Bone-targeted Src kinase inhibitors: Novel pyrrolo- and pyrazolopyrimidine analogues

Article abstract of DOI:10.1016/S0960-894X(03)00647-4

Src tyrosine kinase is a therapeutic target for bone diseases that has been validated by gene knockout studies. Furthermore, in vitro cellular studies implicate that Src has a positive regulatory role in osteoclasts and a negative regulatory role in osteoblasts. The potential use of Src inhibitors for osteoporosis therapy has been previously shown by novel bone-targeted ligands of the Src SH2 (e.g., AP22408) and non-bone-targeted, ATP-based inhibitors of Src kinase. Significant to this study, compounds 2-12 exemplify novel analogues of known pyrrolopyrimidine and pyrazolopyrimidine template-based Src kinase inhibitors that incorporate bone-targeting group modifications designed to provide tissue (bone) selectivity and diminished side effects. Accordingly, we report here the structure-based design, synthetic chemistry and biological testing of these compounds and proof-of-concept studies thereof.

Full text of DOI:10.1016/S0960-894X(03)00647-4

Bioorganic & Medicinal Chemistry Letters 13 (2003) 3063–3066
Bone-Targeted Src Kinase Inhibitors: Novel Pyrrolo-
and Pyrazolopyrimidine Analogues
Raji Sundaramoorthi,* William C. Shakespeare, Terence P. Keenan, Chester A. Metcalf, III,
Yihan Wang, Ukti Mani, Merry Taylor, Shuangying Liu, Regine S. Bohacek,
Surinder S. Narula, David C. Dalgarno, Marie Rose van Schravandijk,
Sheila M. Violette, Shuenn Liou, Susan Adams, Mary K. Ram, Jeffrey A. Keats,
Manfred Weigele and Tomi K. Sawyer
ARIAD Pharmaceuticals, Inc., 26 Landsdowne Street, Cambridge, MA 02139-4234, USA
Received 31 March 2003; accepted 18 April 2003
Abstract—Src tyrosine kinase is a therapeutic target for bone diseases that has been validated by gene knockout studies. Further-
more, in vitro cellular studies implicate that Src has a positive regulatory role in osteoclasts and a negative regulatory role in
osteoblasts. The potential use of Src inhibitors for osteoporosis therapy has been previously shown by novel bone-targeted ligands
of the Src SH2 (e.g., AP22408) and non-bone-targeted, ATP-based inhibitors of Src kinase. Significant to this study, compounds 2–
12 exemplify novel analogues of known pyrrolopyrimidine and pyrazolopyrimidine template-based Src kinase inhibitors that
incorporate bone-targeting group modifications designed to provide tissue (bone) selectivity and diminished side effects. Accord-
ingly, we report here the structure-based design, synthetic chemistry and biological testing of these compounds and proof-of-con-
cept studies thereof.
# 2003 Elsevier Ltd. All rights reserved.
Bone diseases provide significant opportunities for
therapeutic intervention with small-molecule inhibitors
of varying targets, including farnesyl diphosphate syn-
thase, estrogen steroid receptor, and protein kinases.1
Most bone pathology is due to an imbalance in bone
remodeling which is a dynamic process of bone forma-
tion and bone resorption.1,2À4 While a large number of
protein kinases are potential targets for bone disease,
attention has been focused on Src kinase because of the
discovery that the c-Src À/Àmouse had little overt
pathology apart from osteopetrosis.5 The defect in these
mice was traced to a cell-autonomous defect in the
activity of mature osteoclasts, and not in osteoclasto-
genesis.6 Recently, it was shown that the Src activity
also has an overall negative regulatory effect on osteo-
blastic cells.7 The selectivity of the Src À/Àphenotype
for bone tissue and the complementary effects on bone
resorption and bone formation, implicate Src as an
attractive target for the treatment of diseases with
excessive bone loss, such as post-menopausal osteo-
porosis, Paget’s disease, and osteolytic metastases.
Recently, we reported a SH2 inhibitor (AP22408) hav-
ing a bone targeting group incorporated into its struc-
ture.8 By bone-targeting this nonpeptide small molecule,
we were able to achieve tissue selectivity by selectively
localizing it to the bone surface and achieve higher
concentrations for inhibiting osteoclasts. We have
extended the same approach to a series of known Src
tyrosine inhibitors. By introducing the bone targeting
groups to these known inhibitors, we retrieved a wealth
of SAR data to advance important ‘proof-of-concept’
studies. In this paper, we describe the design and
synthesis of bone-targeted-pyrrolo[2,3-d]pyrimidines,
originally reported by the Novartis group9 as well as
bone-targeted-pyrazolo[2,3-d]pyrimidines, reported by
Pfizer group.10 A molecular model of Src kinase was
developed to help deter-mine the precise regiospecificity
of the bone-targeting group that would maximize the
molecular interactions of the inhibitor at the enzyme
active site.
A 3D model of the catalytically-active conformation of
Src kinase was constructed using the crystal structures
of inactive Src kinase, Lck kinase and active insulin
receptor kinase.11 The model was tested by docking
diverse series of Src kinase inhibitors found in the
*Corresponding author. Fax: +1-617-494-8144; e-mail: raji.
sundaramoorthi@ariad.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00647-4

3064
R. Sundaramoorthi et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3063–3066
literature. Adjustments were made until the model could
rationalize the relative Src kinase inhibitory potencies of
such compounds (i.e., both active and inactive). Dock-
ing studies using the FLO software were carried out to
predict how 3b would bind to Src kinase active site (Fig.
1a–c). The pyrrolopyrimidine ring of 3b was determined
to overlay with the purine ring of ATP, including
hydrogen bonds to Met-319 and Ser-320, and the phe-
nol group of 3b interacts with a hydrophobic pocket not
occupied by ATP. Finally, the phenyl ring of 3b
attached to the N9 nitrogen occupies the region at the
solvent interface of the binding site. Modeling studies
suggested bone-targeting group incorporation at this
position, and similar drug design rationale was sug-
gested for bone-targeted pyrazolopyrimidine analogues.
Table 1 shows the IC50 values of compounds 2–12. The
IC50 values of compounds 2c and 2d were similar to that
of their non-bone-targeted parent molecules 1a and 1b.
Molecular modeling studies suggest that the bone-tar-
geting group interacts with the Mg2+ binding pocket of
the Src kinase active site in a manner related, but not
identical, to that of the triphosphate of the ATP. The
Src kinase inhibitory potency of these compounds could
be further improved to provide IC50 values as low as 4
nM by demethylating the O-methyl group (cf. com-
pounds 3a and 3b vs 1a and 1b, respectively, as well as
compounds 4c and 4d vs compounds 2c and 2d, respec-
tively). Molecular modeling studies suggest that the
phenol OH group H-bonds with residues in the hydro-
phobic specificity pocket of Src kinase. Relative to
compounds 5e and 5f, a series of systematically mod-
ified pyrrolopyrimidines were synthesized (Scheme 2) to
explore varying bone-targeting groups. Both aryl and
aliphatic modified, bone-targeting groups were explored
(cf. compounds 7g–n). Src kinase inhibitory potencies
revealed that all were markedly less potent (50- to 100-
fold) than the aforementioned lead compound 4d.
Figure 2 shows the candidate bone-targeting groups and
their abbreviated nomenclature. These groups were
prepared according to the procedures described else-
where13 for incorporation into either pyrrolopyrimidine
or pyrazolopyrimidine templates (examples in Fig. 3). A
few different bone-targeting groups, having varying
hydroxyapatite affinity properties were examined.
Synthesis of compounds 2–4 are depicted in Scheme 1
and 7g–n in Scheme 2. The experimental details of the
synthesis of 2–14 is described elsewhere.13 Reported
compounds 1a and 1b9 were phosphonomethylphos-
phonylated using methylenebis (phosphonic dichloride)
in trimethyl phosphate at 0 ꢀC following a previously
reported procedure14 to give 2c and 2d. Compounds 1a
and 1b were demethylated with boron tribromide in
DCM at rt for 5 h to provide compounds 3a and 3b.
This was then phosphonomethyphosphonylated using
methylenebis (phosphonic dichloride) in trimethyl
phosphate at 0 ꢀC to give 4c and 4d. Compounds 2–12
were tested in vitro using a scintillation proximity assay
(SPA) developed by Amersham.15
Hydroxyapatite adsorption chromatography assay was
developed to experimentally determine the ability of
these bone-targeted compounds to bind to hydroxy-
apatite (mineral component of bone). As expected,
Table 1. Src kinase inhibition of pyrrolopyrimidines and pyr-
azolopyrimidines using the SPA15 kinase assay
Pyrrolopyrimidine
Pyrazolopyrimidine
Compd IC50, mM Compd IC50, mM
Compd
IC50, mM
1a
2c
3a
4c
1b
2d
3b
4d
1.15
0.7
5e
5f
7g
7h
7i
7j
7k
7l
4.0
1.5
2.5
0.87
19
40
0.36
0.25
PP1
8
9
10
11
12
0.12
1.6
0.02
0.40
0.32
0.68
0.04
0.06
0.16
0.31
0.04
0.004
Scheme 1.
Scheme 2.

R. Sundaramoorthi et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3063–3066
3065
compound 2d gave a K value of 2.26 which was slightly
higher than that of H2N-PCP (K=1.8), whereas the
non-bone-targeted analogue 1b gave a K value ꢁ0.
Functionally, compounds 2c and 4c were active in vitro
to inhibit rabbit osteoclast mediated resorption of den-
tine slices,15,16 showing IC50 values in the 1–3 mM range.
In terms of proof-of-concept, these data support the
expected physical and biological properties of a bone-
targeted pyrrolopyrimidine template-based Src kinase
inhibitor.
Although not as detailed relative to the above studies, a
similar effort to test a series of bone-targeted pyr-
azolopyrimide template-based Src kinase inhibitors was
also undertaken. Relative to the known Src kinase inhi-
bitor PP1 (Fig. 4), compounds 8–12 were tested for their
comparative enzyme inhibitory properties using the
SPA kinase assay (Table 1). Noteworthy was compound
9 which showed an Src kinase IC50 of 20 nM and sup-
ported molecular modeling studies that predicted its
bone-targeting group would likely not interfere with
binding to the enzyme active site.
In summary, using structure-based drug design we suc-
cessfully incorporated novel bone-targeting groups into
Fig 2. Candidate bone-targeting groups and abbreviated nomen-
clature.
Fig 3. Incorporation of a bone-targeting group onto a pyrrolopyri-
midine template-based Src kinase inhibitor 1b to give 2d.
Figure 1. (a) A non-hydrolyzable ATP analogue AMP-PNP used in
the crystallographic studies12 docked into the model of Src kinase
illustrating the critical hydrogen bonds; (b) compound 3b docked in
the Src kinase model illustrating the solvent exposed region for placing
the bone-targeting group; (c) incorporation of phosphonomethylpho-
sphonic acid (PCP) moiety into 3b.
Fig 4. Structures of compounds 8–12 and PP1.

3066
R. Sundaramoorthi et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3063–3066
both pyrrolopyrimidine- and pyrazolopyrimidine-based
templates of known Src kinase inhibitors. Compound
4d was determined to be more potent than its parent
non-bone-targeted Src kinase inhibitor 3b. The bone-
targeted Src kinase inhibitors exhibited high affinities
for hydroxyapatite and potent in vitro inhibition of Src-
dependent, osteoclast-mediated bone resorption. Col-
lectively, these studies support the proof-of-concept for
bone-targeted Src kinase inhibitors relative to hetero-
cyclic templates closely resembling ATP.
2000, 97, 9373. Metcalf, C. A.; van Shravandijk, M. R.; Dal-
garno, D. C.; Sawyer, T. K. Curr. Pharma. Des. 2002, 8, 2049.
9. Missbach, M.; Jeshke, M.; Feyen, J.; Muller, K.; Glatt, M.;
Green, J.; Susa, M. Bone 1999, 24, 437. Kniessel, M.; Gamse,
R.; Missbach, M.; Wilder, L.; Beerli, R.; Susa, S.; Green, J.
Calcif. Tissue Int. 1999, 64 (Suppl 1), S75. Missbach, M.; Alt-
mann, E.; Susa, M. Curr. Opin. Drug Discov. Dev. 2000, 3, 54.
10. Hanke, J. H.; Gardner, J. P.; Dow, R. L.; Changelian,
P. S.; Brissette, W. H.; Weringer, E. J.; Pollack, B. A.; Con-
nelly, P. A. J. Biol. Chem. 1996, 271, 695.
11. Xu, W.; Harrison, S. C.; Eck, M. J. Nature 1997, 385, 595.
Yamaguchi, H.; Hendrickson, W. A. Nature 1996, 384, 484.
Hubbard, S. R. EMBO J. 1997, 16, 5572.
12. Xu, W.; Doshi, A.; Lei, M.; Eck, M. J.; Harrison, S. C.
Mol. Cell 1999, 3, 629.
Acknowledgements
We thank Mr. James Biggie and Mr. Richard Brawley for
their help in preparing this manuscript. We thank Aventis
Pharmaceuticals Inc. for the SPA kinase assay data.
References and Notes
1. Rodan, G. A.; Martin, T. J. Science 2000, 289, 1508.
2. Boyce, B. F.; Hughes, D. E.; Wright, K. R.; Xing, L.; Dai,
A. Lab. Invest. 1999, 79, 83.
13. Shakespeare, W. C.; Sawyer, T. K.; Metcalf, C. A., III;
Wang, Y.; Bohacek, R.; Sundaramoorthi, R. US Patent
Application US 19632, 2002; Chem. Abstr. 2001, 138, 56075.
Gray, N. S.; Wodicka, L.; Thunnissen, A. W. H.; Norman,
T. C.; Kwon, S.; Espinoza, F. H.; Morgan, D. O.; Barnes, G.;
LeClerc, S.; Meijer, L.; Kim, S.-H.; Lockhart, D. J.; Schultz,
P. G. Science 1998, 281, 533.
3. Ducy, P.; Schinke, T.; Karsenty, G. Science 2000, 289,
1501.
4. Teitelbaum, S. L. Science 2000, 289, 1504.
5. Soriono, P.; Montgomery, C.; Geske, R.; Bradley, A. Cell
1991, 64, 693.
6. Lowe, C.; Yoneda, E.; Boyce, B. F.; Chen, H.; Mundy,
G. R.; Soriono, P. Proc. Natl. Acad. Sci. U.S.A. 1993, 90,
4485.
7. Marzia, M.; Sims, N. A.; Voit, S.; Migliaccio, S.; Taranta,
A.; Bernardini, S.; Faraggiana, T.; Yoneda, T.; Mundy, G. R.;
Boyce, B. F.; Baron, R.; Teti, A. J. Cell. Biol. 2000, 151, 311.
8. Sawyer, T. K.; Bohacek, R. S.; Eyermann, C. J.; Kawahata,
N.; Metcalf, C. A., III; Shakespeare, W. C.; Sundaramoorthi,
R.; Wang, Y.; Yang, M. G. Mini-Rev. Med. Chem. 2002, 2,
475. Shakespeare, W.; Yang, M.; Bohacek, R.; Cerasoli, F.;
Stebbins, K.; Sundaramoorthi, R.; Vu, C.; Pradeepan, S.;
Metcalf, C.; Haroldson, C.; Merry, T.; Dalgarno, D.; Narula,
S.; Hatada, M.; Lu, X.; van Schravendijk, M. R.; Adams, S.;
Violette, S.; Smith, J.; Guan, W.; Bartlett, C.; Herson, J.;
Iuliucci, J.; Weigele, M.; Sawyer, T. Proc. Natl. Acad. Sci. U.S.A.
14. Hartel, C.; Gobel, M. W. Helv. Chim. Acta 2000, 83, 2541.
Uludag, H. Curr. Pharma. Des. 2002, 8, 1929.
15. Detailed description of the following assays: Anti-resorp-
tive rabbit osteoclast assay, SPA Src kinase assay and hydroxy-
apatite assay can each be found in: Weigele, M.; Dalgarno, D.;
Luke, G. P.; Sawyer, T. K.; Bohacek, R. S.; Shakespeare, W.
C.; Sundaramoorthi, R.; Wang, Y.; Metcalf, C. A.; Vu, C. B.;
Kawahata, N. H. PCT Intl. Patent Application WO O1/44258,
2001.
16. Weigele, M.; Dalgarno, D. C.; Luke, G. P.; Sawyer, T. K.;
Bohacek, R. S.; Shakespeare, W. C.; Sundaramoorthi, R.;
Wang, Y.; Metcalf, C. A.; Vu, C. B.; Kawahata, N. H. US
Patent WO 0144259, 2001; Chem. Abstr. 2001, 135, 46048.