Small molecules with potent osteogenic-inducing activity in osteoblast cells

Article abstract of DOI:10.1016/j.bmcl.2009.01.025

A chemical screen of 45,000 compounds from a diverse collection led to the identification of two series of small molecules with potent osteogenic activity in mouse MC3T3-E1 osteoblast cells. The first chemical group was characterized by an amino benzothia

Full text of DOI:10.1016/j.bmcl.2009.01.025

Bioorganic & Medicinal Chemistry Letters 19 (2009) 1442–1445
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Small molecules with potent osteogenic-inducing activity in osteoblast cells
Chun-Ya E. Han a, Youping Wang a, Longchuan Yu a, David Powers b, Xiaoling Xiong b, Violeta Yu b,
Yen Nguyen b, David J. St. Jean Jr. c, Philip Babij a,
*
a Department of Metabolic Disorders, Amgen Inc., One Amgen Center Drive, Thousand Oaks, CA 91320, USA
b Department of HTS Molecular Pharmacology, Amgen Inc., One Amgen Center Drive, Thousand Oaks, CA 91320, USA
c Department of Small Molecule Chemistry, Amgen Inc., One Amgen Center Drive, Thousand Oaks, CA 91320, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A chemical screen of 45,000 compounds from a diverse collection led to the identification of two series of
small molecules with potent osteogenic activity in mouse MC3T3-E1 osteoblast cells. The first chemical
group was characterized by an amino benzothiazole core (AMG0892 series) and the second group by a
naphthyl amide core (AMG0309 series). Using alkaline phosphatase (ALP), osteocalcin (OCL) and calcium
as markers of osteoblast differentiation and mineralization, both chemical series showed EC50s in the
Received 13 November 2008
Revised 7 January 2009
Accepted 12 January 2009
Available online 15 January 2009
0.01–0.2 lM range and were consistent for all three markers. Compounds inhibited cell proliferation,
Keywords:
had no effect on apoptosis and showed evidence for CREB pathway activity. The present compounds rep-
resent some of the most potent osteogenic small molecules reported to date and provide new tools for
elucidating signaling mechanisms in osteoblasts.
Chemical screen
Small molecules
Osteogenic
Cell proliferation
Differentiation
CREB pathway
Ó 2009 Elsevier Ltd. All rights reserved.
The identification of small molecule bone anabolic agents capa-
ble of increasing bone mass represents a major unmet medical
need in the area of osteoporosis and bone disease. Several small
molecules with osteogenic activity have been described but few
if any showed EC50s in the sub-micromolar range. Previously it
was reported that the osteogenic compound purmorphamine was
ALP and bone nodule formation in human osteoblasts derived from
bone marrow mesenchymal cells.3 Subsequent work on the mech-
anism of action of purmorphamine revealed its role as a smooth-
ened agonist in the hedgehog pathway.4,5
ALP expression is a marker that typically correlates with the
early phase of osteoblast differentiation. To search for compounds
that also potentially affected osteoblast function, we designed a
screen in mouse MC3T3-E1 osteoblast cells that also examined
the mineralization response. A chemically diverse collection of
approximately 45,000 compounds was screened measuring both
ALP activity and calcein fluorescence that was used as a surrogate
for calcium incorporation into mineralizing cultures. The screen
yielded 558 hits of which 2% were positive for both ALP and calcein
fluorescence. Analogs (>98% purity) of selected hits were subse-
quently tested in secondary assays that measured ALP, osteocalcin
(OCL) a marker of late osteoblast differentiation and calcium di-
rectly for evidence of mineralization.
As shown in Figure 1, after 3 weeks in culture, several com-
pounds from two chemical classes stimulated osteogenesis with
EC50s for ALP, OCL and calcium in the 0.01–0.2 lM range. Effects
were evident within 1 week of treatment (data not shown). The
structures of the benzothiazole and naphthyl amide series are
shown in Figure 2. Preliminary structure–activity relationships
(SAR) of examples from both series are shown in Table 1. For the
benzothiazole analogs, fluorine substitution on the phenyl group
of the benzyl amine was well tolerated (AMG0895), while conver-
sion of the amine to an amide (AMG9594) resulted in a significant
active at 1
scribe the identification of two potent chemical series with osteo-
genic activity and cellular EC50s in the 0.01–0.2 M range. The first
l
M in C3H10T1/2 mesenchymal cells.1 Here we de-
l
was characterized by an amino benzothiazole core (AMG0892 ser-
ies) and the second by a naphthyl amide core (AMG0309 series).
Bone formation is driven by the activity of osteoblasts and the
production of a mineralized matrix. However, in bone diseases
such as osteoporosis, the rate of bone resorption by osteoclast cells
exceeds the rate of formation by osteoblasts resulting in a net loss
of bone. Thus there is a major interest in the identification of ther-
apeutic small molecules that can enhance osteoblast activity and
result in a net increase in skeletal bone formation.
A chemical screening approach was used previously to identify
the compound purmorphamine that increased alkaline phospha-
tase (ALP) activity in C3H10T1/2 cells.1 Expression of Runx2 which
is essential for bone formation2 was also increased but less so in
the more differentiated osteoblast progenitor cell line MC3T3-E1.
Purmorphamine in the 1–3 lM range was also shown to increase
* Corresponding author.
E-mail address: pbabij@amgen.com (P. Babij).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.01.025

Chun-Ya E. Han et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1442–1445
1443
Figure 1. Induction of osteogenesis by (A) AMG0892 and (B) AMG0895 (benzothiazole series) and, (C) AMG6168 and (D) AMG0309 (naphthyl amide series). Original hit
(AMG2888, open black circles) and intermediate analog (AMG5581, open black squares).
PGE-551609 at 10–100 l l
M,10 TH at 1 M11 and OIC-A006 at
6 l
M.12 However other studies have shown more potent effects
with the immunosuppressants Rapamycin (50 nM),13 FK506 (10-
100 nM),14,15 and cyclosporin A (10–500 nM).16 In the latter exam-
ple, the effects of cyclosporin A were biphasic, showing osteogenic
effects only at lower concentrations.
Figure 2. Chemical structures of the benzothiazole and naphthyl amide series.
drop in activity. In the naphthyl amide series, incorporation of
other aromatic amines (AMG6168) as well as the corresponding
‘reverse amide’ (AMG4500) resulted in a slight drop in activity rel-
ative to the potent compound AMG0309. Both AMG1051 and
AMG6610 showed little or no activity in the cellular assays (Details
on compound synthesis are shown in Supplementary material).
Alizarin Red and von Kossa staining were also increased in
MC3T3-E1 cultures providing further qualitative evidence for en-
hanced mineralization (data not shown). Compounds were also ac-
tive in 1- to 3-day-old primary rodent calvarial cells that contain a
mixture of osteoprogenitor cells and differentiated osteoblasts
(Supplementary material).
The consistency shown by the compounds in having potent ef-
fects on three independent markers suggests a robust effect on
osteoblast differentiation. Other small molecules in addition to
purmorphamine, have been reported to increase osteogenesis but
in most cases at much lower potencies. For example, TAK-778
Figure 3. Compounds dose-dependently inhibit cell proliferation (A), show no
cytotoxicity except at the highest concentration (B), do not increase apoptosis (C)
and do not inhibit apoptosis in Staurosporine-induced MC3T3-E1 cells (D).
was active at 1–10 l l l
M,6 statins at 0.1–5 M,7,8 91E2 at 5 M,9

1444
Chun-Ya E. Han et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1442–1445
Table 1
To confirm the molecular basis for the osteogenic effects, sev-
eral RNA markers (Mouse Osteogenesis RT2 Profiler PCR Array,
SABiosciences, Frederick, MD) were evaluated after MC3T3-E1 cells
Structure–activity relationship of osteogenic compounds on markers measured in
MC3T3-E1 cells at 3 weeks
Compound Structure
ALP
OCL
M)
Calcium
M)
were treated with AMG0892 and AMG0309 at 0.75 lM or with
(
lM)
(
l
(l
BMP2 at 50 ng/ml as a positive control for up to 21 days. In addi-
tion to ALP, several genes involved in osteoblast differentiation
including BMP3, Dmp1, Phex, and Sclerostin were significantly
changed in a manner that resembled the changes seen for the po-
tent osteogenic molecule BMP2 (Supplementary material). Fur-
N
S
NH
NH
O
O
N
AMG0892
0.09
0.06
0.09
0.09
0.21
O
thermore, there was
a transient 4- to 5-fold increase in
O
expression of the key osteogenic gene Runx2 which was similar
to the upregulation observed with purmorphamime in C3H10T1/
2 cells.1
F
F
N
S
O
N
AMG0895
0.15
0.52
As an initial step in characterizing the mechanism underlying
the osteogenic effects, we examined the ability of the compounds
to regulate cell proliferation and apoptosis. The results in Figure
3 showed that both series of compounds strongly inhibited cell
proliferation measured by BrdU incorporation and did not alter cell
viability measured by Alamar Blue labeling. In addition, the com-
pounds had no effect on apoptosis measured either by an increase
in caspase-3 activity or a decrease in caspase-3 activity following
Staurosporine-induced apoptosis. This finding thus suggests a link
between inhibition of cell growth and enhancement of osteoblast
differentiation.
O
O
O
N
S
NH
O
N
AMG9594
0.41
0.51
O
O
O
N
S
NH
O
N
AMG1051
>20
>20
>20
Since it is known that BMP and Wnt signaling are important for
osteoblast differentiation,17,18 we tested whether the compounds
might have effects via these two pathways. Stable MC3T3-E1 cell
lines expressing BMP219 and Wnt20 response elements linked to
O
O
H
N
O
N
luciferase were treated with compounds at doses up to 3 lM for
O
periods up to 3 weeks. A similar Wnt reporter was used previously
in a chemical screen to identify small molecule modulators of Wnt
signaling.21 Interestingly, we did not observe any effects of the
compounds in any of these assays at any time point, thus raising
the possibility that other pathways may be involved in the osteo-
genic response (data not shown).
AMG0309
0.04
0.03
0.03
O
N
O
O
H
N
O
Surprisingly, however, the osteogenic compounds were ob-
served to activate the CREB pathway in MC3T3-E1 cells as shown
using a CREB response element (Clontech, Mountain View CA)
linked to luciferase (Fig. 4). A dose-dependent increase in luciferase
activity was noted, in contrast to the lack of response from two
negative control compounds (AMG6610 and AMG1051) that were
structurally similar but possessed very low cellular activity.
The cellular activities of the two chemical series provide a con-
venient starting point for exploration of the mechanism of action of
the compounds in a manner similar to that for purmorphamine.1
Moreover, the greater cellular potency and distinct chemical struc-
tures of the current compounds compared to purmorphamine raise
the possibility that the molecular target(s) is also unique. Other
work has shown that tool compounds identified via chemical ge-
netic screens and that demonstrate functional activity can be used
as the basis to identify novel targets.22–26 Thus different com-
pounds have been used in affinity chromatography approaches
AMG6168
0.09
0.03
0.05
O
O
O
N
O
HN
AMG4500
0.12
0.10
0.15
O
N
O
O
H
N
O
N
AMG6610
5.0
>20
>20
O
O
O
N
Figure 4. Osteogenic compounds activate CREB pathway. A compound dose-response at 1:3 dilution was used up to 3
(AMG1051) and naphthyl amide (AMG6610) series were structurally similar but showed very little cellular activity. Forskolin used as positive control at 1:2 dilution up to
M. Means SEM for two experiments in duplicate. *p < 0.05; **p < 0.01; ***p < 0.001 versus control.
lM. Negative control compounds for the benzothiazole
1
l

Chun-Ya E. Han et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1442–1445
1445
8. Maeda, T.; Matsunuma, A.; Kawane, T.; Horiuchi, N. Biochem. Biophys. Res.
Commun. 2001, 280, 874.
9. Choi, Y.; Kawazoe, Y.; Murakami, K.; Misawa, H.; Uesugi, M. J. Biol. Chem. 2003,
278, 7320.
10. Liu, Z.; Shi, W.; Ji, X.; Sun, C.; Jee, W. S. S.; Wu, Y.; Mao, Z.; Nagy, T. R.; Li, Q.; Cao,
X. J. Biol. Chem. 2004, 279, 11313.
11. Ohba, S.; Nakajima, K.; Komiyama, Y.; Kugimiya, F.; Igawa, K.; Itaka, K.; Moro,
T.; Nakamura, K.; Kawaguchi, H.; Takato, T.; Chung, U. Biochem. Biophys. Res.
Commun. 2007, 357, 854.
for target enrichment and protein sequencing by mass spectrome-
try. It is important to note that in the current work, the chemical
hits were not optimized for osteoblast cellular activity or target
specificity but instead provide a starting point for such studies.
Thus the present compounds represent some of the most potent
osteogenic small molecules reported to date and provide new tools
for elucidating signaling mechanisms in osteoblasts.
12. Cai, M.; Liu, X.; Shao, J.; Qi, J.; Wang, J.; Zhu, Y.; Zhou, Q.; Wang, J.; Zhao, Q.; Li,
G.; Liang, J.; Lu, W. W.; Deng, L. Pharmazie 2008, 63, 751.
13. Ogawa, T.; Tokuda, M.; Tomizawa, K.; Matsui, H.; Itano, T.; Konishi, R.;
Nagahata, S.; Hatase, O. Biochem. Biophys. Res. Commun. 1998, 249, 226.
14. Tang, L.; Ebara, S.; Kawasaki, S.; Wakabayashi, S.; Nikaido, T.; Takaoka, K. Cell
Biol. Int. 2002, 26, 75.
Acknowledgments
We greatly appreciate the discussions and assistance from
many other Amgen colleagues during the course of this work.
15. Kugimiya, F.; Yano, F.; Ohba, S.; Igawa, K.; Nakamura, K.; Kawaguchi, H.; Chung,
U. Biochem. Biophys. Res. Commun. 2005, 338, 872.
16. Yeo, H.; Beck, L. H.; McDonald, J. M.; Zayafoon, M. Bone 2007, 40, 1502.
17. Cao, X.; Chen, D. Gene 2005, 357, 1.
18. Johnson, M. L.; Harnish, K.; Nusse, R.; Van Hul, W. J. Bone Miner. Res. 2004, 19,
1749.
19. Korchynskyi, O.; ten Dijke, P. J. Biol. Chem. 2002, 277, 4883.
20. Veeman, M. T.; Slusarski, D. C.; Kaykas, A.; Hallagan Louie, S.; Moon, R. T. Curr.
Biol. 2003, 13, 680.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.01.025.
21. Liu, J.; Wu, X.; Mitchell, B.; Kintner, C.; Ding, S.; Schultz, P. G. Angew. Chem., Int.
Ed. 2005, 44, 1987.
References and notes
22. Ding, S.; Schultz, P. G. Nat. Biotechnol. 2004, 22, 833.
23. Bush, E.; Fielitz, J.; Melvin, L.; Martinez-Arnold, M.; Mckinsey, T. A.; Plichta, R.;
Olson, E. N. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 2870.
24. Jessen, K. A.; English, N. M.; Wang, J. Y.; Maliartchouk, S.; Archer, S. P.; Qiu, L.;
Brand, R.; Kuemmerle, J.; Zhang, H.-Z.; Gehlsen, K.; Drewe, J.; Tseng, B.; Cai, S.
X.; Kasibhatla, S. Mol. Cancer Ther. 2005, 4, 761.
25. Chen, S.; Do, J. T.; Zhang, Q.; Yao, S.; Yan, F.; Peters, E. C.; Scholer, H. R.; Schultz,
P. G.; Ding, S. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 17266.
26. Zhang, Q.; Major, M. B.; Takanashi, S.; Camp, N. D.; Nishiya, N.; Peters, E. C.;
Ginsberg, M. H.; Schultz, P. G.; Moon, R. T.; Ding, S. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 7444.
1. Wu, X.; Ding, S.; Ding, Q.; Gray, N. S.; Schultz, P. G. J. Am. Chem. Soc. 2002, 124,
14520.
2. Karsenty, G.; Wagner, E. F. Dev. Cell 2002, 2, 389.
3. Beloti, M. M.; Bellesini, L. S.; Rosa, A. L. Cell. Biol. Int. 2005, 29, 537.
4. Wu, X.; Walker, J.; Zhang, J.; Ding, S.; Schultz, P. G. Chem. Biol. 2004, 11, 1229.
5. Sinha, S.; Chen, J. K. Nat. Chem. Biol. 2006, 2, 29.
6. Notoya, K.; Nagai, H.; Oda, T.; Gotoh, M.; Hoshino, T.; Muranishi, H.; Taketomi,
S.; Sohda, T.; Makino, H. J. Pharmacol. Exp. Ther. 1999, 290, 1054.
7. Mundy, G.; Garrett, R.; Harris, S.; Chan, J.; Chen, D.; Rossini, G.; Boyce, B.; Zhao,
M.; Gutierrez, G. Science 1999, 286, 1946.