Synthesis and biological evaluation of novel bisphosphonates with dual activities on bone in vitro

Article abstract of DOI:10.1016/j.bmcl.2005.04.062

In an effort to generate novel anti-osteoporosis agents, new bisphosphonates with various benzoyl groups on side chains have been synthesized and evaluated for activities on both bone resorption and bone formation in vitro. Several candidates showed distinct dual activities: promoting bone formation and inhibiting bone resorption.

Full text of DOI:10.1016/j.bmcl.2005.04.062

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3267–3270
Synthesis and biological evaluation of novel bisphosphonates
with dual activities on bone in vitro
Yuli Xie,a Huasheng Ding,b,c Lihui Qian,a Xueming Yan,b
Chunhao Yangb and Yuyuan Xieb,*
aDepartment of Medicine, Columbia University, NY 10032, USA
bState Key Laboratory of Drug Research, Shanghai Institute of Materia Media, Shanghai Institutes for Biological Sciences,
CAS, Shanghai 201203, China
cGraduate School of the Chinese Academy of Sciences, Beijing 100049, China
Received 25 November 2004; revised 22 April 2005; accepted 23 April 2005
Abstract—In an effort to generate novel anti-osteoporosis agents, new bisphosphonates with various benzoyl groups on side chains
have been synthesized and evaluated for activities on both bone resorption and bone formation in vitro. Several candidates showed
distinct dual activities: promoting bone formation and inhibiting bone resorption.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Osteoporosis is a common skeletal disease characterized
by gradual loss of bone mass and disrupted bone archi-
tecture as a result of an imbalance between the bone
resorption activity of osteoclasts and the bone forma-
tion activity of osteoblasts.1 Bisphosphonates, stable
bone-targeted mimics of inorganic pyrophosphate, are
used principally to inhibit osteoclastic bone resorption
for the treatment of osteoporosis.2 Since changes of
bone mass can be induced pharmacologically by regulat-
ing either bone resorption or bone formation, or ideally
both, there has been great interest in the discovery of
new bisphosphonates causing anabolic effects on bone
in addition to their anti-resorptive effects.3 Our group
and othersꢀ attempts4,5 to achieve this goal by conjugat-
ing bisphosphonates with bone formation stimulants
have not resulted in significant success so far, presum-
ably because these conjugates, while inactive themselves,
are difficult to be hydrolyzed to release active moieties in
the bone. We now report the synthesis and biological
evaluation of novel dually active bisphosphonates, able
to both inhibit bone resorption and stimulate bone
formation in cell cultures.
In the development of chelating agents for removing
plutonium from bone, we serendipitously observed that
CBMIDA, an analog of EDTA (Fig. 1), markedly in-
creased osteoid volume in Beagle dogs.6 Further studies
showed that CBMIDA and other aromatic mimics of
EDTA could induce proliferation of rat calvarial-
derived osteoblasts in vitro whereas EDTA could not.
We reason that incorporation of aromatic groups with
simple bisphosphonates might also lead to an anabolic
effect on bone. On the other hand, structure–activity
relationship (SAR) studies of bisphosphonates have
demonstrated that nitrogen-containing bisphosphonates
are more potent inhibitors of bone resorption, and fur-
thermore, introduction of a cyclic group on the side
chain may increase the anti-resorptive potency.7 There-
fore, we designed and synthesized a new type of bisphos-
phonates with various aromatic rings on the side chains
linked by amide bonds (Fig. 2). A hydroxyl group
attached to the geminal carbon atom of the P–C–P
Keywords: Bisphosphonates; Osteoporosis; Osteoclast; Osteoblast.
*
Corresponding author. Tel.: +86 21 54920507; fax: +86 21
50806770; e-mail: yyxie@mail.shcnc.ac.cn
Figure 1.
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.04.062

3268
Y. Xie et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3267–3270
Table 1. Structure of compound 3 and their osteoclast inhibitory
activity
Compound
R
n
Osteoclast inhibitory
ratio mean SE
3a
3b
3c
3d
3e
3f
2,3-2OCH3
3,4-CH2O2
4-CH3
2
2
2
2
2
2
3
3
3
5
5
5
68.1 5.8
60.2 8.4
30.4 4.6
64.4 7.3
26.9 9.5
9.5 2.9
2,3-2OCH3, 5-NO2
3-OCH3
H
Figure 2. The structure of the target compound.
3g
3h
3i
3-OCH3
H
2-CH3
34.2 13
35.6 7.1
30.3 17.7
15.9 5.3
9.4 19.9
39.0 15.0
47.7 10.3
group supposedly increases their affinity for bone min-
eral because bisphosphonates chelate calcium ions more
effectively by tridentate rather than bidentate.2 In addi-
tion, the presence of a lipophilic aromatic group may
also be favorable for oral bioavailability,7 which is
the major disadvantage of the clinically utilized
bisphosphonates.
3j
3-Cl
2-CH3
4-OCH3
3k
3l
Aln
The osteoclast precursor cells were plated at a density 5 · 104 cells/well
in a 96-well plate cultured for 8 days with or without the presence of
10À7 M of bisphosphonates. After culturing, the cells were stained with
tartrate-resistant acid phosphonates (TRAP) to count the number of
positive cells containing three or more nuclei. The data were inhibitory
ratio mean SE, n = 5. The inhibitory ratio (%) = (number of control
cells À number of treated cells)/number of control cells · 100.
The synthesis of target compound 3 is shown in Scheme
1. Starting compound 1 was prepared according to Kie-
czykowskiꢀs method.8 Aromatic acid chloride 2 was pre-
pared from the corresponding benzoic acid. A freshly
prepared benzoyl chloride (2 equiv) in THF was added
to the solution of tetrasodium salt of aminoalkylenebis-
phosphonic acid in 2 N NaOH at room temperature.
After completion of reaction, the solution was acidified
to pH 3–4 with 2 N HCl and the disodium salts of com-
pound 3 were obtained as white crystals. The hygro-
scopic free bisphosphonic acids can be obtained from
ion-exchange chromatography.
range from 9.18% to 68.08%. The overall trend can be
seen that electron-donating substituents and short side
chains (n = 2) provide more effective inhibitors. Three
compounds (3a,b,d) with similar structures (n = 2, two
methyloxy groups or a methylenedioxy group on the
ring) displayed stronger activities (68.08%, 60.18%,
64.43%, respectively) than the reference agent, alendro-
nate (47.74%). This result suggests that alkoxy substitu-
ents might enhance the anti-resorptive activity.
Twelve desired compounds 3 have been synthesized in
which benzoyl rings can have one or more various sub-
stitutes such as methyloxy groups, halogen, nitro, etc.,
and the length of the methylene chain can vary9 (Table
1). Their biological activities were evaluated for both
the inhibition of bone resorption and promotion of bone
formation.
Compounds 3 were also tested for the proliferative
activity in cultures of SD rat calvarial osteoblasts.10 Re-
sults showed that five compounds (3a,b,d,e,g) stimulated
cellular proliferation and this effect was strongly depen-
dent upon the dose and peaked at 72 h after treatment
(Table 2). In particular, 3a markedly increased the cell
number over controls throughout the entire concentra-
tion range tested, with the peak effect, a 74% increase
at 10À7 M. Interestingly, there is some overlap in SAR
between the proliferative effect on bone formation and
the inhibitory effects on bone resorption. Alkylphenolic
moieties and short side chains (n = 2) not only resulted
The effects of compound 3 and alendronate (one of most
potent bone resorption inhibitors), as reference com-
pound on bone resorption, were examined in bone
marrow culture system.10
Their inhibitory potency varied according to the side
chain length and the benzoyl ring substitution in a wide
Scheme 1. Reagents: (a) H3PO3, PCl3; (b) SOCl2; (c) NaOH.

Y. Xie et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3267–3270
3269
Table 2. Effects of bisphosphonates on rat calvaria osteoblast prolif-
eration at different concentrations
ing cell population. While not significantly affecting the
synthesis of Type I collagen, compound 3a remarkably
enhanced osteocalcin expression up to fourfold com-
pared to controls in a dosage-dependent manner. These
results indicate that compound 3a is also a promoter of
osteoblast maturation.
Compound 10À5 M
10À6 M
10À7 M
10À8 M
3a
3b
3d
3e
3g
35.2 8.7
9.3 5.7
47.4 6.0
À8.3 4.8
74.4 5.1 26.6 3.8
À7.3 8.6 À0.2 3.3
21.4 9.8 À0.8 1.4
17.7 15.5 À1.5 3.6
8.9 6.9
0.7 2.4
7.1 1.6
À5.5 2.6
0.6 3.5
3.7 6.7
À5.5 1.9 À12.4 2.4
In summary, we have identified a series of novel bis-
phosphonates that significantly promote osteoblastic
bone formation, aside from their roles as potent inhibi-
tors of osteoclastic bone resorption in cell cultures. This
result suggests that these compounds, especially 3a,
might offer distinct advantages over currently available
bisphosphonates, which selectively target osteoclasts.
The further investigation of their bioavailabilities and
activities in vivo is in progress.
The cells were plated at a density of 2 · 104 cells/well in 24-well plates
and treated with various concentrations of bisphosphonates for 72 h.
Proliferation was determined using MTT colorimetric assay. The data
were proliferative ratio mean SE, n = 3. The proliferative ratio
(%) = (number of treated cells À number of control cells)/number of
control cells · 100.
in more potent inhibitors of osteoclasts, but also are
essential for the proliferative effect on osteoblasts.
Moreover, compound 3a, the most potent promoter
and inhibitor, shares the same catechol-like scaffold with
CBMID, the original lead compound (Fig. 2). Given
that similar structural features frequently appear in nat-
ural anti-osteoporosis compounds such as estrogen (Fig.
3), we believe that phenolic or alkylphenolic moieties
might provide biological benefits to anti-osteoporosis
agents.
References and notes
1. Francis, R. M. Curr. Ther. Res. 1997, 58, 656.
2. Russell, R. G. G.; Rogers, M. J. Bone 1999, 25, 97.
3. Lopez, F. G. Curr. Opin. Chem. Biol. 2000, 4, 383.
4. Gil, L.; Han, Y.; Opas, E. E.; Rodan, G. A.; Ruel, R.;
Seedor, J. G.; Tyler, P. C.; Young, R. N. Bioorg. Med.
Chem. 1999, 7, 901.
5. Page, P. C. B.; Moore, J. P. G.; Mansfield, I.; Mckenzie,
M. J.; Bowler, W. B.; Gallagher, J. A. Tetrahedron 2001,
57, 1837.
6. Fukuda, S.; Iida, H.; Hseih, Y. Y.; Chen, W. Z. Hoken
Butsuri. 1991, 26, 101.
7. Vepsalainen, J. J. Curr. Med. Chem. 2002, 9, 1201.
8. Kieczykowski, G. R.; Jobson, R. B.; Melillo, D. G.;
Reinhold, D. F.; Grenda, V. J.; Shinkai, I. J. Org. Chem.
1995, 60, 8310.
Intrigued by the extraordinary proliferative effect of
compound 3a, we further investigated its effect on the
maturation and differentiation of osteoblasts with bioas-
say of the activity of alkaline phosphatase (ALP) and
the synthesis of Collagen type I carboxyterminal pro-
peptide (CICP) and osteocalcin (OC).10 In our study
(Table 3), alkaline phosphatase levels increased 61%
over controls within 72 h after the treatment of 3a at
10À6 M, indicating enhanced differentiation of the exist-
9. Analytic data for a representative compound 3a: Anal.
Calcd for C12H17O10NP2Na2Æ4H2O: C, 27.96; H, 4.85; N,
Figure 3. The structures of three anti-osteoporosis compounds.
Table 3. Effect of compound 3a on the levels of alkaline phosphatase, osteocalcin and Type I collagen
Compound
Concentration (M)
ALP
OC
CICP
3a
10À6
10À7
10À8
10À9
3.60 0.84*
2.15 0.91
0.75 0.30
0.56 0.41
11.73 1.42***
8.63 0.17***
4.73 0.01**
4.07 0.16
4.06 0.04
4.17 0.09
4.03 0.02
4.27 0.07
Control
2.23 1.50*
3.92 0.32***
4.07 0.27
The cells were plated at a density of 5 · 103 cells/well in 96-well plates and treated with various concentrations of compound 3a for 72 h. The alkaline
phosphatase activity was determined by an established technique with p-nitrophenylphosphate as the substrate. Collagen type I carboxyterminal
propeptide (CICP) and osteocalcin (OC) concentrations were measured from the cell lysates. All the markers were normalized to the relative number
of viable cells as determined directly in the 96-well plates. Data are presented as mean SE (n = 5, *0.1 > P > 0.05; **0.05 > P > 0.01; ***P < 0.01).

3270
Y. Xie et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3267–3270
1
2.72. Found: C, 27.97; H, 4.82; N, 2.81; H NMR (D2O)
diluted with Dulbeccoꢀs Modified Eagles Medium
(DMEM). Then the obtained osteoblast-enriched cell
suspension was seeded in culture flask and cultured at
37 ꢁC in the presence of 5% CO2. Medium was changed
every 3 days. Cell viability was measured using a MTT
Cell Proliferation Assay. Absorbance readings of the
wells were taken of 492 nm. The alkaline phosphatase
activity was determined by an established technique with
p-nitrophenylphosphate as the substrate. T Collagen type
I carboxyterminal propeptide (CICP) concentration was
determined in cell culture supernatants with an ELISA.
Osteocalcin was measured by the use of serum osteocal-
cium radioimmunity kit from the Radioimmunity Insti-
tute of the Liberation Armyꢀs General Hospital. All
the makers of osteoblastic differentiation values were
normalized to the relative number of viable cells as
determined directly in the 96-well plates using the
above-mentioned proliferation assay.
ppm: 2.35–2.50 (m, 2H, CH2—); 3.80–3.89 (m, 2H,
–NHCH2–); 3.91 (s, 1H, –OCH3); 3.98 (s, 1H, –OCH3);
7.24–7.36 (m, 3H, ArH); 13C NMR (D2O) ppm: 35.10
(CH2—); 38.15 (–NHCH2–); 58.49 (–CH3); 64.22 (–OCH3);
75.04 (1-C); 118.21–154.59 (6C, Ar); 171.21 (C@O);
31P NMR (D2O) ppm: 19.49.
10. Materials and methods of bioactivity evaluation: Osteo-
clasts were prepared from thighbone, shoulder bone, and
tibia of 24 h new born rat. The cells were stained with
tartrate-resistant acid phosphonates (TRAP) to count the
number of positive cells. Primary osteoblast-enriched
cultures were prepared from newborn Sprague-Dawley
rat calvaria at axenic circumstances. Parietal bones were
dissected free from the sutures. The periosteal layers
from both sides of the bones were removed carefully. The
bones were cut into chips and digested with pancreas for
30 min. After centrifugation, the precipitate cells were