Benzylideneacetone Derivatives Inhibit Osteoclastogenesis and Activate Osteoblastogenesis Independently Based on Specific Structure-Activity Relationship

Article abstract of DOI:10.1021/acs.jmedchem.9b00270

(E)-3,4-Dihydroxybenzylideneacetone (compound 1) inhibited receptor activator of NF-κB ligand-induced osteoclastogenesis of C57BL/6 bone marrow monocyte/macrophages with IC₅₀ of 7.8 μM (IC₅₀ of alendronate, 3.7 μM) while stimulating the differentiation of MC3T3-E1 osteoblastic cells, accompanied by the induction of Runt-related transcription factor 2, alkaline phosphatase, and osteocalcin. (E)-4-(3-Hydroxy-4-methoxyphenyl)-3-buten-2-one (compound 2c) showed a dramatically increased osteoclast-inhibitory potency with IC₅₀ of 0.11 μM while sustaining osteoblast-stimulatory activity. (E)-4-(4-Hydroxy-3-methoxyphenyl)-3-buten-2-one (compound 2g) stimulated alkaline phosphatase production 2-fold at 50 μM without changing osteoclast-inhibitory activity, compared with compound 1. Oral administration of compounds 1, 2c, and 2g prevented ovariectomy-induced osteoporosis in ddY mice to a degree proportional to their osteoclastogenesis-inhibitory potencies. The administration of 1 (mg/kg)/d compound 2c ameliorated histomorphometry of osteoporotic bone to a degree comparable with 10 (mg/kg)/d alendronate. Conclusively, the in vitro capacity of a few benzylideneacetone derivatives to inhibit osteoclastogenesis supported by independent osteoblastogenesis activation was convincingly reflected in in vivo management of osteoporosis, suggesting a potential novel therapeutics for osteopenic diseases.

Full text of DOI:10.1021/acs.jmedchem.9b00270

Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC
Article
Benzylideneacetone Derivatives Inhibit Osteoclastogenesis
and Activate Osteoblastogenesis Independently
Based on Specific Structure–Activity Relationship
Triveni Pativada, Myung Hwan Kim, Jung-Hun Lee, Seong Su Hong, Chun Whan Choi, Yun-
Hyeok Choi, Woo Jung Kim, Da-Woon Song, Serk In Park, Eun Jung Lee, Bo-Yeon Seo,
Hankyeom Kim, Hong Kyu Kim, Kee-Ho Lee, Sung k Ahn, Jin-Mo Ku, and Gil Hong Park
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b00270 • Publication Date (Web): 19 Jun 2019
Downloaded from http://pubs.acs.org on June 19, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Benzylideneacetone Derivatives Inhibit Osteoclastogenesis and Activate
Osteoblastogenesis Independently Based on Specific Structure–Activity
Relationship
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
#
†#
‡#
‡
‡
Triveni Pativada, Myung Hwan Kim, Jung-Hun Lee, Seong Su Hong, Chun Whan Choi,
‡
‡
‡
†
†
Yun-Hyeok Choi, Woo Jung Kim, Da-Woon Song, Serk In Park, Eun Jung Lee, Bo-Yeon
†
§
∥
⊥
∇
*‡
Seo, Hankyeom Kim, Hong Kyu Kim, Kee Ho Lee, Sung K Ahn, Jin-Mo Ku, and
Gil Hong Park*†
†Department of Biochemistry and Molecular Biology, College of Medicine, Korea Molecular
Medicine and Nutrition Research Institute, Korea University, Seoul 02841, Korea
‡Bio-Center, Gyeonggido Business & Science Accelerator, Suwon 16229, Korea
§Department of Pathology, Korea University Guro Hospital, Seoul 08308, Korea
∥
Department of Surgery, Seoul National University Hospital, Seoul 03080, Korea
⊥
Division of Radiation Cancer Research, Korea Institute of Radiological and Biomedical
Sciences, Seoul 01812, Korea
∇
Statistics, Department of Finance and Management Science, College of Business,
Washington State University, Pullman, WA 99164-4746, United States
ABSTRACT: (E)-3,4-Dihydroxybenzylideneacetone (compound 1), inhibited receptor
activator of NF-kB ligand-induced osteoclastogenesis of C57BL/6 bone marrow
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
monocyte/macrophages with IC50 of 7.8 µM (IC50 of alendronate, 3.7 µM), while stimulating
the differentiation of MC3T3-E1 osteoblastic cells, accompanied by the induction of Runt-
related transcription factor 2, alkaline phosphatase and osteocalcin. (E)-4-(3-Hydroxy-4-
methoxyphenyl)-3-buten-2-one (compound 2c) showed a dramatically increased osteoclast-
inhibitory potency with IC50 of 0.11 µM, while sustaining osteoblast-stimulatory activity. (E)-
4-(4-Hydroxy-3-methoxyphenyl)-3-buten-2-one (compound 2g) stimulated alkaline
phosphatase production 2 folds at 50 µM without changing osteoclast-inhibitory activity,
compared with compound 1. Oral administration of compounds 1, 2c, and 2g prevented
ovariectomy-induced osteoporosis in ddY mice to a degree proportional to their
osteoclastogenesis-inhibitory potencies. The administration of 1 mg/kg/d compound 2c
ameliorated histomorphometry of osteoporotic bone to a degree comparable with 10 mg/kg/d
alendronate. Conclusively, the in vitro capacity of a few benzylideneacetone derivatives to
inhibit osteoclastogenesis supported by independent osteoblastogenesis activation was
convincingly reflected in in vivo management of osteoporosis, suggesting a potential novel
therapeutics for osteopenic diseases.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
Bone is a dynamic organ in which the existing matrix is continually resorbed by osteoclasts of
1
hematopoietic origin, and replaced by osteoblasts of mesenchymal origin. Excessive bone
resorption by osteoclasts aggravates osteopenic diseases such as osteoporosis, thereby
increasing the frequency of arthralgia and fracture. Continuous bone resorption due to
abnormal osteoclastic activity and subsequent abnormal replacement of matrix by osteoblasts
leads to bone deformities which increases vulnerability to fractures; a condition known as
2
Paget’s disease of bone. Currently, the most widely used clinical therapeutics for osteopenic
ACS Paragon Plus Environment

Page 3 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
diseases include osteoclast-inhibitory agents such as bisphosphonates, including alendronate,3
and anabolics represented by parathyroid hormone (PTH) regimens, which act to enhance
4
osteoblast activity. Interactions between bone marrow and cancer cells, results in increased
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
bone turnover that facilitates cancer bone metastasis. Among many cell populations
constituting the bone marrow microenvironment, osteoclasts are key players in the
development of osteolytic lesions for cancer metastasis.6,7 Zoledronic acid, a bisphosphonate
drug which inhibits osteoclastogenesis, preserves the integrity of tumor-bearing bones and
8
increases the survival time of prostate cancer patients. When PTH was administered
concurrently with zoledronic acid, there was a remarkable decrease in the incidence of prostate
cancer metastasis to the skeleton, suggesting that simultaneously suppressing bone destruction
while promoting bone formation resulted in much greater suppression of tumor bone
metastasis.5
Osteoclasts differentiate from bone marrow monocyte/macrophage lineage cells (BMM)
which are under the control of two essential cytokines, macrophage colony-stimulating factor
(
M-CSF) and receptor activator of NF-kB ligand (RANKL).9,10 Additionally, signals arising
from antibodies such as autoimmune antibodies in rheumatoid arthritis that enhance receptor
activator of NF-kB (RANK) signaling, are necessary for complete differentiation of
osteoclasts.11,12 NF-kB activated by RANK in early osteoclastogenesis is one of the crucial
1
3
transcription factors underlying osteoclastogenesis, which stimulates the induction of nuclear
factor of activated T-cell cytoplasmic 1 (NFATc1), a master transcription factor for the
14
terminal differentiation of osteoclasts. Osteoblast-induced bone formation is regulated by
various hormones including PTH, growth factors, and cytokines, as well as by nutrition,
mechanical loading, and aging.15-18 Runt-related transcription factor 2 (RUNX2) is an early-
stage transcription factor strongly associated with osteoblastic differentiation and
1
9
calcification. RUNX2 increases bone formation by stimulating the transcription of bone-
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
forming proteins such as alkaline phosphatase (ALP), an early/intermediate stage marker of
osteoblastogenesis, and osteocalcin (OCN), a late-stage osteoblastic differentiation marker.
2
0
Bone-ALP is an indicator of bone formation, and OCN constitutes a major non-collagenous
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
21
matrix protein in bone. Homeostasis of bone involves tightly regulated interactions between
osteoblasts and osteoclasts, suggesting that osteoblast differentiation promotes osteoclast
2
2
differentiation. Thus, any modification of osteoclast differentiation by disease or drugs tends
to induce parallel changes in osteoblasts, posing challenges for the management of osteopenic
diseases.23,24 However, crosstalk between pathways regulating osteoclastogenesis and
osteoblastogenesis for bone homeostasis, have yet to be fully elucidated.
In this study, we have examined the capacity of benzylideneacetone derivatives for their
ability to independently promote osteoclastogenic inhibition and osteoblastogenic activation,
and determined specific structure-activity relationship (SAR) in an effort to search for key
structural determinants of these biological activities with the aim of discovering compounds
25
with increased potency. Compound 1 is a constituent of Osmunda japonica rhizome, and is
known to inhibit the production of immune mediators such as cytokines (TNF-alpha,
interleukin-1β, interleukin-6), NO, and prostaglandin E2 that play crucial roles in the
development of rheumatoid arthritis,26,27 as well as to activate osteoblast differentiation and
28
2
9
exert cancer-specific cytotoxicity. Osmunda regalis L., a species closely related to Osmunda
japonica, has been used in Spain for hundreds of years as a highly efficient remedy for treating
3
0
rheumatoid arthritis and pain, without clinically significant adverse effects. The rhizome of
ginger (Zingiber officinale) has been used as an herbal drug for the management of osteoporosis
31
in Indian ethnopharmacology as well as food additives, and suppresses RANKL-induced
3
2
osteoclastogenesis. Compound 2g is a pungent constituent of ginger, and compound 2c is its
33
synthetic isomer. The capacities of these compounds to prevent bone loss in vivo was
investigated in ovariectomy (OVX)-induced osteoporotic mice. Furthermore, we investigated
ACS Paragon Plus Environment

Page 5 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the effect of compounds 1, 2c, and 2g on the expression of differentiation markers including
key transcription factors for the differentiation of osteoclasts and osteoblasts during
osteoclastogenesis and osteoblastogenesis.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RESULTS
Synthesis of Novel Benzylideneacetone Derivatives. Derivatives of compound 1 (Figure
1) were designed and synthesized chemically de novo for SAR analysis, and isolated with
purities higher than 95% (Supporting Information 1). Compound 2g (CAS; 1080-12-2) is a
33
33
constituent of ginger. Compounds 2c (CAS; 22214-39-7) and 2h (CAS; 1704417-22-0) are
synthetic compounds reported previously. The remainder (2a, 2b, 2d, 2e, 2f, 3a, and 3b) were
novel compounds. Structures of synthesized compounds were identified by analysis of spectral
data including MS, 1-D and 2-D NMR (Supporting Information 2).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 1. Structures of compound 1 and its structural analogues.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Purity Determination of Compounds 2a−h and 3a, b. For qualitative analysis, an Agilent
1
200 series HPLC instrument (Agilent Technologies, Inc. Santa Clara, CA) was used. Analyses
were carried out on a Waters TC-C18 reversed-phase analytical column (2.1 × 50 mm, 2 μm)
Waters Co., Milford, MA). The detection wavelength was set at 250 nm. The mobile phase
(
consisted of the solvent A (0.1%, v/v solution of formic acid in water) and solvent B (0.1%,
v/v solution of formic acid in acetonitrile), and was subject to gradient elution. As a result, ≥95%
purity was confirmed for each compound tested. Purity and NMR data of the synthesized
compounds are presented in Supporting Information 1, 2.
Compound 1 Inhibits the Differentiation of Preosteoclasts in a Concentration-
Dependent Manner, Independently of Concurrent Proliferation of Co-Existing
Preosteoblasts. To evaluate the effect of compound 1 on osteoclast differentiation, we isolated
BMM (preosteoclasts) from C57BL/6 mice, and induced osteoclastogenesis in the presence of
a range of concentrations of compound 1, or of alendronate as a reference compound. BMM
differentiated into mature multinucleated osteoclasts in 6 d, in the presence of M-CSF and
RANKL, and were identified by tartrate-resistant acid phosphatase (TRAP) staining (Figure
2
A, Positive control). In contrast, BMM grown in the presence of M-CSF only, failed to
differentiate into mature multinucleated osteoclasts (Figure 2A, Negative control). Compound
1 suppressed the proliferation and differentiation of osteoclasts with an IC50 of approximately
7
.8 µM, and almost completely inhibited osteoclastogenesis at 10 µM (Figure 2A, Compound
). Alendronate inhibited osteoclastogenesis with an IC50 of approximately 3.7 µM in the same
1
setting (Figure 2A, Alendronate). To determine if compound 1 inhibits the bone resorption
function of fully differentiated mature osteoclasts, we performed a bone resorption assay with
ACS Paragon Plus Environment

Page 7 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mature multinucleated osteoclasts in the presence of compound 1 at the IC50 concentration.
Under the condition, it suppressed bone resorption of mature osteoclasts up to 59 ± 10% in 3 d
compared to that of untreated mature osteoclasts. Compound 1 inhibited the differentiation of
preosteoclasts without inhibiting the proliferation of co-existing preosteoblasts (MC3T3-E1
murine osteoblastic cells), as shown in the compound 1-treated co-culture of preosteoclasts and
preosteoblasts in the presence of differentiation factors for both cell types (Figure 2B,
Compound 1). Preosteoclasts failed to differentiate into mature multinucleated osteoclasts in 6
d, and preosteoblasts continued to proliferate in the presence of 10 µM compound 1. By
contrast, in the absence of compound 1, both preosteoclasts and preosteoblasts appeared to
proliferate and differentiate in the presence of differentiation factors until day 7 after induction
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(Figure 2B, Positive control). In this setting, differentiated osteoclasts were smaller than those
grown in the absence of osteoblasts, which may be attributed to steric hindrance of cell growth,
due to the high cell concentrations resulting from co-existence of proliferating preosteoclasts
and preosteoblasts. Meanwhile, preosteoclasts and preosteoblasts grown in the presence of M-
CSF only, continued to proliferate, but preosteoclasts failed to differentiate into mature
multinucleated osteoclasts (Figure 2B, Negative control). Collectively, the findings
demonstrated that compound 1 inhibits osteoclast differentiation in a concentration-dependent
manner independent of the concurrent proliferation of co-existing preosteoblasts.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Benzylideneacetone derivatives inhibit RANKL-induced osteoclastogenesis
independent of the proliferation of co-existing osteoblast. (A) Comparison of IC50 to inhibit
osteoclastogenesis between compounds 1 and 2c, and alendronate. C57BL/6 murine BMM was
grown in the presence of M-CSF and RANKL, and a range of concentrations of compound 1
and alendronate (0-10 µM), and compound 2c (0.02-2 µM). On the sixth day, mature
multinucleated osteoclasts were identified by TRAP staining, and observed by an inverted
microscope (Magnification: 40 ×. Scale bar = 50ꢀμm). (B) Compound 1 inhibits osteoclast
ACS Paragon Plus Environment

Page 9 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
differentiation while osteoblasts proliferate in the co-culture of preosteoclast and preosteoblast.
C57BL6 murine BMM and MC3T3-E1 murine osteoblastic cells were co-cultured in the
presence of osteoclast and osteoblast differentiation factors (Positive control). Negative control
co-culture was grown in the presence of M-CSF only as a differentiation factor. To test the
effects of compound 1 on the proliferation and differentiation of osteoclast and osteoblast,
compound 1 was added at a final concentration of 10 µM (Compound 1). Cells were subject to
TRAP staining on the sixth day, and observed by an inverted microscope (Magnification: 40
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
×. Scale bar = 50ꢀμm).
Specific SAR of Benzylideneacetone Derivatives Inhibiting Osteoclastogenesis. We next
compared different synthetic and natural derivatives of benzylideneacetone in order to
elucidate the critical structural determinants necessary for inhibition of osteoclastogenesis
(Table 1). Compound 2c replaced a hydroxyl group with a methoxy group at carbon 4 of
catechol of compound 1, and showed increased osteoclast-inhibitory activity of approximately
70-fold compared to compound 1, with IC50 of approximately 0.11 µM, while sustaining an
osteoblast-stimulatory activity as judged by ALP activation. Intriguingly, the replacement of a
hydroxyl group by a methoxy group at carbon 3 of catechol of compound 1, as exemplified by
compound 2g, did not change the osteoclast-inhibitory activity of compound 1 significantly.
Similarly, the replacement of a hydroxyl group by a methyl group at carbon 4, or by fluorine
(F) at carbon 3 of catechol of compound 1, exemplified by (E)-4-(3-hydroxy-4-
methylphenyl)but-3-en-2-one (2a) and (E)-4-(3-fluoro-4-hydroxyphenyl)-3-buten-2-one (2h),
did not affect the osteoclast-inhibitory activity of compound 1 significantly. By contrast, the
introduction of F or bulky chemical groups at carbon 4 of catechol of compound 1, exemplified
by
(E)-4-(4-fluoro-3-hydroxyphenyl)but-3-en-2-one
(2b),
(E)-4-(3-hydroxy-4-
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
isopropoxyphenyl)but-3-en-2-one (2d), (E)-4-(4-benzyloxy-3-hydroxyphenyl)but-3-en-2-one
(2e), and (E)-4-(4-allyloxy-3-hydroxyphenyl)but-3-en-2-one (2f) decreased the activity by 2-3
folds. The findings indicated that the nature of chemical group bound to carbon 4 of catechol
of compound 1, e.g. hydroxyl, methyl, methoxy, F or bulky chemical groups, determined the
potency to inhibit osteoclastogenesis, suggesting the presence of a specific receptor for a few
benzylideneacetone derivatives on the signaling pathway of osteoclastogenesis. Further, the
removal of a hydroxyl group at carbon 3 of catechol of compound 1 such as (E)-4-
hydroxybenzalacetone, decreased the osteoclast-inhibitory activity 4 folds. Notably, the
absence of both hydroxyl groups at carbons 3 and 4 of catechol, as exemplified by trans-
cinnamic acid and phenylacetic acid, resulted in a complete loss of osteoclast-inhibitory
activity. Meanwhile, natural products (caffeic acid, chlorogenic acid, neochlorogenic acid)
containing additional functional groups (e.g. carboxylic acid, glycosylate ester) on the aliphatic
ketone substructure of compound 1 showed 5- to 11-fold weaker inhibition of
osteoclastogenesis than that of compound 1. Likewise, synthetic carboxamides (3a, 3b)
displayed 2- to 3-fold lower osteoclast-inhibitory potency compared with compound 1.
Similarly, other natural products bearing catechol with a modified aliphatic ketone substructure
of compound 1, e.g. protocatechualdehyde (6), protocatechuic acid, protocatechuic acid methyl
ester, homoprotocatechuic acid, and homoprotocatechuic acid methyl ester, showed
remarkably diminished osteoclast-inhibitory activities. Collectively, the osteoclast-inhibitory
activity of the benzylideneacetone derivatives displayed a highly specific SAR with regard to
the functional group on catechol as well as the aliphatic ketone substructures. Specifically, the
nature of chemical groups bound to carbon 4 of catechol were identified as the key structural
determinants of benzylideneacetone derivatives necessary for inhibition of osteoclastogenesis.
Additionally, the aliphatic ketone substructure was required as an intact side chain for the full
manifestation of the osteoclast-inhibitory activity.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 11 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 1. Osteoclast Inhibition and Osteoblast Activation by Compound 1 Derivatives
%
Activation of ALP activity1
Compounds
TRAP
IC50 (µM)
10 µM
50 µM
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Osmundacetone (1)
7.8 ± 2
5.9 ± 1
120 ± 9
280 ± 60*
130 ± 30
100 ± 20
180 ± 20*
2a
100 ± 20
100 ± 20
100 ± 20
2
b
c
15 ± 1
2
0.11 ± 0.02
2
d
24 ± 3
19 ± 5
18 ± 8
100 ± 20
130 ± 8
120 ± 8
94 ± 10
140 ± 2
140 ± 2
2e
2f
2
g
h
7.5 ± 0.5
8.3 ± 2
15 ± 5
25 ± 2
15 ± 3
32 ± 4
41 ± 7
48 ± 4
86 ± 10
110 ± 30
100 ± 10
110 ± 20
110 ± 10
110 ± 10
110 ± 8
600 ± 70**
100 ± 20
120 ± 40
120 ± 20
140 ± 20
150 ± 30
130 ± 40
170 ± 30*
2
3a
3b
Protocatechualdehyde (6)
E)-4-Hydroxybenzalacetone
(
Caffeic acid
120 ± 30
120 ± 10
Neochlorogenic acid
Chlorogenic acid
110 ± 4
100 ± 6
150 ± 30
140 ± 20
100 ± 20
170 ± 20*
130 ± 30
110 ± 10
110 ± 10
Protocatechuic acid
89 ± 10
78 ± 10
75 ± 20
80 ± 20
>100-
Protocatechuic acid methyl ester
Homoprotocatechuic acid
Homoprotocatechuic acid methyl ester
trans-Cinnamic acid
110 ± 20
120 ± 20
100 ± 20
100 ± 30
120 ± 30
Phenylacetic acid
>100
Alendronate
3.7 ± 1
140±20
PTH-related peptide (1 µM)
1%
Activation of ALP activity was tested using MC3T3-E1 murine osteoblastic cells.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The values are the mean ± standard deviation (SD) of three independent experiments. *
**
P<0.05. P<0.01
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compound 1 Upregulates the Production of ALP and OCN by Preosteoblasts in a
Concentration-Dependent Manner. To assess the osteogenic capability of compound 1, we
tested whether it could induce the production of ALP by preosteoblasts. ALP production in
osteoblast-like MC3T3-E1 cells was increased by 280% by compound 1 at a concentration of
5
0 µM 14 d after administration when compared to control cells (P<0.05) (Table 1), a result
2
8
consistent with a previous report. Compound 1 increased ALP production by preosteoblasts
regardless of its simultaneous inhibition of differentiation of co-existing preosteoclasts (murine
BMM). Preosteoblasts and preosteoclasts were co-cultured in the presence of compound 1 (10
µM) as well as differentiation factors for both, which resulted in complete inhibition of
osteoclast differentiation on day 6 (Figure 2B, Compound 1). The co-culture of both cell types
continued to grow for 21 d. The activation of ALP production by 10 µM compound 1-treated
preosteoblasts at 7, 14, and 21 d after induction of co-culture differentiation, was 100%, 110%,
and 100%, respectively compared with untreated cells, therefore, quite similar to the 120±9%
activation of ALP production that occurred 14 d after induction by 10 µM compound 1-treated
osteoblasts grown in the absence of osteoclasts. Overall, the findings indicated that compound
1
activated osteoblastogenesis, while inhibiting osteoclastogenesis.
OCN is expressed only near or at the time of mineralization, i.e., 14-21 d after the induction
21
of osteoblastogenesis. To investigate the effect of compound 1 on the production of OCN by
osteoblasts, MC3T3-E1 preosteoblastic cells were cultured in the presence of 50 µM compound
1. Western blot analysis revealed that OCN production 7, 14, and 21 d after the induction of
osteoblastogenesis, increased by 1.8-, 4.1- (P<0.01), and 2.0- (P<0.05) fold, respectively,
ACS Paragon Plus Environment

Page 13 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compared with untreated positive control cells (Figure 3A,B). In the preosteoblast and
preosteoclast co-cultures, supplemented with both osteoblast and osteoclast differentiation
factors, compound 1 maintained the capacity to induce production of OCN by osteoblasts,
while inhibiting osteoclast differentiation (Figure 3C,D). When preosteoblasts in co-cultures
were treated with compound 1 (10 µM), OCN production was similar to that seen with
untreated positive control cells. In sum, compound 1 proved capable of increasing the
production of ALP and OCN by osteoblasts, while inhibiting osteoclast differentiation.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Compound 1 increases the production of OCN during osteoblastogenesis while
inhibiting osteoclastogenesis. (A) Compound 1 upregulates the production of OCN by
osteoblasts during differentiation. MC3T3-E1 murine osteoblastic cells were induced to
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
differentiate for 21 d in the presence or absence of compound 1 as described in Experimental
Section. The amounts of OCN in culture media were quantitated by Western blot 7, 14, and 21
d after the induction. NC (negative control), cells grown in the absence of osteoblast
differentiation factors; PC (positive control), cells grown in the presence of osteoblast
differentiation factors; C1, positive control cells grown in the presence of 50 µM compound 1.
(B) Bar graph representation of mean ± SD of three independent experiments performed as in
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(A). *P<0.05, **P<0.01. (C) Compound 1 upregulates the production of OCN by osteoblasts
over the course of differentiation independent of the inhibition of differentiation of co-existing
osteoclasts. MC3T3-E1 murine osteoblastic cells and C57BL6 murine BMM were induced to
differentiate for 21 d in the presence or absence of compound 1 as described in Experimental
Section. The amounts of OCN in culture media were quantitated by Western blot 7, 14, and 21
d after the induction. NC (negative control), cells grown in the presence of M-CSF only as a
differentiation factor; PC (positive control), cells grown in the presence of osteoblast and
osteoclast differentiation factors; C1, positive control cells grown in the presence of 10 µM
compound 1. (D) Bar graph representation of the levels of OCN in (C).
Specific SAR of Benzylideneacetone Derivatives Activating Osteoblastogenesis.
Chemical modifications of the catechol substructure of compound 1 dramatically altered its
ability to activate ALP production by osteoblasts (Table 1). Compound 2g which had a
hydroxyl group replaced by a methoxy group at carbon 3 of catechol of compound 1, displayed
a 2-fold increased capacity to activate osteoblasts to produce ALP at 50 µM, compared with
compound 1 (P<0.01), while showing similar osteoclast-inhibitory potency to that of
compound 1. By contrast, compound 2c with a methoxy instead of a hydroxyl at carbon 4 of
catechol resulted in diminished capacity at ALP induction by osteoblasts. Compounds 2d, 2e,
ACS Paragon Plus Environment

Page 15 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and 2f substituting bulky isopropoxy, benzyloxy or allyloxy groups at carbon 4 of catechol, to
examine the effect of steric hindrance of functional groups on the stimulation of
osteoblastogenesis, displayed remarkably reduced capacities in activating ALP production.
Introducing F at carbons 3 or 4 of catechol like in compounds 2b and 2h, completely
annihilated the capacity of compound 1 to activate ALP production, in contrast with their
sustenance of osteoclast-inhibitory potentials. The findings indicated that the nature of
chemical group bound to carbon 3 or 4 of catechol of compound 1, e.g. hydroxyl, methyl,
methoxy, F or bulky chemical groups, increased or decreased the potency to activate
osteoblastogenesis markedly, suggesting the presence of a specific receptor for a few
benzylideneacetone derivatives on the signaling pathway of osteoblastogenesis. Similarly,
removal of the hydroxyl at carbon 3 of catechol, e.g. (E)-4-hydroxybenzalacetone, also resulted
in loss of functionality. Notably, the absence of hydroxyl groups at both carbons 3 and 4 of
catechol, exemplified by trans-cinnamic acid and phenylacetic acid, resulted in complete loss
of ALP-stimulating ability, likewise to their inability to inhibit osteoclast differentiation.
Meanwhile, any structural modification of the aliphatic ketone substructure of
benzylideneacetone derivatives, i.e., natural products (caffeic acid, chlorogenic acid,
neochlorogenic acid) containing additional chemical groups (e.g. carboxylic acid, glycosylate
ester), as well as other natural products bearing catechol with a modified aliphatic ketone
substructure, e.g. protocatechualdehyde (6), protocatechuic acid, protocatechuic acid methyl
ester, homoprotocatechuic acid, homoprotocatechuic acid methyl ester, and synthetic
carboxamides (3a, 3b) caused remarkable reductions in the capacity to activate ALP production
by osteoblast. Collectively, the capacity of benzylideneacetone derivatives to stimulate ALP
production by osteoblasts displayed a specific SAR, with any change in the molecular structure
remarkably altering the biological activity. Most importantly, the nature of chemical group
bound to carbons 3 and 4 of catechol was identified as the key structural determinant.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Benzylideneacetone derivatives also required the intact aliphatic ketone substructure for the
full manifestation of its biological potency.
Effects of Compounds 1, 2c, and 2g on the Expression of NF-kB and NFATc1. We
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
investigated the effects of compounds 1, 2c, and 2g on the expression of key transcription
factors involved in RANKL-induced osteoclastogenesis, NF-kB13 and NFATc114 (Figure
4
A,B). NF-kB expression began to increase in preosteoclasts induced by M-CSF and RANKL
(positive control, PC) 2 d after induction of osteoclastogenesis, and then returned to a basal
level 5 d after induction (Figure 4A, PC). The preosteoclasts differentiated into mature
multinucleated osteoclasts after 6 d. Compound 1 (10 µM) did not affect the NF-kB expression
profile of positive control preosteoclasts over the course of osteoclastogenesis significantly
(Figure 4A, PC + C1), while markedly inhibiting osteoclastogenesis. Intriguingly, the
replacement of a hydroxyl group at carbon 4 or 3 of catechol of compound 1 by a methoxy
group, compounds 2c (0.5 µM) and 2g (10 µM), increased the NF-kB expression of positive
control preosteoclasts markedly 1 d after induction and continued to increase the expression
during the period of osteoclastogenesis (Figure 4A, PC + 2c, PC + 2g), while simultaneously
inhibiting osteoclastogenesis. NFATc1 expression increased in positive control preosteoclasts
3 d after RANKL-induced differentiation, and continued to increase (Figure 4B, PC).
Compound 1 (10 µM) showed no influence on the NFATc1 expression profile (Figure 4B, PC
+
C1), while inhibiting osteoclastogenesis. Intriguingly, compounds 2c (0.5 µM) and 2g (10
µM) reduced the induction of NFATc1 expression over the course of RANKL-induced
osteoclastogenesis compared with positive control cells and compound 1-treated positive
control cells. Thus, compounds 2c and 2g regulated the expression of NFATc1 as well as NF-
kB in a similar way, but differently from compound 1 during RANKL-induced
osteoclastogenesis.
ACS Paragon Plus Environment

Page 17 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Next, we investigated the effects of compounds 1, 2c, and 2g on the expression of NF-kB
and NFATc1 in proliferating murine BMM induced by M-CSF in the absence of RANKL
(negative control, NC). NF-kB expression started to increase remarkably 2 d after
administration of M-CSF (Figure 4A, NC), and the expression profile was not significantly
affected by the presence of compounds 1, 2g, and 2c (Figure 4A, NC + 1, NC + 2c, NC + 2g).
NFATc1 expression in M-CSF-induced murine BMM increased progressively from day 2 to
day 5 (Figure 4B, NC). By contrast, in the presence of compounds 1, 2g, and 2c, the NFATc1
expression induced by M-CSF was remarkably suppressed (Figure 4B, NC + 1, NC + 2c, NC
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
+
2g). In sum, compound 1 had no effect on the expression of NF-kB and NFATc1 over the
course of RANKL-induced osteoclastogenesis, while still inhibiting osteoclastogenesis.
Whereas, compounds 2c and 2g increased the NF-kB expression remarkably, and diminished
the induction of NFATc1 expression during RANKL-induced osteoclastogenesis, particularly
compound 2c accompanying a tremendously increased potency inhibiting osteoclastogenesis.
Meanwhile, in M-CSF-induced proliferating BMM in the absence of RANKL, compounds 1,
2
c, and 2g all suppressed the induction of NFATc1 expression, while not affecting NF-kB
expression significantly.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 19 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 4. The effects of benzylideneacetone derivatives on the expression of osteoclast
transcription factors, NF-kB and NFATc1, and an osteoblast transcription factor, RUNX2 over
the course of osteoclastogenesis and osteoblastogenesis. (A) Compound 1 did not affect NF-
kB expression, but compound 2c and 2g increased NF-kB expression markedly over the course
of RANKL-induced osteoclastogenesis, while, during M-CSF-induced BMM proliferation, all
of compounds 1, 2c, and 2g did not affect NF-kB expression significantly. C57BL/6 murine
BMM were induced to differentiate, and harvested 0, 1, 2, 3, and 5 d after the induction. NF-
kB expression was quantitated by Western blot. NC (negative control): BMM grown in the
presence of M-CSF alone as a differentiation factor. NC + C1: negative control BMM treated
with 10 µM compound 1. NC + 2c: negative control BMM treated with 0.5 µM compound 2c.
NC + 2g: negative control BMM treated with 10 µM compound 2c. PC (positive control):
BMM grown in the presence of M-CSF and RANKL. PC + C1: positive control BMM treated
with 10 µM compound 1. PC + 2c: positive control BMM treated with 0.5 µM compound 2c.
PC + 2g: positive control BMM treated with 10 µM compound 2g. (B) Compound 1 did not
affect NFATc1 expression, but compound 2c and 2g suppressed the induction of NFATc1
expression markedly over the course of RANKL-induced osteoclastogenesis, while, during M-
CSF-induced BMM proliferation, all of compounds 1, 2c, and 2g suppressed NFATc1
expression. The murine BMM over the course of osteoclastogenesis was prepared as in (A).
NFATc1 expression was quantitated by Western blot. NC, NC + C1, NC + 2c, NC + 2g, PC,
PC + C1, PC + 2c and PC + 2g designate the same as in (A). (C) Compound 1 increased the
expression of RUNX2 in osteoblasts during differentiation. MC3T3-E1 murine osteoblastic
cells were induced to differentiate in the presence or absence of 50 µM compound 1 for 21 d.
The cells were harvested 7, 14, and 21 d after the induction. RUNX2 expression was quantitated
by Western blot. NC (negative control), cells grown in the absence of osteoblast differentiation
factors; PC (positive control), cells grown in the presence of osteoblast differentiation factors;
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
C1, positive control cells grown in the presence of compound 1. (D) Bar graph representation
of mean ± SD of three independent experiments performed as described in (C). *P<0.05.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compound 1 Upregulates the Expression of RUNX2 during Osteoblastogenesis. Next,
we tested if compound 1 increased the expression of RUNX2, a transcription factor and an
early differentiation marker that increases bone formation by stimulating the transcription of
ALP and OCN in preosteoblasts.19 Compound 1 increased the expression of RUNX2 in
preosteoblasts over the course of differentiation by 1.7, 1.4, and 3.2 (P<0.05) folds 7, 14, and
2
1 d after induction, respectively, compared with the corresponding compound 1-untreated
positive control cells (Figure. 4C,D, PC, C1). This finding indicated that compound 1
upregulated and prolonged the induction of RUNX2 expression in preosteoblasts, partly
explaining the molecular mechanism behind the ability of compound 1 to activate
osteoblastogenesis.
Pharmacokinetic Behavior of Compounds 1 and 2c. Next, we investigated the
pharmacokinetic (PK) behavior of compounds 1 and 2c (Table 2). Plasma concentration vs
time plots for compounds 1 and 2c were obtained to extract important PK parameters
(Supporting Information 3). The plasma concentrations represented the sum of unbound
compounds free from plasma proteins and those reversibly bound to plasma proteins, based on
the experimental procedure to measure the concentrations. T1/2 (terminal half-life) is the time
required to reach half the plasma concentration at pseudo-equilibrium, a parameter controlled
by the volume of distribution and plasma clearance mainly dependent on hepatic clearance
(metabolic stability and biliary clearance) and renal clearance, when absorption is not an
3
4
influencing factor, e.g. intravenous (IV) administration. T1/2 of IV-administered compounds
and 2c were in proportion to AUC (area under the plasma concentration-time curve from
1
∞
ACS Paragon Plus Environment

Page 21 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
time zero to infinity, indicating total blood exposure concentration). AUC of compounds 1
∞
and 2c administered IV at 2 mg/kg were 39 and 100 nM∙h, respectively, and T1/2 of those were
0.40 and 1.0 h, respectively, indicating that compound 2c is more stable than compound 1 in
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
blood plasma. The difference between AUC of compounds 1 and 2c might be ascribed to the
∞
different efficiency of plasma clearance system for each compound. On per os (PO)
administration, the rate and extent of absorption is more reflected in T1/2 than the elimination
3
4
process in plasma. T1/2 of compound 1 administered PO at 10 and 50 mg/kg (2.1 and 3.1 h,
respectively) was much higher than T1/2 of the IV administration at 2 mg/kg (0.40 h). The
finding might be due to much higher AUC
mg/kg (140 and 660 nM∙h, respectively) than AUC
39 nM∙h). The high value of AUC of PO-administered compound 1 might be ascribed to the
∞
of compound 1 administered PO at 10 and 50
∞
of compound 1 administered IV at 2 mg/kg
(
∞
high bioavailability (F) of compound 1 (66 and 76% at 10 and 50 mg/kg), indicating
AUCP.O./AUCI.V.. F of compound 1 was 2-3 fold higher than 26-29% of compound 2c in the
dose range of 10-50 mg/kg PO. As might be deduced based on T1/2 on IV administration and
F, AUC∞ of compounds 1 and 2c administered PO at 10 or 50 mg/kg were similar in systemic
circulation, 140 or 660 nM∙h for compound 1, and 140 or 720 nM∙h for compound 2c,
respectively. Likewise, Cmax (maximum plasma concentration) of compounds 1 and 2c
administered PO were not different significantly. Cmax of compound 2c administered PO at 10
and 50 mg/kg was 0.16 and 0.88 µM, respectively, which were higher than IC50 of the
compound (0.11 µM) to inhibit osteoclastogenesis. The pharmacokinetic behavior of
compound 1 via oral administration belongs to two-compartment model, which divides the
body into central and peripheral compartment. The central compartment consists of the plasma
where the distribution of the compound 1 is practically instantaneous (distribution phase). The
peripheral compartment consists of tissues where the distribution of compounds is slower. The
decrease of compound 1 in peripheral compartment is likely attributed to drug metabolism and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
excretion (elimination phase). So we calculated T1/2 of orally administered compound 1 in
elimination phase. However, the pharmacokinetic behavior of compound 1 via IV route showed
single compartment model where the concentration of compound 1 rapidly decreased and
disappeared in plasma.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Pharmacokinetic Parameters of Compounds 1 and 2c
Dose
(mg/kg)
Cmax
AUC
∞
(nM∙h)
Compounds
Route
T1/2 (h)
F (%)
(
nM)
IV
PO
PO
2
10
50
0.40
2.1
-
39 ± 10
140 ± 10
660 ± 200
-
1
200 ± 60
66
76
3.1
1100 ± 300
IV
PO
PO
2
10
50
1.0
1.2
2.2
-
100 ± 5
140 ± 30
720 ± 400
-
2c
160 ± 50
880 ± 300
26
29
T
1/2; terminal half-life. Cmax; maximum plasma concentration. AUC ; area under the plasma
∞
concentration-time curve from time zero to infinity, indicating total blood exposure
concentration. F; bioavailability indicating AUCP.O./AUCI.V. × 100. The values of Cmax and
AUC are the mean ± SD obtained from three male ICR mice.
∞
Preventive Effects of Compounds 1, 2c and 2g on OVX-Induced Osteoporosis. The
potential of benzylideneacetone derivatives as novel anti-osteoporotic agents was tested in
OVX-induced osteoporotic ddY mice. Compound 1 (10 mg/kg/d), compound 2g (10 mg/kg/d)
or compound 2c (1 and 10 mg/kg/d) in 300 µL PBS were administered PO to OVX mice for 4
weeks using gavage, beginning 1 week after OVX. Alendronate (10 mg/kg/d) in 300 µL PBS
was administered PO to OVX mice as a reference compound. PBS as a vehicle was
administered 300 µL/d PO to OVX and sham controls. Body weights of OVX control,
benzylideneacetone derivatives- or alendronate-administered OVX mice increased remarkably
5
weeks after OVX compared with sham control (Supporting Information 4). Femur and tibia
ACS Paragon Plus Environment

Page 23 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of each mouse were subjected to scanning (SkyScan1173 Ver. 1.6, Bruker-CT, Kontich,
Belgium), and were reconstructed using Nrecon Ver. 1.7.0.4 (Bruker, Kontich, Belgium) based
on the Feldkamp algorithm. Representative images of CT scans and 3-D representations of
each group are presented in Figure 5, and those of all mice in all experimental groups are
presented in Supporting Information 5. After reconstruction of the femur images of all mice in
each group, the histomorphometrical indices and regenerated bone volume, including bone
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
mineral density (BMD, g/mm ), percent bone volume (BV/TV, %), crossectional thickness
(Cs.Th, mm), trabecular number (Tb.N, number/mm), trabecular thickness (Tb.Th, mm),
3
trabecular separation (Tb.Sp, mm), and new bone volume (NBV, mm ) were obtained (Figure
6
, Supporting Information 6). OVX mice showed significantly deteriorated histomorphometric
indices of bone including BMD, BV/TV, Cs.Th, Tb.N, Tb.Th, Tb.Sp, and NBV, displaying
osteoporotic features compared with sham control. Oral administration of benzylideneacetone
derivatives or alendronate for 28 d to OVX mice led to the recovery of all the
histomorphometric indices mainly in proportion to their potencies at inhibiting
osteoclastogenesis. The BMD of femur increased significantly (P<0.01) in all
benzylideneacetone derivatives- and alendronate-treated groups with the highest seen with
compound 2c (1 or 10 mg/kg/d) (P<0.001) and alendronate (10 mg/kg/d) (P<0.01) groups
compared with OVX control. BV/TV was recovered significantly (P<0.05) in all
benzlideneacetone derivatives- and alendronate-treated groups with the highest seen in
compound 2c (1 or 10 mg/kg/d) (P<0.001) and alendronate (10 mg/kg/d) (P<0.05) groups
compared with OVX control. Cs.Th was significantly recovered (P<0.05) in compound 2c (1
mg/kg/d) and alendronate (10 mg/kg/d) groups up to the levels comparable with sham control,
and compounds 1, 2c and 2g (10 mg/kg/d, respectively) groups also showed recoveries in Cs.Th
compared with OVX control. Tb.N increased significantly in compound 2c (1 mg/kg/d)
(P<0.01), compound 2c (10 mg/kg/d) (P<0.05) and alendronate (10 mg/kg/d) (P<0.05) groups,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and compounds 1 and 2g (10 mg/kg/d, respectively) groups also showed increases compared
with OVX control. Tb.Th was significantly recovered in compound 2c (10 mg/kg/d) group
(P<0.01), and compound 2c (1 mg/kg/d), compound 1 (10 mg/kg/d), and compound 2g (10
mg/kg/d) groups also showed recoveries compared with OVX control. By contrast, alendronate
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(10 mg/kg/d) failed to increase Tb.Th compared with OVX control. Tb.Sp was significantly
reduced (P<0.05) by compound 2c (1 mg/kg/d) and alendronate (10 mg/kg/d) compared with
OVX control, up to the levels comparable with the sham control, and Tb.Sp of compounds 1,
2
c and 2g (10 mg/kg/d, respectively) groups also decreased compared with OVX control. NBV
increased significantly (P<0.05) with compound 2c (1 or 10 mg/kg/d) compared with OVX
control, and NBV of compounds 1, 2g, and alendronate (10 mg/kg/d, respectively) groups also
increased compared with OVX control.
Collectively, the findings demonstrated that all benzylideneacetone derivatives display the
capacity to prevent bone loss presumably through their capacity to inhibit osteoclastogenesis.
C
max of 0.16 µM for compound 2c administered 10 mg/kg PO was higher than IC50 inhibiting
osteoclastogenesis in vitro, explaining its osteoporosis-preventive efficacy. The in vivo
osteoporosis-preventive effect of 1 mg/kg/d compound 2c implied that the effective therapeutic
concentration of compound 2c was reached in vivo by the administration of 1 mg/kg/d PO to
show the efficacy comparable to alendronate administration of 10-fold more daily dose, which
was ascribed to 30-fold stronger potential of compound 2c for in vitro osteoclast inhibition than
that of alendronate. While Cmax of 0.20 µM for compound 1 administered 10 mg/kg/d PO was
lower than its in vitro IC50 of 7.8 µM against osteoclastogenesis, compounds 1 and 2g (IC50 7.5
µM) exhibited an in vivo osteoporosis-preventive effect up to half the level of improvement by
alendronate (IC50 3.7 µM), which indicated that the in vitro osteoclastogenesis-inhibitory
potencies were reflected proportionally in the degree of in vivo osteoporosis-preventive
efficacy. Even though Cmax of 0.20 µM for compound 1 was much lower than the concentration
ACS Paragon Plus Environment

Page 25 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
substantially inducing osteoblastogenesis in vitro, it was suggested that the osteoblastogenesis-
stimulatory potential of compound 1 may be sustained to preserve normal osteoblastic activity
in vivo while inhibiting osteoclast, based on the fact that compound 1 caused higher Tb.Th than
alendronate despite its IC50 of half the level of alendronate inhibiting osteoclastogenesis.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Representative CT and 3-D images of femur and tibia of mice in different
experimental groups. Beginning a week after OVX surgery, mice received 10 mg/kg/d of
compounds 1, 2g, or alendronate, or 1 or 10 mg/kg/d of compound 2c in 300 μL of 0.5% (v/v)
DMSO-containing PBS PO once daily using gavage. Sham and OVX control received 300 μL
of 0.5% (v/v) DMSO-containing PBS once daily. After 28 d. mice were euthanized, and femur
and tibia of each mouse were fixed and subject to CT analysis to obtain CT and 3-D images.
The results of anti-osteoporotic effects of the benzylideneacetone derivatives were compared
with those of alendronate. A. Sham control. B. OVX control. C. Alendronate 10 mg/kg/d. D.
Compound 1 10 mg/kg/d. E. Compound 2g 10 mg/kg/d. F. Compound 2c 10 mg/kg/d. G.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Compound 2c 1 mg/kg/d. a. CT images of a region of interest, b. 3-D images of a region of
interest. OVX = ovariectomy.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Histomorphometrical indices and regenerated bone volume after reconstruction of
the CT images of femur of each mouse in different experimental groups. The CT images of
femur of each mouse in all groups were reconstructed using a computer program, Nrecon Ver.
ACS Paragon Plus Environment

Page 27 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
.7.0.4 based on the Feldkamp algorithm to determine histomorphometrical indices in defect
areas. The reconstruction by NRecon was performed using the same applied scan and
reconstruction parameters for all specimens. NBV (new bone volume) was calculated within
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
the regions of interest. BMD (bone mineral density, g/mm ), BV/TV (percent bone volume, %),
Cs.Th (crossectional thickness, mm), Tb.N (trabecular number, number/mm), Tb.Th
3
(trabecular thickness, mm), Tb.Sp (trabecular separation, mm), NBV (mm ). Compounds 1, 2c
and 2g = test compounds. Alendronate = a reference compound. Sh = sham control mice. OVX
=
ovariectomy control mice. Alen = alendronate 10 mg/kg/d. C1 = compound 1 10 mg/kg/d.
2
g = compound 2g 10 mg/kg/d. 2c∙10 = compound 2c 10 mg/kg/d. 2c∙1 = compound 2c 1
mg/kg/d. Values are expressed as mean ± SD of four mice. *P<0.05, **P<0.01, ***P<0.001.
Cancer-Specific Cytotoxicity of Benzylideneacetone Derivatives. Any compound with
3
5
LD50 less than 100 µM is considered cytotoxic and antiproliferative. Accordingly, compound
1
did not exhibit cytotoxicity against normal cell lines based on 3-(4,5-dimethylthiazol-2-yl)-
,5-diphenyltetrazolium bromide (MTT) assay used to estimate cellular growth and survival
2
(Supporting Information 7).36 Specifically, compound 1 displayed negligible cytotoxicity
against human adipose-derived mesenchymal stem cells, HaCaT human epidermal
keratinocytes, and NIH3T3 mouse embryo fibroblasts with LD50 values of 2,800, 3500, and
>5,000 µM, respectively. It did however, show weak antiproliferative activity on RAW264.7
murine macrophage cell line with LD50 of 500 µM in the absence of RANKL, suggesting that
compound 1 may suppress the proliferation of a few phagocytic cells at high concentrations.
This supports a report showing that compound 1 suppressed the production of immune
2
6
mediators. It exhibited considerable cytotoxicity against cancer cell lines including AGS
human gastric adenocarcinoma, PC3 human prostate adenocarcinoma, and B16F10 mouse
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 66
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
melanoma with LD50 values of 60, 66, and 65 µM, respectively, supporting a cancer-specific
2
9
cytotoxicity that has been reported. Compounds 2c and 2g, did not exert an antiproliferative
activity on preosteoclastic RAW264.7 macrophage cells with LD50 of >5,000 µM in the
absence of RANKL in spite of a strong potency to inhibit RANKL-induced osteoclastogenesis,
suggesting that compound 2c and 2g may not have any antiproliferative activity on
macrophages that are not undergoing osteoclastogenesis. Compound 2c displayed a relatively
high antiproliferative activity on a few cancer cell lines including T24 human bladder
carcinoma cells with LD50 of 180 µM, and compound 2g on HCT116 human colon carcinoma
cells with LD50 of 310 µM in line with its specific cytotoxicity against HT-29 human colon
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
3
cancer cells. Overall, compounds 1, 2c and 2g showed negligible anti-proliferative effects
against normal cell lines tested, but exhibited significant cytotoxicity against a few cancer cell
lines.
DISCUSSION AND CONCLUSIONS
We demonstrated that a few benzylideneacetone derivatives strongly inhibit osteoclastogenesis
and activate osteoblastogenesis independently, which stands in contrast to earlier observations
that the differentiation of osteoclasts and osteoblasts tends to occur in parallel.23,24 In particular,
it is worth noting that the in vitro capacities of a few benzylideneacetone derivatives were
reflected in the management of OVX-induced osteoporosis mainly in proportion to their in
vitro potencies to inhibit osteoclastogenesis. Compound 2c administered 1 mg/kg/d PO
displayed a preventive effect on the development of osteoporosis in OVX mice comparable to
that of 10 mg/kg/d alendronate. The observation was consistent with the finding that IC50 of
compound 2c to inhibit in vitro osteoclastogenesis of C57BL/6 murine BMM was 30-fold
stronger than 3.7 µM of alendronate. The IC50 value of alendronate in our setting was
ACS Paragon Plus Environment

Page 29 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compatible with the previous finding that IC50 of alendronate was 460 nM against recombinant
human farnesyl diphosphate synthase, 1.7 µM against isopentenyl diphosphate isomerase and
farnesyl diphosphate synthase in a liver cytosolic extract, and 15 µM at murine osteoclast
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
37
cellular level to inhibit protein prenylation and mevalonate-derived lipid synthesis. Cmax of
compounds 1 and 2c administered a single dose of 10 mg/kg PO was 200 and 160 nM,
respectively, and thus Cmax of compound 2c administered 1 mg/kg/d PO might be much lower
than in vitro IC50 against osteoclastogenesis. However, the finding that 1 mg/kg/d compound
2
c PO exhibited expected pharmacological effects comparable to those of 10 mg/kg/d
alendronate may indicate that very low target engagement is required for desired therapeutic
effects of compound 2c. Intriguingly, alendronate displayed osteoclastogenesis-inhibitory
effects at in vivo Cmax concentrations much lower than in vitro IC50 at enzymatic as well as
cellular levels. Cmax of alendronate was 227−260 nM after the single 70 mg dose of three
generic and reference products, which was slightly higher than Cmax of compound 1 and 2c
38
(
200 and 160 nM at single dose of 10 mg/kg PO, respectively). In other study, healthy males
were administered single 70 mg dose of alendronate sodium regimens, reference and
bioequivalent test drugs, where Cmax of each was 234 and 199 nM, respectively, which were
3
9
slightly higher than Cmax of single dose of 10 mg/kg PO compound 1 and 2c. Thus, we suggest
the possibility that compounds 1 and 2c may also exert desired therapeutic effects toward the
management of osteoporosis at Cmax concentrations much lower than in vitro IC50, and that low
target engagement may be required at the molecular level inside cells to block
osteoclastogenesis to occur, to similar degrees to alendronate.
SAR analysis conclusively demonstrated that any slight modification in catechol and
aliphatic ketone substructures of benzylideneacetone derivatives affect differentially the
potencies to inhibit osteoclastogenesis and activate osteoblastogenesis. Indeed, the replacement
of a hydroxyl group at carbon 4 of catechol of compound 1 increased IC50 against
ACS Paragon Plus Environment