Synthesis of novel 5,6-dehydrokawain analogs as osteogenic inducers and their action mechanisms

Article abstract of DOI:10.1016/j.bmcl.2017.04.016

An imbalance between bone resorption by osteoclasts and bone formation by osteoblasts can cause bone loss and bone-related disease. In a previous search for natural products that increase osteogenic activity, we found that 5,6-dehydrokawain (1) from Alpinia zerumbet promotes osteoblastogenesis. In this study, we synthesized and evaluated series of 5,6-dehydrokawain analogs. Our structure-activity relationships revealed that alkylation of para or meta position of aromatic ring of 1 promote osteogenic activity. Among the potential analogs we synthesized, (E)-6-(4-Ethylstyryl)-4-methoxy-2H-pyran-2-one (14) and (E)-6-(4-Butylstyryl)-4-methoxy-2H-pyran-2-one (21) both significantly up-regulated Runx2 and Osterix mRNA expression at 10?μM. These osteogenic activities could be mediated by bone morphogenetic protein (BMP) and activation of p38 MAPK signaling pathways. Compounds 14 and 21 also inhibited RANKL-induced osteoclast differentiation of RAW264 cells. These results indicated that novel 5,6-dehydrokawain analogs not only increase osteogenic activity but also inhibit osteoclast differentiation, and could be potential lead compounds for the development of anti-osteoporosis agents.

Full text of DOI:10.1016/j.bmcl.2017.04.016

Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of novel 5,6-dehydrokawain analogs as osteogenic inducers
and their action mechanisms
b
a
a
a
Momochika Kumagai a,b, , Keisuke Nishikawa , Takashi Mishima , Izumi Yoshida , Masahiro Ide ,
⇑
Keiko Koizumi a, Munetomo Nakamura a, Yoshiki Morimoto b
a Department of Research and Development, Japan Food Research Laboratories, Osaka 567-0085, Japan
b Department of Chemistry, Graduate School of Science, Osaka City University, Sumiyoshi-ku, Osaka 558-8585, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
An imbalance between bone resorption by osteoclasts and bone formation by osteoblasts can cause bone
loss and bone-related disease. In a previous search for natural products that increase osteogenic activity,
we found that 5,6-dehydrokawain (1) from Alpinia zerumbet promotes osteoblastogenesis. In this study,
we synthesized and evaluated series of 5,6-dehydrokawain analogs. Our structure-activity relationships
revealed that alkylation of para or meta position of aromatic ring of 1 promote osteogenic activity. Among
the potential analogs we synthesized, (E)-6-(4-Ethylstyryl)-4-methoxy-2H-pyran-2-one (14) and (E)-6-
(4-Butylstyryl)-4-methoxy-2H-pyran-2-one (21) both significantly up-regulated Runx2 and Osterix
mRNA expression at 10 mM. These osteogenic activities could be mediated by bone morphogenetic pro-
tein (BMP) and activation of p38 MAPK signaling pathways. Compounds 14 and 21 also inhibited RANKL-
induced osteoclast differentiation of RAW264 cells. These results indicated that novel 5,6-dehydrokawain
analogs not only increase osteogenic activity but also inhibit osteoclast differentiation, and could be
potential lead compounds for the development of anti-osteoporosis agents.
Received 6 February 2017
Revised 3 April 2017
Accepted 5 April 2017
Available online xxxx
Keywords:
5,6-Dehydrokawain
Kavalactone
MC3T3-E1
Mineralization
Osteoblast
Ó 2017 Elsevier Ltd. All rights reserved.
Osteoporosis, characterized by systemic dysfunction in bone
mass and strength that leads to fragility fractures, is a growing
medical and socioeconomic issue. Osteoblastic bone formation
and osteoclastic bone resorption are tightly coordinated during
the remodeling cycle and loss of bone mass results from an imbal-
ance of these fundamental processes.1,2 Recent therapies for osteo-
porosis have mainly focused on the suppression of bone resorption
by osteoclasts.2 For example, bisphosphonates and selective estro-
gen receptor modulators have been used clinically to prevent bone
loss.1,2 However, by the time they initiate such therapies, many
osteoporotic patients have already lost substantial amounts of
bone. Therefore, anabolic treatments that enhance osteogenesis
are needed.3 In this regard, the ability of some chemical compo-
nents in food and medicinal plants to promote osteogenesis has
been studied.4–7 In particular, the ability of isoflavones such as
daidzein and genistein, which abundant in soybeans, to prevent
bone loss has been extensively investigated using various experi-
mental models and in clinical studies.8–10 Recently, the synthesis
and osteogenic activity of daidzein analogs was reported.11,12
These analogs are expected to be lead compounds for osteoporosis
treatments.
As a part of our exploration of natural products that enhance
osteogenic activities, we identified the kavalactone 5,6-dehy-
drokawain (1) from Alpinia zerumbet,13 and showed that it has
osteogenic activity (Fig. 1).14 To develop 5,6-dehydrokawain ana-
logs that have more potent osteogenic activity than the natural
parent products, and to reveal structure–activity relationships,
OMe
Abbreviations: ALP, alkaline phosphatase;
aMEM, alpha minimum essential
5
medium; BMP, bone morphogenetic protein; DMSO, dimethyl sulfoxide; HRMS,
high-resolution mass spectrometry; MAPK, mitogen-activated protein kinase;
ODM, osteoblast differentiation medium; RANKL, receptor activator of nuclear
8
O
O
6
7
factor-
j
B ligand; TRAP, tartrate-resistant acid phosphatase.
⇑
1
Corresponding author at: Department of Research and Development, Japan
Food Research Laboratories, Osaka 567-0085, Japan.
E-mail address: kumagaim@jfrl.or.jp (M. Kumagai).
Fig. 1. Structure of 5,6-dehydrokawain (1).
http://dx.doi.org/10.1016/j.bmcl.2017.04.016
0960-894X/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Kumagai M., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.04.016

2
M. Kumagai et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
here we synthesized a series of chemically modified 5,6-dehy-
drokawain analogs and evaluated their osteogenic activity using
the mouse osteoblast cell line MC3T3-E1. For those compounds
that had potent activities, we investigated the molecular mecha-
nisms underlying the osteogenic activity and the effects against
osteoclastogenesis using RANKL-induced RAW264 cells.
The synthesis of 5,6-dehydrokawain analogs was performed
using an aldol reaction (Scheme 1).15,16 Commercially available
4-hydroxy-6-methylpyrone was methylated by dimethyl sulfate,
and condensed with a variety of aldehydes by magnesium methox-
ide in anhydrous methanol. Only 3-ethylbenzaldehyde was synthe-
sized as previously described.17 Other aldehydes were obtained
from commercial suppliers and used without further purification.
For condensation, the mixtures were heated at reflux for 3–6 h
and then cooled. The solvent was removed in vacuo and the crude
product was purified by preparative HPLC.
OH
OMe
OMe
a
b
To prepare pyrones 27–29, commercially available pyrones
were reacted with each alkylating reagent and then condensed
with benzaldehyde (Scheme 2). The detailed method, synthetic
yields, melting points, 1H and 13C NMR, HRMS data, purity data
are described in Supplementary data.
To evaluate the effects of these compounds on osteoblast differ-
entiation, MC3T3-E1 cells were cultured with the synthesized
derivatives in osteoblast differentiation medium (ODM) containing
ascorbic acid and b-glycerophosphate which were essential for the
expression of osteoblast phenotype.18 After 4 and 10 days, alkaline
phosphatase activity and osteoblastic mineralization, respectively,
were assessed. Alkaline phosphatase activity is an early-stage
osteogenesis marker, whereas mineralization characterizes late-
stage osteogenesis.19
O
O
O
O
O
O
R
1, 4-22
3
2
Scheme 1. Synthesis of 5,6-dehydrokawain and its analogs. Reagents and condi-
tions: (a) Me2SO4, K2CO3, DMSO, rt; (b) RCHO, Mg(OMe)2, MeOH, 60 °C.
2
a (R =Me)
2
OR2
b (R =Et)
OH
2
n
R1
O
c (R = Pr)
R1
d
27-29
O
O
O
1
1
2
R =Me (23)
R =H (2)
R =R = Me (24)
1
1
2
R =H, R = Et (25)
1
2
n
R =H, R = Pr (26)
We first synthesized compounds 1 and 4–14 as a small-scale
library for modification of the phenyl ring in 5,6-dehydrokawain
(1) and evaluated osteogenic activities in an initial screen (Fig. 2A).
Synthetic 1 (60 mM) enhanced ALP activity by approximately 1.25-
Scheme 2. Synthesis of 5,6-dehydrokawain analogs 27–29. Reagents and condi-
tions: (a) dimethyl sulfate, K2CO3, DMSO, rt; (b) diethyl sulfate, K2CO3, DMSO, rt; (c)
dipropyl sulfate, K2CO3, DMSO, rt; (d) benzaldehyde, Mg(OMe)2, MeOH, 60 °C.
Fig. 2. Structures of synthesized compounds 1 and 4–14 (A), and their effects on ALP activity (B) and mineralization (C) in MC3T3-E1 cells. The cells were cultured with or
without test compounds in an osteoblast differentiation medium containing ascorbic acid (50 mg/mL) and b-glycerophosphate (10 mM) for 4 days (ALP activity) and 10 days
(Mineralization). Mineralization was visualized by Alizarin Red S staining and photographed. The data represent means SD of triplicate analysis. The photographs show
representative views of triplicate analyses. *P < 0.05 versus the vehicle control by Dunnett’s test.
Please cite this article in press as: Kumagai M., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.04.016

M. Kumagai et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
3
fold relative to vehicle control (DMSO), which was similar to that
A
seen for the natural products we previously reported (Fig. 2B).14
Furthermore, 60 mM compound 1 enhanced matrix mineralization
(Fig. 2C). Although analogs 5, 6, 7, 10, 12 did not increase ALP activ-
ity and matrix mineralization at 30 mM, six of eleven 5,6-dehy-
drokawain analogs (14, 11, 8, 9, 13 and 4) at 30 mM increased
ALP activity by 1.32- to 2.87-fold relative to vehicle control
(Fig. 2B). This concentration was lower than that at which com-
pound 1 had osteogenic activity. At 30 mM these derivatives also
facilitated matrix mineralization by the cells (Fig. 2C) These results
suggested that six of the 5,6-dehydrokawain analogs promote both
early- and late-stage osteogenesis. Interestingly, compound 4,
wherein the phenyl ring of 1 was changed to a cyclohexyl ring,
had enhanced osteogenic activity, suggesting that the aromatic
ring was not necessary for osteogenic activity. Compounds 14, 11
and 8 were highly active inducers of osteogenic differentiation,
indicating that attachment of hydrophobic substituents at para
positions on the phenyl ring is important to enhance osteogenic
activity.
B
When the substituent was an ethyl group, marked enhance-
ment of osteogenic activity was observed. To examine how substi-
tutions at various positions affected ALP activity, we next
synthesized 5,6-dehydrokawain analogs bearing an ethyl or a
methyl group at the ortho, meta, and para positions (Fig. 3A). Com-
pared with the ALP activity seen for cells treated with 14, cells
treated with the meta derivative 15 had similar ALP activity, but
ortho derivative 16-treated cells had significantly decreased ALP
activity (Fig. 3B). We saw a similar trend for the mineralization
assay. The dramatically increased mineralization induced by 14
was slightly lower in cells treated with 15, and significantly
decreased in cells treated with 16 (Fig. 3C, D). The methylated ana-
logs 17–19 also showed a similar trend, although ALP activity and
mineralization of 17-treated cells were significantly lower than
those of 14-treated cells (Fig. 3). These results demonstrated that
the presence of substituents at the para or meta position are
needed for potent osteogenic activity. These findings may thus be
helpful for future studies of biological mechanisms of bone forma-
tion/resorption using 5,6-dehydrokawain analogs that remain
bioactive.
To investigate how alkyl chain length affects the activity of 5,6-
dehydrokawain analogs, we further synthesized derivatives 20–22
with n-propyl, n-butyl, and tert-butyl groups, respectively (Fig. 4A).
The effects on osteogenic activity of these compounds were depen-
dent on the alkyl length (Fig. 4B, C). Of five alkylated 5,6-dehy-
drokawain analogs, 21, which had the longest alkyl chain, had
the most potent promotion of ALP activity and enhancement of
mineralization. When comparing the activities of 21 and 22, 21
showed significantly stronger ALP and mineralization activity than
22 (at 10 mM), thus highlighting the importance of straight chain
expansion for osteogenic activity.
C
D
We also synthesized and evaluated compounds 27–29, which
were pyrone ring-modified compounds of 1 (Fig. 5A). These mole-
cules did not increase mineralization at 30 mM (Fig. 5B), and at con-
centrations of 0.1–60 mM they had no better osteogenic activity
than 1 (data not shown). These results suggested that the 4-meth-
oxypyran-2-one structure is important for osteogenic activity.
The effects of all synthesized compounds on cell proliferation
were also evaluated. After treating cells with the compounds for
4 days, the cell numbers ranged from 0.85 to 1.25-fold of control
values, except for compound 29 (Supplementary Fig. 1). From these
results, we concluded that the concentrations of the compounds
used in this study (30 mM) were suitable for comparison of osteo-
genic activities.
Fig. 3. Structures of synthesized compounds 14–19 (A) and their effects on ALP
activity (B) and mineralization in MC3T3-E1 cells (C and D). The experimental
procedure is the same as that in Fig. 2. For quantification of mineralization, cells
were solubilized with cetylpyridinium chloride and the absorbance of the solution
was measured at 550 nm. The data represent means SD of triplicate analysis. The
various letters indicate significant differences in Turkey-Kramer test results
(P < 0.05).
as well as Alpl and Bglap, were analyzed by quantitative PCR.
Runx2 and Osterix are essential transcription factors that govern
osteoblast differentiation20,21 and regulate the transcription of
the osteogenic marker genes Alp1 and Bglap.22 Relative to
untreated MC3T3-E1 cells, cells treated with ODM supplemented
To investigate the mechanisms by which the potentially active
analogs 14 and 21 facilitate osteogenic differentiation of MC3T3-
E1 cells, osteogenic mRNA expression levels of Runx2 and Osterix,
Please cite this article in press as: Kumagai M., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.04.016

4
M. Kumagai et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
A
B
C
Fig. 6. Effects of potentially active derivatives 14 and 21 on osteoblast-related
genes in MC3T3-E1 cells. The cells were incubated for 2 days with compounds in
the differentiation medium. Total RNA was isolated and mRNA levels of the genes
were measured by real-time PCR and normalized to Actb. The data represent
means SD of three independent experiments. *P < 0.05 versus the vehicle control
by Dunnett’s test.
Fig. 4. Structures of synthesized analogs 14, 17 and 20–22 (A), and their effects on
ALP activity (B) and mineralization in MC3T3-E1 cells (C). The experimental
procedure is the same as that in Fig. 2. The various letters indicate significant
differences in Turkey-Kramer test results (P < 0.05).
duct 1.14 These results demonstrated the higher potency of 14
and 21 as osteogenic inducers relative to 1.
In the qPCR analysis, we revealed that 14 and 21 enhanced
mRNA expression of Osterix, which is known to be induced by a
bone morphogenetic protein (BMP) via Smad and p38 mitogen–ac-
tivated protein kinase (MAPK) pathways.23,24 To demonstrate the
involvement of these signaling pathways in the osteogenic activity
of 14 and 21, we investigated phosphorylation of p38 and
Smad1/5/8 by western blot analysis. Both 14 and 21 activated
p38 MAPK but not Smad/1/5/8 (Fig. 7). On the other hand, BMP-2
used as a positive control activated both p38 MAPK and Smad1/5/8
Fig. 5. Structures of synthesized compounds 27–29 (A) and their effects on
mineralization in MC3T3-E1 cells (B). The experimental procedure is the same as
that in Fig. 2. The photographs show representative views of triplicate analysis.
with 14 or 21 for two days showed dose-dependent increases in
the expression levels of these four genes. Cells treated with
10 mM 14 expressed higher levels of Runx2 (1.50-fold), Osterix
(2.61-fold), Alpl (2.23-fold), and Bglap (1.92-fold) relative to con-
trol cells (Fig. 6). Meanwhile, cells treated with 10 mM 21 also
expressed higher concentrations of these 4 genes (Runx2, Osterix,
Alp1, Bglap, 1.42-, 2.41-, 2.09-, 1.33-fold, respectively) relative to
control cells (Fig. 6). The expression levels of Runx2 and Osterix
in cells treated with 14 and 21 at 10 mM were higher than the pre-
viously reported values for cells treated with 40 mM natural pro-
Fig. 7. Effects of potentially active derivatives (14 and 21) on phosphorylation of
p38 MAPK and Smad1/5/8 in MC3T3-E1 cells. Cells treated with osteoblast
differentiation medium (ODM) and test compounds (30 mM) for 5, 15, and 30 min
were lysed and analyzed by western blotting. Similar results were obtained in three
independent experiments and representative results are shown. The levels of
phosphorylated p38 and Smad1/5/8 proteins are shown as a ratio to p38 and Actb
respectively.
Please cite this article in press as: Kumagai M., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.04.016

M. Kumagai et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
5
(Supplementary Fig. 2). These results suggest that p38 MAPK acti-
vation is important for the osteogenic activities of 14 and 21.
To reveal participation of BMP signaling pathways in 14 and 21-
mediated effects on osteogenesis, MC3T3-E1 cells were treated
with specific inhibitors, treated with 14 and 21, and mineralization
was examined. Treatment with noggin (BMP antagonist) and LDN-
193189 (type1 BMP receptor kinase inhibitor) reduced 14 and 21-
promotion of MC3T3-E1 mineralization (Fig. 8). The same results
were observed for BMP-2 treated cells. These results suggest that
the osteogenic actions of 14 and 21 are mediated by activation of
BMP signaling pathways. Furthermore, treatment of cells with
SB203580 (p38 MAPK inhibitor) also reduced 14 and 21-induced
mineralization (Fig. 8). The activity of p38 kinase in osteogenesis
was reported to be related to its capacity to phosphorylate and
increase the activity of fundamental osteogenic transcription fac-
tors, including Runx2, Osterix, and Dlx5.25 Thus, p38-mediated
phosphorylation of these transcription factors by p38 may be
involved in the effect of 14 and 21 on osteogenesis.
Previously, we showed that natural 5,6-dehydrokawain did not
inhibit RANKL-induced TRAP-activity at 20 mM in RAW264 cells.
In this study, synthetic 1 also failed to inhibit TRAP activity, but
interestingly, analogs 14 and 21 did show inhibitory activity. In
particular, 21 had a stronger inhibitory effect that was nearly the
same as that seen for ipriflavone, which is used for clinical applica-
tions (Fig. 9A). Analogs 14 and 21 also suppressed formation of
RANKL-induced multinucleated cells (Fig. 9B). These results sug-
gested that the 5,6-dehydrokawain analogs 14 and 21 not only
increase osteogenic activities but also inhibit osteoclastogenesis.
In conclusion, we designed and synthesized 23 5,6-dehy-
drokawain analogs as potential anabolic agents. Similarly to the
natural version, synthetic 5,6-dehydrokawain promoted ALP activ-
ity and matrix mineralization. Among synthesized compounds, 14
analogs promoted higher levels of ALP activity and mineralization
than 5,6-dehydrokawain. The structure-activity relationships of 1
suggested that alkylation at the para and meta positions of the
In our previous report, we showed the activation of p38 MAPK
by treatment with 1.14 Our subsequent research revealed that the
osteogenic activity of 1 could be abolished by noggin, LDN-
193189, and SB203580 (data not shown). These results suggest
that 5,6-dehydrokawain (1) and its analogs 14 and 21 act on
MC3T3-E1 cells via a similar mechanism. The increased hydropho-
bicity of these molecules relative to parent compound may
enhance cell permeability, and in turn promote osteogenic activity.
Some skeletal disease is induced through excess osteoclastic
activity.2 Osteoclasts differentiate from macrophage cells in
response to RANKL stimulation.26 Thus, we examined the inhibi-
tory activities of 14 and 21 against RANKL-induced osteoclast dif-
ferentiation using the murine macrophage cell line RAW264.
Fig. 9. Effects of 5,6-dehydrokawain (1), derivatives 14, 21 and Ipriflavone (Ipri) on
RANKL-induced osteoclast differentiation from RAW264 cells. RAW264 cells were
cultured with the test compounds in the presence of RANKL (10 ng/mL) for 3 days.
After incubation, TRAP activities and cell numbers were measured. The data
represent means SD of triplicate analysis. *P < 0.05 and **P < 0.01 versus the
vehicle control by Dunnett’s test (A). For microscopic analysis, cells were fixed and
stained by TRAP. Arrows show TRAP-positive multinucleated cells containing more
than five nuclei. Each picture is representative of at least three independent
cultures (B). The scale bar shows 200 mm.
Fig. 8. Effects of BMP-2 inhibitors (Noggin and LDN-193189) and p38 inhibitor
(SB203580) on MC3T3-E1 cell mineralization induced by BMP-2 and compounds 14
and 21. MC3T3-E1 cells were pretreated for 0.5 h with inhibitors and then treated
for 10 days with test compounds in an osteoblast differentiation medium. Miner-
alization was visualized by Alizarin Red
S staining. The photographs show
representative views of triplicate analyses. Similar results were obtained in three
independent experiments and representative results are shown.
Please cite this article in press as: Kumagai M., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.04.016

6
M. Kumagai et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
phenyl ring was associated with strong osteogenic activity. Among
the synthesized analogs, the mechanistic action of analogs 14 and
21 were investigated in detail. These two compounds at 10 mM
both up-regulated mRNA expression of Runx2 and Osterix. Further
specific inhibitor experiments suggested that the osteogenic action
of 14 and 21 was mediated by the BMP signaling pathway and p38
MAPK activation. Interestingly, 14 and 21 also inhibited osteoclas-
togenesis at 20 mM despite the lack of 5,6-dehydrokawain-medi-
ated inhibition of this process. Although further research is
needed to define their action in vivo, 5,6-dehydrokawain analogs
could be promising therapeutic lead compounds for treatment of
osteoporosis.
References
1. Rachner TD, Khosla S, Hofbauer LC. Lancet. 2011;377:1276.
2. Rodan GA, Martin TJ. Science. 2000;289:1508.
3. Mohan S, Kutilek S, Zhang C, et al. Bone. 2000;27:471.
4. Yonezawa T, Lee JW, Akazawa H, et al. Bioorg Med Chem Lett. 2011;21:3248.
5. Ito T, Nakamura M, Nakamichi H, Ando M, Tsukamasa Y, Furuichi Y. Biosci
Biotech Biochem. 2014;78:92.
6. Watanabe A, Kumagai M, Mishima T, et al. PLoS ONE. 2015;10:e0127158.
7. Swioklo S, Watson KA, Williamson EM, Farrimond JA, Putnam SE, Bicknell KA. J
Nat Prod. 2015;78:2598.
8. Fonseca D, Ward WE. Bone. 2004;35:489.
ˇ
´
9. Filipovic´ B, Šošic´-Jurjevic´ B, Ajdzanovic V, et al. Osteoporos Int. 2010;21:1609.
10. Marini H, Minutoli L, Polito F, et al. Ann Intern Med. 2007;146:839.
11. Strong AL, Jiang Q, Zheng S, et al. ACS Med Chem Lett. 2014;5:143.
12. Strong AL, Ohlstein JF, Jiang Q, et al. Stem Cell Res Ther.. 2014;5:105.
13. Smith RM. Phytochemistry. 1983;22:1055.
Acknowledgments
14. Kumagai M, Mishima T, Watanabe A, et al. Biosci Biotech Biochem.
2016;80:1425.
15. Tanaka A, Hamada N, Fujita Y, et al. Bioorg Med Chem. 2010;18:3133.
16. McCracken ST, Kaiser M, Boshoff HI, Boyed PDW, Copp BR. Bioorg Med Chem.
2012;20:1482.
17. Nishikawa K, Fukuda H, Abe M, et al. Phytochemistry. 2013;96:132.
18. Quarles LD, Yohay DA, Lever LW, Caton R, Wenstrup RJJ. Bone Miner Res.
1992;7:683.
The authors would like to thank Mr. Masatoshi Watai and Dr.
Kazuhiro Fujita at the Japan Food Research Laboratories and Dr.
Yoshimitsu Tachi at the Osaka City University for their useful com-
ments and suggestions. The authors also thank Dr. Akio Watanabe
at Chubu University for his technical advices.
19. Stein GS, Lian JB, Owen TA. FASEB J. 1990;4:3111.
20. Komori T, Yagi H, Nomura S, et al. Cell. 1997;89:755.
21. Nakashima K, Zhou X, Kunkel G, et al. Cell. 2002;108:17.
22. Hughes FJ, Turner W, Belibasakis G, Martuscelli G. Periodontology.
2006;2000:48.
Supplementary data
23. Matsubara T, Kida K, Yamaguchi A, et al. J Biol Chem. 2008;283:29119.
24. Ulsamer A, Ortuno MJ, Ruiz S, et al. J Biol Chem. 2008;283:3816.
25. Rodríguez-Carballo E, Gámez B, Ventura F. Front Cell Dev Bio. 2016;4:40.
26. Wei S, Teitelbaum SL, Wang MWH, Ross FP. Endocrinology. 2001;142:1290.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2017.04.
016.
Please cite this article in press as: Kumagai M., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.04.016