Methylsulfonylmethane Inhibits RANKL-Induced Osteoclastogenesis in BMMs by Suppressing NF-κB and STAT3 Activities

Osteoclast differentiation is dependent on the activities of receptor activator NF-kB ligand (RANKL) and macrophage colony-stimulating factor (M-CSF). Given that RANKL plays a critical role in osteoclast formation and bone resorption, any new compounds found to alter its activity would be predicted to have therapeutic potential for disorders associated with bone loss. Methylsulfonylmethane (MSM) is a naturally occurring sulfur compound with well-documented anti-oxidant and anti-inflammatory properties; currently its effects on osteoclast differentiation are unknown. We sought to investigate whether MSM could regulate osteoclastogenesis, and if so, its mechanism of action. In this study, we investigated the effects of MSM on RANKL-induced osteoclast differentiation, together with STAT3’s involvement in the expression of osteoclastic gene markers. These experiments were conducted using bone marrow derived macrophages (BMMs) and cell line material, together with analyses that interrogated both protein and mRNA levels, as well as signaling pathway activity. Although MSM was not toxic to osteoclast precursors, MSM markedly inhibited RANKL-induced TRAP activity, multinucleated osteoclast formation, and bone resorptive activity. Additionally, the expression of several osteoclastogenesis-related marker genes, including TRAF6, c-Fos, NFATc1, cathepsin K, and OSCAR were suppressed by MSM. MSM mediated suppression of RANKL-induced osteoclastogenesis involved inhibition of ITAM signaling effectors such as PLCγ and Syk, with a blockade of NF-kB rather than MAPK activity. Furthermore, MSM inhibited RANKL-induced phosphorylation of STAT3 Ser727. Knockdown of STAT3 using shRNAs resulted in reduced RANKL-mediated phosphorylation of Ser727 STAT3, and TRAF6 in cells for which depletion of STAT3 was confirmed. Additionally, the expression of RANKL-induced osteoclastogenic marker genes were significantly decreased by MSM and STAT3 knockdown. Taken together, these results indicate that STAT3 plays a pivotal role in RANKL-induced osteoclast formation, and that MSM can attenuate RANKL-induced osteoclastogenesis by blocking both NF-kB and STAT3 activity.


Introduction
Bone remodeling describes the restructuring of existing bone, which is a delicately controlled balance between bone formation by osteoblasts and resorption by osteoclasts [1]. An imbalance in these processes can lead to excessive osteoclast-induced bone resorption, which causes rheumatoid arthritis and osteoporosis, and can encourage cancer metastases to the bone [2]. Osteoclasts are specialized bone-resorbing cells regulated by osteoblast through the synthesis of macrophage colony-stimulating factor (M-CSF) and receptor activator of NF-κB ligand (RANKL) [2,3]. RANKL-induced activation of RANK causes TNF receptor-associated factor 6 (TRAF6) recruitment in osteoclast precursor cells [4] and the sequential activation of mitogenactivated protein kinases (MAPKs) involving extracellular signaling-related kinase (ERK), p38, and Jun N-terminal kinase (JNK), and transcription factors such as nuclear factor-kappa B (NF-κB), activating protein 1 (AP-1), nuclear factor of activated T cells (NFATc1), and c-Fos [5]. The activation of these signaling effectors induces the expression of osteoclastic genes such as tartrate-resistant acid phosphatase (TRAP), cathepsin K (Cts K), and matrix metalloproteinase 9 (MMP-9), whose activities result in the development of multinucleated bone-resorbing osteoclasts [5,6].
The family of signal transducer and activator of transcription proteins (STATs) play a pivotal role in growth factor, prolactin, and various cytokine signaling pathways [7]. Recent evidence suggests that STATs, particularly STAT5b, play a central role in growth hormone (GH) signaling and osteoblast differentiation [8]. This finding is supported by our recent in-vitro studies showing that methylsulfonylmethane (MSM) enhanced GH-induced osteoblast differentiation via persistent activation of the Jak2-STAT5b signaling pathways [8]. Many studies have demonstrated the importance of STAT3 in bone physiology, with RANKL-mediated osteoclastogenesis diminished by the protein inhibitor of activated STAT3 (PIAS3) [9]. Indeed, recent data demonstrated a dual role for STAT3 depending on cell type (osteoblast or osteoclast) and its phosphorylation status [10].
Sulfur is an essential mineral needed for the biosynthesis of sulfur-containing amino acids, oxygen transport, and in the biosynthesis of various structural and functional proteins including collagen. MSM is an organic sulfur compound found in various fruits, vegetables, grains, and animals including humans [11]. MSM is bioavailable form of dietary sulfur; hence it can resolve the sulfur deficiencies and improve cartilage formation. Nevertheless, the effect of MSM on RANKL-induced osteoclastogenesis has yet to be determined.
In this study, we intended to clarify the anti-osteoclastogenic effect of MSM on RANKLinduced osteoclastogenesis in bone marrow macrophages (BMMs). In addition, we investigated whether STAT3 is directly involved in RANKL-induced osteoclastic marker gene expression. Our study provides important insights into the involvement of MSM-dependent STAT3 in RANKL-induced osteoclastogenesis.

Ethics Statement
This study was approved by the Institutional Animal Care and Use Committee, Konkuk University (Seoul, Korea); all procedures and routine animal care were in accordance with institution guidelines.

Osteoblast Differentiation Protocol
Primary mesenchymal stem cells were prepared from femur and tibia of 6-week-old BALB/c mice (Orient Bio, Gyeonggi-Do, Korea). Bone marrow was flushed out, then cultured, in α-MEM containing 10% FBS. Osteoblastic differentiation was performed as described [8].

Preparation of Bone Marrow-Derived Macrophages, Osteoclast Differentiation, TRAP Activity Assays and Staining
All mice were sacrificed under the CO 2 chamber, tibiae and femurs of 6-week-old BALB/c mice were used to obtain bone marrow cells. Residual cells, following the removal of red blood cells, were plated in 100-mm culture dishes and cultured in α-MEM containing 10% FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin. After incubation at 37°C for 1 day, the media for non-adherent cells was supplemented with 30 ng/ml M-CSF. 3 days later, adherent cells (now bone marrow macrophages (BMMs)) were harvested. BMM were seeded into a 96-well plate and cultured with 100 ng/ml RANKL and 30 ng/ml M-CSF for three to five days, with or without MSM. After OC differentiation, the cells were washed and then fixed with 4% paraformaldehyde, and stained for TRAP using leukocyte acid phosphatase cytochemistry kit (Sigma-Aldrich, MO, USA). Mature osteoclasts were categorized as TRAP-positive multinucleated cells with three or more nuclei. Osteoclast counts were conducted using a microscope. To measure TRAP activity, permeabilized cells were treated with 100 μl of citrate buffer containing 10 mM sodium tartrate and 5 mM p-nitrophenyl phosphate. After incubation at room temperature, cells mixes were transferred to fresh plates containing 50 μl of 0.1 N NaOH and absorbance read at 405 nm.

Cell Viability Assay
BMMs (1×10 4 cells/well) were cultured in 96-well plates for 24 h following incubation with 30 ng/ml M-CSF. Cells were exposed to the indicated concentrations of MSM (25-100 mM) in the presence of 30 ng/ml of M-CSF and 100 ng/ml of RANKL. Cell viability was determined over 4 days using the MTT (5 mg/ml) and Alamar blue assay. All experiments were performed in triplicates and measured three times.

Bone Resorption Assay
BMMs (1×10 4 cells/well) were plated on the fluoresceinamine-labeled chondroitin sulfate (FACS)-labeled CaP-coated plates (COSMO BIO Co., Japan). Cells were cultured for 5 days in α-MEM supplemented with 30 ng/ml of M-CSF and 100 ng/ml of RANKL, with or without MSM. On day 5, 100 μl of the conditioned medium was transferred from each well into the wells of a 96-well plate (with a black plate used for fluorescence measurements). Bone resorption assay buffer was added to each well and mixed. An excitation wavelength of 485 nm, with emission at 535 nm was used for fluorescence measurements.

EMSA
BMMs were stimulated by the addition of RANKL (100 ng/ml) for 10 min with or without MSM pretreatment for 1 h. Nuclear extracts were prepared using nuclear extract kit. NF-κB DNA binding activity was detected by EMSA, in which a DNA probes, used to bind active NF-κB protein in nuclear extracts. The treated and untreated nuclear extracts were incubated with a biotin-labeled transcription factor (TF-NF-κB) probe and then resolved on a non-denaturing 6% PAGE gel. Following this, the proteins were transferred to a nylon membrane and detected using chemiluminescence.
Transfection of STAT3 Short Hairpin RNA (shRNA) RAW 264.7 cells were transfected with 1 μg of STAT3 or non-target shRNA plasmid (Santa Cruz Biotechnology) using transfection reagent (Santa Cruz Biotechnology), in accordance with manufacturer's instructions. Two days later, cells were stimulated with 100 ng/ml of RANKL for 15 min, with or without MSM pretreatment for 1 h. The cells were harvested for western blot and real-time PCR.

Statistical Analyses
All data values were expressed as mean ± SEM. Statistical analysis was done with the student's t-test or analysis of variance (ANOVA) followed by Duncan's multiple range test using the SAS 9.3 software. A value of P < 0.05 was considered as significant.

MSM Suppresses RANKL-Induced Osteoclastogenesis in BMMs
BMMs were exposed to various concentrations of MSM for 96 h, then cytotoxicity assessed by MTT assay. As shown in Fig 1A, MSM did not affect the viability of osteoclast precursors. To investigate the direct role of MSM in osteoclast differentiation, we examined its effect on RANKL-induced osteoclast formation using mouse BMMs. BMMs exposed to RANKL and M-CSF efficiently differentiated into multinuclear osteoclasts, while MSM exposure dosedependently diminished the formation of TRAP-positive multinucleated cells, inhibiting osteoclast formation (Fig 1B). Similarly, MSM decreased RANKL-induced TRAP activity of osteoclasts in a dose-dependent manner (Fig 1D). We then investigated the effect of MSM on RANKL-induced bone resorption during osteoclastogenesis in BMMs. 100 mM MSM suppressed RANKL-induced bone resorption by 75% (Fig 1C). These data indicate that MSM exerts an inhibitory effect on RANKL-induced osteoclastogenesis.

MSM Suppresses RANKL-Induced Osteoclast Marker Gene and Protein Expression
To further elucidate the role of MSM in osteoclast differentiation, we examined its effect on the expression of osteoclast marker genes during RANKL-induced osteoclastogenesis by western blot and RT-PCR analyses. Our data revealed that MSM suppressed RANKL-induced TRAF6, c-Fos, NFATc1, and Cts K protein concentration in a dose-dependent manner, and at most time points (Fig 2A and 2B). Furthermore, MSM inhibited mRNA levels for TRAP, c-Fos, NFATc1, and OSCAR (Fig 2C), suggesting that MSM inhibits osteoclast formation, which is regulated by the RANKL and Ca 2+ signaling pathway. An increased RANKL/osteoprotegerin (OPG) ratio promotes the differentiation of osteoclasts and drives bone resorption. As shown in Fig 2D, the RANKL/OPG ratio in the MSC culture medium supplemented with MSM was reduced in comparison to controls. MSM inhibited the expression of RANKL mRNA in osteoblasts, while promoting the expression of OPG.

MSM Inhibits the RANKL-Induced Signaling Pathway in BMMs
We determined the effects of MSM on the activation of MAPK by RANKL in BMMs. MSM inhibited RANKL-induced phosphorylation of ERK in a dose-dependent manner but failed to diminish either JNK or p38 phosphorylation (Fig 3A). We then investigated the effect of MSM on the activation of PLCγ and Syk by RANKL in BMMs. RANKL-induced activation of both PLCγ and Syk were decreased by MSM in a dose-dependent manner (Fig 3B). These results suggested that MSM suppressed RANKL-induced osteoclastogenesis through blocking the expression of ITAM signaling molecules such as PLCγ and Syk.
To determine whether MSM suppresses the RANKL-induced activity of transcription factors by blocking NF-κB, we examined the effects of MSM on RANKL-induced NF-κB activation. As shown Fig 3C, MSM reduced RANKL-induced IKK phosphorylation and IκBα degradation in a dose-dependent manner. We also found that MSM significantly reduced RANKL-induced NF-κB signaling with diminished DNA binding of NF-κB as revealed by EMSA. These results demonstrated that MSM inhibited RANKL-stimulated osteoclastogenesis by blocking the activation of NF-κB, an essential factor for osteoclast differentiation.

MSM Attenuates RANKL-Induced Osteoclastic Marker Gene Expression by Blocking STAT3 Activity
Many studies have demonstrated the importance of STAT3 in bone physiology [12]. To investigate the effect of MSM on RANKL-induced phosphorylation of STAT3 we quantified the phosphorylation of Ser 727 STAT3 by western blot. As expected, MSM inhibited RANKL-induced phosphorylation of Ser 727 STAT3 (Fig 4A). To examine whether STAT3 is involved in RANKLinduced osteoclastogenesis we then used shRNA to target STAT3. As shown in Fig 4B, the shRNA construct significantly decreased the expression of STAT3, with diminished RANKLinduced phosphorylation of Ser 727 STAT3, and TRAF6.
We then explored the role of STAT3 in RANKL-induced osteoclast marker gene expression using control and STAT3 specific shRNAs. As shown in Fig 4C, the mRNA levels of STAT3, together with various osteoclastogenic marker genes substantially decreased by the shRNAmediated STAT3 knockdown. Collectively, these results demonstrate that STAT3 plays a pivotal role in RANKL-induced osteoclast formation and that MSM attenuated RANKL-induced osteoclastic marker gene expression by blocking STAT3 activity.

Discussion
MSM is a low molecular weight organic sulfur compound with anti-oxidant and anti-inflammatory activities [13]. We recently found that MSM enhances osteoblast differentiation in MSCs through activation of STAT5b. Moreover, in osteoblast-like cells MSM induced GH signaling through the Jak2/STAT5b pathway [8]. However, the effects of MSM have yet to be reported for osteoclasts or their differentiation. Our results showed that MSM inhibits RANKL-induced osteoclastogenesis by suppressing NF-κB and STAT3 activities in BMMs.
In order to further investigate the inhibitory effect of MSM in BMM, we tested the influence of MSM on viability and osteoclast differentiation in-vitro. Our results showed that MSM incubated with or without MSM for 1 h and then either exposed (or not) to RANKL (100ng/ml) for 10 min. Cell lysates were then blotted and immunostained with p-STAT3 and STAT3 antibodies. (B) RAW264.7 cells were transfected with STAT3 shRNA or a non-targeting shRNA for 48 h, then stimulated with RANKL (100 ng/ml) for 10 min. Cell lysates were prepared for western blot with antibodies as indicated. The relative levels of protein were determined using densitometry and normalized to β-actin. (C) RAW264.7 cells were transfected as in (B) and then stimulated with RANKL (100 ng/ml) for 24 h, with total RNA isolated using Qiagen. Expression of osteoclastic marker genes and STAT3 were examined using real-time PCR with GAPDH used as an internal control. Data shown are representative of three independent experiments. Asterisks indicate a significant increase by t-test (**p <0.01, ***p < 0.001). inhibits RANKL-induced osteoclast differentiation without causing any significant decrease in viability of BMMs. Therefore, MSM exerted an inhibitory effect on RANKL-induced osteoclastogenesis.
In RANKL-induced signaling, the cytoplasmic domain of RANK recruits adaptor molecules such as the TRAF6 to initiate a signaling cascade [14]. TRAF6 is also involved in the activation of transcription factors such as NF-κB, NFATc1, and c-Fos [15]. Intriguingly, MSM significantly suppressed RANKL-induced expression of osteoclast marker genes, including TRAF6, c-Fos, NFATc1, and Cts K. MSM also inhibited the expression of OSCAR, which is induced by NFATc1.
The MAPKs (ERK, JNK, and p38) have been reported to be activated by RANKL stimulation and to be associated with osteoclastogenesis [4]. In this study, we evaluated the effects of MSM on the activation of MAPKs and identified a dose-dependent suppression of ERK but not p38 or JNK phosphorylation. ERK is known to induce c-Fos during osteoclastogenesis with its inhibition shown to decrease osteoclast formation [16]. These results tentatively suggest that MSM may contribute to the suppression of NF-κB and calcium signaling primarily, rather than MAPK activity. In addition to RANKL-induced activation of TRAF6, ITAM-activated co-stimulatory signals regulate osteoclastogenesis via cross-talk with RANK-induced signaling [17]. Phosphorylated ITAMs (induced by RANKL) serve as docking sites for the SH2 containing signaling molecule Syk, which then activates the PLCγ-calcium pathway, eventually leading to activation of NFATc1 [18]. As expected, MSM inhibited both Syk phosphorylation and PLCγ, which are critical for the activation of calcium signaling. Furthermore, MSM-induced suppression of osteoclastogenesis would also appear to occur, at least in part, through inhibition of the adaptor molecule Gab2, which is rapidly phosphorylated upon RANK stimulation and associates with the cytoplasmic tail of RANK to regulate osteoclastogenesis [19].
MAPKs and NF-κB are the two main effectors in activating the NFATc1 promoter to encourage NFATc1 expression [20]. The classic NF-κB signaling pathway involves the activation of the inhibitor of κB (IκB) kinase (IKK) complex that then phosphorylates IκBs, targeting them for ubiquitin-dependent degradation [20]. Our results showed that MSM suppressed the phosphorylation of IKK, leading to reduced cytoplasmic degradation of IκB and the prevention of NF-κB's DNA-binding activity. Our results also indicated that inhibition of the NF-κB pathway suppressed RANKL-stimulated induction of NFATc1.
STAT3 is critical to the growth, survival, and differentiation of cells. It was reported that for the induction of RANKL and osteoclast formation, STAT3 activation in osteoblastic cells is required [21]. Although the role of STAT3 in osteoclast biology is somewhat controversial, the protein inhibitor of activated STAT3 (PIAS3) has been shown to negatively regulate RANKLmediated osteoclastogenesis [22]. Furthermore during RANKL induced osteoclastogenesis, both NFATc1 expression and STAT3 activation were inhibited by AG490 (Jak2 inhibitor) [12]. These results support our hypothesis that MSM inhibits RANKL-induced phosphorylation of STAT3 Ser 727 , showing that STAT3 plays a pivotal role in RANKL-induced osteoclast formation.
We previously reported that MSM induced osteoblast differentiation via the Jak2/STAT5b pathways in MSCs [8]. On the other hand, MSM reduced the osteoclastic differentiation of BMMs, as shown from Figs 1 to 4. RANKL promotes bone resorption whereas OPG is a "decoy receptor" that binds and neutralizes RANKL, thus inhibiting bone resorption [4]; osteoblast/ stromal cells express both of these genes [23]. We know that, the osteoblast precursor cells produce RANKL and the mature osteoblasts secrete OPG. Concurrent to this, in this study also MSM induced the secretion of OPG and inhibited RANKL production in osteoblasts (Fig 2C). This suggests that MSM regulate osteoclastogenesis indirectly via MSCs, besides its direct regulation through inhibiting STAT3/TRAF6 signaling axis, and the differentiation and function of osteoclasts.
We demonstrated that MSM suppressed RANKL-induced osteoclastogenesis in BMMs by inhibiting the activation of NF-κB. MSM reduces RANKL-induced osteoclastic marker gene expression by blocking STAT3 activity. We verified that RANKL-induced osteoclastogenesis is dependent on the coordinated mechanisms of NF-κB and STAT3, as mediated by MSM. These data elaborate MSM's mechanism of action in altering osteoclastogenesis and identify MSM as a potential therapeutic candidate for the treatment of disorders associated with bone loss.