Low toxicity and unprecedented anti-osteoclast activity of a simple sulfur-containing gem-bisphosphonate: A comparative study

Article abstract of DOI:10.1016/j.ejmech.2013.04.032

Bisphosphonates (BPs) are key drugs for the treatment of bone resorption diseases like osteoporosis, Paget's disease and some forms of tumors. Recent findings underlined the importance of lipophilic N-containing BPs to ensure high biological activity. Her

Full text of DOI:10.1016/j.ejmech.2013.04.032

European Journal of Medicinal Chemistry 65 (2013) 448e455
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Low toxicity and unprecedented anti-osteoclast activity of a simple
sulfur-containing gem-bisphosphonate: A comparative study
,
b
,
,
,1
Donatella Granchi a , Alessandro Scarso 1, Giulio Bianchini b 1, Andrea Chiminazzo b
,
*
Alberto Minto b 1, Paolo Sgarbossa c, Rino A. Michelin c, Gemma Di Pompo a, Sofia Avnet a,
,
, ,
1
Giorgio Strukul b
*
a Laboratorio di Fisiopatologia Ortopedica e Medicina Rigenerativa, Istituto Ortopedico Rizzoli, via di Barbiano 1/10, I-40136 Bologna, Italy
b Dipartimento di Scienze Molecolari e Nanosistemi, Università Ca’ Foscari di Venezia, Dorsoduro 2137, I-30123 Venezia, Italy
c Dipartimento di Processi Chimici dell’Ingegneria, Università degli Studi di Padova, via F. Marzolo 9, I-35131 Padova, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Bisphosphonates (BPs) are key drugs for the treatment of bone resorption diseases like osteoporosis,
Paget’s disease and some forms of tumors. Recent findings underlined the importance of lipophilic N-
containing BPs to ensure high biological activity. Herein we present some unprecedented results con-
cerning the low toxicity and good anti-osteoclast activity of low molecular weight hydrophilic S-con-
taining BPs. A series of S and N-containing BPs bearing aromatic and aliphatic substitution were prepared
through Michael addition reaction between vinylidenebisphosphonate tetraethyl ester and the proper
nucleophile under basic catalysis. S-containing BPs showed a generally low toxicity, determined with the
neutral-red assay using the L929 cell line, and, in particular for an aliphatic one, a good biological activity
assessed on primary cultures of human osteoclasts.
Received 22 January 2013
Received in revised form
8 April 2013
Accepted 11 April 2013
Available online 11 May 2013
Keywords:
Bisphosphonates
S-containing drugs
Osteoclast
Ó 2013 Elsevier Masson SAS. All rights reserved.
Apoptosis
Cytotoxicity
1. Introduction
In 2011 BPs celebrated 40 years of application in medicinal
chemistry [3], but still the correlation between their chemical
Since 1970s, gem-bisphosphonates (BPs) 1 (Scheme 1) have
been employed as drugs for the treatment of bone disorders,
hypocalcaemia and osteoporosis [1] thanks to their structural
similarity with pyrophosphate 2 a major constituent of hydroxy-
apatite (HAP) present in the mineral portion of bones. This property
ensures that bisphosphonate-based drugs are targeted only to
bones (magic bullets) [2] and do not accumulate on other organs or
tissues. The mineral portion of bone is a mix of inorganic HAP,
carbonatoapatite (a carbonated form of the stoichiometric HAP),
and proteins, mainly collagen. The continuous equilibrium action
between osteoclasts, involved in resorption of the mineral portion
of bones, and osteoblasts, involved in mineral formation, ensures
health to the skeleton of mammals. Deregulation of osteoclast ac-
tivity is associated with osteoporosis, metastasis-induced osteol-
ysis, Paget’s disease, rheumatoid arthritis, and periodontal disease.
structure and biological activity is a highly debated topic [4,5].
Despite their successful use, a critical feature of BPs is their low oral
bioavailability (1e2%) since these derivatives are highly hydrophilic
and extensively charged and consequently poorly absorbed from
the gastrointestinal tract after oral administration thus limiting
transcellular permeation across the epithelial cells [6].
Two enzymes, both involved in the mevalonate pathway that
provide lipids such as cholesterol and isoprenoids to the cell, are
targets of BPs in the inhibition of osteoclasts, namely farnesyldi-
phosphate synthase (FPPS) and geranylgeranyl diphosphate syn-
thase (GGPPS) [7]. This feature allows also to achieve good
inhibition properties toward a wide range of parasitic protozoan
diseases like those caused by Plasmodium, Leishmania and Toxo-
plasma species [8,9].
FPPS is an enzyme that converts dimethylallyldiphosphate in a
longer farnesyldiphosphate containing
fragment while GGPPS converts farnesyldiphosphate into a longer
20-diphosphate moiety. Inhibition of these enzymes turns out into
osteoclast inactivation and apoptosis. Farnesyldiphosphate (FPP)
biosynthesis proceeds via carbocationic transition state/reactive
intermediates and this explains the high activity of the latest
a polyunsaturated C10
C
* Corresponding authors.
E-mail addresses: donatella.granchi@ior.it (D. Granchi), strukul@unive.it
(G. Strukul).
1
Tel.: þ39 041 2348569; fax: þ39 041 2438517.
0223-5234/$ e see front matter Ó 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2013.04.032

D. Granchi et al. / European Journal of Medicinal Chemistry 65 (2013) 448e455
449
deal with S-containing BPs. Some describe exclusively the synthesis
of such class of BPs in the ester protected form [16], and only a few
report the formation of the corresponding bisphosphonic acids [17]
and their potential effects on osteoclast’s activity [18]. Tiludronate
5, like other non-nitrogenous BPs can be incorporated into mole-
cules of adenosine triphosphate (ATP), creating non hydrolysable
analogs of ATP molecules called cytotoxic ATP. Nevertheless,
tiludronate 5 has low cellular potency and low mineral affinity in
comparison to the more potent N-containing BPs, and the clinical
use is limited to the treatment of Paget’s disease and in veterinary
orthopedics [19]. Worth of mention is a very recent contribution
where sulphonylamino derived BPs 6 bearing sulphone moiety in
O
P
O
P
R2
O
P
O
P
HO
HO
OH
OH
HO
HO
OH
OH
O
R1
1
2
O
P
O
P
O
P
O
P
O
P
O
P
OH
OH
HO
HO
OH
OH
HO
HO
OH
OH
HO
HO
OH
OH
( )n
S
N
N
NH2
3b
Cl
(n = 1),
(n = 2)
O
3a
4
5
the
a or b-position proved to be powerful bone resorption in-
O
P
hibitors as assessed by cell line J774 and HepG2 survival assay, thus
suggesting that chemical structure of S-containing BPs could have a
pivotal role in enhancing their therapeutic effectiveness [20].
Herein we present a comparative study on the toxicity and
osteoclast inhibition activity between some known N-containing
BPs and one new S-containing BPs (8e12, Scheme 2), either con-
taining an aromatic portion or an aliphatic substituent on position
R,1 in all cases lacking the presence of the gem-hydroxyl moiety R2
which has been demonstrated not to be crucial for the biological
activity of this class of molecules [10]. The aim of the work is to
address whether small S-containing BPs provide good anti-
osteoclast activity while maintaining a low toxicity level which is
an extremely important aspect for further drug development.
HO
P
OH
O
P
O
P
OH
HO
O
O
O
NH
O
O
S
O
7
6
R3
Scheme 1. Chemical structure of bisphosphonates 1, pyrophosphoric acid 2, pami-
dronic 3a, alendronic 3b, zoledronic 4, tiludronic 5 acid, sulfonyalmino bisphosphonic
acids 6 and vinylidenebisphosphonate tetraethyl ester (VBP) 7.
generation bisphosphonates bearing pyridinium residues. The
active site of FDPS accommodates the BP moiety via coordination to
a Mg(II) metal center and the most potent inhibitors, containing
basic N atoms on C4 that are protonated at physiological pH, pro-
vide strong hydrogen bonding recognition with specific aminoacids
(Lys200 and Thr201). GGPPS is inhibited by more hydrophobic BPs
having long aliphatic or aromatic residues because in this case the
active site of the enzyme is more hydrophobic and does not contain
hydrogen bonding aminoacids. The hydrophobic character of BPs is
a common aspect that greatly influences their activity. Very
recently it has been demonstrated that such enzymes can be effi-
ciently inhibited by more hydrophobic BPs lacking the gem-OH
moiety and containing a positive N center and a long aliphatic
chain [6b,10]. Similarly, hydrophobic BPs bearing rigid polycyclic
bile acid moieties showed enhanced biological activity compared to
commercial BPs drugs like neridronate [11].
Although BPs are the most potent of the currently approved anti-
resorptive agents, their effectiveness is related strictly to drug
adherence, both in terms of compliance (i.e., proportion of the total
treatment time that is covered by the dispensed medication) and
persistence (i.e., continuation of treatment over time). Epidemio-
logic studies have shown that independent predictors of poor
adherence to BP therapy include more frequent dosing, adverse
upper gastrointestinal effects, treatment cost, lack of disease
symptoms, and practical difficulties with the administration
regimen [12,13]. BPs with R2 ¼ OH and containing nitrogen de-
rivatives in R1, like pamidronate 3a, alendronate 3b and zoledronate
4, are by far the most studied class of anti-resorption commercial
drugs [14]. On the other hand, N-containing aromatic species are
characterized by high toxicity that favors the non-adherence to
treatment thus limiting the therapeutic benefit [15]. However, the
BP moiety is intrinsically poorly toxic and the development of new
BPs characterized by low toxicity and good anti-osteoclast activity is
a very important issue.
2. Results and discussion
2.1. Chemistry: synthesis of BP derivatives bearing N and S
substituents
We prepared five BPs reported in Scheme 2 that can be classified
into four classes depending on the presence of heteroatoms in R1
like N (8e10) or S (11e12) and on the presence of an aromatic ring
(9e10 and 12) or an aliphatic moiety (8 and 11). The synthesis of the
BPs was accomplished by reaction of the proper heteroatomic
nucleophile with vinylidenebisphosphonate tetraethyl ester 7 [21]
under triethylamine catalysis, with an improved procedure
compared to those reported in the literature [22e25] providing
almost quantitative yields for all the corresponding BP tetraethyl
ester precursors 13e17. These were treated with trimethyl-
bromosilane for 18 h at room temperature [26], further reacted
with water for 4 h in order to remove the ester moiety providing
the free bisphosphonic acids 8e12 in quantitative yields
O
P
O
P
O
P
O
P
O
P
O
P
HO
HO
OH
OH
HO
OH
OH
HO
HO
OH
OH
HO
N
NH
NH
O
N
8
9
10
O
P
O
P
O
P
O
P
HO
HO
OH
OH
HO
HO
OH
OH
OH
S
S
We turned our attention to S-containing BPs that are an under-
investigated class of BPs. One example is the commercially available
tiludronate 5 (Scheme 1), known under the commercial name
11
12
Scheme 2. Chemical structure of the bisphosphonates synthesized and tested:
aliphatic N-containing BP 8, aromatic N-containing BPs 9 and 10, aliphatic 11 and ar-
omatic 12 S-containing BPs.
SkelidÒ, that contains a S atom in
b
-position. On the other hand in
the more recent literature only a limited number of contributions

450
D. Granchi et al. / European Journal of Medicinal Chemistry 65 (2013) 448e455
(Scheme 3). The products were characterized by 1H and 31P NMR
spectroscopy; in all cases a single sharp resonance was observed in
the 31P spectra in D2O indicating high water solubility and absence
of aggregation phenomena that would provide broad resonances.
Electrospray MS analysis of the BP acids 8e12 were performed in
methanol observing good matching between the [M ꢀ H]ꢀ ion and
the ESI spectra on negative ion, thus confirming the structure of the
BP products.
Table 1
Amount of bound BPs 8e12 in the affinity measurements.
a
BP
%
Adsorbed BP
8
9
10
65
66
94
82
70
34
Zoledronate 4
11
12
[BPs 8e12] ¼ 7 mM; solvent H2O, 10 mL; HAP ¼ 100 mg.
2.2. Hydroxyapatite affinity measurements
a
Determined by 31P NMR integration with respect to an
external H3PO4 standard.
Determination of relative binding affinities of newly synthesized
BPs to hydroxyapatite (HAP), that constitutes the mineral portion of
bones, can be performed with different techniques. Among others,
31P NMR [27] spectroscopy allows a rapid and reliable response.
Even though it is not always true that best BPs in terms of biological
activity are also those with the highest HAP affinity, such affinities
are still important parameters since bone targeting properties and
bioavailability are both heavily influenced by it. Therefore, we
performed a series of experiments incubating in a buffer solution
fixed amounts of BP 8e12 in the presence of a given amount of
nano-powder HAP, monitoring the residual amount of free BP in
solution with respect to an external standard and observing, at the
same time, the release of phosphate in solution by displacement
(see Experimental part). Affinity measurements for compounds 8e
12 along with data for zoledronate 4 taken as a comparison are
reported in Table 1.
Data reported in Table 1 clearly show that the presence of amino
and alcohol groups in the side chain of BPs, even though they tend
to increase the hydrophilicity of the molecule, at the same time
ensure higher HAP affinity. In fact, 12 turned out to be the more
weakly bound BP, while 8, 9 are characterized by higher binding.
Product 11 bearing an OH group in the side chain increases slightly
the affinity for HAP while 10, that bears two N atoms in close
reciprocal proximity is the best BP in terms of binding affinity, even
higher than zoledronate 4 that contains two N atoms in the side
chain and a gem-OH moiety often considered to be an important
structural element to ensure high bone targeting.
commercially available N-containing BP (pamidronate 3a) which
was included as control of the experimental setting. Next, the
pharmacological activity of compounds 8e12 was tested by eval-
uating their capability to inhibit osteoclastogenesis, to induce
apoptosis of newly formed osteoclasts, and to inhibit the
osteoclast-mediated bone resorption [28].
2.3.1. Cytotoxicity testing
The cytotoxicity of compounds was measured by using the
neutral-red (NR) assay on the L929 cell line [11]. The NR assay is a
colorimetric test based on the ability of viable cells to incorporate
and bind neutral-red within lysosomes. Xenobiotics may induce
changes of the cell membrane and lead to lysosomal fragility, with
the inevitable consequence of a decreased uptake and binding of
NR. Therefore, the number of viable cells and the amount of dye
incorporated into cells are directly correlated. The drug concen-
trations for in vitro experiments were chosen on the basis of the
therapeutic dose of pamidronate 3a, which is used for systemic
administration at single dose of 30 mg [29]. The systemic admin-
istration allows the complete adsorption of the drug which is
diluted in 5 L of blood [30]. Taking into account that the skeletal
retention of BPs may be highly variable [31], the cytotoxicity testing
of the new compounds has been performed using a large range of
concentrations (i.e., from 0.0001 mM to 100 mM).
Fig. 1 shows the proportion of viable cells after 72 h of culture
which is expressed as ratio between the neutral-red captured by
L929 exposed to compounds and the negative control. We observed
that pamidronate 3a and pyridine-containing BP 10 affected cell
viability following a doseeresponse relationship. At the higher
2.3. Biological evaluation
The biological activity of the nitrogen- and S-containing BPs 8e
12 was evaluated in two steps. Firstly, a toxicological screening was
performed to compare the new synthesized compounds to a
concentrations, i.e. 100 mM and 10 mM, aliphatic N-containing BP 8,
aromatic N-containing BP 9, aliphatic S-containing BP 11, and
O
P
O
P
O
P
O
P
NuH, CHCl3, rt
TEA (3 mol%)
O
O
O
O
O
O
O
O
Nu
7
13-17
O
P
O
P
i) BrSi(CH3)3, rt
ii) H2O, rt
HO
HO
OH
OH
Nu
8-12
NH
NH
S
S
N
Nu =
Fig. 1. Cytotoxicity of newly synthesized BPs (8e12) in comparison to BPs of estab-
lished activity (pamidronate 3a). The graph shows the ratio between the OD measured
in L929 exposed to compounds and OD measured in cells cultured in standard con-
dition, after 72 h of culture. Mean ꢁ SEM of 5 experimental settings which were
performed in triplicate wells. The asterisks mean significant differences as explained in
the text. According to BonferronieDunn correction, comparisons were considered
significant if the corresponding P value was less than 0.0024.
O
N
OH
Scheme 3. Synthesis of free bisphosphonic acids 8e12 by reaction between different
nucleophiles on 7 followed by deprotection of the ester moiety with BrSi(CH3)3 and
water.

D. Granchi et al. / European Journal of Medicinal Chemistry 65 (2013) 448e455
451
aromatic S-containing BP 12 were significantly less cytotoxic than
the above mentioned compounds (P < 0.0001). However, at
100 M, aliphatic N-containing BP 8 and aromatic N-containing BP
m
9 affected cell viability more than aliphatic S-containing BP 11 and
aromatic S-containing BP 12 (P < 0.0001).
2.3.2. Inhibition of osteoclast activity
The biological activity of the nitrogen- and S-containing BPs was
evaluated on primary cultures of osteoclasts. A more restricted
range of concentrations was employed: 10
therapeutic dose of BPs [30], 0.1 M mimics the absorption rate of
conventional BPs corresponding to the 1% of the therapeutic dose
[3], and 0.001 M was chosen because the culture system did not
allow the excretion of bioactive agents so that the biological effect
may be exacerbated.
Fig. 2 shows that the formation of new osteoclasts which were
obtained by cultivating peripheral blood mononuclear cells with
proper differentiating agents was completely inhibited with the
mM corresponds to the
m
m
Fig. 3. Apoptosis induced by newly synthesized BPs (8e12) and pamidronate 3a.
Mononuclear cells were cultured for seven days with osteoclast differentiation me-
dium and then exposed to the active compounds in order to highlight the pro-
apoptotic activity of drugs. The graph shows the percentage of apoptotic osteoclasts
calculated on the total number of multinucleated cells. Mean ꢁ SEM of 4 cultures
obtained from two different donors. The asterisks mean statistically significant dif-
ferences as explained in the text. According to BonferronieDunn correction, compar-
isons were considered significant if the corresponding P value was less than 0.0018.
major concentration (10 mM) of pyridine-containing BP 10. Pamidr-
onate 3a and aromatic N-containing BP 9 affected the osteoclast
differentiation, and the inhibitory effect was statistically significant
at all concentrations (P < 0.0001).
(P < 0.0001), pyridine-containing BP 10 (P < 0.0001), and aliphatic
S-containing BP 11 (P < 0.0001). Interestingly, the pro-apoptotic
In spite of the low cytotoxicity which was observed at 10 mM,
aliphatic N-containing BP 8, aromatic N-containing BP 9, and
aliphatic S-containing BP 11 decreased significantly the formation
of multinucleated and TRAcP positive cells (P < 0.0001), and the
pharmacological effect was comparable to that of pamidronate 3a
(Fig. 2). On the contrary, the aromatic S-containing BP 12 was
significantly less active than pamidronate 3a (P ¼ 0.0006), and the
osteoclast number was similar to that counted in control cultures.
In order to highlight the capability of active compounds in
promoting the apoptosis of newly formed osteoclasts, mononuclear
cells were cultured for seven days with differentiating agents, and
then the osteoclast cultures were exposed to the drugs. The strong
adhesion to substrate of mature osteoclasts and the lack of
resorption lacunae hampering BP uptake were taken into account
[32]. Therefore, the drug concentrations used in this experimental
setting were higher than those used for evaluating the osteoclas-
activity of 3a disappeared at 10 mM, whilst it was significantly
higher in cultures treated with pyridine-containing BP 10 and
aliphatic S-containing BP 11 than in control cultures (P < 0.0001
and P ¼ 0.0007, respectively).
Our results showed that toxicity and biological activity of newly
synthesized BPs depend on type and concentration of compound,
but significance following statistical analysis was reliable only
when very large differences were observed. In this regard, we
conceived a scoring system for grading objectively the performance
of the tested compounds in comparison to the control cultures, that
were maintained in standard conditions without treatments. The
score was assigned assuming that “positive” values indicate “good
performance” and “negative” values “poor performance”. The sum
of the score values obtained for each compound and concentration
in experiments of cytotoxicity, inhibition of the osteoclastogenesis,
induction of apoptosis, and inhibition of the osteoclast-mediated
bone resorption are shown in Table 2.
togenesis inhibition (i.e., 100 mM, 10 mM, 0.1 mM).
As shown in Fig. 3, the percentage of apoptotic osteoclasts was
very low in control cultures (16.2 ꢁ 2%), but increased significantly
when cells were exposed to the highest concentration of 3a
With regard to the cytotoxicity, the best total score was
observed for aliphatic S-containing BP 11, followed by aromatic S-
containing BP 12, aliphatic N-containing BP 8 and aromatic N-
containing BP 9. The pyridine-containing BP 10 showed a negative
sum of score values, thus confirming a medium degree of toxicity of
this compound at the highest doses. As expected, pamidronate 3a
showed the lowest negative score.
With regard to the anti-osteoclast activity, the best total score
was observed for the pyridine-containing BP 10, which was less
toxic than pamidronate 3a, but exhibited the most potent anti-
resorptive activity. Aliphatic N-containing BP 8 and aromatic N-
containing BP 9 inhibited the osteoclastogenesis, but failed in
inducing the apoptosis of new formed osteoclasts, while the aro-
matic S-containing BP 12 did not show any anti-resorptive activity.
Noteworthy, the aliphatic S-containing BP 11, while not having
potent activity of osteoclastogenesis inhibition, seemed more active
than pamidronate 3a in promoting the apoptosis of osteoclasts. The
compounds which showed a positive sum of the scores related to the
anti-osteoclast activity, i.e. pyridine-containing BP 10, aliphatic S-
containing BP 11, and pamidronate 3a, passed to the next step which
aimed to evaluate the resorptive function of mature osteoclasts.
OsteoLyseÔ provides a quantitative assessment of in vitro
osteoclast-mediated degradation of human Type I bone collagen by
directly measuring the release of fluorescent-labeled collagen
fragments. Fig. 4 demonstrates that the addition of all BPs in culture
Fig. 2. Inhibition of osteoclastogenesis induced by newly synthesized BPs (8e12) in
comparison to BPs of established activity (pamidronate 3a). The active compounds
were added together with the osteoclast differentiation medium to highlight the
ability of drugs in interfering the osteoclast formation. The graph shows the per-
centage of multinucleated and TRAcP positive cells counted after exposure to BPs and
related to the number of osteoclasts formed in untreated cultures (positive
control ¼ 100%). Mean ꢁ SEM of 4 cultures which were obtained from two different
donors. The asterisks mean statistically significant differences as explained in the text.
According to BonferronieDunn correction, comparisons were considered significant if
the corresponding P value was less than 0.0018.

452
D. Granchi et al. / European Journal of Medicinal Chemistry 65 (2013) 448e455
Table 2
medium affected the bone collagen degradation in comparison to
the untreated control cultures, and statistically significant differ-
ences were observed at the highest concentration (P < 0.0001).
The pharmacological activity of aromatic N-containing BPs 10
Comparative study on the biological activity of N-containing BPs and new S-con-
taining BPs: synopsis of results based on the score system.
BP and concentration
a
b
c
d
e
persisted at 10
m
M (P < 0.0001) and 0.1
m
M (P ¼ 0.0011). At 10
mM
Aliphatic N-containing BP (8)
100
m
M
0
16
14
17
47
n.t.
2
ꢀ3
ꢀ4
ꢀ5
ꢀ2
ꢀ3
ꢀ5
n.t.
ꢀ10
n.t.
n.t.
n.t.
n.t.
n.t.
the aliphatic S-containing BP 11 interfered the collagen degrada-
tion, but not significantly, and at the lowest concentration the
pharmacological activity was no more evident. The anti-resorptive
10 mM
0.1
m
M
0.001 mM
Total sum
activity of pamidronate 3a disappeared at 0.1 mM. With regard to
ꢀ15
their capability to interfere with the collagen degradation,
pyridine-containing BP 10 totalized the best score followed by 3a
and aliphatic S-containing BP 11 (Table 2).
Aromatic N-containing BP (9)
100
m
M
ꢀ3
8
16
24
45
n.t.
6
4
4
14
ꢀ9
ꢀ9
ꢀ8
n.t.
ꢀ26
n.t.
n.t.
n.t.
n.t.
n.t.
10 mM
The above results confirm that the chemical structure of BPs
plays a major role in determining their pharmacological activity.
The potency of the compounds varies according to the features of
the R1 side chains. Overall, the results observed can be summarized
as follows: i) aromatic N-containing BPs are more active than
aliphatic ones, but exhibit a noticeable cytotoxicity, ii) S-containing
BPs showed a generally low toxicity, and iii) the S-containing BP 11
is both not toxic and provides good anti-osteoclast activity in terms
of inhibition of osteoclastogenesis, induction of apoptosis and in-
hibition of collagen degradation by osteoclasts.
The uptake of N and non-N-containing BPs occurs by a similar
endocytic mechanism. The more dominant reason for differences
between their pharmacological activity resides in their biochemical
targets within the osteoclast. N-containing BPs act on the meval-
onate pathway and affect the post-translational modification
(prenylation) of small GTPases which are important signaling pro-
teins which regulate a variety of cell processes important for
osteoclast function, including cell morphology, cytoskeletal
arrangement, membrane ruffling, trafficking of vesicles and
apoptosis. S-containing BPs, like other non-N BPs, are metabolized
in the osteoclast cytosol to ATP analogs which can disrupt phos-
phorylation, thereby impairing the activation and deactivation of
regulatory enzymes leading to the apoptosis of the osteoclast [31].
The lack of toxicity of the aliphatic S-containing BPs coupled with
the good anti-osteoclast activity suggests that the activity of this
compound mainly depends on specific mechanisms. More recently,
the knowledge on the activity of S-containing BPs has been
expanded. The commercially available tiludronate 5 inhibits the
ATP-dependent proton pumps located in the plasma membrane of
the osteoclast [33]. Proton pumps are needed to acidify the space
between the osteoclast and the bone surface, thus favoring the
activity of collagenolytic metalloproteinases.
0.1
m
M
0.001
m
M
Total sum
ꢀ12
19
2
Aromatic N-containing BP (10)
100
m
M
ꢀ27
ꢀ24
13
11
ꢀ27
n.t.
12
0
0
12
6
4
ꢀ3
n.t.
7
12
17
13
n.t.
42
10
m
M
0.1
m
M
0.001
m
M
Total sum
Aliphatic S-containing BP (11)
100
m
M
15
18
12
9
n.t.
4
3
1
0
ꢀ3
n.t.
ꢀ2
8
5
ꢀ5
n.t.
8
10
m
M
0.1
m
M
0.001
m
M
ꢀ3
Total sum
54
4
Aromatic S-containing BP (12)
100
m
M
13
12
9
15
49
n.t.
ꢀ1
ꢀ1
ꢀ6
ꢀ8
ꢀ1
ꢀ3
ꢀ5
n.t.
ꢀ9
n.t.
n.t.
n.t.
n.t.
10 mM
0.1 mM
0.001
m
M
Total sum
ꢀ17
Pamidronate 3a
100
m
M
ꢀ89
ꢀ52
12
34
ꢀ95
n.t.
6
4
3
13
9
24
11
4
n.t.
39
10
m
M
ꢀ9
ꢀ8
n.t.
ꢀ8
0.1
m
M
0.001
m
M
Total sum
5
[a]: Cytotoxicity; [b]: inhibition of osteoclastogenesis [c]: induction of apoptosis;
[d]: sum of b þ c; [e]: inhibition of the osteoclast-mediated bone resorption, [n.t.]:
not tested.
3. Conclusion
In this contribution a comparative study on the HAP affinity,
toxicity and anti-osteoclast activity of a series of aromatic and
aliphatic N- and S-containing BPs showed that the latter class of
molecules provides new contributions for the development of more
potent and better tolerated BPs. This observation well complements
recent finding on other classes S-containing BPs that showed
promising properties as antitumor agents for breast, cervix, liver
and colon cancer diseases [18b,c]. Affinity measurements to HAP
measured by means of 31P NMR showed a certain degree of
matching with the biological activity of BPs 8e12, even though
toxicity properties turned out to be crucial. In fact, S-containing BPs
showed a generally low toxicity and in particular the aliphatic BP 11
provided good anti-osteoclast activity in terms of inhibition of
osteoclastogenesis, induction of apoptosis and inhibition of collagen
degradation by osteoclasts. While the low toxicity of S-containing
BPs 11 and 12 is not unexpected, the good anti-osteoclast activity of
11 is absolutely unprecedented since common knowledge, and
recent contribution to the field confirmed, that higher biological
Fig. 4. Inhibition of bone collagen degradation induced by newly synthesized BPs (10,
11) and pamidronate 3a. The experimental setting was the same used for the induction
of apoptosis. The graph shows the ratio between the RFU measured in osteoclast
cultures exposed to compounds and RFU measured in cultures growing in osteoclast
differentiation medium (untreated control ¼ 1). Mean ꢁ SEM of 2 experiments (two
different healthy donors) which were performed in quadruplicate wells. The asterisks
mean statistically significant differences as explained in the text. According to Bon-
ferronieDunn correction, comparisons were considered significant if the corre-
sponding P value was less than 0.0083.

D. Granchi et al. / European Journal of Medicinal Chemistry 65 (2013) 448e455
453
activity is observed predominantly with N-containing BPs bearing
4.2.3. Tetraethyl [2-(pyridin-2-ylamino)ethane-1,1-diyl]bis(phospho
nate) 15
1H NMR (300 MHz, CDCl3),
long aliphatic alkyl chains [10] that impart a high degree of hydro-
phobicity to the BPs. The simple structure of the S-containing BPs
studied in this work suggests further development of the class of
aliphatic S-containing BPs, also in the chiral enantiopure form. This
task is currently underway in our laboratories.
d
7.87 (d, J ¼ 3.9 Hz, 1H), 7.17 (m, 1H),
6.36 (m, 1H), 6.25 (d, J ¼ 8.4 Hz, 1H), 5.33 (t, J ¼ 6.3 Hz, 1H), 3.99 (m,
8H), 3.75 (td, J ¼ 15.2,12.6 Hz, 2H), 2.70 (tt, J ¼ 23.1, 6.3 Hz,1H),1.32 (t,
J ¼ 7.2 Hz,12H) ppm. 31P NMR (300 MHz, CDCl3),
d 20.54 (s, 2P) ppm.
4. Experimental section
4.2.4. Tetraethyl {2-[(2-hydroxyethyl)sulfanyl]ethane-1,1-diyl}bis
(phosphonate) 16
1H NMR (300 MHz, CDCl3),
d 3.99 (m, 8H), 3.90 (br s, 2H), 3.54 (t,
4.1. Reagents and materials
J ¼ 6.3 Hz, 2H), 3.49 (t, J ¼ 6.3 Hz, 2H), 2.87 (td, J ¼ 16.8, 6.0 Hz, 2H),
General. 1H NMR, 31P{1H} NMR, spectra were run on a Bruker
Avance 300 spectrometer operating at 300.15, (121.5), MHz,
2.48 (m, 5H), 1.34 (t, J ¼ 8.4 Hz, 1H), 1.15 (t, J ¼ 7.2 Hz, 12H) ppm. 31
P
NMR (300 MHz, CDCl3),
d 20.26 (s, 2P) ppm.
respectively, at 298 K, unless otherwise stated.
d values in ppm are
relative to Si(CH3)4, 85% H3PO4. GCeMS analyses were performed
on a GC Trace GC 2000 coupled with a quadrupole MS Thermo
Finnigan Trace MS with Full Scan method. Experimental conditions
are reported in the following Table 3.
Vinilydenebisphosphonate tetraethyl ester 7 was prepared as
reported in the literature [21]. BPs 8e12 were prepared following a
modified procedure with respect to what reported in the literature
[22e24,34]. All the synthetic work was carried out with the
exclusion of atmospheric oxygen under nitrogen atmosphere using
standard Schlenk techniques. Solvents were dried and purified
according to standard methods.
4.2.5. Tetraethyl [2-(phenylsulfanyl)ethane-1,1-diyl]bis(phosphonate)
17
1H NMR (300 MHz, CDCl3),
d
7.15 (m, 5H), 4.11 (m, 8H), 3.41 (td,
J ¼ 16.7, 6.3 Hz, 2H), 2.69 (tt, J ¼ 24.0, 6.3 Hz, 1H), 1.25 (t, J ¼ 7.2 Hz,
12H) ppm. 31P NMR (300 MHz, CDCl3),
19.87 (s, 2P) ppm.
d
4.3. Typical experimental procedure for the deprotection of the BP
tetraethylesters
The tetraethyl ester of the bisphosphonic acid (0.2 mmol) was
introduced in a vial under anhydrous condition. To this 12 equiv-
alents of bromotrimethylsilane (2.4 mmol) were added. The vial
was then sealed and stirred under inert atmosphere at rt for 18 h.
The crude mixture was concentrated under reduced pressure,
diluted with double-distilled water, stirred for 4 h at room tem-
perature and dried. The free bisphosphonic acid was obtained in
quantitative yield and was characterized by 1H, 31P NMR and ESI.
4.2. Typical experimental procedure for the synthesis of BP
tetraethylesters
Vinilydenebisphosphonate tetraethyl ester 7 (128
was added to a vial followed by 1 eq. of the N- or S-containing
nucleophile (0.5 mmol), triethylamine (3.48 L, 5 mol%) and 0.5 mL
mL, 0.5 mmol)
m
of chloroform. After stirring overnight at 70 ꢂC, the crude mixture
was evaporated to dryness, then 5 mL of dichloromethane were
added and the solution was extracted three times with water. The
organic phase was evaporated providing the final product that was
characterized by 1H, 31P NMR and GCeMS analyses.
4.3.1. [2-(Morpholin-4-yl)ethane-1,1-diyl]bis(phosphonic acid) 8
1H NMR (300 MHz, D2O),
d 4.13e3.96 (br s, 2H), 3.68 (t,
J ¼ 12.2 Hz, 2H), 3.58e3.33 (br s, 4H), 3.27e3.04 (br s, 2H), 2.64 (tt,
J ¼ 21.3, 8.4 Hz, 1H) ppm. 31P NMR (300 MHz, D2O),
d 13.90 (s, 2P)
ppm. Elemental analysis: calc: C 26.19%, H 5.50%, N 5.09%; found C
25.74%, H 5.58%, N 4.99%.
4.2.1. Tetraethyl [2-(morpholin-4-yl)ethane-1,1-diyl]bis(phosphonate)
13
4.3.2. [2-(Phenylamino)ethane-1,1-diyl]bis(phosphonic acid) 9
1H NMR (300 MHz, CDCl3),
d 3.91 (m, 8H), 3.39 (m, 4H), 2.62 (td,
1H NMR (300 MHz, D2O),
d
7.47 (m, 5H), 3.69 (td, J ¼ 21.6, 7.5 Hz,
2H), 2.37 (tt, J ¼ 21.3, 7.4 Hz, 1H). 31P NMR (300 MHz, D2O),
d
13.90
14.7, 6.3 Hz, 2H), 2.36 (tt, J ¼ 24.0, 6.3 Hz,1H), 2.22 (t, J ¼ 4.5 Hz, 4H),
1.07 (t, J ¼ 7.2 Hz, 12H) ppm. 31P NMR (300 MHz, CDCl3),
d 20.88 (s,
(s, 2P) ppm. Elemental analysis: calc: C 34.18%, H 4.66%, N 4.98%;
found C 33.96%, H 4.71%, N 4.91%.
2P) ppm.
4.2.2. Tetraethyl [2-(phenylamino)ethane-1,1-diyl]bis(phosphonate)
4.3.3. [2-(Pyridin-2-ylamino)ethane-1,1-diyl]bis(phosphonic acid)
14
10
1H NMR (300 MHz, CDCl3),
d
7.04 (t, J ¼ 8.4 Hz, 2H), 6.59 (t,
1H NMR (300 MHz, D2O),
d 8.11e7.45 (br s, 5H), 7.13e6.74 (br s,
4H), 3.08 (br t, J ¼ 14.3 Hz, 2H), 2.44 (br s, 1H) ppm. 31P NMR
J ¼ 7.4 Hz, 1H), 6.53 (m, 2H), 4.18 (m, 8H), 3.74 (td, J ¼ 15.9, 6.0 Hz,
2H), 2.70 (tt, J ¼ 22.8, 5.7 Hz, 1H), 1.33 (t, J ¼ 7.2 Hz, 12H) ppm. 31
P
(300 MHz, D2O),
d 16.28 (s, 2P) ppm. Elemental analysis: calc: C
NMR (300 MHz, CDCl3),
d 20.74 (s, 2P) ppm.
29.80%, H 4.29%, N 9.93%; found C 29.53%, H 4.38%, N 9.71%.
4.3.4. {2-[(2-Hydroxyethyl)sulfanyl]ethane-1,1-diyl}bis(phosphonic
Table 3
acid) 11
Experimental conditions for GCeMS analysis.
1H NMR (300 MHz, D2O),
d
3.64 (t, J ¼ 6.6 Hz, 2H), 2.94 (td,
J ¼ 15.9, 6.0 Hz, 2H), 2.66 (t, J ¼ 6.3 Hz, 2H), 2.42 (tt, J ¼ 22.5, 6.3 Hz,
1H) ppm. 31P NMR (300 MHz, D2O),
18.24 (s, 2P) ppm. Elemental
Capillary column
Initial T, ꢂC
Rate, C/min
Final T, ꢂC
HP5-MS 30 m, 0.25 mm ꢃ 0.25
80 ꢂC for 5 min
30 ꢂC/min
mm
d
280 ꢂC for 30 min
analysis: calc: C 18.05%, H 4.54%, S 12.05%; found C 17.65%, H 4.70%,
S 11.76%.
Injector T (split), ꢂC
Gas carrier flow, mL/min
Injected volume, mL
Solvent delay, min
Mass range, amu
Detector voltage, V:
Interface T, ꢂC
280 ꢂC
0.8 mL/min
0.8e1 mL
4.3.5. [2-(Phenylsulfanyl)ethane-1,1-diyl]bis(phosphonic acid) 12
4 min
1H NMR (300 MHz, D2O),
d
7.25 (m, 5H), 3.31 (td, J ¼ 15.6, 6.3 Hz,
35e500 amu
350 V
280 ꢂC
200 ꢂC
2H), 2.43 (tt, J ¼ 22.8, 6.3 Hz, 1H) ppm. 31P NMR (300 MHz, D2O),
d
18.08 (s, 2P) ppm. Elemental analysis: calc: C 32.22%, H 4.06%, S
Source T, ꢂC
10.75%; found C 31.89%, H 4.17%, S 10.60%.

454
D. Granchi et al. / European Journal of Medicinal Chemistry 65 (2013) 448e455
4.4. Typical experimental procedures for hydroxyapatite affinity
of supernatant of confluent SH-SY5Y cell line (human neuroblas-
measurements
toma, ATCC, CRL-2266) [36].
In a 20 mL capped vial, magnetically stirred at 25 ꢂC, approxi-
mately 0.07 mmol of BP and 100 mg of nano-powder HAP
(<200 nm, ALDRICH) were dissolved in 10 mL of Trizma buffer
solution (pH ¼ 7.5). Subsequently, 1 mL of suspension was taken-up
at 0 and 120 h. The samples were centrifuged (5000 rpm for 5 min)
and the solutions were transferred into 5 mm NMR tubes con-
taining a capillary filled with a solution of H3PO4 in D2O (external
standard) and analyzed via 31P NMR. Affinities were determined as
% of adsorbed BP by comparison with the integration of the
standard.
4.6.2. Inhibition of osteoclastogenesis
This assay allowed to highlight the activity of the new com-
pounds in inhibiting the differentiation of PBMC into osteoclasts.
For this purpose, three concentrations of drugs (10
mM, 0.1 mM,
0.001 M) were added together with to the osteoclast differentia-
m
tion medium. As positive control of the osteoclastogenesis cells
were maintained in osteoclast differentiation medium. After seven
days, the differentiation of PBMC into tartrate-resistant acid
phosphatase (TRAcP) positive multinucleated cells was evaluated.
The cells were fixed with 3% paraformaldehyde and 2% sucrose for
30 min, permeabilized with Triton 0.5% in HEPES for 5 min at room
temperature, and stained for TRAcP using naphthol AS-BI phos-
phoric acid and tartrate solution (Acid Phosphatase kit, Sigmae
Aldrich) for 60 min at 37 ꢂC. Sample and control cultures were
tested in duplicate. Data were expressed as percentage of osteo-
clasts counted after exposure to BPs related to the number of os-
teoclasts counted in positive control. The number of osteoclast
formed in positive control represented the maximum ability (100%)
to generate multinucleated and TRAcP positive cells in presence of
osteoclast differentiation medium.
4.5. Cytotoxicity screening
4.5.1. Cell cultures
The L929 cells (mouse fibroblasts, ATCC CCLl, NCTC clone 929)
were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM)
with Earle’s salts, containing 10% fetal calf serum (FCS), 20 mM
glutamine, 1% non-essential aminoacids penicillin (100 U/mL) and
streptomycin (100 m
g/mL) (SigmaeAldrich, Milan, Italy), at 37 ꢂC in
a humidified atmosphere of 5% CO2e95% air. Confluent cells were
detached by trypsin (SigmaeAldrich) and 0.8 ꢃ 103 cells were
seeded in flat bottomed 96 microplate wells in 0.2 mL of culture
medium (Costar, Cambridge, MA, USA). The cells were incubated for
24 h to obtain a complete cell adhesion. Then, the culture medium
was discarded and replaced with 0.2 mL of complete DMEM con-
4.6.3. Induction of osteoclast apoptosis
The assay tested the ability of the new compounds in inducing
the apoptosis of new formed osteoclasts, thus reducing the number
of active bone-resorbing cells. For this purpose, after seven days
from seeding, the osteoclast cultures were exposed to three con-
taining serial dilution of the tested compounds (from 100
mM to
0.001 M). A commercially available N-containing BPs was included
m
centrations of drugs (100 mM, 10 mM, 0.1 mM). After the next seven
as control of the experimental setting (pamidronate 3a, Sigmae
Aldrich). Stock solutions (10 mM) were obtained dissolving pow-
ders in phosphate buffer saline (PBS) at pH 7.4. As control, cultures
were added with 0.2 mL of DMEM (negative control).
days, the osteoclasts were fixed and permeabilized as described
above. Then they were incubated for 30 min with phalloidin-TRITC
0.3 mg/mL (SigmaeAldrich) and the nuclei were counterstained
with 2.25 mg/mL of Hoechst 33258 (SigmaeAldrich), for 30 min at
room temperature. The multinucleated cells exhibiting F-actin ring
and with apoptotic nuclear bodies were counted across the whole
well surface (0.81 cm2) (magnification of 40ꢃ). Sample and control
cultures were tested in duplicate. Data were expressed as per-
centage of apoptotic osteoclasts calculated on the total number of
multinucleated cells.
4.5.2. Neutral-red uptake assay
The plates were incubated for 72 h and at the end points, the
medium was aspirated from the monolayer and replaced with
0.2 mL per well of 0.4% neutral-red diluted 1:80 in DMEM. After
incubation for 2 h at 37 ꢂC the dye-containing medium was dis-
carded, the wells were washed twice with saline and added with
0.1 mL of extractant solution (50% ethanol, 1% acetic acid). Micro-
plates were shaken gently onto a microplate rotator for 2 min and
the absorbance of the extracted dye was detected on a spectro-
photometer microplate reader equipped with a 540 nm filter
(Spectra SLT, Tecan, Crailsheim, Germany). Results from triplicate
samples were recorded as optical density units (OD) and averaged
after blank subtraction. Cytotoxicity data were expressed as ratio of
the OD measured in sample exposed to compounds and OD
measured in negative control.
4.6.4. Inhibition of bone collagen degradation
The OsteoLyseÔ Bone Resorption Assay Kit (Lonza Walkersville,
Inc., Walkersville, USA) provides a quantitative assessment of
in vitro osteoclast-mediated degradation of human Type I bone
collagen by directly measuring the release of Europium-labeled
collagen fragments via time-resolved fluorescence. Briefly, mono-
nuclear cells were seeded onto the collagen-coated surface of a 96-
wells microplate, cultured in osteoclast differentiating medium,
and after 7 days exposed to three drug concentrations (100
mM,
10 M, 0.1 M). Cells cultured in the absence of differentiating
m
m
4.6. Evaluation of the biological activity on osteoclast cultures
medium served as “undifferentiated” controls. At the end of incu-
bation time (14 days), the cell culture supernatants were collected
4.6.1. Osteoclast isolation
and centrifuged, and 10 of the 200 mL were added to 200 mL of
Osteoclast cultures were obtained from peripheral blood of
healthy donors [35]. Mononuclear cells (PBMC) were isolated on
Ficoll-Hystopaque gradient (SigmaeAldrich), washed with PBS, re-
suspended in Dulbecco’s Modified Eagle’s Medium e High Glucose
(DMEM) (Euroclone, Milan, Italy) and supplemented with 10% fetal
bovine serum (FBS) (Euroclone), and plated in 4-well chamber
slides (0.81 cm2/well) at the density of 3 ꢃ 106 cells/cm2. PBMC
were incubated for 1 h at 37 ꢂC in a humidified atmosphere of 5%
CO2e95% air in order to obtain the monocyte adhesion. After 1 h,
non-adherent cells were removed and the osteoclast differentiation
medium was added, consisting on complete DMEM containing 25%
Fluorophore Releasing Reagent. The fluorescence of each well was
determined in a time-resolved fluorescence fluorimeter (Wallac
Victor, Waltham, MA) with excitation at 340 nm and emission at
615 nm. Data were expressed as percentage of inhibition of the
osteoclast resorption, i.e. the ratio between the Relative Fluores-
cence Units (RFU) released in culture medium of treated and un-
treated osteoclasts.
4.6.5. Statistical analysis
Statistical analysis was performed using StatView 5.01 for Win-
dows (SAS Institute Inc, Cary, NC). Quantitative data were expressed

D. Granchi et al. / European Journal of Medicinal Chemistry 65 (2013) 448e455
455
[9] D. Mukkamala, J. Hwan No, L.M. Cass, T.-K. Chang, E. Oldfield, J. Med. Chem. 51
(2008) 7827e7833.
[10] (a) A.J. Wiemer, D.F. Wiemer, R.J. Hohl, Clin. Pharm. Ther. 90 (2011) 804e
812;
as arithmetic mean plus or minus the standard error of the mean
(SEM). We hypothesized that the biological activity of compounds
could be influenced by multiple independent variables, i.e. the type
of compound and the drug concentration. The analysis of variance
(ANOVA) was applied to detect the effects of the independent var-
iables on the quantitative results, and the post hoc BonferronieDunn
test was applied to highlight the differences among the groups. A
scoring system was conceived for grading the performance of new
compounds in comparison to the control cultures. The calculation
was performed using a “MicrosoftÒ Excel” file, and a pivot table was
obtained to summarize the sum of the of score values obtained in
each experiment for each compound and concentration. The score
values were assigned according to the following criteria:
(b) Y. Zhang, R. Cao, F. Yin, M.P. Hudock, R.T. Guo, K. Krysiak, S. Mukherjee,
Y.G. Gao, H. Robinson, Y. Song, J.H. No, K. Bergan, A. Leon, L. Cass,
A. Goddard, T.K. Chang, F.Y. Lin, E. Van Beek, S. Papapoulos, A.H. Wang,
T. Kubo, M. Ochi, D. Mukkamala, E. Oldfield, J. Am. Chem. Soc. 131 (2009)
5153e5162.
[11] O. Bortolini, G. Fantin, M. Fogagnolo, S. Rossetti, L. Maiuolo, G. di Di Pompo,
S. Avnet, D. Granchi, Eur. J. Med. Chem. 52 (2012) 221e229.
[12] J.S. Sampalis, J.D. Adachi, E. Rampakakis, J. Vaillancourt, A. Karellis,
C. Kindundu, J. Bone Miner. Res. 27 (2011) 202e210.
[13] I.I. Imaz, P. Zegarra, J. Gonzalez-Enriquez, B. Rubio, R. Alcazar, J.M. Amate,
Osteoporos. Int. 21 (2010) 1943e1951.
[14] L. Widler, K.A. Jaeggi, M. Glatt, K. Mueller, R. Bachmann, M. Bisping, A.-R. Born,
R. Cortesi, G. Guiglia, H. Jeker, R. Klein, U. Ramseier, J. Schmid, G. Schreiber,
Y. Seltenmeyer, J.R. Green, J. Med. Chem. 45 (2002) 3721e3738.
[15] P.D. Papapetrou, Hormones (Athens) 8 (2009) 96e110.
[16] A. Bulpin, S. Masson, A. Sene, Tetrahedron Lett. 31 (1990) 1151e1154.
i) cytotoxicity:
2 ¼ ratio treated-to-untreated cells is ꢄ1;
[17]
b
-Thiosubstituted BP S.H. Szajnman, G.G. Linares, P. Moro, J.B. Rodriguez, Eur.
J. Org. Chem. (2005) 3687e3696;
-thiosubstituted BP W.M. Abdou, N.A.F. Ganoub, Y.O. El-Khoshnieh, Synlett
(2003) 785e790;
-thiosubstituted BP R.G. Kultyshev, J. Liu, S. Liu, W. Tjarks, A.H. Soloway,
S.G. Shore, J. Am. Chem. Soc. 124 (2002) 2614e2624.
[18] For an example of -thio BPs: (a) R.A. Nugent, S.T. Schlachter, M. Murphy,
1 ¼ ratio treated-to-untreated cells < 1 and ꢄ0.9;
0 ¼ ratio treated-to-untreated cells < 0.9 and ꢄ0.75;
ꢀ1 ¼ treated-to-untreated cells < 0.75 and ꢄ0.5;
ꢀ2 ¼ treated-to-untreated cells < 0.5 and ꢄ0;
ii) inhibition of the osteoclastogenesis:
g
d
g
C.J. Dunn, N.D. Staite, L.A. Galinet, S.K. Shields, H. Wu, D.G. Aspar, K.A. Richard,
J. Med. Chem. 37 (1994) 4449e4454. and for thiazole substituted BPs;
(b) W.M. Abdou, N.A. Ganoub, A. Geronikaki, E. Sabry, Eur. J. Med. Chem. 43
(2008) 1015e1024;
(c) A.A. Kamel, A. Geronikaki, W.M. Abdou, Eur. J. Med. Chem. 51 (2012) 239e
249. for cleavable d-thioester BP;
(d) K.S.E. Tanaka, T.J. Houghton, T. Kang, E. Dietrich, D. Delorme, S.S. Ferreira,
L. Caron, F. Viens, F.F. Arhin, I. Sarmiento, D. Lehoux, I. Fadhil, K. Laquerre,
J. Liu, V. Ostiguy, H. Poirier, G. Moeck, T.R. Parr Jr., A. Rafai Far, Bioorg. Med.
Chem. 16 (2008) 9217e9229.
ꢀ2 ¼ ratio treated-to-untreated cells is ꢄ100%;
ꢀ1 ¼ ratio treated-to-untreated cells < 100% and ꢄ90%;
0 ¼ ratio treated-to-untreated cells < 90% and ꢄ75%;
1 ¼ treated-to-untreated cells < 75% and ꢄ50%;
2 ¼ treated-to-untreated cells < 50% and ꢄ25%;
3 ¼ treated-to-untreated cells < 25% and ꢄ0%;
iii) induction of apoptosis:
3 ¼ percentage of apoptotic osteoclasts is ꢄ90%;
2 ¼ percentage of apoptotic osteoclasts is <90% and ꢄ75%;
1 ¼ percentage of apoptotic osteoclasts is <75% and ꢄ50%;
0 ¼ percentage of apoptotic osteoclasts is <50% and ꢄ25%;
ꢀ1 ¼ percentage of apoptotic osteoclasts is <25% and ꢄ0%;
iv) inhibition of the osteoclast-mediated bone resorption:
ꢀ1 ¼ ratio treated-to-untreated cells is ꢄ1;
[19] L. Kamm, W. McIlwraith, C. Kawcak, J. Equine Vet. Sci. 28 (2008) 209e214.
[20] Example of a
a and b-thio substituted BPs M.T. Rubino, M. Agamennone,
C. Campestre, P. Campiglia, V. Cremasco, R. Faccio, A. Laghezza, F. Loiodice,
D. Maggi, E. Panza, A. Rossello, P. Tortorella, ChemMedChem 6 (2011) 1258e1268.
[21] R.A. Nugent, M.Murphy, S.T. Schlachter, C.J. Dunn, R.J. Smith, N.D. Staite,
L.A. Galinet, S.K. Shields, D.G. Aspar, K.A. Richard, N.A. Rohloff, J. Med. Chem.
36 (1993) 134e139.
[22] D.W. Hutchinson, D.M. Thornton, J. Organomet. Chem. 346 (1988) 341e348.
[23] S. Ghosh, J.M.W. Chan, C.R. Lea, G.A. Meints, J.C. Lewis, Z.S. Tovian,
R.M. Flessner, T.C. Loftus, I. Bruchhaus, H. Kendrick, S.L. Croft, R.G. Kemp,
S. Kobayashi, T. Nozaki, E. Oldfield, J. Med. Chem. 47 (2004) 175e187.
[24] M.B. Martin, J.S. Grimley, J.C. Lewis, H.T. Heath III, B.N. Bailey, H. Kendrick,
V. Yardley, A. Caldera, R. Lira, J.A. Urbina, S.N.J. Moreno, R. Docampo, S.L. Croft,
E. Oldfield, J. Med. Chem. 44 (2001) 909e916.
0 ¼ ratio treated-to-untreated cells < 1 and ꢄ0.9;
1 ¼ ratio treated-to-untreated cells < 0.9 and ꢄ0.75;
2 ¼ treated-to-untreated cells < 0.75 and ꢄ0.5;
3 ¼ treated-to-untreated cells < 0.5 and ꢄ0;
[25] D. Simoni, N. Gebia, F.P. Invidiata, M. Eleopra, P. Marchetti, R. Rondanin,
R. Baruchello, S. Provera, C. Marchioro, M. Tolomeo, L. Marinelli, V. Limongelli,
E. Noveliono, A. Kwaasi, J. Dunford, S. Buccheri, N. Cacamo, F. Dieli, J. Med.
Chem. 51 (2008) 6800e6807.
Acknowledgments
[26] C.E. Mckenna, M.T. Higa, N.H. Cheung, M.C. Mckenna, Tetrahedron Lett. 2
(1977) 155e158.
The authors acknowledge MIUR for funding PRIN project 2008.
L. Sperni is gratefully acknowledged for GCeMS analysis.
[27] (a) S. Lenevich, M.D. Distefano, Anal. Biochem. 408 (2011) 316e320;
(b) S. Mukherjee, Y. Song, E. Oldfield, J. Am. Chem. Soc. 130 (2008) 1264e1273;
(c) S. Josse, C. Faucheux, A. Soueidan, G. Grimandi, D. Massiot, B. Alonso,
P. Janvier, S. Laïba, P. Pilet, O. Gauthier, G. Daculsi, J. Guicheux, B. Bujolia, J.-
M. Bouler, Biomaterials 26 (2005) 2073e2080;
References
(d) W. Jahnke, C. Henry, ChemMedChem 5 (2010) 770e776.
[28] V. Breuil, F. Cosman, L. Stein, W. Horbert, J. Nieves, V. Shen, R. Lindsay,
D.W. Dempster, J. Bone Miner. Res. 13 (1998) 1721e1729.
[29] D. Merlotti, L. Gennari, G. Martini, F. Valleggi, V. De Paola, A. Avanzati, R. Nuti,
J. Bone Miner. Res. 22 (2007) 1510e1517.
[30] B. Frediani, A. Spreafico, C. Capperucci, F. Chellini, D. Gambera, P. Ferrata,
F. Baldi, P. Falsetti, A. Santucci, L. Bocchi, R. Marcolongo, Bone 35 (2004)
859e869.
[31] R.G.G. Russell, N.B. Watts, F.H. Ebetino, M.J. Rogers, Osteoporos. Int. 19 (2008)
733e759.
[32] K. Thompson, M.J. Rogers, F.P. Coxon, J.C. Crockett, Mol. Pharmacol. 69 (2006)
1624e1632.
[33] P. David, H. Nguyen, A. Barbier, R. Baron, J. Bone Min. Res. 11 (1996) 1498e
1507.
[34] J. Mao, S. Mukherjee, Y. Zhang, R. Cao, J.M. Sanders, Y. Song, Y. Zhang,
G.A. Meints, Y. Gui Gao, D. Mukkamala, M.P. Hudock, E. Oldfield, J. Am. Chem.
Soc. 128 (2006) 14485e14497.
[1] A. Corrado, F.P. Cantatore, Reumatismo 57 (2005) 142e153.
[2] S. Zhang, G. Gangal, H. Uluda&gbreve, Chem. Soc. Rev. 36 (2007) 507e531.
[3] R.G.G. Russell, Bone 49 (2011) 2e19.
[4] F.H. Ebetino, A.-M.L. Hogan, S. Sun, M.K. Tsoumpra, X. Duan, J.T. Triffitt,
A.A. Kwaasi, J.E. Dunford, B.L. Barnett, U. Oppermann, M.W. Lundy, A. Boyde,
B.A. Kashemirov, C.E. McKenna, R.G.G. Russell, Bone 49 (2011) 20e33.
[5] M.J. Rogers, J.C. Crockett, F.P. Coxon, J. Mönkkönen, Bone 49 (2011) 34e41.
[6] (a) B.J. Gertz, S.H. Holland, W.F. Kline, B.K. Matuszewski, A. Freeman, H. Quan,
K.C. Lasseter, J.C. Mucklow, A.G. Porras, Clin. Pharmacol. Ther. 58 (1995) 288e
298. for an example of lipophilic prodrug containing the BP moiety that more
readily enter cells see;
(b) L.W. Shull, A.J. Wiemer, R.J. Hohl, D.F. Wiemer, Bioorg. Med. Chem. 14
(2006) 4130e4136.
[7] (a) J.M. Rondeau, F. Bitsch, E. Bourgier, M. Geiser, R. Hemming, M. Kroemer,
S. Lehmann, P. Ramage, S. Reiffel, A. Strauss, J.R. Green, W. Jahnke, Chem-
MedChem 1 (2006) 267e273;
(b) K.L. Kavanagh, K. Guo, J.E. Dunford, X. Wu, S. Knap, F.H. Ebetino,
M.J. Rogers, R.G. Russell, U. Oppermann, Proc. Natl. Acad. Sci. U.S.A. 103 (2006)
7829e7834.
[35] D. Granchi, I. Amato, L. Battistelli, G. Ciapetti, S. Pagani, S. Avnet, N. Baldini,
A. Giunti, Biomaterials 26 (2005) 2371e2379.
[36] D. Granchi, I. Amato, L. Battistelli, S. Avnet, S. Capaccioli, L. Papucci,
M. Donnini, A. Pellacani, M.L. Brandi, A. Giunti, N. Baldini, Int. J. Cancer 111
(2004) 829e838.
[8] J.H. No, F. de Marcedo Dossin, Y. Zhang, Y.-L. Liu, W. Zhu, X. FEng, J. Anny Yoo,
E. Lee, K. Wang, R. Hui, L.H. Freitas-Junior, E. Oldfield, Proc. Natl. Acad. Sci.
U.S.A. 109 (2012) 4058e4063.