Inhibitory Effect of Chrysanthemum zawadskii Herbich var. latilobum Kitamura Extract on RANKL-Induced Osteoclast Differentiation

Chrysanthemum zawadskii Herbich var. latilobum Kitamura, known as “Gujulcho” in Korea, has been used in traditional medicine to treat various inflammatory diseases, including rheumatoid arthritis. However, these effects have not been tested on osteoclasts, the bone resorbing cells that regulate bone metabolism. Here, we investigated the effects of C. zawadskii Herbich var. latilobum Kitamura ethanol extract (CZE) on osteoclast differentiation induced by treatment with the receptor activator of NF-κB ligand (RANKL). CZE inhibited osteoclast differentiation and formation in a dose-dependent manner. The inhibitory effect of CZE on osteoclastogenesis was due to the suppression of ERK activation and the ablation of RANKL-stimulated Ca2+-oscillation via the inactivation of PLCγ2, followed by the inhibition of CREB activation. These inhibitory effects of CZE resulted in a significant repression of c-Fos expression and a subsequent reduction of NFATc1, a key transcription factor for osteoclast differentiation, fusion, and activation in vitro and in vivo. These results indicate that CZE negatively regulates osteoclast differentiation and may be a therapeutic candidate for the treatment of various bone diseases, such as postmenopausal osteoporosis, rheumatoid arthritis, and periodontitis.


Introduction
Bone remodeling and metabolism are maintained by a sophisticated regulation between osteoblasts, bone matrixforming cells, and osteoclasts, bone-resorbing cells [1,2]. Imbalance between these cells is implicated in the development of bone diseases accompanied by low bone mineral density and bone destruction, such as postmenopausal osteoporosis, periodontitis, and rheumatoid arthritis (RA), which are caused by excessive differentiation and activation of osteoclasts [3][4][5].
Many previously published studies indicate that inflammatory cytokines, including TNF-, IL-1, IL-17, IFN-, and IL-4, which are produced during successful T-cell-based immune responses, directly regulate RANKL expression on osteoblasts as well as osteoclastogenesis and that inflammation affects bone metabolism [1,3]. Although CZE has an anti-inflammatory activity, the effect of CZE on bone metabolism has rarely been reported, with the exception that linarin, a component of CZE, prevents hydrogen peroxideinduced dysfunction in osteoblastic MC3T3-E1 cells [23]. However, its effect on osteoclasts still remains unclear.
In this study, we investigated the inhibitory effect of CZE on osteoclastogenesis and provided basic mechanisms and possibilities for the use of CZE as a traditional remedy against bone diseases, including osteoporosis, RA, and periodontitis. were purchased from Hyclone (Rockford, IL, USA). Soluble recombinant mouse RANKL was purified from insect cells as described previously [24], and recombinant human M-CSF was supplied by T Kim (KIOM, Daejeon, Korea). Antibodies against p-ERK, p-JNK, p-p38, p-I B , p-PLC 2, p-CREB, ERK, JNK, p38, I B , PLC 2, and CREB were purchased from Cell Signaling Technology (Danvers, MA, USA). Anti-NFATc1 and anti-c-Fos antibodies were purchased from Santa Cruz Biotechnology (Dallas, Texas, USA).

Cell Viability
Assay. Cell viability assays were performed using the EZ-Cytox Enhanced Cell Viability Assay Kit, (Itsbio, Korea), following the manufacturer's instructions. Briefly, bone-marrow-derived macrophages (BMMs), which act as osteoclast precursors, were plated in 96-well culture plates at a density of 1 × 10 4 cells per well with various concentrations of CZE (0, 2, 5, 10, 25, and 50 g/mL) for 1 day, or they were cultured with 25 g/mL of CZE under M-CSF treatment for 4 days. Cells were incubated with EZ-Cytox reagent for 4 h at 37 ∘ C. After incubation, the optical density was measured using an ELISA reader (Sunrise, Tecan, Switzerland) at 450 nm.

In Vitro Osteoclast Differentiation.
Murine osteoclasts were prepared from bone marrow cells (BM) as previously described [25]. BMs were collected from the tibiae and femora of 6-8-week-old mice by flushing the marrow space with phosphate-buffered saline (PBS). BMs were cultured with M-CSF (30 ng/mL) for 3 days in -minimal essential medium ( -MEM) containing 10% FBS, and attached cells were harvested and used as osteoclasts precursors (BMMs). To generate osteoclasts, BMMs were cultured with M-CSF (50 ng/mL) and RANKL (100 ng/mL) for 4 days. Fresh -MEM containing M-CSF and RANKL was replaced on day 3. Cells were fixed with 10% formalin and stained for TRAP. TRAP positive-multinuclear cells (TRAP + MNCs) containing more than three nuclei were counted as osteoclasts. In some experiments, total TRAP activity using p-nitrophenyl phosphate (Sigma, USA) as a substrate was measured at an absorbance of 405 nm as previously described [25].

Real-Time Quantitative PCR.
BMMs treated with or without CZE were cultured with M-CSF (30 ng/mL) and RANKL (100 ng/mL) for 4 days as described above. Total RNA was extracted from cultured cells by using Trizol reagent (Invitrogen, USA) on the indicated days. Then, 1 g of the total RNA was transcribed to first strand cDNA with random primers using Maxima Reverse Transcriptase (Thermo Scientific, IL, USA) according to the protocol provided by the supplier. Real-time PCR was performed using the VeriQuest SYBR Green qPCR Master Mix (Affymetrix, USA) and StepOnePlus Real-Time PCR Systems (Applied Biosystems, USA). To control for variation in mRNA concentrations, all results were normalized to the GAPDH housekeeping gene. Relative quantitation was performed using the comparative ΔΔ method, according to the manufacturer's instructions. Primers used in this study are listed in Table 1. Table 1: Nucleotide sequences of the primers used for real-time PCR in this study.

Measurement of Ca 2+ -Oscillation.
Ca 2+ -oscillation in osteoclasts by RANKL stimulation was measured as previously described with minor modification [27]. BMMs were cultured on the cover slips with RANKL in the presence or absence of CZE (25 g/mL). After 24 h of RANKL stimulation, intracellular Ca 2+ mobilization was measured using the fluorescence Ca 2+ indicator, Fura-2-acetoxymethyl ester (Fura-2AM, 5 M; TEFLabs, USA). In some cases, BMMs were cultured with RANKL in the absence of CZE for 1 day, and then treated with CZE to verify the acute effects of CZE at the indicated times. Cells were loaded with Fura-2AM for 50 min at room temperature and placed on a chamber connected with a perfusion system. Unloaded fluorescent dye was washed out with bath solution (10 mM HEPES, pH 7.4, 140 mM NaCl, 5 mM KCl, 1 mM MgCl 2 , 1 mM CaCl 2 , and 10 mM glucose; 310 milliosmole). With continuous perfusion of bath solution (37 ∘ C), the intracellular fluorescence intensity was measured using two excitation wavelengths (340 and 380 nm), and the emitted fluorescence (510 nm) was captured using a CCD camera. Collected images were digitized and analyzed by MetaFluor software (Ratio = F340/F380).

Statistical
Analysis. Data were analyzed using the Student's two-tailed -test and are presented as mean ± SD values, as indicated. A value of < 0.05 was considered statistically significant. All experiments were repeated at least twice and representative data are shown.

Effect of CZE on Cell Viability.
To assess the cytotoxicity of CZE on osteoclast precursors, BMMs were treated with various concentrations of CZE (0, 2, 5, 10, 25, or 50 g/mL) for 1 day. Various concentrations of CZE, up to 50 g/mL, did not affect the viability of BMMs (Figure 1(a)). All tested concentrations of CZE were shown to have viability levels comparable with that of control. In addition, BMMs were treated with 25 g/mL of CZE for 4 days. Cell viability was measured daily. There was no significant difference of viability between the control and CZE (25 g/mL)-treated cells during the 4 days of culture (Figure 1(b)).  inhibited the formation of mature osteoclasts from BMMs in a dose-dependent manner (Figures 2(a) and 2(b)). When the CZE exceeded 25 g/mL, mature osteoclasts were rarely formed. In addition, total TRAP activity from mono-, di-and multinuclear osteoclasts was significantly decreased as the CZE concentration increased (Figure 2(c)). These data suggest that the role of CZE is to repress osteoclast differentiation from the precursor cells to mature osteoclasts, and that it is not involved with the osteoclast fusion and activation steps.

Inhibitory
To confirm the inhibitory effect of CZE on osteoclast differentiation, the expression of osteoclast differentiation marker genes (Acp5, Oscar, CtsK, Tm7sf4, and Atp6v0d2) and a master transcription factor for osteoclast differentiation, Nfatc1, were measured during RANKL-induced osteoclast differentiation. As shown in Figure 3, CZE significantly inhibited the expression of all tested marker genes and Nfatc1. Together with Figure 1, these results indicate that CZE has an inhibitory effect on RANKL-induced osteoclast differentiation and formation without cytotoxicity.

Suppression of c-Fos and NFATc1 Expression via ERK Inactivation by CZE.
In RANKL-induced osteoclast differentiation, RANKL/RANK signaling induces the activation of NF-B and mitogen-activated protein kinases (MAPKs), followed by c-Fos expression, which results in the induction of NFATc1, a key transcription factor for osteoclastogenesis [1]. Therefore, we investigated the effect of CZE on the regulation of RANKL-induced signaling pathways. First, the BMMs were treated with or without CZE (25 g/mL) under RANKL and M-CSF treatment, and then, the activation of MAPKs and I B was measured by western blot analysis. As shown in Figure 4(a), the phosphorylation of ERK diminished in the CZE-treated cells compared to that of control cells. However, the activation of I B and other MAPKs (JNK and p38) was not significantly changed. Next, we measured the expression levels of c-Fos and NFATc1. When CZE was treated, the expression of c-Fos was dramatically repressed 6 h after treatment in osteoclast differentiation (Figure 4(b)). In addition, the induction of NFATc1 was significantly inhibited by CZE treatment (Figure 4(c)), coinciding with mRNA expression patterns (Figure 3). These results indicate that CZE inhibited the expression of c-Fos via the inactivation of ERK in RANKL-induced osteoclast differentiation. This inactivation leads to the repression of the expression of NFATc1, which regulates all the steps involved in osteoclast differentiation, fusion, and activation.

Breakdown of Intracellular Ca 2+ -Oscillation and Inhibition of PLC 2 and CREB Activation by CZE.
In addition to MAPK and NF-B activation, RANKL/RANK signaling also activate phospholipase C gamma 2 (PLC 2) and induces Ca 2+ -oscillation, followed by CREB activation [14,25,28]. CREB is critical for RANKL-stimulated NFATc1 and c-Fos induction in osteoclast precursors [14]. We first examined whether CZE affects the induction of Ca 2+ -oscillation by RANKL stimulation. BMMs were cultured with or without CZE under RANKL stimulation for 24 h. Intracellular Ca 2+ concentration was measured as described previously. Control cells exhibited typical Ca 2+ -oscillation as shown in Figure 5(a). However, CZE-treated cells showed an irregular Ca 2+ -oscillation pattern with significantly increased intensity, but without increased frequency ( Figure 5(b)). In addition, we acutely added CZE on control cells showing typical Ca 2+ -oscillation and then measured Ca 2+ mobilization. As shown in Figure 5(c), Ca 2+ -oscillation was defective in these cells, with a large Ca 2+ influx peak after CZE treatment. It seems that CZE may interact with some Ca 2+ channels, which Evidence-Based Complementary and Alternative Medicine contributes to a substantial Ca 2+ influx into the cells. We next examined whether RANKL-stimulated PLC 2 activation is affected by CZE treatment. Phosphorylation of PLC 2 in CZE-treated cells was significantly inhibited. In addition, the activation of CREB by RANKL was dramatically suppressed during osteoclast differentiation in CZE-treated cells. Collectively, these data demonstrate that CZE regulates not only MAPKs and NF-B activation, but also PLC 2 activation and RANKL-induced Ca 2+ -oscillation, which are important for CREB activation and c-Fos and NFATc1 induction in RANKL-stimulated osteoclast differentiation ( Figure 6).

Discussion
Although C. zawadskii Herbich var. latilobum Kitamura has routinely been used as a traditional remedy against several inflammatory diseases and the mechanisms for its antiinflammatory effects have been studied [16,17], comparatively little is known about its effect against inflammationrelated bone diseases such as RA and periodontitis or on bone cells (osteoclasts and osteoblasts). Only the effect of linarin on osteoblastic MC3T3 cells has been reported that it inhibits cytotoxicity and oxidative damage, and restores the mineralization function of hydrogen peroxide-treated osteoblasts. Control CZE (f) Figure 3: Effects of CZE on the expression of osteoclast differentiation marker genes. BMMs were cultured with RANKL and M-CSF treatment in the presence or absence of CZE (25 g/mL) for 4 days. The expression of osteoclast differentiation marker genes was measured by real-time PCR. The expression of mRNA levels was normalized with GAPDH and described as fold change of mRNA level. Data are expressed as the mean ± SD and are representative of at least three independent experiments. * < 0.05, and † < 0.01 versus control (0 g/mL CZE).
Linarin also suppresses RANKL expression induced by hydrogen peroxide and appears to have antiresorptive activity [23]. Here, we have elucidated an inhibitory effect of CZE via the reduction of NFATc1 expression in the differentiation and formation of osteoclasts, which cause bone destruction associated with inflammation-related bone diseases.
Previously, many studies have established that NFATc1 is a critical transcription factor for RANKL-mediated osteoclast differentiation, fusion, and activation. When BMMs are stimulated by RANKL, the expression of NFATc1 is induced through c-Fos and autoamplification by NFATc1 [9,10]. NFATc1-deficient embryonic stem cells do not form mature osteoclasts by RANKL treatment and overexpression of ectopic ca-NFATc1 in BMMs appropriately induces osteoclast differentiation from BMMs even in the absence of RANKL [9,25,29]. Recently, we had reported that NFATc1 is a key regulator of osteoclast fusion, which is an essential step for efficient bone resorption, via upregulation of ATP6v0d2 and dendritic cell-specific transmembrane protein (DC-STAMP), which are known as osteoclast fusion molecules as confirmed by genetic experiments [29]. Moreover, several reports showed that NFATc1 is implicated in the regulation of osteoclast function. The expression of TRAP, Cathepsin K, c-Src, and 3 integrin, which are involved in osteoclast-mediated bone resorption, is regulated by NFATc1 [9,10,12]. Furthermore, acidosis and RANKL signals in osteoclasts stimulate bone resorption via activation of Ca 2+ /calcineurin/NFAT pathway [30]. These results indicate that NFATc1 is a master key regulator of osteoclastogenesis. Therefore, in order to regulate excessive osteoclasts activity, which causes severe bone destruction in bone diseases, it is efficient and essential to control the expression of NFATc1 as a therapeutic target. In this study, our data demonstrate that CZE suppresses the expression of c-Fos and NFATc1 via inactivation of ERK, which contribute to RANKL-induced osteoclast differentiation ( Figure 4).
Previous studies elucidated that Ca 2+ -oscillation for NFATc1 induction is an essential process for osteoclastogenesis, and that ablation of Ca 2+ -oscillation causes impairment of osteoclastogenesis [12,31,32]. RANKL-stimulated Ca 2+ -oscillation is initiated approximately 24 h after RANKL treatment and is maintained until the formation of mature osteoclasts [15]. It seems that long-term Ca 2+ -oscillation is helpful for sustaining NFATc1 in the nucleus and for ensuring the transcriptional activation of NFATc1 required for terminal osteoclast differentiation [33]. Intracellular Ca 2+ originates from the extracellular space through plasma membrane channels or intracellular organelles such as the ER [34]. IP 3 receptors (IP 3 Rs), especially IP 3 R2 and IP 3 R3 in osteoclasts, mediate Ca 2+ release from the ER to the cytosol in response to IP 3 binding [13]. Recently, it has been reported that the Ca 2+ influx during osteoclastogenesis is regulated by plasma membrane-localized Ca 2+ channels, such as Orai1, TRPV4, and TRPV5 [35][36][37]. In Figure 5, BMMs treated with RANKL for 1 day produced typical Ca 2+ -oscillation. However, when BMMs were treated with CZE under RANKL stimulation for 1 day, the typical Ca 2+ -oscillation was interrupted; instead, abnormal biphasic pick, which is thought to be mediated by Ca 2+ influx, was induced. In addition, when RANKL-stimulated BMMs showing typical Ca 2+ -oscillation were acutely treated with CZE, an abnormal biphasic pick, implicating dramatically increased intracellular Ca 2+ concentration, [Ca 2+ ] was produced, and the typical Ca 2+ -oscillation by RANKL disappeared. Although it is unknown how CZE induces abnormal Ca 2+ influx in osteoclasts, CZE definitely ablated the typical Ca 2+ -oscillation induced by RANKL stimulation. In addition, CZE inhibited PLC 2 and CREB activation, which are important for osteoclast differentiation, followed by the repression of c-Fos and NFATc1 expression. These results indicate that CZE is an effective regulator of osteoclast differentiation and formation via the control of two main pathways induced by RANKL/RANK binding. In further studies, it will be interesting to explore how CZE regulates Ca 2+ influx and which Ca 2+ channel is affected by CZE. Detailed investigations of this mechanism will provide more insights into the effect of CZE on osteoclastogenesis and its relationship with the pathologies of bone diseases, such as osteoporosis, periodontitis, and rheumatoid arthritis.

Conclusion
Our results clearly demonstrate that the inhibitory effect of CZE on RANKL-stimulated osteoclastogenesis is mediated  by the repression of c-Fos and NFATc1 expression, which are critical for osteoclastogenesis, via ERK and PLC /Ca 2+ -oscillation/CREB signaling in osteoclasts. These findings reveal CZE as a traditional therapeutic agent against inflammatory bone diseases, such as rheumatoid arthritis and periodontitis.