Attenuation of murine sclerodermatous models by the selective S1P1 receptor modulator cenerimod

Sphingosine-1-phosphate (S1P), a lipid mediator, regulates lymphocyte migration between lymphoid tissue and blood. Furthermore, S1P participates in several physiological phenomena including angiogenesis, inflammation, immune regulation, and neurotransmitter release. Moreover, S1P/S1P receptor signaling involves in systemic sclerosis (SSc) pathogenesis. This study aimed to investigate whether the selective S1P1 receptor modulator cenerimod attenuates murine sclerodermatous models. Cenerimod was orally administered to murine sclerodermatous chronic graft versus host disease (Scl-cGVHD) mice, either from day 0 to 42 or day 22 to 42 after bone marrow transplantation. Bleomycin-induced SSc model mice were administered cenerimod from day 0 to 28. Early cenerimod administration inhibited, and delayed cenerimod administration attenuated skin and lung fibrosis in Scl-cGVHD mice. Cenerimod suppressed the infiltration of CD4+ T cells, CD8+ T cells, and CD11b+ cells into the inflamed skin of Scl-cGVHD mice as opposed to control mice. In contrast, cenerimod increased the frequency of regulatory T cells in the spleen and skin of Scl-cGVHD mice. Additionally, cenerimod attenuated the mRNA expression of extracellular matrix and fibrogenic cytokines in the skin. Furthermore, cenerimod attenuated bleomycin-induced fibrosis in the skin and lung. Hence, the selective S1P1 receptor modulator cenerimod is a promising candidate for treating patients with SSc and Scl-cGVHD.

. Oral treatment of cenerimod attenuates Scl-cGVHD severity and fibrosis. Recipients were given vehicle (0.25% methylcellulose, 0.05% Tween 80 in water) or were orally administered cenerimod (10 mg/kg/ day) from day 0 to day 42, or from day 22 to day 42. (A) The respective photographs were taken 42 days after BMT. Skin score (B) and body weight change (C) were monitored every 3 days (n = 5-6 per group; *p < 0.05, **p < 0.01 for vehicle vs cenerimod day 0 to day 42. † p < 0.05, † † p < 0.01 for vehicle vs cenerimod day 22 to day 42). All data are representative of two independent experiments.
ScIentIFIc RepoRTS | (2019) 9:658 | DOI: 10.1038/s41598-018-37074-9 18 to day 33, compared with the control group (day18, 21: p < 0.05, day24, 27, 30, 33: p < 0.01, Fig. 1C); however, there was no recovery in body weight loss in the preventive model from day 36 to day 42 compared with the control group. There was no significant difference in body weight loss between cenerimod therapeutic model and the control group. On histopathological analysis, dermal thickness, histopathologic score, trichrome area, and collagen content of the skin were significantly lower in cenerimod-treated groups than in the control group (p < 0.001, Fig. 2A-C,E,F). Furthermore, the fibrotic area and collagen content in the lung were significantly lower in cenerimod-treated groups than in the control group (lung collagen content vehicle vs cenerimod therapeutic model: p < 0.05, lung collagen content vehicle vs cenerimod preventive model: p < 0.01, trichrome area of lung: p < 0.001, Fig. 2D,G). Therefore, cenerimod attenuates skin and lung fibrosis in Scl-cGVHD.

Microarray data suggest that cenerimod regulates inflammatory cytokines and Treg cells.
Microarray analysis was performed to determine the effects of cenerimod on the Scl-cGVHD model. Differentially expressed probes (DEP) were more numerous in the skin of the Scl-cGVHD than in the lung, thereby indicating a robust effect on the skin of the Scl-cGVHD mice. Additionally, cenerimod counter-regulated 477 DEP in the skin of Scl-cGVHD mice (Fig. 3A). Pathway analysis via Metacore TM and Ingenuity Pathway analysis revealed interesting effects of cenerimod on cell adhesion and migration, pathways linked to systemic lupus erythematosus and lupus nephritis, and IFN-α or IFN-β signaling (Fig. 3B,C). In addition, Ingenuity Pathway analysis revealed that cenerimod demonstrated a potent counter-modulation of key genes representing TNF-α, INF-γ, and TLR4 activity (Fig. 3D). Furthermore, down-regulation of IL10RA in the Scl-cGVHD mouse model was normalized by cenerimod (Fig. 3D). Correlation Engine analysis revealed a positive correlation between Scl-cGVHD skin disease and the Scl-cGVHD regulatory T cells (Treg cells) data set ( Supplementary Fig. 1A). These results suggest that cenerimod counter-regulates the effect of Scl-cGVHD on Treg cells, evident from the negative correlation ( Supplementary Fig. 1B).

Cenerimod treatment reduces immune cell infiltration into the skin.
To evaluate immune cell infiltration into the skin with murine Scl-cGVHD, the inflamed dorsal skin samples obtained 42 days after BMT were immunostained with anti-CD4, anti-CD8, anti-CD11b, anti-B220 and anti-phosphorylated Smad3 mAbs. Immunohistochemical staining revealed that infiltration of CD4 + T cells, CD8 + T cells, and CD11b + monocyte/ macrophages into the skin were significantly reduced in the cenerimod-treated group (both preventive and therapeutic model) compared with the control group (p < 0.01, Fig. 4A). In addition, preventive treatment inhibited the infiltration of CD4 + T cells and CD8 + T cells into the skin more potently than therapeutic treatment (p < 0.01, Fig. 4A). However, B220-positive cell infiltration was only observed at low levels in either groups, with no significant difference between them (data not shown). Smad3 phosphorylation was significantly higher in the skin in the allogeneic group than in the syngeneic group 42 days after BMT. Moreover, both of preventive and theraputic cenerimod treatment significantly decreased Smad3 phosphorylation in comparison with that in the control group (p < 0.01, Fig. 4A). FACS analysis also revealed that infiltration of CD4 + T cells, CD8 + T cells, and CD11b + monocyte/macrophages into the skin were significantly reduced in the cenerimod-treated group 14 days after BMT compared with the control group (CD8 + T cells: p < 0.05, CD4 + T cells, CD11b + cells: p < 0.001, Fig. 4B). Therefore, the modulation of S1P 1 receptor significantly decreased immune cell infiltration into the skin of Scl-cGVHD mice.

Cenerimod increases B cells and regulatory T cells.
We enumerated the CD4 + T cells, CD8 T + cells, CD11b + cells, and B220 + B cells in the spleen 14 days after BMT. The number of B220 + B cells in the spleen was significantly higher in the cenerimod-treated group than in the control group (p < 0.05, Fig. 5A), while there was no significant difference in the number of CD4 + T cells, CD8 T + cells, and CD11b + cells between cenerimod-treated group and the control group (Fig. 5A). We enumerated Treg cells (CD4 + CD25 + FoxP3 + cells) in the spleen and skin of allogeneic BMT mice 14 days after BMT. The percentage of Treg cells both in the spleen and skin was significantly higher in the cenerimod-treated group than in the control group (spleen: p < 0.05, skin: p < 0.01, Fig. 5B,C). Furthermore, cenerimod also increased the percentage of splenic Treg cells in the naïve BALB/c mice compared with BALB/c mice administered with vehicle (p < 0.01, Fig. 5D). Thus, cenerimod treatment increased the Treg cells in not only Scl-cGVHD mice but also naïve mice.
Cenerimod downregulates cytokine mRNAs in the skin of Scl-cGVHD mice. Expression of cytokines, type I collagen gene proα2 (COL1A2) and fibronectin 1 mRNA was assessed in the skin of Scl-cGVHD mice via real-time PCR analysis (Fig. 6A). Fibrogenic cytokines including IL-1β, IL-6, and IL-13 mRNA were significantly downregulated in the cenerimod-treated group compared to the control group (p < 0.05, Fig. 6A). TNF-α and IFN-γ mRNA expression were slightly but not significantly lower in the cenerimod-treated group than in the control group. IL-10 and TGF-β mRNA expression levels did not change with cenerimod treatment. COL1A2 and fibronectin 1 mRNA were also significantly downregulated in the skin of cenerimod-treated mice compared with the control mice (p < 0.05, Fig. 6A). Together, cenerimod treatment downregulated mRNA of fibrogenic cytokines and ECM proteins. To assess cytokine production in skin T cells, we isolated T cells from the skin of Scl-cGVHD mice. Cenerimod treatment significantly downregulated IL-6 and IL-13 mRNA in the skin T cells (p < 0.05, Fig. 6B). TNF-α and IFN-γ mRNA expression in the cenerimod-treated group tended to be slightly but not significantly downregulated in comparison with that in the control group (Fig. 6B).

Cenerimod inhibits collagen production in fibroblasts.
To directly investigate the effects of cenerimod on fibroblasts, skin fibroblasts were cultured and treated with cenerimod. Consequently, soluble collagen content of fibroblast culture supernatant decreased after cenerimod treatment (p < 0.05, Fig. 6C). Furthermore, COL1A2 and Smad3 mRNA were significantly downregulated in the cenerimod-treated group compared to that in the vehicle treatment group (p < 0.05, Fig. 6C). Therefore, collagen production of fibroblasts was suppressed by S1P 1 modulation. Cenerimod attenuates fibrosis in the bleomycin-induced scleroderma model. To investigate whether cenerimod attenuates fibrosis in an another mouse model of SSc, we used a bleomycin-induced scleroderma model induced via intradermal administration of bleomycin, which then developed skin and   lung fibrosis 25 . Histopathological analysis revealed that dermal thickness was significantly lesser in the cenerimod-treated group than in the control group (p < 0.001, Fig. 7A,B). Furthermore, the fibrosis area in the skin and lung were significantly lesser in cenerimod-treated group than in the control group (p < 0.001, Fig. 7C). Moreover, IL-6 mRNA level in the skin of the bleomycin model mice was significantly downregulated in the cenerimod-treated group compared to that in the control group (p < 0.05, Fig. 7D). IL-13 mRNA in the skin was also downregulated to below the measurement sensitivity upon cenerimod treatment, although IL-13 mRNA was detected in the control treatment group (data not shown). Thus, cenerimod also suppressed fibrosis in a mouse model of bleomycin-induced scleroderma.

Discussion
This study is the first to evaluate the therapeutic effects of the selective S1P 1 receptor modulator cenerimod on a murine Scl-cGVHD model and a bleomycin-induced scleroderma model. The orally administered cenerimod not only prevented but also attenuated the development of skin and lung fibrosis in Scl-cGVHD mice. Furthermore, the S1P 1 receptor modulator attenuated skin and lung fibrosis in a mouse model of bleomycin-induced scleroderma. Infiltration of CD4 + T cells, CD8 + T cells, and CD11b + cells into the skin were significantly lower in the cenerimod-treated group than in the control group. Additionally, cenerimod increased the number of splenic and skin Treg cells in Scl-cGVHD mice. Furthermore, cenerimod reduced the mRNA expression of fibrogenic cytokines and ECM proteins in the skin and decreased the fibrogenic cytokines in the skin T cells of Scl-cGVHD mice. In addition, collagen synthesis from fibroblasts was directly inhibited by cenerimod treatment. Hence, the selective S1P 1 receptor modulator cenerimod is a promising therapeutic agent for human sclerodermatous chronic GVHD and SSc. S1P/S1P receptor modulators constitute promising therapy for autoimmune diseases, since S1P receptor modulators have a strong immunosuppressive effect. FTY720, a S1P 1, 2, 3, 4, 5 receptors modulator, is being used to treat multiple sclerosis. Furthermore, the effect of FTY720 will be canceled immediately irrespective of whether therapy is discontinued. However, FTY720 has serious side effects including bradycardia and bronchoconstriction owing to inhibition of S1P 2, 3, 4, 5 receptors. Cenerimod is a highly selective S1P 1 receptor modulator along with the absence of adverse events such as vasoconstriction and bronchoconstriction 24 . Thus, compared to FTY 720, cenerimod is safer and beneficial for treating autoimmune diseases. Activation of the S1P 1 receptor by S1P causes lymphocyte outflow from lymphoid tissues into the tissue stream and accelerates lymphocyte migration. Conversely, circulating lymphocytes are decreased upon functional antagonism of the S1P 1 receptor with cenerimod 23,24 . Donor-derived immune cell infiltration into the skin has been observed in early Scl-cGVHD mice 10 . FTY720 inhibits Scl-cGVHD via blockade of lymphocyte migration 22 . Administration of cenerimod to Scl-cGVHD mice reduced the infiltration of CD4 + T cells, CD8 + T cells, and macrophages in the skin. This finding indicates that T cell and macrophage infiltration in the skin were also reduced upon selective S1P 1 receptor modulation. Hence, cenerimod exerts immunosuppressive effects by reducing T cells and macrophages invading the immune reaction site, since the circulating lymphocytes are suppressed.
Cytokines have essential functions in fibrosis in SSc pathogenesis 5,26,27 . Based on previous results, cytokine balance is thought to be inclined towards Th2 in SSc 28,29 . Th2 cytokines, such as IL-4, IL-6, and IL-13 stimulate collagen production in human fibroblasts in vitro and play an essential role in SSc fibrosis 8,[30][31][32][33] . Cenerimod downregulated fibrogenic Th2 cytokine and collagen mRNA in the skin of Scl-cGVHD mice. The present findings indicate that collagen production from fibroblasts was suppressed by decreasing these Th2 cytokines. Furthermore, cenerimod altered T cell phenotype in the skin. These results indicate that cytokine reduction in the skin is due to decreased T cell and macrophage infiltration and alteration of T cell function. Additionally, collagen production from fibroblasts was significantly inhibited by cenerimod in vitro. Although TGF-β mRNA expression in the skin was not changed with cenerimod treatment, cenerimod treatment significantly attenuated Smad3 phosphorylation in the skin and Smad3 mRNA expression of fibroblast in vitro. Thus, the current study suggests that cenerimod directly inhibited collagen production from fibroblasts through down-regulation of TGF-β-Smad signaling pathway. Together, attenuation of fibrosis is not only due to the reduction in Th2 cytokines, but also due to the direct inhibition of fibroblasts by cenerimod.
Treg cells suppress T cell activation, and essentially maintain immune tolerance and suppression of autoimmune development 34 . The impairment of Treg cell function is associated with the development of autoimmune diseases 35 . Treg frequency is increased in early SSc [36][37][38] and decreased in late SSc [39][40][41] . Furthermore, the imbalance of Treg cells and effector T cells accelerate the pathogenesis of SSc 38,42,43 . Additionally, Treg cells are an important factor in suppressing GVHD. The beneficial effects of treating GVHD mice with FTY720 or other therapeutic agents depends on an increase in the number of Treg cells 44,45 . In the current study, microarray data revealed that cenerimod activity is inversely correlated with the modulation of the effects of Treg cells in Scl-cGVHD. Treg cell number and function are enhanced in S1P 1 receptor-deficient mice and conversely, Treg cell function is attenuated in S1P 1 receptor transgenic mice 46 . Moduation of S1P 1 receptors by FTY720 enhances Treg cell number and function 46,47 . Cenerimod also increases the percentage of Treg cells both in the spleen and skin. Collectively, the beneficial effect of cenerimod also depends on an increase in the splenic and skin Treg cells in a mouse model of Scl-cGVHD.
In summary, this study shows that the selective S1P 1 receptor modulator cenerimod attenuates skin and lung fibrosis in a mouse model of Scl-cGVHD and that of bleomycin-induced scleroderma. Cenerimod apparently exerts an immunosuppressive effect with attenuation of immune cell infiltration into the skin, decreased fibrogenic cytokine environment, accompanied by an increase in splenic and skin Treg cell number. The present results indicate that the selective S1P 1 receptor modulator cenerimod can be a promising candidate to treat human Scl-cGVHD and SSc.
The mice were housed in a specific pathogen-free barrier facility. All studies and procedures were approved by the Committee on Animal Experimentation of Kanazawa University Graduate School of Medical Science. All mouse experiments were conducted based on ethical guidelines of Kanazawa University.
Bone marrow transplantation. 8-12-week-old male B10.D2 and female BALB/c mice were used as donors and recipients, respectively. Bone Marrow (BM) was T cell-depleted (TCD) with anti-Thy1.2 microbeads (Miltenyi Biotec, Auburn, CA). BALB/c recipients were irradiated with 400 cGy twice a day (MBR-1520R; Hitachi, Tokyo, Japan) at one day before transplantation and were injected via the tail vein with 10 × 10 6 TCD-BM and 10 × 10 6 splenocytes in 0.5 mL phosphate-buffered saline (PBS) to generate Scl-cGVHD (allogeneic BMT). A control syngeneic group of female BALB/c mice received male BALB/c TCD-BM and splenocytes (syngeneic BMT) as we described previously in our study 22 . Intradermal bleomycin treatment. Bleomycin was dissolved in PBS at a concentration of 1 mg/ml. C57BL/6 mice received intradermal injections of either bleomycin or PBS (300 μl, administered using a 27-gauge needle) into their shaved backs every other day for 4 weeks, as described previously 25 . Reagents. Cenerimod was provided by Idorsia Pharmaceuticals Ltd. 24 It was orally administered to allogenic recipients or bleomycin injected mice at a dose of 10 mg/kg/day. Scl-cGVHD mice were administered cenerimod from day 0 to day 42 ( mice were administered cenerimod from day 0 to day 28. Control group mice received vehicle. The vehicle was 0.25% methylcellulose (Sigma-Aldrich, St. Louis, MO) and 0.05% Tween 80 (Sigma-Aldrich) in water.
GVHD skin score. Mice were weighed every 3 days after BMT and scored for skin lesion as previously described 48 . The scoring was as follows: healthy appearance, 0; skin lesions with alopecia equal to or less than 1 cm 2 in area = 1; skin lesions with alopecia 1 to 2 cm 2 in area = 2; skin lesions with alopecia 2 to 5 cm 2 in area = 3; skin lesions with alopecia 5 to 10 cm 2 in area = 4; skin lesions with alopecia 10 to 15 cm 2 in area = 5; skin lesions with alopecia 15 to 20 cm 2 in area = 6; skin lesions with alopecia more than 20 cm 2 in area = 7. The skin lesion area was traced on a paper, and the traced area was scanned and measured using the ImageJ software (http://rsb. info.nih.gov/ij). Furthermore, animals were assigned 0.4 points for skin disease (lesions or scaling) on the tail and 0.3 points each for lesions on the ears and paws. The minimum and maximum scores were 0 and 8, respectively.
Histological analysis. The skin and lung samples were fixed in 10% formalin and embedded in paraffin.
Sections (6 µm in thickness) were stained with hematoxylin and eosin (H&E) and Masson's trichrome. Skin histopathology score was determined on the basis of epidermal interface change, dermal collagen thickness, mononuclear cell inflammation, subdermal fat loss, and follicular dropout 49 . The scores were evaluated from 0 to 2 for each category. The minimum score was 0, and the maximum score was 10. It was assessed independently by a dermatopathologist in a blinded manner (T.K. and M.T.). Dermal thickness was measured from the upper dermis to the lower dermis by microscopy. The blue-stained area on Masson's trichrome staining, representing collagen, was quantified using ImageJ software, as we described previously in our study 22 .

Measurement of collagen content in tissue samples. The skin and lung samples were fixed in 10%
formalin and embedded in paraffin. Sections (15 µm in thickness) were deparaffinized after incubation with xylol, xylol:ethanol (1:1), ethanol, water: ethanol (1:1), and water. Individual samples were placed in small test tubes and covered with 0.2 ml of a saturated solution of picric acid in distilled water that contained 0.1% fast green FCF and 0.1% sirius red F3BA. The samples were rinsed several times with distilled water until the fluid was colorless. One milliliter of 0.1 N NaOH in absolute methanol (1:1 volume:volume) was added, and the eluted color was read on a spectrophotometer at 540 nm and 605 nm, as we described previously in our study 50 . The method used is based on the selective binding of sirius red F3BA and fast green FCF to collagen and noncollagenous protein, respectively 51 . Immunohistochemical staining of the skin. The skin samples from Scl-cGVHD mice were removed and frozen in liquid nitrogen using embedding medium for frozen tissue specimens (Tissue-Tek OCT compound; Sakura Finetek, Tokyo, Japan) and stored at −70 °C until use. As we described previously in our study 22 , Frozen sections (5µm-thick) were immediately fixed in cold acetone and were incubated with rat anti-mouse CD4 monoclonal antibody (mAb) (RM4-5 clone; BD Biosciences, San Jose, CA), rat anti-mouse CD8 mAb (53-6.7 clone; BD Biosciences), rat anti-mouse CD11b mAb (M1/70 clone; BD Biosciences), antibody against Smad3 phosphorylated at Ser425 (NBP1-72209; Novus Biologicals, Littleton, CO). The paraffinized skin sections were applied to slides. Before B220 immunostaining, the slides were deparaffinized in xylene and 3% H 2 O 2 , hydrated through graded alcohols, and washed in distilled water. After that, the slides were put in the protease K (S3020; DAKO, Santa Clara, CA) for 20 min and were blocked by incubating the slides for 30 min in 5% skim milk with TBS. The sections were incubated with rat anti-mouse B220 mAb (RA3-6B2 clone; BD Biosciences). CD4 mAb, CD8mb, CD11b mAb and B220 mAb sections were then incubated sequentially with a biotinylated goat anti-rabbit IgG secondary antibody (BD Biosciences). The antibody against Smad3 phosphorylated at Ser425 sections were incubated sequentially with a rabbit IgG-heavy and light chain cross-absorbed secondary antibody (Bethyl Laboratories, Montgomery, TX). After those, all sections were incubated with horseradish peroxidaseconjugated avidin-biotin complex (Vectastain ABC method; Vector Laboratories, Burlingame, CA). All sections were washed 3 times with PBS between incubations, developed with 3,3ʹ-diaminobenzidine tetrahydrochloride and H 2 O 2 , and then counterstained with hematoxylin. Positive cells were counted in five high-power fields (HPF) and the average/HPF was calculated.

Gene expression analysis and biological interpretation.
Preparation of skin cell suspensions. As we described previously in our study 53 , a 1 × 3 cm piece of depilated back skin was minced and then digested in 5 mL of RPMI-10% fetal bovine serum containing 2.5 mg/mL collagenase D (Roche, Basel, Switzerland), 1.5 mg/mL hyaluronidase (Sigma-Aldrich), and 0.03 mg/mL DNase I (Roche) at 37 °C for 90 min. Digested cells were then passed through a 70-μm cell Falcon cell strainer (BD Biosciences) to generate single-cell suspensions. The cell suspension was centrifuged at 300 × g for 10 min. The pellet was resuspended in 70% Percoll solution (GE Healthcare, Uppsala, Sweden), and then overlaid by 37% Percoll solution followed by centrifugation at 500 × g for 20 min at room temperature. Cells were aspirated from the Percoll interface and passed through a 70-μm cell strainer. Subsequently, the cells were harvested by centrifugation and washed.  Fibroblast culture. The C57BL/6 mouse Primary Deremal Fibroblasts were purchased from Cell Biologics (Chicago, IL). As we described previously in our study 53 , fibroblasts were cultured in DMEM (Invitrogen, Carlsbad, CA) containing 10% heat-inactivated FCS, 100 U/mL penicillin (Invitrogen), and 100 µg/mL streptomycin (Invitrogen) at 37 °C in a humidified 5% CO 2 atmosphere. Outgrowing fibroblasts were detached by brief trypsin treatment and recultured in the medium. Confluent cultures of fibroblasts were serum starved for 12 h and then pretreated with cenerimod for one hour. The cells were stimulated with 10 ng/mL TGF-β2 (BioLegend) and incubated for another 24 h. The supernatant was harvested, and the monolayers were washed. The cells were used immediately in experiments, as indicated. All experiments involved fibroblasts between passages 2 and 5, depending on the number of cells obtained initially from the tissue samples. Cultured fibroblasts adhered to the dish and maintained their typical spindle-shaped morphology. In each experiment, all cell lines were examined simultaneously under the same conditions (e.g., cell density, passage, days after plating, etc.) as we described previously 53 . The concentration of soluble collagen in cultured fibroblast supernatants was determined using the Quickzyme Soluble Collagen Assay Kit (QuickZyme Biosciences, Leiden, The Netherlands).

Statistics.
All data are shown as mean ± SEM. The significance of differences between sample means was determined by the Student's t test. Bonferroni test was used for multiple comparisons. P values less than 0.05 was considered statistically significant.