**3. Discussion**

Several pharmacotherapeutic drugs such as bisphosphonate, estrogen, and anti-RANKL antibodies are used to treat osteoporosis. However, these drugs are commonly associated with side effects such as medication-related osteonecrosis of the jaw, atrial fibrillation, and esophageal cancer [38–40]. Moreover, osteoporosis is poorly treated globally despite therapeutic advancements [41]. The proportion of patients with a hip fracture who were prescribed bone-protective medication decreased from 40% to 21% between 2001 and 2011 [42]. This undertreatment and poor adherence to drugs may be due to fear of adverse effects [41]. Therefore, therapeutic agents with no or minimal side effects are required. The protective role of carotenoids on bone resorption has been recently gaining attention. Carotenoids present in fruits and vegetables, such as lycopene and β-cryptoxanthin, and marine carotenoid astaxanthin show inhibitory effects on osteoclastogenesis [19,20,43–45].

Prior studies investigated the therapeutic effect of FX—a marine carotenoid—on osteoclastogenesis. Kose et al. [21] investigated the therapeutic effect of FX on alveolar bone resorption in a ligature-induced periodontitis mouse model. FX treatment significantly reduced the number of RANKL-positive osteoclasts located in the resorption lacunae and increased serum bone-specific alkaline phosphatase levels. Das et al. [22] demonstrated that the treatment of osteoclasts differentiated from RAW264.7 cells with 2.5–5 μM FX inhibited RANKL-induced osteoclast differentiation by inducing apoptosis. However, FX did not exert cytotoxic effects in osteoblast-like cells at 2.5 μM. Although not the pure FX compound, extracts containing FX from *Sargassum fusiforme* suppressed osteoclast differentiation, promoted osteoblast formation [46], and exhibited antiresorptive effects in an ovariectomized mice model. These studies sugges<sup>t</sup> that FX functions as a bone-protective agen<sup>t</sup> in osteoclast-mediated skeletal disorders. Here, we showed that FX inhibits both the canonical RANKL-induced osteoclastogenesis and the RANKL-independent TNF α/IL-6-induced differentiation of osteoclasts (Figure 2). Moreover, FX significantly attenuated osteoclastic bone resorption pit in a dose-dependent manner (Figure 3). Unlike the results of Das et al., our results demonstrated that the inhibitory effect of FX at ≤5 μM is not mediated by the apoptosis of osteoclast precursor cells (Figure 1B).

To our knowledge, there is no study focusing on the signaling pathway underlying the bone-protective activity of FX. Osteoclast differentiation is a multistep process that involves cell proliferation, commitment, fusion, and activation [2]. During this process, RANKL interacts with RANK to recruit TNF receptor-associated factor (TRAF) adaptor protein and induce downstream targets such as NF-κB, MAPK (JNK, ERK, and p38), PI3K, and Akt [47,48]. Previous studies also reported that TNFα and IL-6 can independently promote osteoclastogenesis in vitro of RANKL [7,8,49]. Moreover, TNFα can activate various signaling pathways, including p38 MAPK, ERK, and NF-κB [48,50]. Our study confirmed that FX downregulates ERK and p38 in both RANKL-dependent and -independent pathways, but not JNK or PI3K (Figure 6). The pharmacological action of nitrogen-containing bisphosphonates is also mediated by the inhibition of the MEK/ERK pathway [51].

NFATc1 is characterized as a master molecule of RANKL-induced osteoclast differentiation and can autoamplify its own expression [52]. Previous studies demonstrated that NFATc1 binds to the promoter region of *MMP-9* and *DC-STAMP* in osteoclasts, and increases mRNA expression [53,54]. In line with these findings, our study demonstrated a significant decrease in *DC-STAMP* mRNA expression (Figure 4B) and production of MMP-9 (Figure 5) upon FX treatment, in addition to NFATc1 downregulation.

Nrf-2 is a transcription factor expressed in various cell types, and is known as a regulator of cytoprotective genes against oxidative and chemical injuries [55]. Under normal quiescent conditions, Nrf2 is tethered to cytoplasmic protein Keap1. However, in cells exposed to stressful stimuli, Nrf2 is released from Keap1 and activated via phosphorylation. Nrf2 phosphorylation is important for its stabilization, and phospho-Nrf2 preferentially translocates to the nucleus [55,56]. Nrf-2 can also regulate osteoclast formation and activity. The overexpression of Nrf2 suppressed RANKL-induced osteoclast differentiation by increasing the level of antioxidant enzymes and locally inducing nuclear Nrf2-attenuated osteoclastogenesis [33,34,55]. In the present study, FX treatment promoted the nuclear translocation and phosphorylation of Nrf2 in both RANKL-dependent and -independent pathways (Figure 7). Consistent with this finding, nuclear translocation and phosphorylation of Nrf2 by FX was reported in studies of human keratinocytes and ischemia/reperfusion-induced neuron cells [35,36]. As Park et al. reported, the induction of Nrf2 dramatically suppresses the transcriptional activity of NFATc1 [57], the activation of Nrf2 by FX in our findings suggests the decreased expression of NFATc1 by FX (Figure 4A). Conversely, previous studies using HepG2 cells under oxidative-stress conditions demonstrated that ERK or p38 kinase activation is required for drug-mediated Nrf2 translocation [58,59]. However, this study and a previous study with NRK-52E cells [60] showed that the treatment of cells with FX reduced phosphor-ERK/p38 levels and induced

the nuclear translocation of phospho-Nrf2. The interaction between MAPK and Nrf2 pathways may vary depending on cell type, drug, or cell environment.

Limitations exist in this study. Although murine RAW264.7 cells are frequently used as in vitro models of osteoclast differentiation, it is necessary to confirm the beneficial effect of FX in human osteoclast precursors. Furthermore, in vivo studies are required to examine the therapeutic potential of FX in disease models. Nonetheless, to our knowledge, this is the first study to clarify the molecular regulatory mechanisms of FX in osteoclast differentiation.

We demonstrated that FX attenuates both RANKL-dependent and -independent osteoclastogenesis by downregulating ERK and p38 expression, and promoting the nuclear translocation of phospho-Nrf2 in RAW264.7 cells, as summarized in Figure 8. FX is confirmed to have no side effects and can be easily extracted from marine macro/microalgae. Therefore, FX represents a safe and inexpensive candidate drug for the treatment of various diseases accompanying the imbalance between osteoclasts and osteoblasts.

**Figure 8.** Signaling pathways and effects of FX during osteoclastogenesis. Inhibitory effect of FX is mediated by blocking the activation of ERK and p38, promoting Nrf2 nuclear translocation and phosphorylation, and subsequently downregulating NFATc1. Nrf2 induction was previously reported to suppress NFATc1 transcriptional activity [57].

#### **4. Materials and Methods**

#### *4.1. Cell Lines and Reagents*

RAW264.7, a murine macrophage cell line, was purchased from the American Type Culture Collection (ATCC; Rockville, MD, USA). Human lung adenocarcinoma cell line NCI-H3122 was a kind gift from Professor Jong-Seok Lee (Seoul National University College of Medicine, Seoul, South Korea). Human peripheral blood mononuclear cells (PMBCs) were obtained from 4 anonymous donors (Koma Biotech, Seoul, Korea). Dulbecco's Modified Eagle's Medium (DMEM) and Minimum Essential Medium Eagle-Alpha Modification (α-MEM) were purchased from Welgene (Daegu, Korea). Fetal bovine serum (FBS) was obtained from Atlas Biologicals (Fort Collins, CO, USA) and penicillin–streptomycin from Gibco (Carlsbad, CA, USA). Recombinant mouse TNFα, mouse IL-6, and human sRANKL were purchased from PeproTech (Rocky Hill, NJ, USA). CD14 MACS® MicroBeads were purchased from Miltenyi Biotec Inc. (Auburn, CA, USA).

Antibodies against procaspase-3, caspase-3, PARP, cleaved-PARP, ERK, phospho-ERK, p38, phospho-p38, JNK, phospho-JNK, PCNA, PI3K, and phospho-PI3K were purchased

from Cell Signaling Technology (Danvers, MA, USA). Anti-β-actin antibody was purchased from Enogene Biotech (New York, NY, USA), and anti-NFATc1, anti-phospho-p65, anti-Nrf2, and phospho-Nrf2 antibodies were from ABcam (Cambridge, UK).

ELISA kits for MMP-9 were obtained from R&D Systems (Minneapolis, MN, USA). The TRAP staining kit was acquired from Takara (Shiga, Japan), and the bone resorption assay kit from Cosmo Bio (Tokyo, Japan). The MTT assay kit was purchased from Sigma– Aldrich (St. Louis, MO, USA).

#### *4.2. Cell Viability Test Using MTT Assay*

Cell viability was determined using the MTT assay kit as per the manufacturer's protocol. Cells were seeded into 48 well tissue culture plates at a density of 2 × 10<sup>3</sup> per well in growth medium. After 24 h, logarithmic phase cells were incubated with different concentrations of FX for 5 days. Thereafter, MTT (5 mg/mL in PBS) was added to each cell. After 4 h, the medium was removed, and dimethylsulfoxide was added to solubilize MTT for an additional 4 h. After extraction with dimethylsulfoxide, optical density (OD) was measured at 495 nm. Percentage viability was calculated as (OD of drug-treated sample/OD of control) × 100.

#### *4.3. Culture and Differentiation of Cell Lines*

RAW264.7 cells were seeded in 48 well plates (2 × 10<sup>3</sup> per well), and cultured in α-MEM containing 10% FBS and 1% penicillin/streptomycin at 37 ◦C in 5% CO2/95% O2 in a humidified cell incubator. The culture medium was replenished every 3 days. Osteoclast differentiation was induced either by stimulation with sRANKL (50 ng/mL) for 5 days or costimulation with TNFα (50 ng/mL) and IL-6 (50 ng/mL).

Cryopreserved human PMBCs were thawed and washed, and CD14-positive cells were isolated using anti-CD14 antibody-coated microbeads. CD14+ monocytes were seeded at 1.5 × 10<sup>5</sup> cells/well in a α-MEM medium containing 10% FBS and M-CSF (50 ng/mL). After confirming that the cells remained attached the following day, they were treated with sRANKL (50 ng/mL) or TNFα (50 ng/mL)/IL-6 (50 ng/mL) under conditions of 0, 1, 2.5, and 5 μM FX. The culture medium was replaced every 4 days, and multinucleated TRAP-positive cells were counted after 17 days.

#### *4.4. Osteoclast Differentiation from RAW264.7 Cells and Osteoclast Activity Assays*

To evaluate the direct effects of FX on osteoclast differentiation, RAW264.7 cells were treated with 0, 1, 2.5, and 5 μM FX under sRANKL or TNF/IL-6 stimulation. Cells were stained with TRAP after 4 days of stimulation, and TRAP-positive multinucleated cells were enumerated under a light microscope.

Osteoclast activity was determined by measuring the area of resorption pits using calcium phosphate-coated 48 well plates according to the manufacturer's recommendation. RAW264.7 cells were washed once with α-MEM containing 10% FBS and seeded onto 48 well plates (2 × 10<sup>3</sup> per well). The following day, cells were treated with FX under sRANKL or TNF/IL-6 stimulation. On Day 5, the pit area was measured after adding 5% sodium hydrochlorite along the plate wall to remove RAW264.7 cells, and after washing with water. After air drying, microscopic images of all fields were acquired, and the resorbed pit area per well was measured using ImageJ software (NIH, Bethesda, MD, USA).
