LY3214996

α‐Mangostin promotes apoptosis of human rheumatoid arthritis fibroblast‐like synoviocytes by reactive oxygen species‐dependent activation of ERK1/2 mitogen‐activated protein kinase

Xiaoyun Sheng1 | Jun Li2 | Chao Zhang3 | Lianggong Zhao1 | Laiwei Guo1 | Tianen Xu1 | Jiaxin Jin1 | Meng Wu1 | Yayi Xia1,4

Abstract

α‐Mangostin (α‐M) is a commonly used traditional medicine with various biological and pharmacological activities. Our study aimed to explore the effects and mechanism of α‐M in regulating apoptosis of rheumatoid arthritis fibroblast‐like synoviocytes (RA‐FLS). α‐M of 10 to 100 μM was used to treat RA‐FLS for 24 hours, followed by measuring cell viability and apoptosis. The involvement of reactive oxygen species (ROS) and mitogen‐activated protein kinases was detected. Treatment of α‐M promoted apoptosis and reduced viability of RA‐FLS in a dose‐dependent manner. The mitochondrial membrane potential in RA‐FLS was remarkably reduced by α‐M treatment, accompanied by the cytochrome c accumulation in the cytosol and increased activities of caspase‐3 and caspase‐9. Moreover, we found that α‐M treatment promoted ROS production and extracellular signal‐regulated kinase 1/2 (ERK1/2) phosphorylation. The proapoptotic activity of α‐M in RA‐FLS was markedly reversed by the co‐induction with the ERK1/2 inhibitor LY3214996 or ROS scavenger N‐acetyl‐L‐cysteine. In conclusion, our studies found that α‐M had remarkable proapoptotic activities in RA‐FLS, which is regulated by the induction of ROS accumulation and ERK1/2 phosphorylation. α‐M may thus have potential therapeutic effects for rheumatoid arthritis.

KEYWOR DS
apoptosis, extracellular signal‐regulated kinase 1/2, rheumatoid arthritis, rheumatoid arthritis fibroblast‐like synoviocytes, α‐mangostin

1 | INTRODUCTION

Rheumatoid arthritis (RA) is a systemic inflammatory disease that primarily affects the joints. In RA, the immune system damages the lining of the joints and synovium, which leads to the destruction of bone and articular cartilage. Mechanism of the disease is highly complex and the underlying mechanisms are still being elucidated. However, it is known that fibroblast‐like synoviocytes (FLS) play a critical role in the pathogenesis of RA.1 As the disease progresses, the activation of FLS leads to the production of various proinflammatory mediators that signal to recruit, retain, and activate local immune cells. These cells, along with resident tissue cells, cause tissue destruction via an unregulated immune response.2 The synovium, usually consisting of an inner lining and three layers of adjacent cells, is expanded to form multiple layers of proliferative and destructive tissue in RA. Known as pannus, this is a direct result of increased numbers of FLS, which actively and aggressively remodel the tissue structure of the joint via inflammatory processes. These cells can contribute to local cartilage degradation and promote synovial inflammation by releasing chemokines and cytokines, and metalloproteases that break down the extracellular cellular matrix.3 In addition, the release of pro‐inflammatory molecules and growth factors can activate FLS to secrete interleukins and prostanoids to further propagate the unhealthy immune response.4 Furthermore, additional studies have indicated that FLS directly interact with infiltrating T and B cells via LFA‐1/ ICAM‐1 to secrete inflammatory cytokines.5
Inactivation or removal of FLS presents a therapeutic approach for treating RA, based upon the pivotal role played by these cells in disease initiation and progression via inflammatory processes. α‐Mangostin (α‐M) is a naturally occurring xanthone with a range of biological activities and the potential to be used as a treatment for many diseases, including cancers6,7 and obesity.8 α‐M is isolated from the mangosteen tree and displays activity as an antineoplastic agent, an antioxidant, an antiproliferative, and an inducer of apoptosis.9 α‐M was shown to induce cervical cancer cell apoptosis by activating ASK1/p38 signaling and promoting the production of reactive oxygen species (ROS).10 Many studies have demonstrated the anti‐ inflammatory properties of the compound and it has been shown to reduce edema through anti‐inflammatory effects.11 Regarding arthritis, the compound reduced immunocytoadherence in animal studies and inhibited primary and secondary responses to adjuvant‐induced disease.12 Other studies have also shown that xanthones can ease experimental arthritis in mice by inhibiting proliferation of FLS through modulation of mitogen‐ activated protein kinase (MAPK) signaling.13,14
Previous studies have shown that the therapeutic effect of α‐M in RA is mainly via immunosuppressive effects, cytokine regulation, and antioxidant activity. However, more recent studies have indicated that α‐M is also able to induce the apoptosis of human rheumatoid arthritis fibroblast‐like synoviocytes (RA‐FLS) by increasing ROS accumulation and the ratio of Bax/Bcl‐2,15 which suggests that α‐M may act via additional, previously unknown pathways. The current study aims to uncover additional mechanisms related to the apop- tosis‐inducing abilities of α‐M by studying the effects of α‐M on apoptosis of RA‐FLS. The results obtained show that α‐M is able to significantly inhibit the viability of RA‐FLS and increase the proportion of apoptotic cells in a concentration‐dependent manner. We also show that these remarkable proapoptotic activities are regulated by the accumulation of ROS and extracellular signal‐ regulated kinase 1/2 (ERK1/2) phosphorylation. Data provide new insights into the mechanisms of RA progression and indicate that α‐M may be a promising treatment for RA by reducing FLS survival.

2 | MATERIALS AND METHODS

2.1 | Cell treatment

The medical ethics committee of our hospital (LUSH‐ 2016‐0042; approval date: 23 May 2016) approved this study, and all patients provided informed consent. Five female patients with RA (aged 48‐69 years) who were treated with synovectomy or joint replacement donated synovial tissues. Healthy control synovial tissue samples were obtained from five emergent trauma amputation patients consists of three males and two females. All healthy control specimens were free from RA or osteoarthritis. To isolate synoviocytes, synovial tissues were cut and subjected to collagenase digestion in Dulbeccoʼs modified Eagleʼs medium (DMEM)/F12 medium at 37°C for 2 hours. DMEM/F12 medium supplemented with 10% fetal bovine serum was used as a cell culture medium and cell culture conditions were 37°C and 5% CO2. α‐M (Sigma‐Aldrich, MO) of different concentrations (10‐100 μM) were used to treat the cells for 24 hours. In inhibition experiments, ERK1/2 inhibitor N‐acetylcysteine (NAC) (5 mM) or LY3214996 (1 mM) was used to treat cells for 1 or 2 hours before exposure to α‐M.

2.2 | Cell viability measurement

Cells were transferred to a 96‐well plate with 104 cells in each well. Cells were cultivated under conditions of 37°C and 5% CO2 overnight and 0.1% dimethyl sulfoxide (DMSO) and α‐M (various dosages) were added to treat cells for 24 hours. After that, MTT was added and the final concentration was 5 mg/mL. Cells were cultivated for additional 4 hours, and DMSO was added. Finally, optical density (OD) values were measured at 570 nm.

2.3 | Flow cytometry

Cells were transferred to a six‐well plate with 105 cells per well. After cell culture for 24 hours, incubation with 0.1% DMSO or α‐M was performed for 72 hours. After that, cells were harvested and mixed with binding buffer, followed by treatment with FITC‐conjugated PI and annexin V (Becton Dickinson Biosciences, San Jose, CA) for 20 minutes in the dark. Finally, FACSCalibur flow cytometer (Becton Dickinson Biosciences) was used to detect apoptotic cells.

2.4 | Mitochondrial membrane potential assay

Rhodamine‐123 (Rho‐123) dye was used to measure mitochondrial membrane potential. When 60% confluence was reached, cells were treated with 100 μM α‐M or 0.1% DMSO in a six‐well plate with 106 cells per well. After 24 hours, cells were washed and stained with Rho‐123 (1 mL/L) at 37°C for 30 minutes in the dark. Rho‐123 fluorescence was quenched by mitochondrial energization and the fluorescence emissions were analyzed by flow cytometry. The rate of fluorescence decay was represented as the loss of Δψm.

2.5 | Western blot analysis

Cells (106) were mixed with 1 mL radioimmunoprecipita- tion assay buffer to extract total protein and cells concentrations were measured using BCA protein assay kit (Pierce Chemical Co, Rockford, IL). After that, 10% sodium dodecyl sulfate‐polyacrylamide gel electrophor- esis was performed, followed by gel transfer to nitrocellulose membranes. After incubation in 5% nonfat milk for 1.5 hours, incubation with primary antibodies was performed overnight at 4°C as follows: rabbit monoclonal to p38 MAPK, 1:1000; rabbit monoclonal to p‐p38 MAPK, 1:500; mouse monoclonal to p‐ERK1/2, 1:500; mouse monoclonal to ERK1/2, 1:1000; mouse monoclonal to p‐JNK, 1:500; mouse monoclonal to JNK, 1:1000; rabbit monoclonal to cytochrome c, 1:500; rabbit monoclonal to β‐actin, 1:2000. After washing, horseradish peroxidase conjugated secondary antibodies were added to the membrane. ECL (GE Healthcare, Madison, WI) was used to develop signals, which were processed using the Quantity One image software (Pierce Biotechnology Inc., Rockford, IL).

2.6 | Caspase activity

After treatments aforementioned, cells were mixed with caspase assay buffer, followed by incubation on ice for 30 minutes. The mixtures were then centrifuged for 10 minutes at 1000g to collect the supernatant. Colorimetric substrates of c caspase‐3 (Ac‐DEVD‐pNA) or caspase‐9 (Ac‐LEHD‐pNA) was then used to incubate with the supernatant for 90 minutes. OD values were measured to reflect caspase activity.

2.7 | ROS measurement

After treatments aforementioned, oxidative conversion of 2′,7′‐dichlorofluorescin diacetate (DCFH‐DA) to fluorescent dichlorofluorescein in the cells was measure to quantify intracellular ROS levels. Following this, cells were treated with DCFH‐DA for 30 minutes and the ROS levels were quantified by flow cytometry using the excitation wave- length of 488 nm and the emission wavelength of 525 nm.

2.8 | Statistical analysis

Mean values were from three biological replicates. One‐ way analysis of variance and Tukeyʼs multiple comparison test were used for statistical analyses. P < 0.05 was statistically significant. 3 | RESULTS 3.1 | α‐M inhibits the cell viability and promotes apoptosis in RA‐FLS To examine the effects of α‐M (Figure 1A) treatment on RA‐FLS, cell viability and apoptosis were measured. As shown in Figure 1B, α‐M of 10 to 100 μM significantly attenuated the cell viability of RA‐FLS in a concentration‐ dependent manner compared to untreated cells (P < 0.05 or P < 0.01). Compared to untreated cells, RA‐FLS treated with α‐M also demonstrated increased apoptosis (at 100 μM, P < 0.01) (Figure 1C and 1D). α‐M of 100 μM was used for the further experiments. Next, we measured the effect of α‐M on cell viability and apoptosis in normal FLS from healthy individuals. It was shown that α‐M at the concentrations from 10 to 100 μM did not affect the cell viability (Figure 1E) and cell apoptosis (Figure 1F and 1G) in normal FLS. 3.2 | α‐M triggers mitochondrial apoptotic cascade in RA‐FLS To investigate the effects of α‐M on the mitochondria‐ dependent apoptosis, the changes of Δψm and cytochrome c in RA‐FLS were examined. As shown in Figure 2A and 2B, α‐M treatment resulted in a significant reduction of Δψm in RA‐FLS (P < 0.05). The accumulation of cyto- chrome c in the cytosol was significantly increased by α‐M treatment as revealed by western blot analysis (P < 0.05, Figure 2C and 2D). To elucidate the involvement of caspase‐3 and caspase‐9 in α‐M‐induced apoptosis, the activities of caspase‐3 and caspase‐9 in RA‐FLS were measured. We show that the activities of caspase‐3 and caspase‐9 were significantly increased by α‐M treatment (P < 0.05, Figure 2E). 3.3 | α‐M treatment leads to activation of ERK1/2 in RA‐FLS Western blot analysis revealed that α‐M significantly increased the phosphorylation of ERK1/2 in RA‐FLS (P < 0.05, Figure 3A and 3B). Meanwhile, no remark- able changes were observed on the phosphorylation status of p38 MAPK or JNK. To determine the role of ERK1/2 signaling in α‐M‐induced apoptosis in RA‐FLS, the ERK1/2 inhibitor LY3214996 was used to block the phosphorylation of ERK1/2. As shown in Figure 3C and 3D, pretreatment with LY3214996 significantly reversed the α‐M‐induced apoptosis in RA‐FLS (P < 0.05, compared to α‐M treatment alone). 3.4 | ROS generation is involved in α‐M‐induced ERK1/2 activation and apoptosis in RA‐FLS Next, we investigated the effects of α‐M on cellular ROS level in RA‐FLS. α‐M treatment resulted in a significant increase in cellular ROS level compared to untreated cells (P < 0.05, Figure 4A). Finally, we examined the involve- ment of ROS in the α‐M‐induced ERK1/2 activation. We found that pretreatment with ROS scavenger NAC significantly reversed the upregulation of ROS produc- tion induced by α‐M (Figure 4A, P < 0.05), compared to α‐M treatment alone. Moreover, NAC significantly reversed the apoptosis (Figure 4B and 4C) and ERK1/2 phosphorylation (Figure 4D and 4E) induced by α‐M (P < 0.05), compared to α‐M treatment alone. 4 | DISCUSSIONS Many studies have highlighted a role for FLS in the initiation and progression of RA and it was known that RAFLS promotes joint destruction via proinflammatory processes.16 However, most therapies approved for treating RA do not target FLS themselves and therapies that are capable of eliminating these destructive cells may provide better clinical outcomes.17 α‐M has the potential to fulfill this unmet clinical need since it has previously been shown to induce apoptosis of FLS via ROS accumulation and modulation of Bax:Bcl‐2.15 The phamacokinetics, bioavailability, and clinical efficacy of the compound have also been improved in recent years via new lipid‐soluble formulations and micro emulsions.18,19 In this study, we show that α‐M treatment induces significant apoptosis of RA‐FLS in a dose‐dependent manner and that it does not affect healthy FLS. This is in line with earlier studies, which have demonstrated that α‐M can induce apoptosis via several molecular mechanisms.20 However, α‐M at the same concentration range did not affect the viability and apoptosis of normal FLS. Similarly, previous study also revealed little cytotoxicity of α‐M in normal cells and normal animals.21 Here we demonstrate that the process involves caspase‐3 and caspase‐9 because these were significantly increased by the induction of α‐M. Both of these caspases are known to either initiate or execute programmed cell death.22,23 ERK phosphorylation can promote proapoptotic functions under certain circumstances24 and α‐M was also shown to significantly increase the phosphorylation of ERK1/2 solely in RA‐ FLS. ERK activity can stimulate either intrinsic or extrinsic apoptotic pathways by inducing the release of mitochondrial cytochrome c or via caspase‐8 activa- tion.25 Our finding that levels of cytochrome c were increased in the cytoplasm of treated cells indicates that apoptosis was stimulated by cytochrome c release from the respiratory chain. This process is dependent on the presence of ROS26 and we observed significant increases in ROS in RA‐FLS following treatment. The ROS‐dependency of the process was confirmed following the addition of the ROS scavenger NAC, which was able to significantly reverse α‐M‐mediated apoptosis. Taken together, these results indicate that α‐M induces ROS‐dependent apoptosis via the ERK1/2 signaling pathway in RA‐FLS, in a comparable mechanism to that observed in human osteosarcoma cells.27 ROS accumulation and cytochrome c release has been previously observed in α‐M‐treated HL60 cells, which indicates that α‐M most likely targets mitochondria in the early phase, resulting in the initiation of apoptosis via cytochrome c release.28 Significant RH‐FLS death was observed following treatment with α‐M, whereas no effect was seen in non‐RA cells. This result demonstrates that the com- pound could offer a new therapy for RA because these cells promote synovial proliferation and hyperplasia that forms the driving force for the chronic inflammation and joint destruction observed in RA patients. These destruc- tive synovial cells produce excessive proinflammatory cytokines and matrix metalloproteinases that directly damage the tissue and extracellular matrix. These changes are supported by a cycle of remodeling of the rheumatoid pannus,26 which is driven by synovial angiogenesis. Inhibition of synovial angiogenesis and synovial hyperplasia is potentially possible through control of FLS by α‐M, which could prevent the development and progression of RA. 5 | CONCLUSIONS In summary, we show that α‐M induces apoptosis via the ROS/ERK1/2 signaling pathway in RA‐FLS. Since these cells drive RA progression then α‐M could offer a new treatment for RA through the abolition of these damaging cells. Recent improvements in drug delivery methods leading to increased bioavailability of α‐M support this development; however, future studies in animal models and humans are still required to determine the safety and clinical efficacy of the compound for treating RA. REFERENCES 1. Sweeney SE, Firestein GS. Rheumatoid arthritis: regulation of synovial inflammation. Int J Biochem Cell Biol. 2004;36:372‐378. 2. Edwards JC. Fibroblast biology. Development and differentiation of synovial fibroblasts in arthritis. Arthritis Res. 2000;2:344‐347. 3. Mor A, Abramson SB, Pillinger MH. The fibroblast‐like synovial cell in rheumatoid arthritis: a key player in inflammation and joint destruction. Clin Immunol. 2005;115:118‐128. 4. Abeles AM, Pillinger MH. The role of the synovial fibroblast in rheumatoid arthritis: cartilage destruction and the regulation of matrix metalloproteinases. Bull NYU Hosp Jt Dis. 2006;64:20‐24. 5. Nakatsuka K, Tanaka Y, Hubscher S, et al. Rheumatoid synovial LY3214996 fibroblasts are stimulated by the cellular adhesion to T cells through lymphocyte function associated antigen‐1/intercellular adhesion molecule‐1. J Rheumatol. 1997;24:458‐464.
6. Kumazaki M, Shinohara H, Taniguchi K, et al. Understanding of tolerance in TRAIL‐induced apoptosis and cancelation of its machinery by alpha‐mangostin, a xanthone derivative. Oncotarget. 2015;6:25828‐25842.
7. Kwak H‐H, Kim I‐R, Kim H‐J, Park B‐S, Yu S‐B. Mangostin induces apoptosis and cell cycle arrest in oral squamous cell carcinoma cell. Evid‐Based Complement Alternat Med. 2016;2016: 5352412.
8. Choi YH, Bae JK, Chae H‐S, et al. α‐Mangostin regulates hepatic steatosis and obesity through SirT1‐AMPK and PPARγ pathways in high‐fat diet‐induced obese mice. J Agricult Food Chem. 2015;63:8399‐8406.
9. Chen G, Li Y, Wang W, Deng L. Bioactivity and pharmacolo- gical properties of α‐mangostin from the mangosteen fruit: a review. Expert Opin Ther Pat. 2018;28:415‐427.
10. Lee C‐H, Ying T‐H, Chiou H‐L, et al. Alpha‐mangostin induces apoptosis through activation of reactive oxygen species and ASK1/p38 signaling pathway in cervical cancer cells. Oncotar- get. 2017;8:47425‐47439.
11. Chen L‐G, Yang L‐L, Wang C‐C. Anti‐inflammatory activity of mangostins from Garcinia mangostana. Food Chem Toxicol. 2008;46:688‐693.
12. Gopalakrishnan C, Shankaranarayanan D, Kameswaran L, Nazimudeen SK. Effect of mangostin, a xanthone from Garcinia mangostana Linn. in immunopathological & inflam- matory reactions. Indian J Exp Biol. 1980;18:843‐846.
13. Zuo J, Xia Y, Li X, Chen JW. Xanthones from Securidaca inappendiculata exert significant therapeutic efficacy on adjuvant‐ induced arthritis in mice. Inflammation. 2014;37:908‐916.
14. Zuo J, Xia Y, Li X, Ou‐Yang Z, Chen JW. Selective modulation of MAPKs contribute to the anti‐proliferative and anti‐inflammatory activities of 1,7‐dihydroxy‐3,4‐dimethoxyxanthone in rheumatoid arthritis‐derived fibroblast‐like synoviocyte MH7A cells. J Ethno- pharmacol. 2015;168:248‐254.
15. Zhang J, Chen Q, Wang S, et al. α‐Mangostin, a natural xanthone, induces apoptosis and ROS accumulation in human rheumatoid fibroblast‐like synoviocyte MH7A cells. Curr Mol Med. 2017;17:375‐380.
16. Bustamante MF, Garcia‐Carbonell R, Whisenant KD, Guma M. Fibroblast‐like synoviocyte metabolism in the pathogenesis of rheumatoid arthritis. Arthritis Res Ther. 2017;19:110.
17. Wilsdon TD, Hill CL. Managing the drug treatment of rheumatoid arthritis. Aust Prescr. 2017;40:51‐58.
18. Xu WK, Jiang H, Yang K, Wang YQ, Zhang Q, Zuo J. Development and in vivo evaluation of self‐microemulsion as delivery system for alpha‐mangostin. Kaohsiung J Med Sci. 2017;33:116‐123.
19. Zhao Y, Tang G, Tang Q, et al. A method of effectively improved alpha‐mangostin bioavailability. Eur J Drug Metab Pharmacokinet. 2016;41:605‐613.
20. Lee HN, Jang HY, Kim HJ, et al. Antitumor and apoptosis‐ inducing effects of alpha‐mangostin extracted from the pericarp of the mangosteen fruit (Garcinia mangostana L.) in YD‐15 tongue mucoepidermoid carcinoma cells. Int J Mol Med. 2016;37:939‐948.
21. Ibrahim MY, Hashim NM, Mohan S, et al. α‐Mangostin from Cratoxylum arborescens: an in vitro and in vivo toxicological evaluation. Arabian J Chem. 2015;8:129‐137.
22. Porter AG, Janicke RU. Emerging roles of caspase‐3 in apoptosis. Cell Death Differ. 1999;6:99‐104.
23. Wurstle ML, Laussmann MA, Rehm M. The central role of initiator caspase‐9 in apoptosis signal transduction and the regulation of its activation and activity on the apoptosome. Exp Cell Res. 2012;318:1213‐1220.
24. Lu Z, Xu S. ERK1/2 MAP kinases in cell survival and apoptosis. IUBMB Life. 2006;58:621‐631.
25. Cagnol S, Chambard J‐C. ERK and cell death: mechanisms of ERK‐induced cell death—apoptosis, autophagy and senes- cence. FEBS J. 2010;277:2‐21.
26. Dong J, Ramachandiran S, Tikoo K, Jia Z, Lau SS, Monks TJ. EGFR‐independent activation of p38 MAPK and EGFR‐dependent activation of ERK1/2 are required for ROS‐induced renal cell death. Am J Physiol Renal Physiol. 2004;287:F1049‐F1058.
27. Huang K, Chen Y, Zhang R, et al. Honokiol induces apoptosis and autophagy via the ROS/ERK1/2 signaling pathway in human osteosarcoma cells in vitro and in vivo. Cell Death Dis. 2018;9:157.
28. Matsumoto K, Akao Y, Yi H, et al. Preferential target is mitochondria in alpha‐mangostin‐induced apoptosis in human leukemia HL60 cells. Bioorg Med Chem. 2004;12: 5799‐5806.