Identification and structure-activity relationship (SAR) of chemical constituents from Daemonorops draco (Willd.) Blume and selected commercial flavonoids on anti-osteoclastogenesis activity

Fitoterapia 138 (2019) 104280

Contents lists available at ScienceDirect

Fitoterapia journal homepage: www.elsevier.com/locate/fitote

Identification and structure-activity relationship (SAR) of chemical constituents from Daemonorops draco (Willd.) Blume and selected commercial flavonoids on anti-osteoclastogenesis activity Xiaoyu Wanga, Irmanida Batubarab, Kosei Yamauchia, Tohru Mitsunagaa, a b

T

⁎

The United Graduate School of Agricultural Science, Gifu University, Gifu, Japan Department of Chemistry, Faculty of Mathematics and Natural Sciences, Tropical Biopharmaca Research Center, Bogor Agriculture University, Bogor 16128, Indonesia

A R T I C LE I N FO

A B S T R A C T

Keywords: Daemonorops draco (Willd.) Blume Anti-osteoclastogenesis Flavonoid SAR

Osteoclastogenesis-related bone diseases including osteoporosis, rheumatoid arthritis, Paget's disease and periodontitis are worldwide occurred and cause severe health problems including bone fracture and bone cancer. However, A few studies have shown that Daemonorops draco (Willd.) Blume may decrease bone destruction and relieve bone cancer pain. In this research, we isolated and purified four known and two novel compounds from D. draco and investigated their anti-osteoclastogenesis activity using RAW264.7 cells. Among them, com.1 exhibited the most effective inhibitory activity on osteoclastogenesis with 78% inhibition at 10 μM and identified to be a novel natural flavan; and com.2 displayed a bit slighter inhibition (50% at 10 μM), indicating that the methylation of 7-hydroxyl group increased the anti-osteoclastogenesis activity. Moreover, nineteen commercial flavonoids were also performed in this study to investigate their inhibitory activity on osteoclastogenesis, and furtherly develop the SAR profile in flavonoid skeleton combined with the information of isolated compounds. Interestingly, the absence of substituents in B-ring and (3R)-hydroxyl group seems to play a crucial role in increasing anti-osteoclastogenesis activity.

1. Introduction Excessive osteoclastic bone resorption plays a crutial role in the pathogenesis of many bone diseases, such as osteoporosis, rheumatoid arthritis, rickets (children) / osteomalacia (adults) Paget's disease and periodontitis [1]. The delicate balance of bone resorption and bone synthesis involved in bone remodeling is a physiological bone process that is regulated by osteoclasts and osteoblasts, respectively [2,3]. Excessive formation of osteoclasts is caused by usual activation of osteoclastogenesis, resulting in bone destruction and inflammation. Osteoclastogenesis progress through multiple stages, including differentiation, fusion, and activation (maturation) related by various factors, including cytokines, hormones and other cells in the bone microenvironment [4]. Dragon's blood (DB) is a non-specific name for bright red resinous exudations which is obtained from different species of a number of distinct plant genera: Dracaena and Daemonorops, Pterocarpus and Croton [5]. It had antiviral, antibacterial and antifungal properties which have been known since long time ago [5] and widely used for wound healing, curing ulcers, diarrhea, as an analgesic, etc. [6].

⁎

Besides, DB extract has antioxidant properties and used to produce antiageing skin creams. Studies also indicated that it has anti-cancer properties as well [7–9]. Moreover, few studies have been issued that DB may alleviate bone destruction and bone cancer pain [10]. In this research, six compounds 1–6 were isolated from Daemonorops draco (Arecaceae). Compound 1 and compound 2 identified as flavans which are an unusual class of natural-source flavonoids exhibited the most effective inhibition on osteoclastogenesis. Up to date, a structure-activity relationship (SAR) study for flavonoids with reference to their osteoclastogenesis activity has not been reported. The aim of the present study is to investigate the biological constituents of D.draco as well as nineteen commercial flavonoids that contribute inhibitory osteoclastogenesis activity. 2. Materials and methods 2.1. Plant materials Dragon's blood (D. draco) was collected from Bogor, West Java, Indonesia, and identified by Research Center for Biology, Indonesia

Corresponding author at: Faculty of Applied Biological Science, Department of Applied Life Science, Gifu University, 1-1 Yanagido, Gifu 501-1112, Japan. E-mail address: [email protected] (T. Mitsunaga).

https://doi.org/10.1016/j.fitote.2019.104280 Received 4 July 2019; Received in revised form 30 July 2019; Accepted 30 July 2019 Available online 31 July 2019 0367-326X/ © 2019 Elsevier B.V. All rights reserved.

Fitoterapia 138 (2019) 104280

X. Wang, et al.

elution of MeOH: 0.05% trifluoroacetic acid (TFA) in H2O [10:90 (v/v) to 60:40 (v/v) for 40 min] to yield compound 3 (37.8 mg) and compound 6 (3.1 mg). Fr.5 was separated by Sephadex LH-20 column (44 mmφ × 405 mml) eluted with EtOH: H2O (100:0–85:15–0:100, v/v) and furtherly purified by PHPLC (20 mmø × 250 mmL) with a gradient elution of MeOH: 0.05% trifluoroacetic acid (TFA) in H2O [5:95 (v/v) to 100:0 (v/v) for 40 min] to yield compound 5 (33.1 mg).

Institute of Sciences, Cibinong, Jakarta. Voucher specimens (No BMK0101082016) were deposited in Tropical Biopharmaca Research Center, Bogor Agricultural University, Indonesia. 2.2. Commercial flavonoids Baicalein, quercetin, (−)-catechin (Wako, Japan), luteolin, myricetin, robinetin (Extrasynthese, France), kaempferol, 3-hydroxyflavone, 7- hydroxyflavone (TCI, Tokyo, Japan), apigenin, galangin, flavanone (Aldrich Chemical Company Inc., Milwaukee, Wisconsin, USA), chrysin, taxifolin, (−)-epicatechin, hesperetin (Sigma Aldrich, USA), flavone, naringenin (Nacalai Tesque, Inc., Kyoto, Japan) and morin (Merck, Germany) were provided by the Laboratory of Natural Products Chemistry, Gifu University, Japan.

2.5. Cell culture The murine macrophage RAW264.7 cell line was purchased from Riken BioResource Research Center (BRC) and cultured in minimal essential medium alpha (purchased from Wako) supplemented with 100 U/mL of penicillin, 100 U/mL of streptomycin, 1% non-essential amino acids (NEAA) and 10% fetal bovine serum (FBS). The cells were seeded at a density of 5 × 104 cells/well in 24-well plates and then incubated overnight at 37 °C in humidified atmosphere of 5% CO2.

2.3. General experimental procedure Optical rotations were measured by using a JASCO P-2300 spectropolarimeter (JASCO, Easton, MD, USA) equipped with a 10 cm pathlength cell. Nuclear magnetic resonance (NMR) spectroscopy was performed in chloroform‑d or methanol‑d4 by using a JEOL ECA-600NMR (JEOL, Tokyo, Japan). Matrix-assisted laser desorption/ionization timeof-flight mass spectrometry (MALDI-TOF-MS) was performed on a SHIMADZU AXIMA-Resonance spectrometer (Shimadzu Corporation, Kyoto, Japan). High-performance liquid chromatography (HPLC, JASCO Corporation, Tokyo, Japan) was performed by using an Inertsil® ODS-3 column (4.6 mmφ × 250 mm, GL sciences Inc., Tokyo, Japan). Preparative HPLC (PHPLC) was performed by using an Inertsil® ODS-3 column (20 mmφ × 250 mm, GL Sciences Inc., Tokyo, Japan). Analytical thin layer chromatography (TLC) was performed using precoated silica gel 60 GF254 plates (Merck KGaA, Darmstadt, Germany). Column chromatogra-phy was performed on a silica gel (Fuji Silysia Chemical Ltd., Aichi, Japan) and Sephadex LH-20 (GE Healthcare Biosciences AB, Upp-sala, Sweden). Receptor activator of nuclear factor kappa-B ligand (RANKL) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Osteoclasts were observed using a fluorescence microscope (BZ-X710, KEYENCE, Japan).

2.6. Cell viability measurement RAW264.7 cells were seeded in 24-well plates at a density of 5 × 104 cells/well and incubated with test compounds at the indicated concentrations and DMSO as control at 37 °C in a humidified atmosphere of 5% CO2 for 48 h. The test compounds were dissolved in DMSO (stock solution) and diluted with medium to a final DMSO concentration of 0.2%. MTT solution (50 μl, 5 mg/mL in phosphate-buffered saline (PBS)) was added to each well and incubated for additional 4 h. Then the medium was removed and 1.0 mL of isopropyl alcohol (containing 0.04 N HCl) was added to each well to dissolve the formazan crystals that formed, and the 150 μl was transferred to a 96-well plate. The absorbance was measured at 590 nm using a microplate reader (Tecan Spark 10 M, Switzerland). Cell viability was calculated as a percentage of the viability of control cells treated with DMSO [11]. 2.7. Measurement of anti-osteoclastogenesis Anti-osteoclastogenesis activity was measured based on previously research, with slight modifications [12–14]. RAW 264.7 cells were seeded in 96-well plate at a density of 6 × 103 cells/well supplied with 100 ng/ml RANKL solution. After 24 h incubation, the adherent cells were treated with tested compounds at indicated concentrations in refreshed medium with RANKL (100 ng/ml). The cells were incubated for 4 days and the medium was refreshed every 2 days with RANKL and tested compounds. Oleic acid (50 μg/ml) was used as a positive control [15]. In order to determine the number of osteoclasts and the number of nuclei per osteoclast at the end of incubation, cell were fixed by 3.7% paraformaldehyde phosphate buffer solution for 20 min and then stained with TRAP staining solution containing 5 mg Naphthol AS-MX Phosphate, 500 ml N-N-Dimethylformamide and 30 mg Fast Red Violett LB Salt in 50 mL TRAP buffer (40 mM sodium-acetate and 10 mM sodium-tartrate in PBS) for 60 min. Cells with a positive staining for TRAP containing 3 or more nuclei were counted as osteoclasts. For quantification, the number of TRAP-positive multinucleated cells in each well was counted by the average number of five representative fields of view (magnification 100×) using electronic microscope. Osteoclastogenesis activity was calculated as a percentage of that measured in control treated with DMSO and RANKL solution only.

2.4. Isolation and identification Powders of D.draco (289 g) were extracted three times (24 h × 3) with methanol (MeOH) on a rotary shaker (NR-20, TAITEC, Saitama, Japan) at room temperature. The solvent were removed under reduced pressure to give the MeOH extract (10.023 g). The MeOH extracts (5319 mg) was subjected on open silica gel column chromatography (75 mmφ × 510 mml) eluted with benzene:MeOH (7:3, v/v) to obtain seven fractions (Fr.1–7). Fr.1 was then subjected on PHPLC (20 mmφ × 250 mml) with a gradient elution of MeOH: 0.05% trifluoroacetic acid (TFA) in H2O [80:20 (v/v) to 100:0 (v/v) for 60 min] to yield compound 1 (7.1 mg). Fr.2 was separated by open silica gel column chromatography (55 mmφ × 520 mml) eluted with n-hexane:chloroform (5:95, v/v, 12 h) and then n-hexane:chloroform:acetone (15:80:5, v/v, 48 h) to obtain 6 fractions (Fr.2–1–2–6). Then, Fr.2–1 and Fr.2–3 were subjected to PHPLC (20 mmφ × 250 mml) with the gradient elution of MeOH: 0.05% trifluoroacetic acid (TFA) in H2O [80:20 (v/v) to 100:0 (v/v) for 60 min and 80:20 (v/v) to 81:19 (v/v) for 40 min, respectively] to yield compound 1 (51.6 mg, totally up to 58.7 mg) and compound 2 (28.1 mg) separately; besides, Fr.2–6 was separated on open silica gel column chromatography (44 mmφ × 405 mml) eluted with chloroform:MeOH (1:1, v/v) and then purified by PHPLC (20 mmφ × 250 mml) with a gradient elution of MeOH: 0.05% trifluoroacetic acid (TFA) in H2O [50:50 (v/v) to 100:0 (v/v) for 40 min] to give compound 4 (9.1 mg). Moreover, Fr.4 was subjected to Sephadex LH-20 column (44 mmφ × 405 mml) eluted by chloroform:MeOH (5:5–2:8, v/v) and furtherly purified by PHPLC (20 mmø × 250 mmL) with a gradient

2.8. Statistical assay All experiments were performed in triplicate (n = 3) and data are expressed as mean values and standard deviation. Student's t-test was used to detect difference: p-value of ⁎p < .05 and ⁎⁎ < 0.01 were considered statistically significance. 2

Fitoterapia 138 (2019) 104280

X. Wang, et al.

Fig. 1. Structures of six isolated compounds 1–6 from D.draco.

3. Results and discussion 3.1. Isolation and identification of chemical constituents from D. draco Four known and two novel compounds were isolated from D.draco (Fig. 1). (2S)-5,7-dimethoxy-6-methylflavan (1) was obtained as yellow crystal and determined to have molecular formula C18H20O3 by MALDITOF-MS (positive ion mode) m/z: 285.1138 [M + H]+, implying 9 degrees of unsaturation; optical rotation measurement (JASCO P-2300 system): [α]20 D-3.50 (c 0.3, CHCl3). Optical rotation of 1 was compared to that of 2, indicated that 1 has the same (2S) configuration as 2. Combined analysis of the 1H and 13C NMR (Table 1) plus the HMQC spectroscopic data revealed the presence three singlet regarding one methyl and two methoxy groups (δH 2.09, δH 3.71 and δH 3.76), two sp3 methylenes, one oxymethine, six aromatic protons (δH 6.30, δH 7.41, δH 6.70 and δH 7.31). These data accounted for all NMR resonances of 1 and four of the nine unsaturations, indicating that 1 include two aromatic rings. Analysis of 1He1H COSY spectrum established one partial structure of H-2/H2–3 and H2–3/ H2–4. According to the HMBC spectrum of compound 1, long-range correlations were observed between H-2 and C-4, C-1’,C-2’ and C-6’; H2-4 and C-2, C-3C-5, C-4a and C-8a; H6-Me and C-6 and C-7; H-7-OMe and C-7; H-5-OMe and C-5; H-8 and C4a, C-7, C-6 and C-8a; H-4’ and C-2’and C-6’; H-3’/5’ and C-1’; H-2’/6’ and C-4’ (Fig. 2). And NOESY data shows the correlations between H-5OMe and H2–4, H-6-Me; H-7-OMe and H-8; H-2’/6’ and H2–3 (Fig. 3). Thus, compound 1 was identified as (2S)-5,7-dimethoxy-6-methylflavan, a novel compound. (2S)-5-methoxy-6-methylflavan-7-ol (2) was obtained as colorless crystal and determined to have molecular formula C17H18O3 by MALDI-

Fig. 2. 1He1H COSY and HMBC correlations of 1 and 4.

Fig. 3. Selected NOESY correlations of 1 and 4.

TOF-MS (positive ion mode) m/z: 271.5289 [M + H]+, implying 9 degrees of unsaturation; optical rotation measurement (JASCO P-2300 system): [α]20 D-9.30 (c 2.1, CHCl3)·1H NMR spectrum (CDCl3, 600 MHz): δ (ppm) 1.99–2.01 (1H, m,H-3), 2.17–2.20 (1H, m,H-3), 2.12 (3H, s, H-6-Me), 2.76–2,81 (2H, m, H-4), 3.71 (3H, s, H-5-OMe), 4.65 (1H, s, H-7-OH), 4.97 (1H, dd, J = 2.04, 9.96 Hz, H-2), 6.22 (1H, s, H-8), 7.31 (1H, dd, J = 2.04, 7.56 Hz, H-4’), 7.37 (2H, m, H-2’,6’), 7.39 (2H, m, H-3’,5’); 13C NMR spectrum (CDCl3, 150 MHz): δ (ppm) 19.84(C-4), 29.86(C-3), 55.63(C-5-OMe), 60.09(C-6-Me), 77,92(C-2), 95.78(C-8), 107.55(C-4a), 111.47(C-7), 126.12(C-2’,6’), 127.95(C-4’), 128.60(C-3’,5’), 141.80(C-1), 154.06(C-8a), 157.15(C-6), 157.50(C-5). Spectral data were coincided to that of published report [16]. 4-hydroxybenzoic acid (3) was obtained as colorless solid and determined to have molecular formula C7H6O3 by MALDI-TOF-MS (positive ion mode) m/z: 139.1261 [M + H]+, implying 5 degrees of unsaturation.1H NMR spectrum (CDCl3, 600 MHz): δ (ppm) 6.79–6.80 (2H, d, J = 8.22 Hz, H-3,5), 7.85–7.86 (2H, d, J = 8.94 Hz, H-2,6); 13C NMR spectrum (CDCl3, 150 MHz): δ (ppm) 114.70(C-3,5), 121.40(C-1), 131.68(C-2,6), 162.01(C-4), 168.84(C-7). Spectral data were coincided to that of published report [17]. (2R)-3-(4-hydroxy-6-((2-(4-hydroxyphenyl)-5-methoxy-6-methylchroman-7-yl)oxy)-2-methoxy-3-methylphenyl)-1-phenylpropan-1-one (4) was obtained as brown crystal and determined to

Table 1 1 H and 13C NMR spectral data for compound 1 in CDCl3 (1H NMR: 600 MHz, 13C NMR: 150 MHz, J in Hz, δ in ppm). Position

δH (ppm)

J(Hz)

δC (ppm)

2 3

4.99dd 1.98–2.05 m 2.18–2.21 m 2.75–2.87 m

2.04, 10.32

77.92 29.86

4 4a 5 6 7 8 8a 1’ 2’,6’ 3’,5’ 4’ 5-OMe 6-Me 7-OMe

6.30s

7.41 m 7.38 m 7.31dd 3.71 s 2.09 s 3.76 s

2.04, 7.56

19.84 107.55 157.15 111.47 157.50 95.78 154.06 141.80 126.12 128.60 127.95 55.63 8.57 60.09

3

Fitoterapia 138 (2019) 104280

X. Wang, et al.

4 include four aromatic rings. Analysis of 1He1H COSY spectrum established several partial structures: H2–3/H2–4 and H2–3/H-2, H-7”/H8”, H-(3’, 5’)/H-(2’, 6’) as well as H-(3”’, 5”’)/H-(2”’, 6”’) and H-(3”’, 5”’)/H-4”’(Fig. 2). According to the HMBC spectrum of compound 5, long-range correlations were indicated as follows: (Fig. 2). And NOESY data shows the correlations between H-2’/6’ and H-2, H-2’/6’ and H-3, H-2 and H-4, H-4 and H-5-OMe and H-8”, H-8” and H-2”/6” (Fig. 3). Thus, compound 4 was identified as (2R)-3-(4-hydroxy-6-((2-(4-hydroxyphenyl)-5-methoxy-6-methylchroman-7-yl)ox -y)-2-methoxy-3methylphenyl)-1-phenylpropan-1-one, a novel compound. 3,4-dihydroxybenzoic acid (5) was obtained as colorless solid and determined to have molecular formula C7H6O4 by MALDI-TOF-MS (positive ion mode) m/z: 154.8848 [M + H]+, implying 5 degrees of unsaturation.1H NMR spectrum (CDCl3, 600 MHz): δ (ppm) 6.76–6.78 (1H, d, J = 7.56 Hz, H-6), 7.39–7.41 (1H, dd, J = 2.04, 3.45 Hz, H-5), 7.41–7.42 (1H, d, J = 2.04 Hz, H-2); 13C NMR spectrum (CDCl3, 150 MHz): δ (ppm) 114.46(C-6), 116.40(C-2), 121.78(C-1), 122.61(C5), 144.71(C-3), 150.19(C-4), 168.93(C-7). Spectral data were coincided to that of published report [18]. (+)-Afzelechin (6) was obtained as white crystal and determined to have molecular formula C15H14O5 by MALDI-TOF-MS (positive ion mode) m/z: 275.0027 [M + H]+, implying 9 degrees of unsaturation; optical rotation measurement (JASCO P-2300 system): [α]20 D + 0.52 (c 0.09, Acetone:EtOAc = 9:1).1H NMR spectrum (methanol‑d4, 600 MHz): δ (ppm) 2.48 (1H, dd, J = 8.94, 16.17 Hz, H-4ax), 2.86 (1H, dd, J = 5.46, 15.81 Hz, H-4 eq), 3.94–3.98 (1H, m, H-3), 4.57 (1H, d, J = 7.56 Hz, H-2), 5.81 (1H, d, J = 2.04 Hz, H-6), 5.90 (1H, d, J = 2.10 Hz, H-8), 6.76 (2H, d, J = 8.94 Hz, H-3’/5’), 7.20 (2H, d, J = 8.94 Hz, H-2’/6’); 13C NMR spectrum (methanol-d4, 150 MHz): δ (ppm) 27.59(C-4), 67.31(C-3), 81.54(C-2), 94.13(C-6), 94.95(C-8), 99.55(C-4a), 114.69(C-3’,5’), 128.28(C-2’,6’), 130.15(C-1’), 155.66(C8a), 156.24(C-5), 156.53(C-7), 157.08(C-4’). Spectral data were coincided to that of published report [19].

Table 2 1 H and 13C NMR spectral data for compound 4 in CDCl3 (1H NMR: 600 MHz, 13C NMR: 150 MHz, J in Hz, δ in ppm). Position

δH (ppm)

J(Hz)

δC (ppm)

2 3

4.88dd 1.92–2.00 m 2.12–2.15 m 2.70–2.83 m

2.04,8.00

77.56 29.67

4 5 6 7 8 4a 8a 1’ 2’,6’ 3’,5’ 4’ 1” 2” 3” 4” 5” 6” 7” 8” 9” 1”’ 2”’,6”’ 3”’,5”’ 4”’ 5-OMe 6-Me 4”-Me 5”-OMe

6.20s

7.23d 6.83d

9.66 8.22

6.26 s

2.93 t 3.41 t

6.18 6.18

7.96d 7.41d 7.54d 3.70s 2.10s 2.07 s 3.74 s

7.56 8.22 7.56

20.00 157.32 109.51 153.77 99.55 107.77 154.16 133.49 127.62 115.48 156.01 153.66 100.78 154.33 109.40 158.22 113.54 17.35 39.99 203.07 136.23 128.47 128.68 133.80 60.05 8.48 8.93 60.64

have molecular formula C34H34O7 by MALDI-TOF-MS (positive ion mode) m/z: 577.1743 [M + Na]+, implying 18 degrees of unsaturation; optical rotation measurement (JASCO P-2300 system): [α]20 D + 6.6233 (c 0.6, CHCl3). Combined analysis of the 1H and 13C NMR (Table 2) plus the HMQC spectroscopic data revealed the presence three singlet regarding two methyls and two methoxy groups (δH 2.10, δH 2.07, δH 3.70 and δH 3.74), four sp3 methylenes, one oxymethine, eleven aromatic protons (δH 6.20, δH 6.26, δH 6.83, δH 7.23, δH 7.41, δH 7.54, and δH 7.96), as well as twenty-four aromatic carbons and one ketone carbon (δC 203.07). These data accounted for all NMR resonances of 4 and sixteen of the eighteen unsaturations, indicating that

3.2. Biological results and SAR establishment As shown in Fig. 4, com.1 was identified as a novel flavan and exhibited the most potent inhibition on osteoclastogenesis, showed 78% inhibition on osteoclastogenesis at 10 μM without any cytotoxicity and completed inhibition on osteoclastogenesis at 100 μM remaining 75% cell viability. Com.2 also showed fairly good inhibition on osteoclastogenesis, with 50% inhibition on osteoclastogenesis activity and no cytotoxicity at 10 μM. Furtherly compared to positive control, both

Fig. 4. Osteoclastogenesis activity and cell viability of isolated compounds 1–6 (concentrations at 10/100 μM). Assays were measured in triplet experiment. Each value is expressed as the mean ± SD of triplet determinations. ⁎p < .05, ⁎⁎p < .01 versus the control group. (Control: DMSO; Oleic acid was used as positive control). 4

Fitoterapia 138 (2019) 104280

X. Wang, et al.

Fig. 5. Osteoclast differentiation of isolated compounds 1–6 (concentrations at 10/100 μM). Assays were measured in triplet experiment. Each value is expressed as the mean ± SD of triplet determinations. ⁎p < .05, ⁎⁎p < .01 versus the control group. (Control: DMSO; Oleic acid was used as positive control).

5

Fitoterapia 138 (2019) 104280

X. Wang, et al.

B-ring provide more cell toxicity and less osteoclastogenesis activity, indicating that the absence of hydroxyl groups on B-ring is important for anti-osteoclastogenesis activity. However, previous studies also showed good inhibition of bacalein, luteolin and quercetin respectively [20–22]. Furtherly, comparing baicalein (no cytotoxicity, completed anti-osteoclastogenesis activity) to galangin (no cytotoxicity, 73% antiosteoclastogenesis activity) and chrysin (no cytotoxicity, 52% anti-osteoclastogenesis activity) indicated that 6-hydroxyl group is more important than 3-hydroxyl group providing more effective inhibition on osteoclastogenesis. These results are corresponding to other researchers works, which showed inhibition of galangin on osteoclastogenesis [23] and promotion of chrysin on osteogenesis [24]. As discussed above, comparison of com.1 (no cytotoxicity, 78% inhibition on osteoclastogenesis) and com.2 (no cytotoxicity, 50% inhibition on osteoclastogenesis) demonstrated that 7-methoxyl group provides more anti-osteoclastogenesis activity than 7-hydroxyl group, indicating that methylation might be important for anti-osteoclastogenesis activity. Besides, it shows no much difference between 3-hydroxyflavone, 7hydroxyflavone, and flavone, all remains about 60% osteoclstogenesis activity without any cytotoxicity, which indicated that neither 7-hydroxyl group nor 3-hydroxyl group could be important on anti-osteoclastogenesis activity in flavone skeleton. Interestingly, we found that (3R)-hydroxyl group provide much more effective activity on osteoclastogenesis inhibition than (3S)-hydroxyl group by comparing (−)-epicatechin (no cytotoxicity, 64% anti-osteoclastogenesis activity) and (−)-catechin (no cytotoxicity, 18% anti-osteoclastogenesis activity); if furtherly compared to (+)-afzelechin (6), it seem that (2S) configuration might slightly increase inhibition compare to (2R) configuration. Moreover, the osteoclastogenesis inhibition of apigenin (slight cytotoxicity, 49% anti-osteoclastogenesi activity) is almost 3 times compare to naringenin (no cytotoxicity, only 19% anti-osteoclastogenesis activity), indicated that double bond at C-2 and C-3 position increase inhibition on osteoclastogenesis. Well, the inhibitory activity of apigenin and naringenin on osteoclastogenesis were also reported in previous researches [25–27]. This is the first report to study and develop an early SAR profile of flavonoid skeleton on anti-osteoclastogenesis activity, as summarized in Fig. 7.

Table 3 Cell viability and osteoclastogenesis activity of 19 commercial flavonoids. Compound

Cell viability (%)

Osteoclastogenesis activity (%)

Baicalein Luteolin Kaempferol Quercetin Myricetin Apigenin Galangin Chrysin Taxifolin 3-hydroxyflavonone 7-hydroxyflavonone Robinetin (−)-Epicatechin (−)-Catechin Flavone Flavanone Morin Naringenin Hesperetin Oleic acid

100.61 ± 5.94 50.80 ± 5.38⁎ 105.87 ± 3.04⁎ 73.00 ± 4.68⁎⁎ 93.48 ± 4.01⁎⁎ 88.20 ± 1.86⁎ 122.13 ± 8.96 133.52 ± 8.19 106.98 ± 6.42 96.60 ± 5.02 111.93 ± 5.58 103.47 ± 8.68 93.15 ± 5.42 96.17 ± 5.83 95.36 ± 6.94 83.45 ± 5.34 94.83 ± 5.94 100.35 ± 5.38 93.76 ± 3.04 103.99 ± 3.53

0.00 ± 0.00⁎⁎ 54.08 ± 30.85 101.02 ± 11.45 34.69 ± 8.03⁎ 60.20 ± 31.45 51.02 ± 24.51 27.55 ± 6.61⁎ 47.96 ± 26.02 67.06 ± 4.99⁎⁎ 41.18 ± 11.65⁎⁎ 45.88 ± 24.62⁎ 57.65 ± 12.00⁎⁎ 36.47 ± 7.25⁎⁎ 82.35 ± 24.51 51.76 ± 16.64⁎ 38.82 ± 12.56⁎⁎ 68.18 ± 7.42 81.06 ± 22.14 83.33 ± 28.83 57.65 ± 14.79⁎

Assays were measured in triplet experiment. Each value is expressed as the mean ± SD of triplet determinations. ⁎p < .05, ⁎⁎p < .01 versus the control group. (Control: DMSO; Oleic acid was used as positive control).

com.1 and com.2 exhibited more inhibition on osteoclastogenesis, which provide as potent candidate drugs of osteoclastogenesis-related bone disease. Com.3 displayed 75% inhibition on osteoclastogenesis with no cytotoxicity at 100 μM; however, com.5 showed no osteoclastogenesis activity at both concentrations. The only difference of com.3 compared to com.5 is the replacement of hydrogen by hydroxyl group at 3-position, indicating that 3-OH decrease anti-osteoclastogenesisn activity in benzoic acid skeleton. Com.4 and com.6 showed slight inhibition, with 27% and 35% inhibition on osteoclastogenesis at 10 μM respectively. Based on these results, the absent of 3-hydroxyl group of benzoic acid was important for anti-osteoclastogenesis activity. Moreover, the presence of 7-methoxyl group in flavan structure was crucial for anti-osteoclastogenesis (Fig. 5). As shown in Table 3 and Fig. 6, comparing baicalein (no cytoxicity, 100% anti-osteoclastogenesis activity) to luteolin (50% cytotoxicity and 46% anti- osteoclastogenesis activity) and quercetin (27% cytotoxicity, 35% anti- osteoclastogenesis activity), it seems that hydroxyl groups on

Fig. 6. Structures of 19 commercial flavonoids. 6

Fitoterapia 138 (2019) 104280

X. Wang, et al.

Fig. 7. Summary of SAR in flavonoid skeleton for anti-osteoclastogenesis activity.

4. Conclusion [8]

Totally, four known compounds and two novel compounds were isolated from D.draco. Among them, (2S)-5,7-dimethoxy-6-methylflavan (1) was identified as a novel natural flavan, which is a rarely discovered structure from natural-resource product, and display the most potent inhibition on osteoclastogenesis (78% inhibition at 10 μM); (2S)-5-methoxy-6-methylflavan-7-ol (2) exhibited less inhibition (50% inhibition at 10 μM) compare to com.1, indicated that replacement of hydroxyl group by methoxyl group in 7-position increase anti-osteoclastogenesis activity. Moreover, combined with the information of 19 commercial flavonoids, we found that no substituents on B-ring and (3R)-hydroxyl group contribute to increasing anti-osteoclastogenesis activity. Thus, these valuable informations may provide new ideas for the development and treatment of osteoclastogenesis-related bone disease.

[9]

[10]

[11]

[12]

[13]

Declaration of Competing Interest

[14]

The authors declare no competing financial interest. [15]

Acknowledgement [16]

We greatly acknowledge the Draco Industrial Agribusiness (D.I.A) Group for support of preparing tested sample of D. draco.

[17]

References [18] [1] A. Bhansali, Metabolic bone disease: newer perspectives, Indian J. Endocrinol. Metab. 16 (2012) S140–S141, https://doi.org/10.4103/2230-8210.104023. [2] D.A. Corral, M. Amling, M. Priemel, E. Loyer, S. Fuchs, P. Ducy, R. Baron, G. Karsenty, Dissociation between bone resorption and bone formation in osteopenic transgenic mice, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 13835–13840, https://doi.org/10.1073/pnas.95.23.13835. [3] G. Karsenty, E.F. Wagner, Review reaching a genetic and molecular understanding of skeletal development patterning were not really involved in the control of cell differentiation. Thus, the development of the skeleton encompasses, Dev. Cell 2 (2002) 389–406, https://doi.org/10.1016/S1534-5807(02)00157-0. [4] J. Liu, J. Shiono, Y. Tsuji, K. Shimizu, R. Kondo, Methyl ganoderic acid DM : a selective potent osteoclastogenesis inhibitor, Open Bioact. Compd. J. 7 (2009) 37–42, https://doi.org/10.2174/1874847300902010037. [5] D. Gupta, B. Bleakley, R.K. Gupta, Dragon’s blood: botany, chemistry and therapeutic uses, J. Ethnopharmacol. 115 (2007) 361–380, https://doi.org/10.1016/j. jep.2007.10.018. [6] A. Termentzi, N. Fokialakis, A. Leandros Skaltsounis, Natural resins and bioactive natural products thereof as potential anitimicrobial agents, Curr. Pharm. Des. 17 (2011) 1267–1290, https://doi.org/10.2174/138161211795703807. [7] D. Rossi, R. Bruni, N. Bianchi, C. Chiarabelli, R. Gambari, A. Medici, A. Lista, G. Paganetto, Evaluation of the mutagenic, antimutagenic and antiproliferative

[19] [20]

[21]

[22]

[23]

[24]

7

potential of Croton lechleri (Muell. Arg.) latex, Phytomedicine. 10 (2003) 139–144, https://doi.org/10.1078/094471103321659843. M.I.L.E. Lopes, J. Saffi, S. Echeverrigaray, J.A.P. Henriques, M. Salvador, Mutagenic and antioxidant activities of Croton lechleri sap in biological systems, J. Ethnopharmacol. 95 (2004) 437–445, https://doi.org/10.1016/j.jep.2004.08.025. G.F. Gonzales, L.G. Valerio, Medicinal plants from Peru: a review of plants as potential agents against cancer, Anti Cancer Agents Med. Chem. 6 (2006) 429–444, https://doi.org/10.2174/187152006778226486. Y. Bao, Y. Gao, M. Du, W. Hou, L. Yang, X. Kong, H. Zheng, W. Li, B. Hua, Topical treatment with xiaozheng zhitong paste (XZP) alleviates bone destruction and bone cancer pain in a rat model of prostate cancer-induced bone pain by modulating the RANKL/RANK/OPG signaling, evidence-based complement, Altern. Med. 2015 (2015) 1–14, https://doi.org/10.1155/2015/215892. S. Choodej, K. Pudhom, T. Mitsunaga, Inhibition of TNF- α -induced inflammation by sesquiterpene lactones from saussurea lappa and semi-synthetic analogues, Planta Med. 84 (2018) 329–335, https://doi.org/10.1055/s-0043-120115. J. Keller, S. Brink, B. Busse, A.F. Schilling, T. Schinke, M. Amling, T. Lange, Divergent resorbability and effects on osteoclast formation of commonly used bone substitutes in a human in vitro-assay, PLoS One 7 (2012), https://doi.org/10.1371/ journal.pone.0046757. C. Vincent, M. Kogawa, D.M. Findlay, G.J. Atkins, The generation of osteoclasts from RAW 264.7 precursors in defined, serum-free conditions, J. Bone Miner. Metab. 27 (2009) 114–119, https://doi.org/10.1007/s00774-008-0018-6. P. Collin-Osdoby, P. Osdoby, RANKL-mediated osteoclast formation from murine RAW 264.7 cells, Methods Mol. Med. 80 (2003) 153–166, https://doi.org/10.1385/ 1-59259-366-6:153. H. Tan, S. Furuta, T. Nagata, K. Ohnuki, T. Akasaka, B. Shirouchi, M. Sato, R. Kondo, K. Shimizu, Inhibitory effects of the leaves of loquat (eriobotrya japonica) on bone mineral density loss in ovariectomized mice and osteoclast differentiation, J. Agric. Food Chem. 62 (2014) 836–841, https://doi.org/10.1021/jf402735u. G. Cardillo, L. Merlini, G. Nasini, P. Salvadori, Constituents of Dragon’s blood. Part I. structure and absolute configuration of new optically active flavans, J. Chem. Soc. C Org. (1971) 3967–3970, https://doi.org/10.1039/J39710003967. S. Sukari, I.M. Said, Phenolic compounds from the fruits of Orania sylvicola, Malays. J. Anal. Sci. 17 (2013) 276–280. N. Syafni, D.P. Putra, D. Arbain, 3,4-dihydroxybenzoic acid and 3,4-dihydroxybenzaldehyde from the fern Trichomanes chinense L.; isolation, antimicrobial and antioxidant properties, Indones. J. Chem. 12 (2012) 273–278, https://doi.org/10. 22146/ijc.21342. A. Saraswathy, B. Vidhya, (+)-afzelechin from the rhizome of Calamus rotang Linn, Indian Drugs 49 (2012) 49–50. M.H. Kim, S.Y. Ryu, M.A. Bae, J.S. Choi, Y.K. Min, S.H. Kim, Baicalein inhibits osteoclast differentiation and induces mature osteoclast apoptosis, Food Chem. Toxicol. 46 (2008) 3375–3382, https://doi.org/10.1016/j.fct.2008.08.016. J.W. Lee, J.Y. Ahn, S.I. Hasegawa, B.Y. Cha, T. Yonezawa, K. Nagai, H.J. Seo, W.B. Jeon, J.T. Woo, Inhibitory effect of luteolin on osteoclast differentiation and function, Cytotechnology. 61 (2009) 125–134, https://doi.org/10.1007/s10616010-9253-5. A. Wattel, S. Kamel, C. Prouillet, J.P. Petit, F. Lorget, E. Offord, M. Brazier, Flavonoid quercetin decreases osteoclastic differentiation induced by RANKL via a mechanism involving NFκB and AP-1, J. Cell. Biochem. 92 (2004) 285–295, https://doi.org/10.1002/jcb.20071. J.E. Huh, I.T. Jung, J. Choi, Y.H. Baek, J.D. Lee, D.S. Park, D.Y. Choi, The natural flavonoid galangin inhibits osteoclastic bone destruction and osteoclastogenesis by suppressing NF-κB in collagen-induced arthritis and bone marrow-derived macrophages, Eur. J. Pharmacol. 698 (2013) 57–66, https://doi.org/10.1016/j.ejphar. 2012.08.013. W. Zeng, Y. Yan, F. Zhang, C. Zhang, W. Liang, Chrysin promotes osteogenic

Fitoterapia 138 (2019) 104280

X. Wang, et al.

T. Tang, A. Qin, Z. Zhu, The inhibition of RANKL-induced osteoclastogenesis through the suppression of p38 signaling pathway by naringenin and attenuation of titanium-particle-induced osteolysis, Int. J. Mol. Sci. 15 (2014) 21913–21934, https://doi.org/10.3390/ijms151221913. [27] V.D. La, S. Tanabe, D. Grenier, Naringenin inhibits human osteoclastogenesis and osteoclastic bone resorption, J. Periodontal Res. 44 (2009) 193–198, https://doi. org/10.1111/j.1600-0765.2008.01107.x.

differentiation via ERK/MAPK activation, Protein Cell. 4 (2013) 539–547, https:// doi.org/10.1007/s13238-013-3003-3. [25] S. Bandyopadhyay, J.M. Lion, R. Mentaverri, D.A. Ricupero, S. Kamel, J.R. Romero, N. Chattopadhyay, Attenuation of osteoclastogenesis and osteoclast function by apigenin, Biochem. Pharmacol. 72 (2006) 184–197, https://doi.org/10.1016/j.bcp. 2006.04.018. [26] W. Wang, C. Wu, B. Tian, X. Liu, Z. Zhai, X. Qu, C. Jiang, Z. Ouyang, Y. Mao,

8