Orally available pyridinylpyrimidine derivatives as novel RANKL-induced osteoclastogenesis inhibitors

Article abstract of DOI:10.1016/j.bmcl.2012.06.087

An HTS campaign led to the identification of 4-pyrroldino-2-(pyridin-2-yl) pyrimidine compound 1 as an RANKL-induced osteoclastogenesis inhibitor. The compound 1 showed high clearance values in microsomes, however. Modification of the pyrrolidino group to a benzylamino group improved human microsomal stability with a slight loss of in vitro activity. Substitution at the ortho position of the benzyl group ameliorated in vitro activity, and further fluorination of the benzyl group improved microsomal stability in rodents. Representative members of this series, compounds 20 and 23, exhibited efficacy in RANKL-induced osteopenic mice when administered orally at 0.3 mg/kg.

Full text of DOI:10.1016/j.bmcl.2012.06.087

Bioorganic & Medicinal Chemistry Letters 22 (2012) 5681–5684
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Orally available pyridinylpyrimidine derivatives as novel RANKL-induced
osteoclastogenesis inhibitors
Junji Miyata a, , Chiyoshi Kasahara a, Toru Asano a, Shinji Ito a, Norio Seki a, Yasuko Kato a,
⇑
Noriyuki Morikawa a, Kazuyoshi Nozaki a, Kouji Nishimura b, Hisashi Akamatsu b,
Yusuke Taguchi b, Tomonori Yamaguchi b, Yoshito Abe a, Mitsuru Ohkubo a, Toshihiro Watanabe a,
Mitsuaki Ohta a, Makoto Takeuchi a
a Drug Discovery Research, Astellas Pharma Inc., 21, Miyukigaoka, Tsukuba, Ibaraki 305-8585, Japan
b Drug Discovery Laboratory, Wakunaga Pharmaceutical Co., Ltd., 1624 Shimokotachi, Koda-cho, Akitakata-shi, Hiroshima 739-1195, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
An HTS campaign led to the identification of 4-pyrroldino-2-(pyridin-2-yl)pyrimidine compound 1 as an
RANKL-induced osteoclastogenesis inhibitor. The compound 1 showed high clearance values in micro-
somes, however. Modification of the pyrrolidino group to a benzylamino group improved human micro-
somal stability with a slight loss of in vitro activity. Substitution at the ortho position of the benzyl group
ameliorated in vitro activity, and further fluorination of the benzyl group improved microsomal stability
in rodents. Representative members of this series, compounds 20 and 23, exhibited efficacy in RANKL-
induced osteopenic mice when administered orally at 0.3 mg/kg.
Received 22 May 2012
Revised 27 June 2012
Accepted 27 June 2012
Available online 6 July 2012
Keywords:
RANKL
Inhibitor
Ó 2012 Elsevier Ltd. All rights reserved.
Osteoclastogenesis
Pyrimidine
Benzyl
Osteoclasts are multinucleated cells derived from a monocyte
lineage which express tartrate-resistant acid phosphatase (TRAP)
on the cell surface as a differentiation marker, and play an essential
role in bone resorption.1 Osteoclasts and the bone forming cells,
osteoblasts, cooperate in bone remodeling to maintain the
mechanical competence of bone. Osteoblasts supply a critical fac-
difficult to treat.6 Estrogen replacement therapy and selective
estrogen receptor modulators (SERMs) also suppress bone resorp-
tion,1b but these therapies function only in the prevention and
treatment of osteoporosis associated with estrogen deficiency in
women, and estrogen replacement therapy increases the risk of
cancer. These problems with current treatments emphasize the
need for novel anti-resorptives.
tor, receptor activator of nuclear factor-jB (RANK) ligand (RANKL),
which binds to its cellular acceptor, RANK, and triggers a signal
transduction and subsequent gene expression. This in turn results
in differentiation of the osteoclast precursor into TRAP-positive,
multinucleated bone-resorptive cells.2 Aging or diseases, including
osteoporosis, bone metastasis and rheumatoid arthritis, cause
excessive RANKL expression, which leads to abnormally increased
bone resorption.3,4
Inhibition of osteoclastic activity is a primary therapeutic ap-
proach to bone loss.1b Bisphosphonates (BPs) are the major anti-
resorptive therapy currently available.5 These agents accumulate
almost exclusively in skeletal sites and induce apoptosis in osteo-
clasts. The efficacy of BPs has been recognized, but gastrointestinal
disorders are reported as common adverse effects. In addition, BP-
related osteonecrosis of the jaw is a severe side effect which is
The RANKL/RANK signaling pathway is a promising target for
the development of effective osteoclastogenesis inhibitors. Deno-
sumab, a humanized anti-RANKL antibody, is a new agent for the
treatment of osteoporosis and other bone diseases.4 In addition
to biologics, which are administered parenterally, orally available
small-molecule inhibitors might also be valuable in the treatment
of bone loss, and several groups have in fact recently reported
small-molecule inhibitors of RANKL-induced osteoclastogenesis.7,8
Identification of potent small molecules of practical use has been
challenging, however; efficacy through oral administration is not
disclosed in most cases, and no clinical trial of small-molecule
RANKL-induced osteoclastogenesis inhibitors has been reported.
Here, we report the modification of a high throughput screening
(HTS) hit (1) which resulted in the discovery of a series of orally
bioavailable anti-resorptive agents. An HTS campaign was
performed using a RANKL-induced TRAP promoter-dependent
reporter gene assay.8 We successfully identified 4-pyrrolidino-2-
(pyridin-2-yl)pyrimidine derivative 1 as an HTS hit with an IC50
⇑
Corresponding author. Tel.: +81 29 829 6173; fax: +81 29 852 5387.
E-mail address: junji.miyata@astellas.com (J. Miyata).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.06.087

5682
J. Miyata et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5681–5684
6
R'
5
OMe
N
OMe
N
HN
R
1
OMe
N
N
TRAP promoter assay: IC50= 1.8 µM
TRAP staining assay: IC50= 0.15 µM
N
N
6'
N
R'
N
N
4
N
Cl
N
K2CO3, MeCN
heat, (33-96%)
R
4
1
6-9, 11, 12 and 14-23
Figure 1. Structure and anti-osteoclastogenic activity of HTS hit 1.
Scheme 2. Synthesis of compounds 6–9, 11, 12 and 14–23.
value of 1.8
against RANKL-induced osteoclastogenesis was confirmed with
an IC50 of 0.15 M in a TRAP staining assay in RAW264 cells, an
lM in the promoter assay (Fig. 1). Inhibitory activity
Zn Br
OMe
Cl
OMe
l
N
N
Pd(t-Bu3P)2
immortalized mouse macrophage-like cell line possessing the abil-
ity to differentiate into osteoclasts.9 However, compound 1 showed
high clearance values in liver microsomes of mouse, rat and human
(abbreviated as MLM, RLM and HLM, respectively) in vitro. We
therefore conducted a preliminary search for the structure–activity
relationships (SAR) of in vitro activity against osteoclastogenesis,
and also investigated metabolic pathways in microsomes. The fol-
lowing results were obtained:
N
N
N
N
THF, microwave
(21%)
4
13
Scheme 3. Synthesis of compound 13.
Table 1
Potency in TRAP staining assay and microsomal stability of 4-amino/4-cyclopentyl
1. The 2-pyridinyl group on the pyrimidine ring was essential for
inhibitory activity.
pyrimidinesa
2. Substitution at either the 6-position of the pyrimidine ring or
the 60 position in the 2-pyridinyl group attenuated potency.
3. The pyrrolidine ring of 1 was identified as the main site of
metabolism in both RLM or HLM in vitro by LC–MS/MS analysis
(Scheme 1). One of the C–N bonds in the pyrrolidine ring was
cleaved to generate two oxygenated metabolites 2 and 3.10
OMe
N
N
N
R
Compd
R
TRAP staininga
IC50 (M)
Clint,
(ml /min/kg)b
in vitro
MLM
RLM
HLM
458
657
On the basis of these preliminary results, we embarked on a
modification of the pyrrolidine ring to improve both potency and
metabolic stability.
1
6
N
N
0.15
0.10
>1000
N.T.
987
>1000
N.T.
The general synthetic route for various 4-amino derivatives was
efficient in only one step, as illustrated in Scheme 2. A commer-
cially available chloropyrimidine 4 was reacted with amines 5 to
afford the compounds 6–9, 11, 12 and 14–23 in moderate to high
yields. The cyclopentyl compound 13 was prepared by Negishi cou-
pling reaction of the chloropyrimidine 4 with cyclopentylzinc bro-
mide under microwave irradiation as shown in Scheme 3.11 The
synthesized compounds were evaluated in vitro to investigate both
their inhibitory activity against RANKL-induced osteoclastogenesis
and microsomal stability, as shown in Table 1. The piperidine 6 and
N-methyl-N-cyclopentylamine 9 showed equipotent activity with
1, although clearance values in liver microsomes were increased.
The azabicyclic compound 8 retained activity and stability, while
conversion of the pyrrole group into azepane (7) decreased activ-
ity. These results suggested that replacement of the pyrrolidine
7
8
0.69
0.24
N.T.
N.T.
N.T.
512
N
N
734
c-Pent
N
N
9
0.18
N.T.
>1000
775
Me
10
11
O
>2
N.T.
N.T.
210
N.T.
178
F
1.0
>1000
N
N
F
F
12
1.4
>1000
395
310
13
14
c-Pent
1.8
N.T.
>1000
202
749
210
N
H
c-Pent
Ph
0.57
>1000
N
H
RLM
or
HLM
15
0.42
911
387
67
OMe
N
OMe
N
N
a
OH
N
N
Inhibitory activity of RANKL-induced osteoclastogenesis using RAW264 cells (n
P3).
N
N
N
H
b
Average of two experiments. N.T.: Not tested.
1
2
ring with tertiary amino groups possessing simple aliphatic sub-
stituents might not improve metabolic stability sufficiently while
retaining inhibitory activity. With the aim of blocking metabolism,
introduction of electro-negative atoms and depletion of nitrogen
were implemented. Disappointingly, embedding an oxygen atom
in the ring as morpholine (10) resulted in the complete loss of
activity. Fluorination of the pyrrolidine ring (11, 12) and replace-
ment of the nitrogen with a carbon (13) also reduced inhibitory
activity, although the fluorination showed a tendency to improve
+
OMe
N
OH
N
N
N
H
O
3
Scheme 1. Major metabolites of compound 1 in both RLM or HLM.

J. Miyata et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5681–5684
5683
microsomal clearance. These observations indicated that the elec-
tro-donating ability of the nitrogen atom is an important factor
for activity. Next, the potency of the secondary amino groups
was investigated. The cyclopentylamino derivative 14 showed de-
creased activity in comparison to 9, implying that the presence of
two substituents of tertiary amino groups affected the inhibitory
activity positively. The benzylamino compound 15 slightly de-
creased activity, but the microsomal stability of 15 was improved
sufficiently in human and moderately in rodents.
**
100
50
0
**
**
**
*
Compd 15
**
Compd 20
Compd 22
Compd 23
Based on these results and the feasibility of benzyl modification,
we selected the benzyl derivative 15 as a lead and planned a strat-
egy to improve inhibitory activity. To compensate for the loss of
one substituent on the nitrogen atom compared to tertiary amino
groups, a substituent was introduced into the benzyl group to ren-
der the proximity of the nitrogen atom sterically congested. The re-
sults of this modification of the benzyl group are shown in Table 2.
Introduction of a methyl group at the benzyl position of 15 atten-
uated activity (16), whereas ortho substitution at the phenyl ring
enhanced potency while retaining metabolic stability (17).
Replacement of the methyl group on the phenyl ring of 17 with a
trifluoromethyl group improved stability in rodents (18). Encour-
aged by these results, we investigated halogen substituents at
the ortho position. In contrast to the disappointing result with
the chloro group (19), the fluoro group ameliorated inhibitory
activity and retained microsomal stability (20). Considering the
size of a fluoro group, these SAR results might indicate a preference
for a small substituent at the ortho position. Additional fluorination
of the phenyl ring improved microsomal stability in rodents with-
out any loss of activity (21, 22 and 23).
Given this improved metabolic profile in vitro, we evaluated
in vivo efficacy of the benzyl derivatives in RANKL-induced oste-
openic mice, which were reported as a rapid bone loss model by
Yasuda and co-workers.12 The benzyl compounds 15, 20, 22 and
23 were given to the mice by daily oral administration for 2 days,
concomitantly with intraperitoneal injection of RANKL. In contrast
to the insignificant inhibition of 15 at 3 mg/kg, 20, 22 and 23 inhib-
ited bone loss significantly at 3 mg/kg in a dose-dependent man-
ner, owing to the ameliorated metabolic stability and potency
in vitro (Fig. 2). Compounds 20 and 23, in particular, showed signif-
icant inhibition at 0.3 mg/kg. In the article of Yasuda et al., special
0.01
0.1
1
10
Dose of Drugs (mg/kg)
Figure 2. Anti-resorptive effect of compound 15, 20, 22 and 23 in RANKL-induced
osteopenic mice. Orally administered once daily for 2 days. 10-wk-old female
C57BL/6 N mice (n = 5). Medium was 0.5% methyl cellulose solution. BMC: Bone
mineral content; ⁄p <0.05 versus control; ⁄⁄p <0.01 versus control (Dunnett’s
multiple comparison test).
protocols were prepared for the assessment of BPs and SERMs;
evaluation of BPs required pretreatment with the agents before
the first RANKL injection and consequently a longer total duration
of the procedures, probably to secure sufficient time to accumulate
BPs in bones. Evaluation of raloxifene, a second-generation SERM,
was also reported to require ovariectomy after the last RANKL
injection, followed by administration of raloxifene for 2 weeks.
The efficacy of 20, 22 and 23 with 2 days administration concom-
itantly with the RANKL injection supports the idea that the pyridyl-
pyrimidine derivatives have a different mode of action from BPs
and SERMs.
The pharmacokinetic study of 20, 22 and 23, as described in Ta-
ble 3, showed increased plasma concentration compared to the
HTS hit 1 when administered orally to normal mice. In addition,
a single administration of any of these three substituted benzyl
compounds exhibited comparable exposure in ovariectomized
(OVX) rats, which are the standard model for osteoporosis in post-
menopausal women. Furthermore, these three compounds showed
an acceptable off-target profile of both CYP inhibition (IC50 P5
for 1A2, 2C9, 2C19, 2D6; <10% inhibition of 3A4 at 5 M for either
reversible inhibitory activity or time-dependent inhibitory activ-
ity) and hERG inhibition (Rb efflux assay; IC50 >25 M), in addition
lM
l
Table 2
Potency in TRAP staining assay and microsomal stability of 4-
benzylaminopyrimidines
l
to good physicochemical properties (solubility in pH6.8 phosphate
buffer is >25
l
g/ml; PAMPA permeability is >30 ꢀ 10ꢁ6 cm/s) and
OMe
N
toxicological profiles (3T3 NRU phototoxicity test: negative;
in vivo micronucleus test: negative).13
N
R2
N
NH
R1
R4
R3 R4 TRAP staininga
R3
Clint,
Table 3
Compd R1
R2
(ml/min/kg)
Oral pharmacokinetic profiles of compounds 1, 20, 22, and 23 in normal female mice
b
in vitro
and OVX rats when single dose was administereda
IC50
(
lM)
MLM
RLM
HML
Compd Dose (mg/kg)
Female mouse (10-wk-
old C57BL/6N)
OVX ratb (12-wk-old
Wister)
15
16
17
18
19
20
21
22
23
H
Me
H
H
H
H
H
H
H
H
H
Me
CF3
Cl
F
F
F
CF3
H
H
H
H
H
H
F
H
H
H
H
H
H
H
F
0.42
0.66
0.22
0.23
0.52
0.13
0.11
0.10
0.19
911
570
901
619
704
672
524
486
261
387
420
791
555
827
413
N.T.
280
193
67
65
55
77
55
61
79
53
77
Cmax
AUC0–1
Cmax
AUC0–1
(ng/ml)
(ngꢂh/ml)
15c
(ng/ml)
(ngꢂh/ml)
1
1
3
3
3
7c
N.T.
49
135
156
N.T.
151
542
943
20
22
23
130
98
599
240
188
1022
F
F
H
a
b
c
a
Medium was 0.5% methyl cellulose solution. n = 3.
Ovariectomy was conducted 2 weeks prior to administration.
Male mouse. N.T.: Not tested.
Inhibitory activity of RANKL-induced osteoclastogenesis using RAW264 cells (n
P3).
b
Average of two experiments. N.T.: Not tested.

5684
J. Miyata et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5681–5684
Commun. 2009, 378, 1; (c) Jules, J.; Ashley, J. W.; Feng, X. Expert Opin. Ther.
In summary, we identified a novel class of RANKL-induced
Targets 2010, 14, 923; (d) Allen, J.; Fotsch, C.; Babij, P. J. Med. Chem. 2010, 53,
177.
3. (a) Terpos, E.; Efstathiou, E.; Christoulas, D.; Roussou, M.; Katodritou, E.;
Dimopoulos, M. A. Expert Opin. Biol. Ther. 2009, 9, 465; (b) Mundy, G. R. Nutr.
Rev. 2007, 65, S147.
4. (a) Demper, D. W.; Lambing, C. L.; Kostenuik, P. J.; Grauer, A. Clin. Ther. 2012, 34,
521; (b) Nangia, J. R.; Ma, J. D.; Nguyen, C. M.; Mendes, M. A.; Trivedi, M. V.
Expert. Opin. Biol. Ther. 2012, 12, 491; (c) Tsourdi, E.; Rachner, T. D.; Rauner, M.;
Hamann, C.; Hofbauer, L. C. Eur. J. Endocrinol. 2011, 165, 833; (d) Cohen, S. B.;
Dore, R. K.; Lane, N. E.; Ory, P. A.; Peterfy, C. G.; Sharp, J. T.; van der Heijde, D.;
Zhou, L.; Tsuji, W.; Newmark, R. Arthritis Rhuem. 2008, 58, 1299.
5. Russell, R. G.; Watts, N. B.; Ebetino, F. H.; Rogers, M. J. Osteoporos. Int. 2008, 19,
733.
osteoclastogenesis inhibitors using an HTS campaign. Modification
of the series to improve metabolic stability led to compounds that
are orally bioavailable and efficacious in a murine osteopenia mod-
el, with a different mode of action to BPs and SERMs. These com-
pounds showed lower in vitro intrinsic clearance values in
human than in rodents. These promising compounds are being
evaluated in other bone loss models in rodents, and the results will
be published in due course.
Acknowledgments
6. (a) Conte, P. F.; Guarneri, V. Oncologist 2004, 9, 28; (b) Filleul, O.; Crompot, E.;
Saussez, S. J. Cancer Res. Clin. Oncol. 2010, 136, 1117.
The authors would like to express their profound gratitude to
Mr. Masaharu Yokomoto for his valuable advice and constant sup-
port throughout this research. We deeply appreciate Dr. Katsuhiro
Yamano, Dr. Hidetsugu Murai, Ms. Masako Furutani, Mr. Fumihiro
Nakamori, Dr. Megumi Irie, Dr. Ai Tamura, and Mr. Shoji Takagaki
for their measurements in pharmacokinetic studies and valuable
discussion. We are grateful to Ms. Miwa Okazawa for her conduct
of the HTS, and Mr. Ryutaro Wakayama for his advice on synthesis.
7. (a) Tseng, C. H.; Lin, R. W.; Chen, Y. L.; Wang, G. J.; Ho, M. L.; Tzeng, C. C. J. Med.
Chem. 2011, 54, 3103; (b) Xu, J.; Li, Z.; Luo, J.; Yang, F.; Liu, T.; Liu, M.; Qui, W.;
Tang, J. J. Med. Chem. 2012, 55, 3122; (c) Li, C.; Yang, Z.; Li, Z.; Ma, Y.; Zhang, L.;
Zheng, C.; Qiu, W.; Wu, X.; Wang, X.; Li, H.; Tang, J.; Qian, M.; Li, D.; Wang, P.;
Luo, J.; Liu, M. J. Bone Miner. Res. 2011, 26, 644; (d) Zhu, M.; Kim, M. H.; Lee, S.;
Bae, S. J.; Kim, S. H.; Park, S. B. J. Med. Chem. 2010, 53, 8760; (e) Lee, J. H.; Kim, H.
N.; Yang, D.; Jung, K.; Kim, H. M.; Kim, H. H.; Ha, H. J. Biol. Chem. 2009, 284,
13725; (f) Greig, I. R.; Idris, A. I.; Ralston, S. H.; Van’t Hof, R. J. J. Med. Chem.
2006, 49, 7487.
8. Chen, T.; Knapp, A. C.; Wu, Y.; Huang, J.; Lynch, J. S.; Dickson, J.; Lawrence, R. M.,
Jr.; Feyen, J. H. M.; Agler, M. L. Assay Drug Dev. Technol. 2006, 4, 387.
9. Intensity of RANKL signal could be one of factors causing the 10-fold difference
in potency of 1 in the two in vitro assays. The promoter assay adopted 50 ng/ml
of final RANKL concentration, whereas 8 ng/ml of RANKL was used in the TRAP
staining assay.
Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.
06.087.
10. With the same HPLC program, the major metabolites in microsomes were
identified as compounds 2 and 3, which were synthesized as described in the
Supplementary data. Metabolites
osteoclastogenesis activities in TRAP staining assay, with IC50s of 3.0 and
>10 M, respectively.
2
and
3
showed reduced anti-
l
References and notes
11. Negishi, E. Acc. Chem. Res. 1982, 15, 340.
12. Tomimori, Y.; Mori, K.; Koide, M.; Nakamichi, Y.; Ninomiya, T.; Udagawa, N.;
Yasuda, H. J. Bone Miner. Res. 2009, 24, 1194–1205.
13. Protein kinase panel assay of compounds 22 and 23 showed <40% inhibition
1. (a) Boyle, W. J.; Simonet, W. S.; Lacey, D. L. Nature 2003, 423, 337; (b)
Teitelbaum, S. L. Science 2000, 289, 1504.
against 93 of all tested kinases at 10 lM.
2. (a) Suda, T.; Takahashi, N.; Udagawa, N.; Jimi, E.; Gillespie, M. T.; Martin, T. J.
Endocr. Rev. 1999, 20, 345; (b) Soyas, N. S.; Alles, N. Biochem. Biophys. Res.