Arsenic Trioxide in Synergy with Vitamin D Rescues the Defective VDR-PPAR-γ Functional Module of Autophagy in Rheumatoid Arthritis

Dysregulated autophagy leads to autoimmune diseases including rheumatoid arthritis (RA). Arsenic trioxide (ATO) is a single agent used for the treatment of acute promyelocytic leukemia and is highly promising for other malignancies but is also attractive for RA, although its relationship with autophagy remains to be further clarified and its application optimized. For the first time, we report a defective functional module of autophagy comprising the Vitamin D receptor (VDR), PPAR-γ, microtubule-associated protein 1 light-chain 3 (LC3), and p62 which appears in RA synovial fibroblasts. ATO alleviated RA symptoms by boosting effective autophagic flux through significantly downregulating p62, the inflammation and catabolism protein. Importantly, low-dose ATO synergizes with Vitamin D in RA treatment.


Introduction
Rheumatoid arthritis (RA) is a chronic inflammatory joint disease that leads to cartilage and bone damage as well as disability [1].
Autophagy is defined as a degradation mechanism by which cells recycle cytoplasmic components. Several autophagy-related genes are involved including Beclin , microtubule-associated protein light-chain (LC-), p , and mammalian target of rapamycin (mTOR) [2][3][4]. The assessment of autophagic activity can be achieved by detection of the LC-3 protein. Autophagic flux is a complex process that involves transporting, binding, degrading, and recycling the cytoplasmic components. Nevertheless, increases in the levels of LC-3 can be caused by either the induction of autophagy or inhibited fusion of the autophagosomes with lysosomes; it cannot be used to monitor autophagic flux per se. The p62 protein serves as a link between LC-3 and ubiquitinated substrates and is degraded in autolysosomes. Thus, inhibition of the final step of autophagy correlates with increased levels of p62 and indicates impaired flux [5]. Dysregulated autophagic flux is involved in the pathogenesis of many autoimmune diseases, including RA, by regulating the organism's lifespan and cartilage homeostasis [6][7][8][9]. In addition, the activation of autophagy in immune cells is significantly associated with inflammatory parameters such as interleukin-6 (IL-6) and tumor necrosis factor-(TNF-), which are proinflammatory cytokines involved in the pathogenesis and progression of RA [10].
The Vitamin D receptor (VDR) is expressed in multiple cells and numerous RA and autophagy-related genes, including mTOR, are potential candidate targets of Vitamin D (Vit D) and VDR [11,12]. It maintains "robustness" in unfavorable conditions and serves a "capacitor" in homeostasis in RA biology [13,14]. Peroxisome proliferator-activated receptor-(PPAR-) is a primary target of Vit D [15]. In addition, both VDR and PPAR-are key autophagy regulators [16,17] and play central roles in the development and progression of RA, supporting the possible involvement of autophagy in RA. In vivo and in vitro experiments confirmed that the PPARagonist could downregulate inflammatory cytokines [18,19]. Bone erosion can be alleviated in RA patients by reducing catabolism through PPAR-pathway activation [20]. The regulatory effects of VDR and PPAR-through autophagy have been proven in tumors and other diseases [21][22][23]. Nevertheless, the exact role and involvement of VDR and PPAR-in RA-related autophagy remain to be defined. The search for novel drugs that can manage more than one single age-related disease is encouraged [24]. Among these drugs, arsenic trioxide (As 2 O 3 , ATO) is recognized for treating tumors [25,26] and autoimmune rheumatic diseases by enhancing apoptosis and inhibiting angiogenesis [27][28][29]. ATO has been shown to induce antitumor effects through autophagy [30]. However, the effect of ATO on autophagy in RA is unknown.

PPAR Research
In the present study, we demonstrated for the first time that VDR, PPAR-, and LC-3 participate in a functional module of autophagy (a functional module is a group of genes that are tightly associated through multiple feedback loops) and are significantly upregulated in RA synovial tissues, although autophagic flux was unexpectedly severely impaired. Surprisingly, ATO rescued this defective functional module of autophagy in RA fibroblast-like synoviocytes (FLS) and mice with RA, and the effect was even better when ATO was used with Vit D. Hence, ATO combined with Vit D could be a potential therapeutic strategy against RA.

Materials and Methods
. . Synovial Fibroblast Culture and TNF-Stimulation. RA FLS and normal human (NH) FLS were purchased from Cell Applications (San Diego, CA, USA) and maintained in a synoviocyte growth medium (Cell Applications). Cells were used for the experiments in stages 4-6. RA FLS were pretreated with 50 ng/mL of TNF-for 4 h before the application of ATO and/or Vit D. Cell culture supernatants were used for enzyme-linked immunosorbent assays (ELISAs) 48 h after the addition of the treatments.
. . In Vitro Proliferation Assay. Cell proliferation was evaluated with a Cell Counting Kit-8 (Sigma, St Louis, MO, USA) following procedures described earlier with minor modifications [31]. Briefly, 10 4 cells were seeded in a 96-well plate. After 24 h, different concentrations of drugs or vehicles were added with fresh medium. Cells were incubated at 37 ∘ C for 48 h. The plates were read at 450 nm. The experiments were repeated three times.

. . RNA Preparation and Real-Time Quantitative Polymerase
Chain Reaction Analysis. Total RNA was extracted from FLS with TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and converted to cDNA. Real-time quantitative polymerase chain reaction (PCR) amplification was performed as described previously [28]. The sequences of the primers are shown in Table 1. GAPDH was used as an internal control. All samples were measured in triplicate and the results were evaluated via the 2 −ΔΔ ct method [9].
. . RNA Interference. VDR knockdown was achieved by transfecting specific VDR small interfering RNA (siRNA) into RA FLS. The siRNA sequence was as follows: forward, 5 -GCUGAAGUCAAGUGCCAUUTT-3 ; reverse, 5 -AAUGGCACUUGACUUCAGCTT-3 . RA FLS were plated in 12-well plates at 10 5 cells per well with serum-free DMEM and transfected with siRNA via Lipofectamine 2000 (Invitrogen) as previously described [28]. Six hours after transfection, the culture medium was replaced by DMEM with 10% FBS. Knockdown efficiency was determined by real-time PCR and western blot analysis. The experiments were repeated at least three times.
. . Establishment of Collagen-Induced Arthritis. Specific pathogen-free 6-week-old male DBA/1J mice weighing 18±2 g were purchased from SLAC (Shanghai, China). Animal welfare and experimental procedures were carried out in accordance with the Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee of Harbin Medical University. The collagen-induced arthritis (CIA) mouse model was established as described previously [28]. Mice were randomly assigned to different groups (n=6 per group): normal control group (mice without immunization and injected with saline); the CIA control group (CIA mice treated with saline, CIAsaline); the ATO-treated group (CIA mice treated with ATO at a dose of 1.0 or 2.0 mg/kg/day), Vit D (400 ng/kg/d) or Vit D (400 ng/kg/d) + ATO (2.0 mg/kg/day); and the methotrexate (MTX) group (CIA mice treated with MTX at 2 mg/kg/week as a positive control). Mice were given intraperitoneal injections from day 28 to day 41. The body weight of each mouse was recorded every other day from day 21.
. . Assessment of Arthritis Severity. To evaluate the severity of the arthritis quantitatively, the arthritis score was assessed every 2 days starting from day 21, according to a scoring system used previously [32]. In addition, body weight and the thickness of the two hind paws were measured every other day.
. . Microcomputed Tomography Imaging. To investigate the effects of ATO on the three-dimensional (3D) bone structure, we conducted an assay using microcomputed tomography (micro-CT) imaging (Quantum GX, Perkin Elmer, Waltham, USA). Mice were scanned and reconstructed into a 3D structure via micro-CT imaging 39 days after the initial collagen injection. The voxel size was 72 m, the X-ray tube voltage was 90 KV, the current was 88 A, and the exposure time was 4 min. Mean CT values of the hind paws were calculated with Caliper Analyze software (Analyze Direct, Kansas, USA) to assess bone loss.
. . Histological Analysis. Whole knee joints of the mice were collected and fixed in 10% buffered neutral formalin and decalcified in 10% EDTA for 4 weeks. The paraffin-embedded specimens were stained with hematoxylin and eosin (H&E). Histological changes were scored in a blinded manner by two independent observers based on synovial hyperplasia, joint inflammation, and bone erosion following a scoring system described previously [33,34].
. . Immunohistochemistry Analysis. Immunohistochemistry was performed with specific antibodies for target proteins following a protocol described previously with some modifications [6]. Knee joint sections on slides were incubated with anti-VDR, anti-PPAR-, anti-LC-3, or anti-P62 (Boster, Wuhan, China) antibodies. Subsequently, the sections were stained with a polymer horseradish peroxidase (HRP) detection system (PV9001, ZSGB-BIO, Beijing, China) and visualized with a diaminobenzidine (DAB) peroxidase substrate kit (ZLI-9017, ZSGB-BIO, Beijing, China). Each section was evaluated under a microscope (DMi8, LEICA, Wetzlar, Germany) in three randomly selected areas at a magnification of 20×. Image-Pro Plus 6 (Media Cybernetics, Inc.) was used to analyze the average integrated optical density (IOD) according to a previously described protocol [35].
. . Enzyme-Linked Immunosorbent Assay. Serum samples from the mice and the cell culture supernatant were collected and stored at −80 ∘ C. The concentrations of IL-6, IL-1 , matrix metalloproteinase-3 (MMP-3) and MMP-13 were detected with commercial kits (Elabscience Biotechnology Co., Ltd., Wuhan, China), according to the manufacturer's protocols. All assays were conducted in triplicate.
. . Statistical Analysis. Data are represented as means ± standard error of the mean and were analyzed via Student's t-test or analysis of variance (ANOVA), as appropriate. All analyses were carried out in SPSS 17.0 software. Values of p < 0.05 were considered statistically significant. Afterwards, RA FLS were treated with ATO (2 M) after VDR knockdown. PPAR-and LC-3 were significantly downregulated but p62 was enhanced after silencing VDR, and ATO could reverse these effects (Figure 1(c), p<0.05). Immunohistochemistry was performed to validate the effect of ATO on the VDR-PPAR-autophagy functional module in CIA mice. Strikingly, treatment with ATO (2 mg/kg/d) or MTX (2 mg/kg/week) led to significant increases in the expression of VDR, PPAR-, and LC-3 and a decrease in p62 compared with CIA-saline mice (Figure 2 . . ATO Alleviates Symptoms and Joint Destruction in CIA Mice. We assessed the body weight, degree of paw swelling, and arthritis scores of the CIA and control mice. Both ATO (1 or 2 mg/kg/d) and Vit D (400 ng/kg/d) alone alleviated the symptoms of CIA mice ( Figure 5(c), p<0.01). However, combining ATO and Vit D further reduced the arthritis score but this was not significantly different from ATO alone. The severity of arthritis in CIA mice was assessed by H&E staining of knee joint sections. ATO treatment at 2 mg/kg/d and Vit D (400 ng/kg/d) significantly reduced the histological scores including synovial hyperplasia, cartilage, and bone erosion and joint inflammatory ( Figure 5(b), n=6, p<0.01). Additionally, ATO (2 mg/kg/d) significantly enhanced the PPAR Research   Protein concentration (pg/ml)   .

. ATO Inhibits the TNF--Induced Inflammation Module Representative and Catabolism Module Representative
Release by Regulating the VDR-PPAR-Autophagy Functional Module. Interestingly, ATO (1 mg/kg/d) decreased MMP-13 expression, but key inflammation biomarkers such as IL-6, IL-1 , and IL-8 and catabolism factor module representatives including MMP-3 and MMP-13 were significantly suppressed by ATO (2 mg/kg/d) (Figure 4(a), p<0.01) but were significantly upregulated in CIA-saline mice compared with the normal control mice, as shown by ELISA (Figure 4(a), p<0.01). Vit D (400 ng/kg/d) alone significantly decreased these cytokines and may significantly enhance the effect of ATO when used in combination (Figure 4(a), p<0.05). Therefore, ATO and Vit D downregulated inflammation factors and catabolism factors in CIA mouse serum in vivo.
Furthermore, TNF-treatment (50 ng/mL) significantly induced the inflammation factors and catabolism factors in RA FLS. Stimulation of the VDR-PPAR-functional module by its agonist Vit D and rosiglitazone (50 mol/L) decreased cytokine release, as did ATO with or without Vit D. In contrast, silencing VDR by siRNA and inhibiting PPARwith GW9662 (10 mol/L) increased the secretion of inflammatory and catabolic factors (Figure 4(b), p<0.05); this was reversed by ATO. Furthermore, inflammation and catabolism factors were downregulated by stimulating autophagy with rapamycin (100 nmol/L). Consistently, the autophagy blocker BafA1 induced cytokine release but ATO reversed this effect (Figure 4(b), p<0.05). Therefore, ATO inhibited the release of TNF--induced inflammation factors and catabolism factors in RA FLS by regulating the VDR-PPAR-autophagy module.

Discussion
Autophagy is a highly conserved biological process in eukaryotic cells. Some autophagy-associated genes are closely related to RA [36].
Firstly, we explored the roles of VDR and PPARduring autophagy in RA FLS. To the best of our knowledge, this is the first study reporting that the expressions of VDR, PPAR-, and LC-3 are significantly upregulated, and mTOR is significantly downregulated in RA FLS. Our results showed that p62 in RA FLS was significantly upregulated, indicating autophagic flux obstruction. Since impaired autophagic flux may be involved in a variety of diseases [37], we concluded that defective autophagic flux, rather than other factors causing functional loss of autophagy, leads to RA.
Indeed, PPAR-and VDR are interconnected and modulate the expression of genes in similar ways as a functional module [38,39]. In order to further clarify the relationship between PPAR-and VDR in RA FLS, we applied rosiglitazone to excite PPAR-plaques in RA FLS and found that VDR expression was also elevated, suggesting that PPARmay have a positive feedback effect on VDR.
Anti-TNF biological agents have been proven to be effective in the treatment of RA; however, around 30% of the patients respond poorly or not at all to the biologics. In addition, the treatment is limited by preexisting malignancy and inflammation and further impose an economic burden on patients because of the high costs. ATO has attracted global attention as one of several highly promising effective anticancer drugs. To our knowledge, this is the first report to demonstrate that ATO may rescue impaired autophagic flux beyond its role in activating the VDR-PPAR-autophagy functional module, consequently inhibiting the inflammatory response and joint destruction.
Previously, we and others reported that ATO may relieve the clinical symptoms of patients with leukemia and other rheumatic diseases [27,40,41]. Although we observed no significant effect of ATO on cell proliferation and no toxic effect on CIA mice, several adverse effects have been reported in patients such as liver injury and carcinogenesis [41], which limit the application of ATO in clinical settings. Solution could be sought regarding local drug delivery, drug combinations [42], or chemical modifications [43]. For the first time, we provided evidence that ATO showed a synergistic effect in combination with Vit D both in vivo and in vitro. Since Vit D is safe and easily available, the combination of the two drugs may reduce the dosage of ATO and mitigate its side effects.

Conclusion
In closing, VDR, PPAR-, LC-3, and p62 form a functional module, which is obviously upregulated in RA FLS and CIA mice synovial fibroblasts, but with impaired autophagic flux, as demonstrated by the upregulation of p62.
Although the exact mechanisms are still poorly understood, ATO may enhance the VDR-PPAR-autophagy functional module and rescue impaired autophagic flux both in vivo and in vitro in RA and consequently inhibit the expression of inflammatory and catabolic factors that participate in joint destruction ( Figure 6). Furthermore, we provided evidence that low-dose ATO showed a better effect in combination with Vit D in ameliorating RA symptoms, suggesting novel promising protocols for RA and cancer therapy.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
The authors declare that they have no conflicts of interest.

Authors' Contributions
Weiyan Wang, Chun Ling Li, and Yue Zhang contributed equally to this work. Serving as a regulatory mechanism, autophagy has complex interrelationships with other physiological processes including angiogenesis and apoptosis, which can be regulated by ATO. The solid lines indicate our original data or results from our study that have been corroborated by other labs; dotted lines indicate that research is still ongoing.