Mice were irradiated UVA1 (5 J/cm 2 or 10 J/cm 2 ) five times a week for 4 months. The nonirradiated mice were left untreated after shaving. LE-like skin lesions of mice were macroscopically evaluated. The mortality was also estimated during experimental period.

Measurements of Antinuclear Antibody (ANA) in Sera.
Serum samples, collected at 4 months after the first irradiation, were subjected to ANA analysis by a commercial kit (Fluoro HEPANA test, MBL Co., Ltd, Nagoya, Japan), as described previously [21].

Light Microscopic Observation of the Skin. At 4 months
after the first irradiation, mice were euthanized by the cervical dislocation, and skin specimens were taken from the upper back regions of all mice. The specimens were fixed in 4% formaldehyde buffered with PBS (pH 7.2), embedded in paraffin, and stained with hematoxylin and eosin (HE). Toluidine blue staining was also performed to assess the infiltration of mast cells. For the evaluation of dermal infiltrating mast cells, 5 microscopic fields were randomly selected, and the number of mast cells was counted. The degree of mast cells infiltration was expressed as the average number of the mast cells in the 5 microscopic fields (magnification, ×100). All measurements were performed without prior knowledge of the experimental procedures.

Light Microscopic Observation of the Kidney.
Kidney specimens taken from all MRL/lpr mice were subjected to HE and periodic acid-Schiff (PAS) staining. Then, the degree of glomerulonephritis was evaluated according to previously described methods [22] with a slight modification. All measurements were performed without prior knowledge of the experimental procedures.

Immunohistochemical Study.
Skin specimens were frozen immediately in Tissue-Tek O.C.T. Compound (Sakura Finetechnical Co., Ltd, Tokyo, Japan) to be stored at −80 • C until the use. Cryosections (6 μm thick) were prepared using Cryostat CM1900 (Leica Microsystems, Heidelberg, Germany), and the sections were reacted with anti-H1R pAbs, anti-H2R pAbs, or anti-H3R pAbs. After the incubation of biotinylated secondary Abs, immune No-irradiation Non-lesional skin (a) No-irradiation Skin lesion   Figure 1(b)). On the other hand, macroscopic skin lesions were never seen in all UVA1-irradiated (5 J/cm 2 or 10 J/cm 2 ) mice (Table 1, Figure 1(c)). As reported previously [6][7][8], nonirradiated MRL/lpr mice with macroscopic skin lesions showed characteristic histopathological alterations such as acanthosis with hyperkeratosis, liquefaction, mononuclear cell infiltration into the dermis, and slight extravasation of erythrocytes in the upper dermis (Figure 1(e)). Moreover, even 4 nonirradiated mice without macroscopic skin lesions showed moderate histopathological changes mentioned above, while 5 mice in the UVA1 irradiated groups (5 J/cm 2 : 2 mice and 10 J/cm 2 : 3 mice) showed only LE-like histopathological changes. With respect to the development of LE-like skin lesions, there was a significant difference between the nonirradiated and all UVA1-irradiated groups, and between the nonirradiated and UVA1-irradiated (only 5 J/cm 2 ) groups ( Table 1, P < .05). These observations implied that UVA1 irradiation significantly protected mice against the development of spontaneous LE-like skin lesions.

Effects of UVA1 Irradiation on the Mortality, Renal
Disease, and Serum ANA Levels. To assess the systemic effects of UVA1 irradiation on MRL/lpr mice, we evaluated the mortality, the development of renal disease, and serum ANA levels. The mortality of UVA1-irradiated MRL/lpr mice, especially the 10 J/cm 2 irradiated group, tended to be lower than that of the nonirradiated mice, without statistical significance ( Table 2). MRL/lpr mice almost always showed renal disease as evidenced by proteinuria, which was similar to the human lupus nephritis with SLE. UVA1 irradiation had no influence on proteinuria ( Figure 2). Furthermore, there was no significant difference in renal histopathological alterations among these three groups (data not shown). Similarly, there was no significant difference in the serum ANA levels and immunoglobulin deposits in skin among three groups (data not shown). These data suggested that, in our experimental procedures, the effects of UVA1 irradiation on systemic symptoms of MRL/lpr mice were not significant.
In comparison, some studies on human SLE patients have reported that UVA1 therapy improved the systemic symptoms and parameters [12][13][14][15]. Considering the tendency of improving mortality by UVA1 irradiation, further studies of other parameters reflecting systemic immune functions might show some systemic effects of UVA1 in our system.

Effects of UVA1 Irradiation on Dermal Infiltration of Mast Cells.
In consistent with the previous study [19], we could detect massive infiltration of mast cells in the LE-like skin lesions of MRL/lpr mice, compared with nonlesional skins in both nonirradiated and UVA1-irradiated mice. (Figures 3(b), 4). Furthermore, in nonlesional skins, mast cell recruitment was more evident in UVA1-irradiated mice than in nonirradiated ones (Figures 3(a), 3(c), 4). These observations implied that mast cells might play a role in the protective effects of UVA1 irradiation as well as in the development of spontaneous LE-like skin lesions. Recently, it has been reported that mast cells have both positive and There were no significant differences among three groups. The score of proteinuria Nonirradiated UVA1 5 J/cm 2 UVA1 10 J/cm 2 Figure 2: Degree of proteinuria in MRL/lpr mice of the three groups. There were no significant differences among the three groups. UVA1 irradiation had no effect on the degree of proteinuria. The data represent means ± 1 SD. negative regulatory functions for immunity [23]. Mast cells produce IL-10, eventually playing a negative role in the immunomodulatory mechanisms. Actually, UVB irradiation can induce IL-10 production in mast cells [24]. When we examined the IL-10 localization in the skin samples by immunofluorescence analyses, IL-10-positive signals could be detected in the cytoplasm of mast cells at the LE-like skin lesions of nonirradiated mice ( Figure 5(c), indicated with arrows). Unexpectedly, mast cells recruited on the skin of UVA1-irradiated mice never showed IL-10-positive signals ( Figure 5(f)). Thus, these observations implied at least that UVA1 irradiation could not induce IL-10 production in dermal mast cells in the present study.  inflammatory mediator mainly derived from mast cells, was closely related with the development of LE-like skin lesions of MRL/lpr mice through HRs [19,25]. Tachibana et al. suggested that the impairment of histamine metabolism in MRL/lpr mice would contribute to the development of LE-like skin lesions [25]. Actually, the gene expression of H2R was significantly enhanced in the LE-like skin lesions of MRL/lpr mice, and H2R-expressing mononuclear cells could be observed immunohistochemically [19]. The diverse biological effects of histamine are mediated through four different histamine receptors, H1R, H2R, H3R, and H4R [26,27]. We also examined the expression of the histamine receptors in both the LE-like skin lesion and nonlesional skins of nonirradiated and UVA1-irradiated mice by immunohistochemical and RT-PCR analyses. Similar to our previous results [19], some infiltrating mononuclear cells expressed H2R in the LE-like skin lesions, and the expression of H2R mRNA was significantly up-regulated (data not shown). However, in nonlesional skins of MRL/lpr mice, there was no significant change in H2R mRNA expression between the nonirradiated and UVA1-irradiated mice.

Effects of UVA1 Irradiation on the mRNA Expression of Cytokines and the Immunochemical Examination and mRNA
In addition, we analyzed the mRNA expressions of TNFα, IFN-γ, IL-4, IL-10, and TGF-β. In the LE-like skin lesion, we could observe more remarkable mRNA expressions of TNF-α, IL-10, and TGF-β compared with nonlesional skin of nonirradiated mice, as reported in the previous study [20]. However, in the nonlesional skins, there were no significant changes between the nonirradiated and UVA1-irradiated mice (data not shown). Briefly, within our investigations, we could not detect any significant changes associated with the protective effects of UVA1 irradiation, regarding cytokines and HRs.

Induction of Apoptosis of Dermal Cells with UVA1
Irradiation. The apoptosis of infiltrating cells has been recognized as one of the main mechanisms responsible for UVA1 treatment [16,17,28]. In the present study, TUNEL-positive Furthermore, in the nonlesional skins, there was a significant difference between UVA1-irradiated (10 J/cm 2 ) and nonirradiated mice (P < .05). There was no significant difference between UVA1 5 and 10 J/cm 2 groups. The data represent means ± 1 SD from 5 samples of each group. Our study demonstrated that mast cells might play a role in the development of spontaneous LE-like skin lesions of MRL/lpr mice. Although we also speculated the possibility that UVA1 irradiation induced such mast cells as aggressively inhibited the development of LE-like skin lesions, for example, through the production of IL-10, we could not confirm such mast cells in the skin of UVA1irradiated mice. Thus, this implied that UVA1 irradiation might induce the apoptosis of dermal pathogenic mast cells. In consistent with our results, apoptosis of proliferating mast cells was reported to be induced by UV in vitro, while resting skin mast cells were resistant to UV light-induced apoptosis [29]. However, as shown in Figure 7, UVA1 irradiation also induced the apoptosis of other types of dermal cells, except mast cells. Characterization of these apoptotic cells in the dermis of UVA1-irradiated mice still remained unclear, and we need further investigations, because several recent lines reported the new insights on autoimmunity in MRL/lpr mice through the impaired regulation of Langerhans cells [30] or regulatory T cells [31,32]. Besides mast cells, these various immunocytes are speculated to be involved in the development of spontaneous LE-like skin lesions of MRL/lpr mice, and also the protective effects on them by UVA1 irradiation. In this context, the apoptosis of mast cells by UVA1 irradiation should be considered to contribute, but only in part, to the effects of UVA1.  In human LE, there have been only few reports investigating mast cells in the skin lesions [33]. Studies on 7 Japanese cases with chronic cutaneous LE revealed a 1.5-fold increase in infiltrating mast cells compared with the control skin [34]. Although evidence is lacking, mast cells are considered good candidates to mediate the tissue damage with autoantibodies and immune complexes in SLE [35].

Conclusions
In the present study, we demonstrated that UVA1 irradiation prevented the spontaneous development of macroscopic LE-like skin lesions of MRL/lpr mice. On the contrary, UVA1 irradiation did not have harmful systemic effects. As one of the mechanisms for the protective effects of UVA1 irradiation, we could suggest the induction of apoptosis of dermal pathogenic mast cells. These observations supported the possibility that UVA1 phototherapy would be one of the more rational treatments for skin lesions of SLE patients.