17,20S(OH)2pD Can Prevent the Development of Skin Fibrosis in the Bleomycin-Induced Scleroderma Mouse Model

Systemic sclerosis (SSc; scleroderma) is a chronic fibrotic disease involving TGF-β1. Low serum vitamin D (vit D) correlates with the degree of fibrosis and expression of TGF-β1. This study was designed to determine whether the noncalcemic vit D analog, 17,20S(OH)2pD, suppresses fibrosis and mediators of the TGF-β1 pathway in the bleomycin (BLM) model of fibrosis. Fibrosis was induced into the skin of female C57BL/6 mice by repeated injections of BLM (50 μg/100 μL) subcutaneously. Mice received daily oral gavage with either vehicle (propylene glycol) or 17,20S(OH)2pD using 5, 15, or 30 μg/kg for 21 days. The injected skin was biopsied; analyzed histologically; examined for total collagen by Sircol; and examined for mRNA expression of MMP-13, BMP-7, MCP-1, Gli1, and Gli2 by TR-PCR. Spleen was analyzed for lymphocytes using flow cytometry. Serum was analyzed for cytokines using a multiplexed ELISA. Results showed that all three doses of 17,20S(OH)2pD suppressed net total collagen production, dermal thickness, and total collagen content in the BLM fibrosis model. 17,20S(OH)2pD also increased MMP-13 expression, decreased MCP-1 and Gli-2 expression in vivo, and suppressed serum levels of IL-13, TNF-α, IL-6, IL-10, IL-17, and IL-12p70. In summary, 17,20S(OH)2pD modulates the mediators of fibrosis in vivo and suppresses total collagen production and dermal thickness. This antifibrotic property of 17,20S(OH)2pD offers new therapeutic approaches for fibrotic disorders.

Animal models of fibrosis are used to study SSc disease pathogenesis and can mimic some pathophysiologic characteristics of the disease. Using the C57BL/6 murine model, cutaneous fibrosis can be induced by injecting bleomycin (BLM) subcutaneously into the same skin site on alternate days [12]. This model mimics early inflammatory changes in SSc with fibrosis limited to the area of injection [13].
In the present study, we demonstrate 17,20S(OH) 2 pD administered by oral gavage decreases the development of dermal fibrosis and loss of subcutaneous adipose tissue, and affects the expression of key players in the TGF-β1 pathway in the BLM-induced scleroderma model of fibrosis 2. Results 2.1. 17,20S(OH) 2 pD Decreases Dermal Thickness Figure 1 shows attenuation of dermal thickness by 17,20S(OH) 2 pD induced by BLM injections with quantitative data and representative histology, both H&E and trichrome stains. Sectioned skin biopsies of the BLM injection site from mice treated with 17,20S(OH) 2 pD showed a significant decrease in dermal thickness compared to skin biopsies from vehicle control mice (Figure 1). Sections of skin from the BLM injection site from mice treated with vehicle showed increased collagen deposition and decreased adipose tissue in the subcutaneous layer compared to saline-injected mice ( Figure 1). Treatment with 17,20S(OH) 2 pD not only prevented the development of increased dermal thickness at the BLM injection site but also allowed the maintenance of subcutaneous adiposity ( Figure 1).

17,20S(OH)2pD Suppresses Total Collagen Content in the BLM Fibrosis Model
We assessed the effect of 17,20S(OH)2pD on collagen content at the site of skin that received repeated injections of BLM. Mice that received oral gavage with either 17,20S(OH)2pD 5, 15, or 30 μg/kg in contrast to mice that received oral gavage with vehicle control (PG) had reduced levels of total collagen in the skin at the site of BLM injection. We observed a significant reduction in net total collagen levels at the BLM injection site with increasing doses of 17,20S(OH)2pD ( Figure 2). 17,20S(OH)2pD, at the highest dose administered (30 μg/kg), had no effect on calcium levels in serum (Ca ++ 7 ± 1.30 mg/dL) compared to vehicle-treated mice (7 ± 0.5 mg/dL; p = 0.85).

17,20S(OH) 2 pD Suppresses Total Collagen Content in the BLM Fibrosis Model
We assessed the effect of 17,20S(OH) 2 pD on collagen content at the site of skin that received repeated injections of BLM. Mice that received oral gavage with either 17,20S(OH) 2 pD 5, 15, or 30 µg/kg in contrast to mice that received oral gavage with vehicle control (PG) had reduced levels of total collagen in the skin at the site of BLM injection. We observed a significant reduction in net total collagen levels at the BLM injection site with increasing doses of 17,20S(OH) 2 pD ( Figure 2). 17,20S(OH) 2 pD, at the highest dose administered (30 µg/kg), had no effect on calcium levels in serum (Ca ++ 7 ± 1.30 mg/dL) compared to vehicle-treated mice (7 ± 0.5 mg/dL; p = 0.85).

17,20S(OH)2pD Suppresses Total Collagen Content in the BLM Fibrosis Model
We assessed the effect of 17,20S(OH)2pD on collagen content at the site of skin that received repeated injections of BLM. Mice that received oral gavage with either 17,20S(OH)2pD 5, 15, or 30 μg/kg in contrast to mice that received oral gavage with vehicle control (PG) had reduced levels of total collagen in the skin at the site of BLM injection. We observed a significant reduction in net total collagen levels at the BLM injection site with increasing doses of 17,20S(OH)2pD ( Figure 2). 17,20S(OH)2pD, at the highest dose administered (30 μg/kg), had no effect on calcium levels in serum (Ca ++ 7 ± 1.30 mg/dL) compared to vehicle-treated mice (7 ± 0.5 mg/dL; p = 0.85).

17,20S(OH) 2 pD Modulates the Mediators of the TGF-β1 Pathway
A decrease in the synthesis of MMPs is one of several mechanisms that contribute to fibrosis in SSc. MMP-13 is a well-known collagenase, acknowledged for its involvement in the pathogenesis of fibrotic disorders [36]. We examined whether treatment of mice with 17,20S(OH) 2 pD would increase MMP-13 mRNA expression at the site of BLM injection in the skin. Compared to levels of MMP-13 mRNA in the site of injection of saline in control mice, MMP-13 mRNA levels in the site of BLM injection in the BLM-treated mice significantly decreased ( Figure 3A). Oral gavage treatment of the BLM-treated mice with either of the three doses of 17,20S(OH) 2 pD compared to oral gavage with vehicle was associated with an upregulation of MMP-13 mRNA expression in the skin at the BLM injection site ( Figure 3A). fibrosis in SSc. MMP-13 is a well-known collagenase, acknowledged for its involvement in the pathogenesis of fibrotic disorders [36]. We examined whether treatment of mice with 17,20S(OH)2pD would increase MMP-13 mRNA expression at the site of BLM injection in the skin. Compared to levels of MMP-13 mRNA in the site of injection of saline in control mice, MMP-13 mRNA levels in the site of BLM injection in the BLM-treated mice significantly decreased ( Figure 3A). Oral gavage treatment of the BLM-treated mice with either of the three doses of 17,20S(OH)2pD compared to oral gavage with vehicle was associated with an upregulation of MMP-13 mRNA expression in the skin at the BLM injection site ( Figure 3A).
We next examined whether 17,20S(OH)2pD affected the level of mRNA of MCP-1, a chemokine that can promote fibrosis. BLM increased MCP-1 mRNA expression in BLMinjected mice compared to vehicle-treated mice who received s.c. injections of saline. 17,20S(OH)2pD decreased the expression of MCP-1 mRNA at all three doses ( Figure 3B).
We also assessed effect of 17,20S(OH)2pD on the mRNA levels of fibrogenic mediators BMP-7, Gli-1, and Gli-2 mRNA. Bleomycin increased the expression of BMP7, Gli-1, and Gli-2 mRNA in vehicle-treated mice. The increased mRNA level of Gli-2 induced by BLM at the skin injection site was reduced by 17,20S(OH)2pD treatment ( Figure 3C). 17,20S(OH)2pD did not significantly lower the bleomycin stimulation of BMP-7 or Gli1 mRNA.  We next examined whether 17,20S(OH) 2 pD affected the level of mRNA of MCP-1, a chemokine that can promote fibrosis. BLM increased MCP-1 mRNA expression in BLMinjected mice compared to vehicle-treated mice who received s.c. injections of saline. 17,20S(OH) 2 pD decreased the expression of MCP-1 mRNA at all three doses ( Figure 3B).
We also assessed effect of 17,20S(OH) 2 pD on the mRNA levels of fibrogenic mediators BMP-7, Gli-1, and Gli-2 mRNA. Bleomycin increased the expression of BMP7, Gli-1, and Gli-2 mRNA in vehicle-treated mice. The increased mRNA level of Gli-2 induced by BLM at the skin injection site was reduced by 17,20S(OH) 2 pD treatment ( Figure 3C). 17,20S(OH) 2 pD did not significantly lower the bleomycin stimulation of BMP-7 or Gli1 mRNA.

Discussion
Collagen overproduction is a hallmark feature of SSc. 17,20S(OH) 2 pD decreased the total collagen content of the skin at the site of BLM injection in mice. The suppression of collagen content exhibited a dose-dependent pattern. These findings are consistent with our previous study showing that 17,20S(OH) 2 pD decreased collagen synthesis in cultured human dermal fibroblasts stimulated with TGF-β1 [35]. Using the BLM model of fibrosis, vit D analog 22-oxacalcitriol (OCT) decreased dermal collagen bundle without influencing dermal thickness at the site of BLM injection [31]. In the current study, the skin at the site of injected BLM showed increased dermal fibrosis and reduction in subcutaneous adipose tissue compared with the skin at the site of repeated saline injection. The fibrosis also extended into the subcutaneous fat layer. These findings closely mirror the clinical presentation of SSc. Mice that were subjected to BLM injection and treated with 17,20S(OH) 2 pD had a significant reduction in skin fibrosis and maintenance of the subcutaneous fat layer. This study introduces evidence of BLM-induced dermal thickness reduction by 17,20S(OH) 2 pD. Reduction in dermal thickening is a crucial component of skin softening and overall mobility in patients with scleroderma.
The present study highlights the effect of 17,20S(OH) 2 pD on the mediators in the TGF-β1 pathway. MMP-13 expression is suggested to be a strong indicator of disease severity in fibrosing disorders. Lower expression of MMP-13 was found in patients with generalized morphea compared to normal controls [36]. There is an inverse correlation between serum MMP-13 levels and the degree of body area that is affected by fibrosis. In the present study, we found that 17,20S(OH) 2 pD at 5, 15 and 30 µg/kg increased MMP-13 expression in the BLM model.
MCP-1 is a chemokine that induces inflammation and fibrosis. Mice injected with BLM had a significant increase in MCP-1 production [32]. Yamamoto et al. identified the direct and indirect pro-inflammatory effects exerted by MCP-1 on fibroblast proliferation [33]. In the present study, our findings are consistent in that MCP-1 mRNA expression increased at the site of BLM skin injection and its expression was decreased in BLM injected mice treated with 17,20S(OH) 2 pD. The serum levels of MCP-1 increased in mice that received s.c. injections of BLM; 17,20S(OH) 2 pD decreased serum levels of MCP-1, although nonsignificantly. BMP-7, an anti-fibrotic mediator, acts to antagonize the profibrotic effects TGF-β1 [37]. Current evidence regarding the attenuating effects of BMP-7 on fibrosis is unclear. Using murine dermal papilla cells, Bin et al. were able to show inhibition of the TGF-β1 pathway by BMP-7 [37]. Other studies suggest that BMP-7 is less effective at inhibiting TGF-β1 for both in vivo and in vitro BLM-induced skin and lung fibrosis models, suggesting that perhaps renal models show the best reduction in fibrosis by BMP-7 [38,39]. In the present study, BLM increased BMP-7 expression, and 17,20S(OH) 2 pD had no significant effect on BMP-7 expression in treated mice, suggesting that the 17,20S(OH) 2 pD effect on fibrosis in the BLM skin fibrosis murine model is not mediated by increased BMP-7 production.
Gli-1 and Gli-2 are transcription factors in the hedgehog pathway and are positive mediators of TGF-β1 signaling. Inhibiting Gli-1 transcriptional activity is correlated with decreased lung fibrosis and collagen production [40]. In the present study, BLM increased Gli-1 and Gli-2 but 17,20S(OH) 2 pD did not suppress Gli-1 production but suppressed Gli2 production at 5 and 30 µg/kg, suggesting that 17,20S(OH) 2 pD's mechanism of TGF-β1 inhibition of fibrosis may involve portions of the hedgehog pathway.
Finally, we evaluated the effects of 17,20S(OH) 2 pD on various cytokines, some of which are known to be associated with scleroderma and fibrogenesis. IL-6, IL-13, IL-17, and TNF-α are cytokines recognized to be involved in the pathophysiology of scleroderma [40]. Several studies have implicated each of these cytokines in collagen production, and lung and systemic fibrosis [41][42][43][44][45][46][47][48]. In the present study, serum levels of IL-12p40, IL-12p70, IL-6, IL-13, IL-17, MCP-1, and TNF-α were decreased in mice treated with 17,20S(OH) 2 pD. 17,20S(OH) 2 pD modulated the serum expression of IL-6, IL-10, IL-13, IL-17, and TNF-α independent of the BLM effect on these cytokines, suggesting that 17,20S(OH) 2 pD, as a vit D, analog has properties that can alter these cytokines and effect fibrosis. Serum vascular endothelial growth factor (VEGF) is another critical mediator of inflammation known to play a key role in endothelial dysfunction and is increased in systemic lupus erythematosus patients [49]. Since vitamin D was shown to decrease VEGF in clinical trials, it will be important to determine how well 17,20S(OH) 2 pD decreases VEGF in SSc patients in future studies [50]. Moreover, since the immune system and microbiome are interconnected and vitamin D was shown to be critical for maintaining a healthy microbiome, it will be of interest to study how 17,20S(OH) 2 pD affects the microbiome in scleroderma and other autoimmune patients [51].
In conclusion, this study shows that 17,20S(OH) 2 pD modulates mediators of fibrosis in vivo, and suppresses total collagen production and dermal thickness, wherein a balance will favor a reduction in fibrosis and may offer a new therapeutic approach for treating scleroderma and other disorders of fibrosis.
Our study is limited in that it provides evidence of 17,20S(OH) 2 pD's ability to prevent fibrosis and does not address 17,20S(OH) 2 pD's ability to treat or regress established fibrosis. We will address this important issue in future studies using the experimental approach . This study is also limited in that it addresses 17,20S(OH) 2 pD's ability to prevent skin fibrosis in the bleomycin scleroderma model but does not include a validation study that addresses its ability to prevent fibrosis in another scleroderma model. A second independent validation study using a separate scleroderma mouse model will be a project for future studies to address this limitation.

Materials and Methods
Experiments using mice were approved by the Institutional Animal Care and Use Committee at the University of Tennessee Health Science Center (project #18-092), approved 24 October 2018). Secosteroid 17,20S(OH) 2 pD was synthesized as previously described [29,30].

Histology Analysis
Biopsied skin sections of the site where BLM or saline was injected were formalin-fixed and processed for histopathologic analysis using hematoxylin and eosin and trichrome staining. The quantitative analysis of the dermal thickness between different groups was assessed by light microscopy by two evaluators (ATS and AEP) using an Olympus microscope with 20× magnification. Three skin layers were included, and six sections were measured for each evaluation.

Quantitative Real-Time PCR
Total RNA was extracted from homogenized skin of mice using Trizol reagent (Invitrogen-Thermo Fisher Scientific, Waltham, MA, USA) and reversed-transcribed into cDNA using reverse transcriptase (Life Technologies-Applied Biosystems (ABI), Grand Island, NY, USA) for RT-PCR. RNAs were quantified using NanoDrop-2000 (Thermos, Franklin, NJ, USA). Fifteen nanograms of total RNA was used for these experiments. TaqMan

Analysis of Total Collagen
Sections of biopsied injected skin from each mouse were digested with pepsin ((0.1 mg/mL) and 0.5 M acetic acid) overnight at 4 • C to remove terminal nonhelical telopeptides and to release collagen into solution as previously described [52]. Total solubilized collagen was measured using a Sircol collagen assay kit (Bicolor Ltd., Carrickgergus, Co Antrim, UK) per the manufacturers protocol. Data are presented as the mean µg/mL ± SEM of collagen per each group.

Quantitation of Serum Calcium
A calcium assay kit was employed to quantitate the levels of Ca ++ in mouse sera (Cat# AB102505; Abcam, Cambridge, MA, USA).

Statistical Analyses
Data are presented as mean ± SE and analyzed with Student's t-test for groups of two or ANOVA using Prism 4.00 (GraphPad, San Diego, CA, USA). Differences in values obtained with the vit D hydroxyl derivatives were compared with appropriate controls using ANOVA. p ≤ 0.05 was considered statistically significant.

Conclusions
In the current study, we showed that 17,20S(OH) 2 pD can: (1) suppress net total collagen synthesis, (2) modulate key mediators in the TGF-β1 pathway, and (3) alter expression of inflammatory cytokines in the bleomycin (BLM) model of skin fibrosis.