**4. Discussion**

Osteoclasts are multinucleated cells of hematopoietic origin which are derived from the monocyte/macrophage in their ability to resorb bone, whereas osteoblast are derived from pluripotent mesenchymal stem cells and are involved in bone formation [1,2]. Since excessive bone resorption by osteoclasts causes an imbalance in bone regeneration and induces osteolytic diseases, osteoclasts are considered prime targets for the managemen<sup>t</sup> and treatment of bone diseases [2,3]. RANKL is a pro-osteoclastogenic cytokine and plays a crucial role in promoting osteoclastogenesis from osteoclast progenitor cells [4,5]. As has been well established in many earlier studies, RANKL binds to RANK expressed on the plasma membrane of osteoclast precursors and activates complex signaling cascades including NF-κB and NFATc1 for osteoclast differentiation [9,10]. Differentiation through activation of these signal transduction systems by RANKL is characterized by the formation of multinucleated giant cells [5,7]. This is a preliminary step in the maintenance, formation and function of the F-actin loop structure, which plays an important role in seal zone formation and resorption of bone mineral matrix in osteoclasts by activated TRAP [37,38]. According to the present findings, FST effectively inhibited RANKL-induced TRAP activation and F-actin ring formation without causing any significant cytotoxicity in RAW 264.7 cells, implying that FST suppressed osteoclast differentiation from osteoclast precursors at an early stage.

NF-κB, a transcription factor that plays a key role in inducing osteoclast differentiation, complexes with cytoplasmic IκB-α in the absence of osteoclastogenic induction signals and keeps it in an inactive form that tightly regulates its transcriptional activity for osteoclast differentiation [4,37]. However, the interaction of RANKL and RANK promotes the activation of the IκB kinase (IKK) complex, which phosphorylates IκB-α leading to ubiquitin-dependent degradation [6,7]. As a result, free NF-κB translocates to the nucleus and activates transcription of various genes involved in osteoclastogenesis [5]. Our results demonstrated that RANKL promoted the degradation of IκB-α in the cytoplasm and induced the translocation of NF-κB into the nucleus, both of which are essential for the activation of NF-κB but that these changes were completely inhibited by FST.

In the early stages of NF-κB activation and osteoclast differentiation, NFATc1 acts as a master regulator that enhances transcription of various osteoclast marker genes, which are highly expressed in the terminal differentiation stage to promote bone resorption [5,39]. In addition to blocking RANKL-induced NF-κB activation, FST inhibited NFATc1 expression in RANKL-treated cells. OPG treatment also completely blocked the expression of NFATc1. As further proof that FST was effectively inhibiting osteoclastogenesis, we showed that it attenuated RANKL-induced up-regulation of osteoclast marker genes such as TRAP, MMP-9, CTSK and OSCAR to levels seen in the control and OPG co-treated groups. Although further experiments are required to determine whether NFATc1 inhibition is the direct result of NF-κB inactivation, the present results indicate that inactivation of the NF-κB signaling pathway and inhibition of the expression of osteoclast marker genes associated with a decrease in NFATc1 expression are involved as important mechanisms in the anti-osteoclastogenic effect of FST.

A number of previous studies have shown that ROS, as specific secondary messengers, play a key role in the initiation of RANKL-stimulated osteoclast differentiation and bone resorption through similar pathways involving the activation of NF-κB and NFATc1 [10,12]. However, the accumulation of excessive ROS due to oxidative stress blocks osteoblast differentiation, suppresses osteoblast survival and acts to promote bone loss [12,13]. It has also been reported that a variety of natural products with antioxidant activity inhibit osteoclast differentiation by inhibiting ROS production [6,10,11]. Therefore, ROS can be considered a potential target for inhibition of osteoclast differentiation and prevention of bone loss. The present results showed that FST significantly suppressed ROS production by RANKL. Moreover, consistent with the results of previous studies [12,40], RANKL-induced osteoclast differentiation was completely inhibited when production of ROS was artificially blocked using NAC, indicating that FST blocks osteoclast differentiation by acting as a scavenger or inhibitor of ROS. In order to reduce the damage from oxidative stress in the face of excess production of ROS in cells, several transcription factors are known to be activated to increase the expression of downstream antioxidant enzymes [41,42]. One of these redox sensitive transcription factors, Nrf2 has recently been reported to attenuate osteoclast differentiation through the regulation of ROS production [42,43]. For example, Nrf2 deficiency improved RANKL-induced osteoclast differentiation [44], whereas local induction of nuclear Nrf2 weakened RANKL-mediated osteoclastogenesis [45]. Under normal conditions, Nrf2 is sequestered by Kelch-like ECH-associated protein 1 (Keap1) to the cytoplasm but becomes separated from Keap1 by oxidative or electrophilic stress and translocated into the nucleus. In the nucleus, Nrf2 binds to the antioxidant response elements to induce the transcription of target antioxidants and detoxifying enzymes including HO-1 and NQO-1 [42,43]. In this study, FST significantly increased expression of Nrf2 and its transcriptional targets, including HO-1 and NQO-1 in RANKL-treated RAW 264.7 cells. We also observed that FST increased phosphorylation and nuclear translocation of Nrf2 compared to the RANKL-alone stimulated group. The results presented, indicate that FST attenuates

osteoclast differentiation by decreasing RANKL-induced oxidative stress in osteoclast precursor cells through the activation of Nrf2 and its downstream genes.
