Citation Nagasaki University ( 長崎大学 ), 博士 ( 歯学 )

Transcription

1 NAOSITE: Nagasaki University's Ac Title Author(s) Inhibitory effects of two ellagitan osteoclastogenesis 岩竹, 真弓 Citation Nagasaki University ( 長崎大学 ), 博士 ( 歯学 ) Issue Date URL Right This document is downloaded

2 Thesis Inhibitory effects of two ellagitannins, castalagin and punicalagin, on osteoclastogenesis Mayumi Iwatake Graduate School of Biomedical Sciences Nagasaki University June 2016 Doctor of Philosophy (Dental Science)

3 List of Journal Publications 1. Mayumi Iwatake, Kuniaki Okamoto, Takashi Tanaka, Takayuki Tsukuba, Castalagin Exerts Inhibitory Effects on Osteoclastogenesis Through Blocking a Broad Range of Signaling Pathways with Low Cytotoxicity, Phytotherapy Research, 29, (2015) DOI: /ptr Mayumi Iwatake, Kuniaki Okamoto, Takashi Tanaka, Takayuki Tsukuba, Punicalagin attenuates osteoclast differentiation by impairing NFATc1 expression and blocking Akt- and JNK-dependent pathways, Molecular and Cellular Biochemistry, 407, (2015) DOI /s

4 Table of Content Background 1 Chapter 1. Effects of castalagin on osteoclastogenesis 1.1 Introduction Materials and Methods Results Discussion Summary 18 Chapter 2. Effects of punicalagin on osteoclastogenesis 1.1 Introduction Materials and Methods Results Discussion Summary 37 Conclusions 38 Acknowledgements 39 References 40

5 Background Bone diseases have been an important public health issue according to the gradual increase in average life expectancy. A representative example of bone diseases is osteoporosis, which is defined as a disease characterized by low bone mass and structural deterioration of bone tissue. The predominant cause of the bone loss is due to imbalance of bone formation by osteoblasts and bone destruction by osteoclasts(ocls) through increase in oxidative stress. In spite of the presence of currently effective therapeutic agents, some agents have severe side-effects such as Bisphosphonate-Related Osteonecrosis of the Jaw termed as BRONJ. Therefore, it is necessary for development of new therapeutic agents having mild side effects such as natural compounds. Among various natural compounds, polyphenols have been reported to prevent chronic diseases including osteoporosis, periodontitis, and rheumatoid arthritis. Polyphenols can be classified into different classes, according to chemical structure (Fig. 1-1). For example, epigallocatechin 3-gallate (EGCG), which is a prototype of catechins distributed mainly in green tea, has inhibitory effects on osteoclastogenesis (Fig. 1-2). To explore a more effective polyphenols that reduces the activity of OCL without affecting their viability, we have screened 20 polyphenols in several groups of polyphenol. As a result of this screening test, we found ellagitannins, which include castalagin, punicalagin, pedunculagin, and, lambertianin A, have powerful inhibitory effects on OCL differentiation. Ellagitannins occur naturally in some fruits (pomegranate, strawberry, blackberry, raspberry), nuts (walnuts, almonds), and seeds. They form a diverse group of bioactive polyphenols with anti-inflammatory, anticancer, antioxidant and antibacterial. So far, there are only a few studies on the effects of ellagitannins on osteoclastogeneis. Based on the research background, the author analyzed two ellagitannins, castalagin and punicalagin, on the effects of OCL differentiation using receptor activator of nuclear factor-kappa B (NF-κB) ligand (RANKL)-induced stimulation method. The effects of castalagin are described in chapter 1, and those of punicalagin are in chapter 2. 1

6 Figure 1-1. The classification of phenolic compounds. DMSO 5 μm EGCG ** ** ** ** 10 μm EGCG 25 μm EGCG Figure 1-2. EGCG suppresses osteoclasts differentiation. 2

7 Chapter 1 Effects of castalagin on osteoclastogenesis

8 1.1 Introduction Castalagin is a rare plant polyphenol that is mainly distributed in oak and chestnut wood, and in the stem bark of Anogeissus leicarpus and Terminalia Avicennoides[6]. Castalagin is part of a particular group of ellagitannins that are composed of a series of highly hydrosoluble C-glucosidic variants [19]. Concerning the bioactivities of castalagin, only a few studies have been reported. A previous study using in vitro systems has reported that castalagin displays a powerful leishmanicidal activity compared with other ellagitanins [17]. Moreover, castalagin has been shown to have anti-cancer effects in vitro. Fridrich et al.[3] have reported that castalagin suppresses phosphorylation of the epidermal growth factor receptor in human colon carcinoma HT29 cells. However, beyond this, whether castalagin has pharmacological effects useful in the prevention of bone diseases such as osteoporosis, periodontitis and rheumatoid arthritis remains unknown. Using an in vitro system, we show here that castalagin has a powerful inhibitory effect on the differentiation of OCLs and their bone resorbing activity with an in vitro system. OCLs are bone resorbing multinucleated cells that are derived from monocyte/macrophage precursor cells [18]. Their differentiation pathway from the precursor cells into OCLs is responsible for the 2 essential cytokines: RANKL and macrophage colony-stimulating factor (M-CSF). When RANKL binds to the receptor, receptor activator of nuclear factor kappa-b (RANK), the key signaling pathways in OCL differentiation is activated: nuclear factor of activated T cells cytoplasmic-1 (NFATc1), nuclear factor kappa B (NF-κB), phosphatidylinositol 3-kinase (PI3K)/Akt, Jun N-terminal kinase (JNK), extracellular signal-regulated kinase (Erk), p38 mitogen-activated protein kinases (MAPKs)[2, 12, 20]. In addition to these signaling pathways, reactive oxygen species (ROS) have been shown to be essential for the OCL differentiation [5,11]. The transiently increased ROS triggers a signaling cascade involving TRAF 6, Rac 1, and NADPH oxidase 1 (Nox 1). In contrast, blocking ROS by addition of the reducing agent N-acetylcysteine or diphenylene iodonium, an inhibitor of Nox, completely inhibits the osteoclastogenesis via various signaling pathways, such as JNK, p38 MAPKs, and Erk [11]. Recently, we demonstrated that the expression of the anti-oxidant enzyme heme-oxigenase 1 (HO-1) by pharmacological or genetic induction 3

9 clearly inhibits OCL differentiation [15]. In this study, we report the inhibitory mechanisms of castalagin on OCL differentiation using in vitro OCL culture systems. 4

10 1.2 Materials and Methods Reagents Human M-CSF was purchased from Kyowa Hakko Kogyo (Tokyo, Japan). Recombinant human soluble RANKL was prepared as described previously [9]. Antibodies were purchased as follows: β-actin (Sigma-Aldrich, St. Louis, MO, USA), Src (Upstate Biotechnology, Lake Placid, NY, USA); anti-c-fms, anti-rank,, anti-cfos, and anti-nfatc1 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Abs specific for NF-κB, Erk1/2, phospho-erk1/2, Akt, phospho-akt, JNK, phospho-jnk, p38 MAPK, phospho-p38 MAPK, and inhibitor of nuclear factor kappa B alpha (IκBα), phospho-iκbα were purchased from Cell Signaling Technology (Danvers, MA, USA);and cathepsin K antibody was prepared as described previously [10]. The Osteo Assay Plate was purchased from Corning (Corning, NY, USA). All other reagents, including phenylmethylsulfonyl fluoride and the protease inhibitor cocktail, were obtained from Sigma-Aldrich Isolation of castalagin Isolation of castalagin was performed as described previously[17]. Briefly, castalagin was isolated from the wood of Castanea crenata, an aqueous acetone extract of the dried bark was separated by Sephadex LH-20 column chromatography with H 2 O containing increasing proportions of MeOH and finally with H 2 O-acetone (1:1, v/v). The fractions containing castalagin were collected and further separated by Diaion HP20SS chromatography with 10-20% MeOH. Castalagin was obtained as a white crystalline powder from H 2 O and identified by 1 H NMR spectroscopic comparison with an authentic sample. The structure of castalagin is shown in Fig.2. Figure 2. Structure of castalagin. 5

11 1.2.3 Cell culture Five-week-old male BALB/c mice were obtained from CLEA Japan, Inc. (Tokyo, Japan), and handled in our facilities under protocols approved by the Nagasaki University Animal Care Committee. Bone marrow-derived macrophages (BMMs) were isolated as described previously[15]. The BMMs were replated in culture plates and incubated in α-minimal essential medium (α-mem) containing 10% Fetal bovine serum (FBS) with 100 U/mL of penicillin and 100 µg/ml of streptomycin in the presence of M-CSF (30 ng/ml) and RANKL (50 ng/ml) for 72 h until the cells differentiated into multinucleated mature OCLs. The cells were fixed with 4% paraformaldehyde and stained for tartrate-resistant acid phosphatase (TRAP) activity using a previously described method [8]. TRAP-positive red-colored cells with 3 or more nuclei were considered mature OCLs. For bone resorption pit formation, BMMs were seeded onto Osteo Assay Plates coated with thin calcium phosphate films (Corning, NY, USA) and incubated with M-CSF and RANKL for 5 days until the mature OCLs resorbed the calcium phosphate film. Cells were dissolved in 5% sodium hypochlorite. Images of the resorption pit were taken with a reverse phase microscope (Olympus, Tokyo, Japan). The ratios of the resorbed areas to the total areas were calculated using Image J image-analysis software ( as described previously [13] Cell viability assay Cells seeded in 96-well cell culture plates were incubated with the Cell Counting Kit-8 (Dojindo, Kumamoto, Japan) for 1 h, and then the absorbance at 450 nm was measured with a microplate reader (Bio-Rad imark TM, Hercules, CA, USA). Castalagin was added the medium at the beginning Western blot analysis BMMs were stimulated with or without RANKL in the presence of M-CSF for the indicated time. Cells were rinsed twice with ice-cold PBS, and lysed in a cell lysis buffer (50 mm Tris-HCl [ph 8.0], 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 150 mm NaCl, 1 mm phenylmethylsulfonyl fluoride, and 6

12 proteinase inhibitor cocktail). The protein concentration of each sample was measured with BCA Protein Assay Reagent (Thermo Pierce, Rockford, IL, USA). An equal amount of protein (5 µg) was applied to each lane. After SDS-PAGE, proteins were electroblotted onto a polyvinylidene difluoride membrane. The blots were blocked with 3% milk in 5%Tris-buffered saline solution/tween(tbst) for 1 h at room temperature, probed with various antibodies overnight at 4 C, washed, incubated with horseradish peroxidase-conjugated secondary antibodies (Cell Signaling Technology, and Dako, Glostrup, Denmark), and finally detected with ECL-Plus (GE Healthcare Life Sciences, Tokyo, Japan). The immunoreactive bands were analyzed by LAS1000 (Fuji Photo Film, Tokyo, Japan) Immunofluorescence microscopy BMMs were cultured with M-CSF (30 ng/ml) and RANKL (50 ng/ml) for 72 h in the presence of 0-50 μm castalagin. The cells were fixed with 4% paraformaldehyde at 4 C for 30 min, permealized with 0.1% Triton X-100 for 10 min, and then blocked with 0.2% gelatin. For NF- B localization, anti-p65 antibody was diluted 1:100 in 0.2% gelatin and incubated overnight at 4 C, followed by incubation for 2 h at room temperature with Alexa 488 conjugated anti-mouse immunoglobulin diluted in 1:100 in 0.2% gelatin. Microscope images of fluorescence were digitized using a LSM 710 (Carl Zeiss, Germany) confocal microscope Reverse transcription and real-time quantitative PCR Total RNA was extracted using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA). Reverse transcription was performed using an oligo(dt) 15 primer (Promega, Madison, WI, USA) and Revertra Ace (Toyobo, Osaka, Japan). Quantitative real-time polymerase chain reaction(pcr) was performed using a MX3005P QPCR system (Agilent Technology, La Jolla, CA, USA). The cdna was amplified using Brilliant III Ultra-Fast SYBR QPCR Master Mix (Agilent Technology, La Jolla, CA, USA) according to the manufacturer s instructions. 7

13 The following primer sets were used: β-actin: 5'-ACCCAGATCATGTTTGAGAC-3' forward and 5'-GTCAGGATCTTCATGAGGTAGT-3' reverse, HO-1: 5'-CACGCATATACCCGCTACCT-3' forward and 5'-CCAGAGTGTTCATTCGAGCA-3' reverse, Statistical analysis All values are expressed as the means standard deviations for 3 independent experiments. Tukey-Kramer method was used to identify differences between concentrations when ANOVA indicated a significant difference (*P < 0.05 or **P < 0.01). Alternatively, 2-factor ANOVA was used. 8

14 1.3 Results Castalagin inhibits osteoclastogenesis in vitro To evaluate whether castalagin inhibits osteoclastogenesis, we first examined its effects on OCL differentiation from native BMMs treated with M-CSF (30 ng/ml) and RANKL (50 ng/ml). As shown in Fig 3A, TRAP activity staining showed that castalagin inhibited the formation of mononuclear and multinuclear OCLs. The number of TRAP-positive, multinucleated OCLs decreased after castalagin treatment even at the lower concentration of 1µM (Fig. 3B). However, the cell viability was still maintained even at the higher concentrations of 25 and 50 µm (Fig. 3C). These results indicate that castalagin strongly inhibits osteoclastogenesis in vitro in the RANKL-induced culture system at low concentrations, but has scarcely any cytotoxicity, even at higher concentrations Effects of castalagin on the bone resorbing activity of OCLs To further examine whether castalagin prevents the bone resorbing activity of OCLs, we performed a pit formation assay with BMM-derived OCLs after treatment with M-CSF and RANKL. As shown in Fig. 3A, castalagin significantly inhibited bone resorbing activity. When treated with 1µM castalagin, the bone resorbing activity was markedly reduced (Fig. 4A). The calculated resorption area of castalagin-treated OCLs was decreased compared with that of untreated OCLs (Fig. 4B). Thus, castalagin inhibits the physiological bone resorbing activity of OCLs. 9

15 A 0 µm 1 µm 5 µm 10 µm 25 µm 50 µm B C TRAP+ multinucleated cells/well * * * * * Concentration of castalagin (μm) Cell viability (absorbance 450nm) Concentration of castalagin (μm) Fig. 3. Effects of castalagin on osteoclast (OCL) differentiation from bone marrow-derived macrophages (BMMs). A) BMMs were cultured for 72 h with 30 ng/ml macrophage colony-stimulating factor (M-CSF), 50 ng/ml receptor activator of nuclear factor kappa-b ligand (RANKL), and castalagin at the indicated concentrations. Tartrate-resistant acid phosphatase (TRAP) staining was performed. B) The number of TRAP-positive OCLs was counted. C) Cell viability of the BMM-derived-OCLs was analyzed using a Cell Counting Kit. Data are shown as mean ± standard deviation (significance compared with M-CSF and RANKL * P < 0.05, ** P < 0.01). Results are representative of 3 independent experiments. 10

16 A 0 µm 1 µm 10 µm 25 µm 50 µm B 60 Bone Resorption Area (%) ** * ** * ** ** Concentration of Castalagin (μm) Fig. 4. Effects of castalagin on the bone resorbing activity of osteoclasts (OCLs). Bone marrow-derived macrophages (BMMs) were cultured with 30 ng/ml macrophage colony-stimulating factor (M-CSF), 50 ng/ml receptor activator of nuclear factor kappa-b ligand (RANKL), and castalagin at the indicated concentrations for 5 days. A) Castalagin inhibits RANKL-induced osteoclastic bone resorption. B) The resorption area was determined using ImageJ software (significance compared with the control. ** P < 0.01). 11

17 1.3.3 Effects of castalagin on intracellular signaling and expression levels of OCL marker proteins We further investigated the effects of castalagin on RANKL-induced intracellular signaling during the OCL differentiation from BMMs. Since OCL differentiation is known to be mainly regulated by several signaling pathways, including those modulated by NFATc1, NF-κB, PI3K/Akt, JNK, Erk, and p38 MAPKs [1]. Therefore, we examined the effects of castalagin on phosphorylation of Akt, JNK, IκBα, Erk, and p38 MAPKs by western blotting. When BMMs were pre-incubated without or with 10 μm of castalagin for 12 h, and then further incubated for the indicated times (5, 10, 15, and 30 min) after stimulation with RANKL. As shown in Fig. 4, castalagin sufficiently blocked phosphorylation of Akt, Erk, and p38. In addition, castalagin moderately inhibited the phosphorylation of JNK and IκBα (Fig. 5). These results indicate that castalagin interferes with all the 5 signaling pathways such as Akt, Erk, JNK, p38 MAPKs and IκBα-dependent pathways. Control 10 μm Castalagin p-akt AKT p-jnk JNK p-iκb IκB p-erk ERK p-p38 MAPK p38 MAPK (min) Fig. 5. Effects of castalagin on the essential signaling involved in osteoclast (OCL) differentiation. A) Bone marrow-derived macrophages (BMMs) were cultured with 30 ng/ml macrophage colony-stimulating factor (M-CSF) for 12 h in the presence of a vehicle or 10 μm castalagin. The cells were subsequently stimulated with 50 ng/ml receptor activator of nuclear factor kappa-b ligand (RANKL) for the indicated times (0, 5, 10, 15, and 30 min). The cell lysates were subjected to SDS-PAGE, followed by western blotting with antibodies to p-extracellular signal-regulated kinase (Erk), p-protein kinase B (Akt), p-p38, p-jun N-terminal kinase (JNK), p-inhibitor of nuclear factor kappa B alpha (IκBα), and ß-actin. Results are representative of 3 independent experiments. β-actin 12

18 To further analyze the effects of castalagin on osteoclastogenesis, we determined the expression levels of several OCL marker proteins by western blotting. RANK is a RANKL receptor, while c-fms is an M-CSF receptor. NFATc1, NFκB, and c-fos are essential transcriptional factors for OCL differentiation, whereas c-src is a non-receptor-type tyrosine kinase. Cathepsin K is the OCL specific cysteine proteinase. As shown in Fig.5, 5 µm castalagin considerably decreased the expression levels of RANK, c-fms, NFATc1, NFκB, c-src, and cathepsin K in OCLs. However, 50 µm castalagin had weak inhibitory effects on the protein levels of c-fos in OCLs (Fig. 6A). Since the phosphorylated IκBα enhances its degradation and nuclear transport of NF-κB as a transcription factor, we next determined the effects of castalagin on NF-κB activation as nuclear translocation of NF-κB p65. Immunofluorescence microscopy analysis revealed that nuclear localization of NF-κB p65 was observed in RANKL-stimulated control (0 µm) BMMs (Fig. 6B). However, treatment with 25 or 50μM castalagin prevented the nuclear accumulation of NF-κB p65 in the cells (Fig. 6B). These results indicate that castalagin caused significantly down-regulation of many essential factors involved in OCL differentiation, including RANK, c-fms, NFκB, c-src, NFATc1, and cathepsin K. Taken together with the data on signaling pathways, castalagin inhibited 6 major signaling pathways, including NFATc1, NF-κB, PI3K/Akt, JNK, Erk, and p38 MAPK, and it hampered the expression of several OCL marker proteins. 13

19 A B Castalagin (μm) Nucleic Acid Stain Anti-NfκB (p65) merged (DAPI) RANK 0 µm 10 µm c-fms NFATc-1 1 µm NFκB c-fos 5 µm c-src 25 µm Cathepsin K β-actin 50 µm Fig. 6. Effects of castalagin on protein expression of osteoclast (OCL) marker proteins. A) Bone marrow-derived macrophages (BMMs) were cultured with 30 ng/ml macrophage colony-stimulating factor (M-CSF) and 50 ng/ml receptor activator of nuclear factor kappa-b ligand (RANKL) for 72 h in the presence of castalagin at the indicated concentrations (0, 1, 5, 25, and 50 μm). The cell lysates were subjected to SDS-PAGE, followed by western blotting with antibodies specific to receptor activator of nuclear factor kappa-b (RANK), c-fms, nuclear factor of activated T cells cytoplasmic-1 (NFATc1), nuclear factor kappa B (NF-κB), c-fos, c-src, cathepsin K, and ß-actin. B) BMMs were cultured with M-CSF (30 ng/ml) and RANKL(50ng/mL) for 72 h in the presence of 0-50 μm castalagin. For NF- B localization, anti-p65 antibody was used as primary antibody, and followed by Alexa 488 conjugated-second antibody. Microscope images of fluorescence were digitized using LSM710 confocal laser microscope. Bar: 10μm 14

20 1.3.4 Castalagin inhibits HO-1 expression in OCLs Recently, we demonstrated that RANKL-induced suppression of HO-1 is required for osteoclastogenesis [15]. Finally, to determine the molecular mechanisms by which castalagin inhibits OCL differentiation, we analyzed the effects of castalagin on the expression levels of HO-1. Conversely, induction of HO-1 by pharmacological compounds suppresses the OCL differentiation. As shown in Fig.7, the time course of mrna levels of HO-1 in OCLs treated with 10 μm castalagin increased after a 3-h incubation period, but gradually decreased. Taken together, these results indicate that HO-1 induction by castalagin also participates in its inhibition of OCL differentiation. 2- factor ANOVA Time course: P< Castalagin treatment : P< Interactions: P< Relative HO-1 mrna expression a ab a ab a ab No treatment Castalagin treatment Time after treatment (h) Fig. 7. Effects of castalagin on mrna expression of heme oxygenase-1 (HO-1). Bone marrow-derived macrophages (BMMs) were cultured with 30 ng/ml macrophage colony-stimulating factor (M-CSF) and 50 ng/ml receptor activator of nuclear factor kappa-b ligand (RANKL) for the indicated times in the absence or presence of 10 μm castalagin. The mrna expression levels were determined by real time PCR using specific primers for HO-1. β-actin was used as a control. The graph shows the fold induction of each gene as compared to non-treated cells at 0 time point. Data are shown as mean ± standard deviation, and analyzed by 2-factor ANOVA ( a indicates significant difference from corresponding 0 time point. b indicates significant difference between non-treatment and castalagin treatment at the same time point.). Results are representative of 3 independent experiments. 15

21 1.4 Discussion In this study, we demonstrated that castalagin inhibits the OCL differentiation from BMMs into mature OCLs, and markedly prohibited the bone resorbing activity of OCLs. Importantly, cell viability was maintained even at higher castalagin concentrations of 25 and 50 µm. Castalagin interfered with the Akt, Erk, JNK and p38 MAPK-dependent pathways. Castalagin significantly caused down-regulation of many essential factors involved in OCL differentiation, including RANK, c-fms, c-src, NFκB, NFATc1, and cathepsin K. As one of its molecular mechanisms, induction of HO-1 by castalagin also participated in the inhibition of osteoclastogenesis. Thus, this study is the first report, to our knowledge, that castalagin has an inhibitory effect on osteoclastogenesis. The important pharmacological characteristic of castalagin is its inhibitory effects on osteoclastogenesis through a broad range of signaling pathways. Major signaling pathways for RANKL-induced OCL differentiation are known to include the NFATc1, NF-κB, PI3K/Akt, JNK, ERK, and p38 MAPK dependent pathways [2,12,20]. Castalagin inhibits all the 6 of these major signaling pathways, although it strongly blocks phosphorylation of Akt, Erk, and p38 MAPK, and sufficiently reduces phosphorylation of JNK and IκBα. However, in general, most natural compounds inhibiting osteoclastogenesis attenuates only some of these 6 major signaling pathways. For example, our previous studies indicated that kahweol, a coffee-specific diterpene, suppresses OCL differentiation through the abolishment of Erk phosphorylation, and partial inhibition of IκBα and Akt-dependent pathways, despite having no effect on the JNK and p38 MAPKs-dependent pathways [4]. Fisetin, a natural flavonoid, inhibits osteoclastogenesis via intense inhibition of the Erk, Akt and JNK-dependent pathways, and partial inhibition of the IκBα, but not the p38 MAPKs-dependent pathway [16]. Thus, inhibition by castalagin of all 6 major signaling pathways involved in OCL differentiation may lead to powerful inhibitory effects on osteoclastogenesis. Despite its the multiple inhibitory effects on RANKL-induced OCL differentiation signaling, castalagin has low cytotoxicity. Indeed, castalagin had more than a 50 % inhibitory effect on osteoclastogenesis of BMMs at 1 µm, whereas it displayed no cytotoxicity, even at 50 µm (Fig. 2). For example, in our previous studies 16

22 with similar analyses, kahweol has approximately 60 % inhibitory effects on osteoclastogenesis of BMMs at 10 µm, whereas it displays slightly cytotoxicity at 25 µm [4]. Similarly, fisetin also displays approximately 80 % inhibitory effects on osteoclastogenesis of BMMs at 10 µm, but with significant cytotoxicity at 20 µm [16]. Thus, castalagin has a lower cytotoxicity compared with other natural compounds like kahweol and fisetin, although it has a powerful inhibitory effect on the osteoclastogenesis of BMMs. To date, only a few of the ellagitannins have been shown to have inhibitory effects on OCL differentiation. Furosin, a hydrolyzable tannin, suppress RANKL-induced OCL differentiation from BMMs or macrophage cell line RAW264.7 cells via inhibition of JNK and p38-mapks [14]. Recently, geraniin, a well-studied tannin, has been shown to reduce the number of mature OCLs and inhibit bone resorption activity, through inhibition of MMP-9 expression [7]. However, those studies on the effects of furosin and geranin on osteoclastogenesis failed to report cell viabilities [7,14]. Therefore, the cytotoxicity of furosin or geranin on OCLs remains unknown. In this study, the lower concentration of castalagin significantly inhibited osteoclastogenesis, but cell viability was maintained even at the higher concentration, suggesting that it has powerful inhibitory effects on osteoclastogenesis, but low cytotoxicity. At present, the most common therapeutic agents for bone metabolism are bisphosphonates, which have strong inhibitory effects on OCLs but with high toxicity. Therefore, the development of agents with lower toxicity is required. Castalagin and its analogs may provide a new ideal agent for OCL inhibition. This is because high doses or prolonged use of bisphosphonates, which are the most commonly used for osteoporosis, causes several side effects including osteonecrosis of the jaw, atypical femoral fracture, and musculoskeletal pain. In conclusion, castalagin inhibits OCL differentiation through blocking a broad range of signaling pathways, but it has a lower cytotoxicity compared to other agents. The present study indicates that castalagin may have therapeutic effects in the treatment of bone diseases such as osteoporosis, periodontitis and rheumatoid arthritis. 17

23 1.5 Summary Castalagin is a rare plant polyphenol that is classified as a hydrolyzable tannin. Although it has anti-oxidant, anti-tumorigenic, and leishmanicidal effects, the utility of castalagin against bone diseases remain to be elucidated. Here, we investigated the effects of castalagin on the differentiation of osteoclasts, multinucleated bone resorbing cells. After stimulation with receptor activator of nuclear factor kappa-b ligand (RANKL), the formation of OCLs from bone marrow-derived macrophages (BMMs) was significantly inhibited by castalagin even at 1 µm. However, castalagin displayed little cytotoxicity at a higher concentration of 50 µm. The effects of castalagin on intracellular signaling during OCL differentiation showed that castalagin suppresses RANKL-stimulated phosphorylation of major signaling pathways including, Akt, extracellular signal-regulated kinase (Erk), Jun N-terminal kinase (JNK), p38 mitogen-activated protein kinases (MAPKs), and inhibitor of nuclear factor kappa B alpha (IκB). Moreover, following castalagin treatment, the protein levels of nuclear factor of activated T cells cytoplasmic-1 (NFATc1), a master regulator for OCL differentiation, and NF-κB were decreased. Thus, castalagin exerts inhibitory effects on osteoclastogenesis through blockage of a broad range of signaling pathways, but has low cytotoxicity. 18

24 Chapter 2 Effects of punicalagin on osteoclastogenesis

25 2.1 Introduction OCLs are bone-resorbing multinucleated cells derived from hematopoietic precursors of monocyte-macrophage lineage [18]. Bone resorption by OCLs is required for the formation of a resorption lacuna, characterized by the maintenance of an acidic ph and the secretion of specific hydrolases, such as cathepsin K and TRAP. Cathepsin K, an OCL-specific cysteine protease, is essential for bone matrix degradation [22,21,22], while TRAP is required for specialized electrochemical reactions associated with bone matrix resorption [23]. Therefore, resorption of the mineralized bone matrix by OCL is closely related to the regulation of these OCL-specific lysosomal enzymes. OCL differentiation is mainly regulated by two important cytokines, M-CSF and RANKL. M-CSF is a secreted cytokine that promotes the differentiation of hematopoietic stem cells into macrophages and OCLs [24]. RANKL, a member of the tumor necrosis factor superfamily, is a key cytokine that regulates osteoclastogenesis and bone resorption [25,26]. The RANKL-RANK triggers the activation of the essential signaling pathways for OCL differentiation: nuclear factor kappa B (NF-κB), phosphatidylinositol 3-kinase (PI3K)/Akt, Jun N-terminal kinase (JNK), extracellular signal-regulated kinase (Erk), p38 mitogen-activated protein kinase (MAPK) [2,12,20], and NFATc1/calcineurin dependent pathways [27]. Recent studies have shown that in addition to these signaling pathways oxidative stress plays important roles in osteoclastogenesis [11]. Indeed, the induction of heme oxygenase-1 (HO-1), an anti-oxidative stress enzyme, inhibits osteoclastogenesis [28]. We have recently demonstrated that the RANKL-mediated suppression of HO-1 promotes OCL differentiation and, conversely, the induction of HO-1 inhibits osteoclastogenesis[16]. Based on these findings, there is a strong possibility that compounds interfering with OCL differentiation signaling pathways and/or HO-1 induction have inhibitory effects on OCL differentiation. Punicalagin (2,3-(S)-hexahydroxydiphenoyl-4,6-(S)-gallagyl-D-glucose) is an ellagitannin, which is a high-molecular-weight bioactive polyphenol [29]. Punicalagin is mainly found in fruit pomegranates (Punica granatum) or in the leaves of Terminalia catappa. Punicalagin has been shown to have a variety of pharmacological effects, such as anti-oxidative, anti-inflammatory, and anti-tumor effects [30-33]. For 19

26 example, punicalagin attenuates oxidative stress and apoptosis in human placental trophoblasts [33]. In addition, punicalagin inhibits inflammation responses in lipopolysaccharide (LPS)-induced macrophages through Toll-like receptor 4-dependent NF-κB activation [32]. Punicalagin suppresses the proliferation of H-ras-transformed NIH3T3 cells, but only partially affects the proliferation of non-transformed NIH3T3 cells [33]. Interestingly, punicalagin has immunosuppressant activity, including the ability to inhibit nuclear factor of activated T cells (NFAT) in murine splenic CD4+ T cells [34]. Since NFATc1 is an essential factor in OCLs, we hypothesized that punicalagin may inhibit OCL differentiation. Therefore, in this study, we investigated the effects of punicalagin on osteoclastogenesis using in vitro culture systems. 20

27 2.2 Materials and Methods Reagents M-CSF was purchased from Kyowa Hakko Kogyo (Tokyo, Japan). Recombinant RANKL was prepared as described previously [9]. Antibodies (Abs) were purchased as follows: anti-β-actin (Cat. No. A5060, rabbit polyclonal Ab) was purchased from Sigma-Aldrich (St. Louis, MO, USA), Src (Cat. No , mouse monoclonal Ab) was purchased from Upstate Biotechnology (Lake Placid, NY, USA), and anti-c-fms (Cat. No. sc-692, rabbit polyclonal Ab), anti-rank (Cat. No. sc-9072, rabbit polyclonal Ab), and anti-nfatc1 (Cat. No. sc-7294, mouse monoclonal Ab) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Abs specific for phospho-erk1/2 (Cat. No. 9101S, Thr202/Tyr204, rabbit polyclonal Ab), phospho-jnk (Cat. No. 9751S, Thr183/Tyr185, rabbit polyclonal Ab), phospho-p38 (Cat. No. 9211S, Thr180/Tyr182, rabbit polyclonal Ab), phospho-inhibitor of nuclear factor kappa B alpha (IκBα) (Cat. No. 2859S, Ser32), and phospho-akt (Cat. No. 9271S, Ser473, rabbit polyclonal Ab) were purchased from Cell Signaling Technology (Danvers, MA, USA). Cathepsin K Abs were prepared as described previously [10]. The Osteo Assay Plate was purchased from Corning (Corning, New York, NY, USA). All other reagents, including phenylmethylsulfonyl fluoride (PMSF) and the protease inhibitor cocktail, were obtained from Sigma-Aldrich Isolation of punicalagin Isolation of punicalagin was performed as described previously [17]. Briefly, an aqueous acetone extract of the fresh peel of Punica granatum was separated by Sephadex LH-20 column chromatography with H 2 O containing an increasing proportion of MeOH and finally with H 2 O-acetone (1:1, v/v). The fractions containing punicalagin were collected and further separated by Diaion HP20SS chromatography with 20-40% MeOH. Punicalagin was obtained as a yellow amorphous powder and identified by 1 H NMR spectroscopic comparison with an authentic sample. The structure of punicalagin is shown in Fig.8. 21

28 Figure 8. Structure of castalagin Cell culture Five-week-old male BALB/c mice were obtained from CLEA Japan, Inc. (Tokyo, Japan) and handled in our facilities under the approved protocols of the Nagasaki University Animal Care Committee. The isolation of BMMs was performed as described previously [15]. The BMMs were replated in culture plates and incubated in α-minimal essential medium (α-mem) (Wako Pure Chemicals, Code: , bicarbonate buffered with L-glutamine) containing 10% fetal bovine serum (FBS) with 100 U/mL penicillin and 100 µg/ml streptomycin in the presence of M-CSF (30 ng/ml) and RANKL (50 ng/ml) for 72 h until the cells differentiated into multinucleated mature OCLs. The cells were fixed with 4% paraformaldehyde and stained for TRAP activity using a previously described method [8]. TRAP-positive red-colored cells with 3 or more nuclei were considered mature OCLs. Murine monocytic RAW-D cells were kindly provided by Prof. Toshio Kukita (Kyushu University, Japan) and cultured in α-mem containing 10% FBS with RANKL (50 ng/ml) [36]. For bone resorption pit formation, BMMs were seeded onto Osteo Assay Plates coated with thin calcium phosphate films (Corning, New York, NY, USA) and incubated with M-CSF and RANKL for 5 days until mature OCL resorbed the calcium phosphate film. Cells were dissolved in 5% sodium hypochlorite. Images of the resorption pit were taken with a reverse phase microscope (Olympus, Tokyo, Japan). The ratios of the resorbed areas to the total areas were calculated using Image J image-analysis software ( as described previously [13]. 22

29 2.2.4 Cell viability assay Cells seeded in 96-well cell culture plates were incubated with the Cell Counting Kit-8 (Dojindo, Kumamoto, Japan) for 1 h, and then the absorbance at 450 nm was measured with a microplate reader (Bio-Rad imark TM, Hercules, CA, USA) Western blot analysis BMMs were stimulated with or without RANKL in the presence of M-CSF for the indicated amount of time. Cells were rinsed twice with ice-cold phosphate buffered saline (PBS, ph 7.4), and lysed in a cell lysis buffer (50 mm Tris-HCl [ph 8.0], 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 150 mm NaCl, 1 mm PMSF, and proteinase inhibitor cocktail). The protein concentration of each sample was measured with BCA Protein Assay Reagent (Thermo Pierce, Rockford, IL, USA). Lysate protein (5 µg) was applied to each lane. After sodium dodesyl sulfate polyacrylamide gel (SDS-PAGE), proteins were electroblotted onto a polyvinylidene difluoride membrane. The blots were blocked with 5% bovine serum albumin/ Tris buffered saline with 0.1% Tween 20 (TBS-T) for 1 h at room temperature, probed with various primary Abs overnight at 4 C. The concentrations of primary Abs were as followes: anti-β-actin (1:20000, 2.5μl (10 times dilution)/5 ml), anti-src (1:1000, 5μg/5 ml), anti-c-fms (1:1000, 1μg/5ml), anti-rank (1:1000, 1μg/5ml), anti-nfatc1 (1:1000, 1μg/5ml) anti-phospho-erk1/2 (1:1000, 5μl/5ml), anti-phospho-jnk (1:1000, 5μl/5ml), anti-phospho-p38 (1:1000, 5μl/5ml), anti-phospho-iκbα (1:1000, 5μl/5ml), and anti-phospho-akt (1:1000, 5μl/5ml) and anti-cathepsin K Abs (1:1000, 5μl/5ml). Subsequently, the samples were washed, incubated with horseradish peroxidase -conjugated secondary Abs (anti-rabbit IgG, Cat. No. 7074, 1:2000, and anti-mouse IgG, Cat. No. 7076, 1:2000; Cell Signaling Technology), and finally detected with ECL-Plus (GE Healthcare Life Sciences, Tokyo, Japan). The immunoreactive bands were analyzed by LAS-4000mini (Fuji Photo Film, Tokyo, Japan). 23

30 2.2.5 Immunofluorescence BMMs were cultured with M-CSF (30 ng/ml) and RANKL (50 ng/ml) for 72 h in the presence of 0-50 μm punicalagin. The cells were fixed with 4% paraformaldehyde at 4 C for 30 min, permeabilized with 0.1% Triton X-100 for 10 min, and then blocked with 0.2% gelatin. For α-tublin localization, anti-α-tublin antibody (anti-rat, Cat. No. sc-53029; Santa Cruz Biotechnology) was diluted 1:100 in 0.2% gelatin and incubated overnight at 4 C, followed by incubation for 2 h at room temperature with Alexa Fluor 488 conjugated anti-mouse immunoglobulin and Alexa Fluor 594 conjugated phalloidin (Cat. No. A-12381; Invitrogen) diluted 1:100 in 0.2% gelatin. Microscope fluorescence images were digitized using a Carl Zeiss LSM 710 confocal microscope Reverse transcription and real-time quantitative PCR Total RNA was extracted using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA). Reverse transcription was performed using an oligo(dt) 15 primer (Promega, Madison, WI, USA) and ReverTra Ace (Toyobo, Osaka, Japan). Quantitative real-time PCR was performed using a MX3005P QPCR system (Agilent Technology, La Jolla, CA, USA). The cdna was amplified using Brilliant III Ultra-Fast SYBR QPCR Master Mix (Agilent Technology, La Jolla, CA, USA) according to the manufacturer s instructions. The following primer sets were used: β-actin:5'-acccagatcatgtttgagac-3' forward and 5'-GTCAGGATCTTCATGAGGTAGT-3' reverse, HO-1:5'-CACGCATATACCCGCTACCT-3' forward and 5'-CCAGAGTGTTCATTCGAGCA-3' reverse, Statistical analysis All values were expressed as means ± standard deviations (SD) for 3 independent experiments. The data were analyzed by the Tukey-Kramer method when ANOVA indicated a significant difference between concentrations (*P < 0.05 or **P < 0.01). 24

31 2.3 Results Punicalagin inhibits differentiation from macrophage into OCLs. We first investigated the effects of punicalagin on RANKL-induced OCL differentiation in the murine monocytic RAW-D cells, which have a high capacity for differentiation into OCLs. Upon stimulation with RANKL (50 ng/ml), RAW-D cells differentiated into TRAP-positive mutinucleated OCLs. As shown in Fig 2A, TRAP staining revealed that punicalagin inhibited the formation of mononuclear and multinuclear OCLs (Fig 9A). Punicalagin treatment decreased the number of TRAP-positive multinucleated OCLs in a dose-dependent manner (Fig. 9B). The cell viability of OCLs treated with 1 5 µm punicalagin was indistinguishable from that of untreated cells; however, treatment with µm punicalagin significantly decreased the cell viability of OCLs (Fig. 9C). To examine whether similar effects were observed in native cells, we tested the effects of punicalagin on RANKL-mediated osteoclastogenesis in BMMs. TRAP staining revealed that punicalagin dose-dependently suppressed OCL formation (Fig. 10A). Punicalagin also decreased the number of TRAP-positive OCLs in a dose-dependent manner (Fig. 10B). The cell viability of OCLs treated with punicalagin was indistinguishable from that of untreated cells (Fig. 10C). Importantly, punicalagin failed to attenuate cell viability even at the higher concentrations of µm (Fig. 10C). These results indicate that punicalagin inhibits osteoclastogenesis in RANKL-induced macrophages derived from RAW-D cells and native BMMs, although cell viability following punicalagin treatment was different between RAW-D cells and native BMMs. 25

32 A 0 µm 1 µm 5 µm 100 µm 10 µm 25 µm 50 µm B 500 µm 0 µm 1 µm 5 µm 10 µm 25 µm 50 µm C 1000 D 120 TRAP-positive multinucleated cells/well * ** * * Cell viablity (%) * * Concentration of punicalagin (μm) * * * Concentration of punicalagin(μm) Fig. 9. The effects of punicalagin on osteoclast (OCL) differentiation from RAW-D cells (A, B) RAW-D cells were cultured for 72 h with the indicated concentrations of punicalagin in the presence of RANKL(50 ng/ml). TRAP staining was performed after fixation. The data are representative of 3 independent experiments. The number of TRAP-positive multinucleated OCLs at 72 h of culture was counted. (A); Magnified 20x, (B); Magnified 4x. (C) The cell viability of RAW-D cell-derived OCLs at 72 h of culture was analyzed using the Cell Counting Kit. The data shown in panels B and C are the mean ± SD of 3 independent experiments. The asterisks indicate statistical significance compared to the control cells without punicalagin, ** P <

33 A 0 µm 1 µm 5 µm 100 µm 10 µm 25 µm 50 µm B 500 µm 0 µm 1 µm 5 µm 10 µm 25 µm 50 µm C 1200 D 120 TRAP-positive multinucleated cells/well ** ** ** * ** Concentration of punicalagin (μm) Cell viablity (%) Concentration of punicalagin(μm) Fig. 10. The effects of punicalagin on osteoclast (OCL) differentiation from BMMs (A, B) Bone marrow-derived macrophages (BMMs) were cultured for 72 h with the indicated concentrations of punicalagin in the presence of macrophage colony-stimulating factor (M-CSF) (30 ng/ml) and receptor activator of nuclear factor kappa-b ligand (RANKL) (50 ng/ml). Tartrate-resistant acid phosphatase (TRAP) staining was performed after fixation. The data are representative of 3 independent experiments. (A); Magnified 20x, (B); Magnified 4x. (C) The number of TRAP-positive multinucleated OCLs at 72 h of culture was counted. (D) The cell viability of BMM-derived OCLs at 72 h of culture was analyzed using the Cell Counting Kit. The data shown in panels B and C are the mean ± SD of 3 independent experiments. The asterisks indicate statistical significance compared to the control cells without punicalagin, * P < 0.05, ** P <

34 2.3.2 Punicalagin differently induces HO-1 in BMM-derived OCLs and RAW-D-derived OCLs Our recent studies have shown that RANKL induces the suppression of the cytoprotective enzyme HO-1. Therefore, we examined whether punicalagin induces the upregulation of HO-1. Real-time PCR analysis showed that punicalagin treatment of RAW-D cells increased HO-1 mrna levels after 48 h of incubation, but no induction was observed between 1 and 24 h after treatment initiation (Fig. 11). However, in BMMs, punicalagin caused marked increases in HO-1 mrna expression that reached a peak at 3 h post-induction (Fig. 11). These results suggest that the differential time course of HO-1 induction between RAW-D cells and BMMs. Relative mrna expression of HO-1 (RAW-D) ** * ** ** Relative mrna expression of HO-1(BMMs) ** ** ** Time after treatment(h) Time after treatment(h) Fig. 11. The effects of punicalagin on the mrna expression of heme oxygenase-1 (HO-1). RAW-D cells were cultured for 72 h with RANKL (50 ng/ml) in the absence or presence of 10 μm punicalagin. BMMs were cultured with macrophage colony-stimulating factor (M-CSF) (30 ng/ml) and receptor activator of nuclear factor kappa-b ligand (RANKL) (50 ng/ml) for the indicated times in the absence or presence of 10 μm punicalagin. The mrna expression levels were determined by real time PCR using specific primers for HO-1. β-actin was used as a control. The graph shows the fold induction of each gene as compared to non-treated cells at time point 0. Data are shown as the mean ± standard deviation (SD). Significance was determined as compared to treated cells at time point 0. * P < 0.05, ** P < Results are representative of 3 independent experiments. 28

35 2.3.3 Punicalagin impairs the bone-resorption activity of OCLs Next, we studied the effects of punicalagin on the bone-resorption activity of BMM-derived OCLs. Pit formation assay was performed with BMM-derived OCLs. The plastic dish caused a rapid differentiation, but cells on the dentin or bone slice showed a slow differentiation. Therefore, for the pit assay we cultured the cells for 5 days, although we did them on plastic dishes for 3 days. Untreated OCLs had moderate resorbing activity, whereas punicalagin treatment reduced resorbing activity in a dose-dependent manner (Fig. 12A). We measured the calculated resorption area of punicalagin-treated OCLs. Results indicated that OCLs treated with more than 1 µm punicalagin showed significantly decreased resorption compared to that of untreated OCLs (Fig. 12B). These results indicate that punicalagin attenuates the bone-resorption activity of OCLs. A B µm 1 µm 5 µm 10 µm 25 µm 50 µm 500 µm Bone Resorption Area (%) ** ** ** ** Concentration of Punicalagin (μm) Fig. 12. The effects of punicalagin on the bone-resorption activity of OCLs. BMMs were cultured for 5 days with the indicated concentrations of punicalagin in the presence of macrophage colony-stimulating factor (M-CSF) (30 ng/ml) and receptor activator of nuclear factor kappa-b ligand (RANKL) (50 ng/ml). (A) Punicalagin inhibits RANKL-induced osteoclastic bone resorption. The data are representative of 3 independent experiments. (B) The resorption area was determined using Image J software. The data are shown as the mean ± standard deviation (SD) of 3 independent experiments. The asterisks indicate statistical significance compared to the control cells without punicalagin, ** P <

36 2.3.4 Punicalagin reduces multinucleation and actin-ring formation in OCLs During a series of TRAP staining experiments, we noticed that punicalagin treatment kept mononuclear cells, but not multinucleated cells, TRAP-positive. Given the importance of multinucleation in OCL differentiation, we investigated whether punicalagin had effects on multinucleation and actin-ring formation in OCLs. For this purpose, we performed immunofluorescence staining of F-actin and α-tublin in BMM-derived OCLs. In the absence of RANKL, BMM-derived macrophages did not undergo actin-ring formation (Fig. 13). However, RANKL-induced OCLs exhibited multinucleation (DAPI-staining, blue) and actin-ring formation (phalloidin-staining, red) (Fig. 13). Under these conditions, punicalagin dose-dependently reduced multinucleation and inhibited the actin-ring formation of OCLs (Fig. 13). In particular, cells treated with more than 50 µm punicalagin displayed a mononucleated and weaker phalloidin-staining (Fig. 13). Thus, punicalagin prevents multinucleation and actin-ring formation in OCLs. 30

37 DAPI α-tublin phalloidin merge Control (-RANKL) 20 µm Punicalagin 0 µm Punicalagin 1 µm Punicalagin 5 µm Punicalagin 10 µm Punicalagin 25 µm Punicalagin 50 µm Fig.13. The effects of punicalagin on actin-ring formation and multinucleation in osteoclasts (OCLs). The BMM-derived OCLs were fixed on glass cover-slips, permeabilized with 0.1% Triton X-100 in PBS, and then allowed to react with phalloidin for F-actin (red) or antibodies for α-tublin (green). After washing, the samples were incubated with a fluorescence-labeled secondary antibody and then were visualized by confocal laser microscopy. Bar: 50 μm. 31

38 2.3.5 Punicalagin perturbs intracellular signaling and the expression levels of OCL marker proteins To analyze the detailed mechanisms of inhibition of OCL differentiation by punicalagin, we examined the effects of punicalagin on RANKL-induced intracellular signaling pathways during OCL differentiation from BMMs. Since signaling cascades involving p38 MAPK, JNK, IκBα, Erk, and Akt are important for osteoclastogenesis [1], western blotting analysis of the phosphorylation of these pathways was performed. BMMs were pre-incubated with 10 μm punicalagin for 12 h, and subsequently stimulated with RANKL. Punicalagin strongly inhibited the phosphorylation of Akt and JNK (Fig. 14). In addition, punicalagin weakly inhibited the phosphorylation of IκBα, Erk, and p38 (Fig. 14). These results indicate that punicalagin blocks RANKL-stimulated signaling cascades via mainly Akt and JNK-dependent pathways and weakly by IκBα, Erk, and p38-dependent pathways. Next, we analyzed the expression levels of OCL marker proteins by western blotting (Fig. 15). RANK is a type I membrane protein that is expressed on the cell surface receptor for RANKL [26], whereas c-fms is an M-CSF receptor. NFATc1 is a master regulator of OCL differentiation that acts through Ca 2+ /calmodulin-dependent calcineurin [27], while c-fos is the proto-oncogene essential for OCL differentiation. c-src is a non-receptor-type tyrosine kinase that regulates the formation of actin-rich podosomes in OCLs [37], while cathepsin K is a lysosomal cysteine proteinase specifically expressed in OCLs [38]. Punicalagin inhibited the expression of almost all of these marker proteins; the protein expression levels of RANK and cathepsin K were considerably decreased, even at a punicalagin concentration of 1 µm. Moreover, punicalagin suppressed the protein expression levels of c-fms, NFATc1, c-fos, and c-src (Fig. 15). Thus, punicalagin strongly blocked expressions of RANK and cathepsin K, and weakly suppressed a broad range of various OCL marker proteins, including c-fms, NFATc1, c-fos, and c-src. 32