Synthesis of new indole-based bisphosphonates and evaluation of their chelating ability in PE/CA-PJ15 cells

Article abstract of DOI:10.1016/j.ejmech.2015.08.019

Bisphosphonates are the most important class of antiresorptive agents used against osteoclast-mediated bone loss, and, more recently, in oncology. These compounds have high affinity for calcium ions (Ca²⁺) and therefore target bone mineral, whe

Full text of DOI:10.1016/j.ejmech.2015.08.019

European Journal of Medicinal Chemistry 102 (2015) 403e412
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Synthesis of new indole-based bisphosphonates and evaluation of
their chelating ability in PE/CA-PJ15 cells
Carlo A. Palmerini a, Francesco Tartacca a, Michela Mazzoni a, Letizia Granieri a,
,
*
Laura Goracci b, Angela Scrascia b, Susan Lepri b
a Department of Agricultural, Food and Environment Science (Unit Research of Biochemistry and Molecular Biology), University of Perugia, via Del
Giochetto, 06125 Perugia, Italy
b Department of Chemistry, Biology and Biotechnology, University of Perugia, via Elce di Sotto, 8, 06123 Perugia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Bisphosphonates are the most important class of antiresorptive agents used against osteoclast-mediated
bone loss, and, more recently, in oncology. These compounds have high affinity for calcium ions (Ca2þ
)
Received 1 April 2015
Received in revised form
22 July 2015
Accepted 8 August 2015
Available online 12 August 2015
and therefore target bone mineral, where they appear to be internalized selectively by bone-resorbing
osteoclasts and inhibit osteoclast function. They are extensively used in healthcare, however they are
affected by severe side effects; pharmacological properties of bisphosphonates depend on their molec-
ular structure, which is frequently the cause of poor intestinal adsorption and low distribution. In this
work we synthesized six novel bisphosphonate compounds having a variably substituted indole moiety
to evaluate their extra- and intracellular calcium chelating ability in PE/CA-PJ15 cells. Preliminary in silico
and in vitro ADME studies were also performed and the results suggested that the indole moiety plays an
important role in cell permeability and metabolism properties.
Keywords:
Indole based bisphosphonate
Cytosolic calcium
Ene reaction
© 2015 Elsevier Masson SAS. All rights reserved.
Chelating ability assessment
1. Introduction
may alter secretion, contraction, protein activity, apoptosis, prolif-
eration and other critical processes [8].
Bisphosphonates (BPs) are analogs of naturally occurring pyro-
phosphates and represent an important class of drugs, used to treat
conditions such as osteoporosis, Paget's syndrome and some tumors
[1,2]. Similarly to inorganic pyrophosphate, BPs antagonize the
dissolution of hydroxyapatite crystals and the metastatic calcifica-
tion of tissues, however unlike their endogenous analogs they are
not cleaved by pyrophosphatase [3,4]. The biological activity of BPs
largely depends on the Ca2þ-chelating properties of the two phos-
phate groups. As it is widely known, calcium is used in cellular
signaling: in eukaryotic cells the ion is sequestered in cellular or-
ganelles (such as endoplasmic reticulum and mitochondria) and it is
released according to the cell activation [5]. Cytoplasmic Ca2þ con-
centration [Ca2þ]c is very low (z100 nM) compared to that in the
extracellular medium (z1 mM). The maintenance of such a calcium
concentration gradient between the cytosol and other compart-
ments is crucial for cells survival [6]. Calcium homeostasis is finely
tuned through various mechanisms [7]; an uncontrolled Ca2þ signal
BPs can be distinguished in two classes: the non-nitrogen-
containing bisphosphonates (NNBPs) and the nitrogen-containing
bisphosphonates (NBPs) [9]. Etidronate (ETD) and clodronate
(CLD) (Fig. 1) belong to the NNBPs family, and they are known to
block ATP-dependent enzymes able to induce the death of the
osteoclast [10]. Alendronate (ALN), risedronate (RSD), and
pamidronate (PMD) (Fig. 1) belong to the NBPs class, and are in-
hibitors of the farnesyl pyrophosphate synthase [2]. These drugs
inhibit the prenylation of GTPase-dependent proteins (Ras, Rab,
Rho and Rac) that are essential for the osteoclast's survival. The
deterioration of cytoskeleton of these cells does not allow the for-
mation of the ruffled border, causing their apoptosis [11,12].
Survival time, proliferation, adhesion and migration are the
main effects of NBPs on tumor cell lines “in vitro” [13,14] and most
of these effects are due to the inhibition of farnesyl pyrophosphate
synthetase [15]. In addition, the inhibition of protein prenylation
may alter the viability and the proliferation of endothelial cells
in vitro [16]. Studies on mice indicate that NBPs inhibit the forma-
tion of bone metastases and decrease bone mass in vivo [16,17],
suggesting a relationship between their antitumor effects and their
cellular uptake in vivo. NBPs induce a pre-apoptotic action on
* Corresponding author.
E-mail address: susan@chemiome.chm.unipg.it (S. Lepri).
http://dx.doi.org/10.1016/j.ejmech.2015.08.019
0223-5234/© 2015 Elsevier Masson SAS. All rights reserved.

404
C.A. Palmerini et al. / European Journal of Medicinal Chemistry 102 (2015) 403e412
Fig. 1. Chemical structure of commercially available bisphosphonate drugs in their sodium salt form (clodronate, etidronate, alendronate, risedronate, and pamidronate).
neoplastic cancer cells and inhibit the mechanisms regulating the
cellular proliferation and adhesion [14,15].
(IBPs) (compounds 1aef, Fig. 2) were synthesized and investigated
for their ability in chelating extracellular and cytosolic calcium in
the PE/CA P-J15 cells as a function of both the nature and the po-
sition of the substituent in the indole nucleus. In particular, owing
to the interesting role played by fluorine in bioactive molecules
[29e31], four fluorinated compounds were investigated. The PE/
CA-PJ15 cells are obtained from human oral squamous cancer cells;
they are similar to epithelial cells and express laminin and cyto-
keratins [32] and are responsible for about 90% of head and neck
tumors. Our results showed that two out of six indole-
bisphosphonates were significantly more effective than the
commercially available ALN and CLD in chelating Ca2þ. The in silico
evaluation of the passive cell membrane permeation ability sug-
gested that the Ca2þ-binding efficacy of these compounds may be
related to an increased cell permeability. Finally, the metabolic
stability in human liver microsomes was determined for the most
promising compounds, which, conversely to the ALN and CLD
compounds, were rapidly metabolized.
The major drawback of NBP drugs is their high hydrophilicity
and polarity. Thus, many common NBP drugs demonstrate poor
cell-membrane permeability towards tumor cells and other non-
endocytic cells [18,19]. The uptake and binding to bone mineral is
determined by the presence of the bisphosphonates moiety, while
the substituents bound to the PeCeP moiety are thought to be
essential to ensure satisfactory antiresorptive potency. Thus, small
changes in the substituents of the PeCeP moiety may influence the
resorption properties [20].
Another consequence of the high hydrophilicity of NBPs is their
metabolic stability. Indeed, they are not metabolized by CYP450 or
other phase I enzymes [21], and are mostly eliminated unchanged
in urine [20]. On one hand, an extensive metabolism of a drug
might jeopardize its pharmacological activity; on the other hand,
when drugs are substrates for CYP450 enzymes, their half-life, as
well as the pharmacological and toxicity effects, can be modulated
by synthetic modifications.
As mentioned above, residual bisphosphonates which are not
taken up by the bones are excreted in urine without being unme-
tabolized. As a consequence, a common adverse event related to
bisphosphonate drugs is nephrotoxicity. In fact, pharmacokinetic
and pharmacodynamic properties (see Table 1 for pKa, LogP, and
oral bioavailability data), modulate the bisphosphonates accumu-
lation which is related to renal histopathology. Even though a low
dose is usually well tolerated, the high dose required in an oncology
setting is often correlated to nephrotoxicity [22]. In particular, the
major toxicity effects seem correlated to intravenous administra-
tion of BPs. In addition, orally administered BPs are often correlated
with adverse effects on gastrointestinal tract, such as ulceration in
esophagus, stomach and duodenum. Other adverse events such as
acute-phase reactions, and the osteonecrosis of the jaw have also
been observed [23]. Thus, the improvement of the pharmacokinetic
properties of bisphosphonates retaining the desired biological ef-
fect represents a valuable strategy to obtain novel series of safer
bisphosphonate drugs.
2. Results and discussion
2.1. Synthesis of indole-based bisphosphonates
The indole nucleus is a well-established pharmacophore present
in many synthetic and natural drugs with remarkable biological
activities [10] and indole containing bisphosphonates have already
been assessed for their anti-bone resorptive activity in bone
marrow osteoclast culture [33].
As previously reported, 3-methyleneindolines are valuable
building blocks for 3-functionalized indole derivatives due to the
remarkable reactivity of the exocyclic carbonecarbon double bond
through the “ene” reaction [34,35]. These findings led us to
consider their use for the synthesis of new regioselectively
substituted
tetraethyl
[3-(1H-indol-3-yl)propane-1,1-diyl]
bis(phosphonate) derivatives 1aef with Ca2þ chelating ability and
improved cell permeability (Fig. 2). The N-Boc protected 3-
methyleneindolines 5aef were prepared in four steps, starting
from commercially available substituted anilines (Scheme 1). When
not commercially available, o-bromoanilides 3 were obtained by
metalation of the N-Boc-protected anilines 2, followed by treat-
ment of the resulting dilithiated anilide intermediate with CBr4.
NaH-promoted deprotonation of anilide 3 in DMF, followed by the
reaction with propargyl bromide, gave the corresponding N-prop-
argyl derivatives 4 in good yield. Finally, the reductive radical
cyclization of N-bromoaryl-N-propargylcarbamate with tribu-
thyltin hydride in refluxing benzene, in the presence of AIBN
initiator, provided the expected 3-methyleneindoline 5 in satis-
factory yields. 3-Methyleneindolines 5 are relatively stable at room
temperature and can be purified by standard chromatographic
Thus, in this work six novel indole-based bisphosphonates
Table 1
Experimental pKa, LogP and oral bioavailability data of commercial BPs.
a
Comp. pKa1e5
LogPb [24] Oral bioavailability %b [25]
CLD
ETD
ALN
RSD
PMD
1.7, 2.1, 5.7, 8.3 [25]
1.7, 2.5, 7.2, 10.8 [26]
ꢀ2.4
ꢀ3.8
1e2
3e7
0.75
0.65
0.3
1.3, 2.2, 6.4, 11.0, 11.8 [27] ꢀ4.3
1.6, 2.2, 5.9, 7.1, 11.7 [28]
ꢀ3.6
1.2, 1.9, 6.0, 10.2, 12.1 [27] ꢀ4.7
a
Related to respective acid form.
Referred to commercial salt form.
b

C.A. Palmerini et al. / European Journal of Medicinal Chemistry 102 (2015) 403e412
405
PO(OEt)2
PO(OEt)2
PO(OEt)2
PO(OEt)2
PO(OEt)2
PO(OEt)2
F
F3C
N
N
N
H
H
H
1a
1b
1c
PO(OEt)2
PO(OEt)2
PO(OEt)2
PO(OEt)2
PO(OEt)2
PO(OEt)2
N
N
N
N
H
F3C
F
H
H
1d
1e
1f
Fig. 2. Chemical structure of novel indole-based bisphosphonates (IBPs, 1aef).
Scheme 1. Synthesis of regiosubstituted 3-methyleneindolines.
methods without any isomerization to the corresponding 3-
methylindoles.
Table 2
Predicted pKa and LogP for compounds 1aef.
Owing to the presence of two powerful electron-withdrawing
phosphoryl groups, alkyl vinylidene gem-bisphosphonates (VBP)
[36] are very electrophilic species. They have been employed in the
synthesis of several alkyl bisphosphonates as Michael-type nucle-
ophile acceptors [37] as well as in Lewis acid-catalyzed Friedele-
Crafts reactions with activated aromatics [38e40] and seldom as
dienophile in DielseAlder cycloaddition. However, to our knowl-
edge, VBP has never been used as an enophile in ene-type reactions.
Indeed, according to NMR spectra (see Supporting Information),
indolylalkylbisphosphonate (IBPs) 1aef can be easily obtained in
50e70% yield by simple heating an equimolar mixture of the
suitable Boc-protected 3-methyleneindoline and VBP [41] at 100 ꢁC
for 2e4 h in solvent-free conditions (Scheme 2). The protecting
tert-butoxycarbonyl group is easily removed by simply rising the
temperature to 150e190 ꢁC until complete nitrogen deprotection to
give final compounds 1aef (Scheme 2). The structures of the
bisphosphonates 5aef and 1aef were confirmed by spectroscopic
methods. In particular, the presence of the PeCHeP moiety was
a
Comp.
pKa
LogPb
1a
1b
1c
1d
1e
1f
>15
>15
>15
>15
>15
2.9
3.1
3.8
3.8
3.1
1.8
7.96
a
Predicted using MoKa 2.6 [46].
Predicted using VolSurfþ [47].
b
J z 130 Hz) in the 13C NMR spectra. The molecular formulas were
obtained by ESI HRMS, and all compounds were fully characterized
by 1H, 13C, 19F and 31P NMR spectroscopy (See Supplementary In-
formation). As an example, 2D NMR spectra and NMR assignments
for compound 1a are reported in Supplementary Information.
The predicted pKa and LogP for the synthesized compounds
1aef are reported in Table 2. It is immediately evident that the
synthesized compounds possess very different physicochemical
supported by the presence of a typical triplet peak (d z 35 ppm,
Scheme 2. aSynthesis of regiosubstituted substituted IBPs by “ene” reactions.

406
C.A. Palmerini et al. / European Journal of Medicinal Chemistry 102 (2015) 403e412
properties with respect to commercial BPs. Indeed, they lack the
acid features, although compounds 1aef could act as prodrugs with
subsequent ester cleavage. Other attempts to synthesized BPs
prodrugs have been proposed [25] suggesting that several groups
(such as carboxylic esters, amides, anhydrides) can be used to
functionalize the acid phosphate in order to improve the com-
pound's oral bioavailability. The presence of the indole scaffold and
the esterification of the bisphosphonate moiety also results in a
significant increase of the LogP value. For compound 1f, a basic pKa
of 7.96 is also predicted for the azaindole scaffold.
According to literature protocols [40], few attempts to obtain the
free bisphosphonic acid derivatives have been made to experi-
mentally evaluate the effect of the ester group (e.g. Scheme SI1).
However, under this condition, the expected acid was obtained as
minority product (14%), while the 2,2-coupled dihydrodimer was
observed (see Supporting Information for further details). Such
couplings are known for simple indole, skatole, and other 3-
substituted indoles under acidic condition [42e45].
Fig. 3. Ca2þ/FURA complex fluorescence drop due to the addition of 1ae1f, ALN, and
CLD (30 M). Data are the average of 10 experiments SEM.
m
2.2.2. BPs chelating ability of cytosolic calcium ([Ca2þ]c) in PE/CA-
PJ15 cells in Ca2þ-free medium
2.2. Biological evaluation of indole-based bisphosphonates
The cytosolic Ca2þ ([Ca2þ]c) in the PE/CA-PJ15 cells was deter-
mined with the FURA-2 AM method [49]. The chelating ability of
The Ca2þ chelating ability of indole-based bisphosphonates 1aef
was tested using PE/CA-PJ15 cells. This cell line is obtained from
squamous cell tongue carcinoma, which accounts for about 90% of
head and neck tumors. These cells, similar to epithelial cells, ex-
press laminin and cytokeratins [32,48]. In this study, we investi-
[Ca2þ
] was assessed with 10 and 30 m
c
M BPs, in Ca2þ free HBSS
buffer. The BPs' chelating effectiveness was evident after the first
addition of BPs in the incubation medium and was confirmed by a
second addition made 150 s later (e.g. Fig. SI2).
gated the impact of synthesized compounds 1aef on [Ca2þ
]
c
The BPs ability to chelate cytosolic calcium ion was expressed as
homeostasis in the oral cancer PE/CA-PJ15 cell model to better
characterize a biochemical/biological link that may have implica-
tions in cell signaling events associated with cell viability and
apoptosis in human oral squamous carcinoma.
D
[Ca2þ]c (nM) measured after 50 s from the addition (Fig. 4). The
D
2.2.1. Bisphosphonates (BPs) chelating ability of extracellular
calcium ion
The chelating ability of each novel IBP toward extracellular
calcium ions was evaluated using the fluorescent indicator Fura-2,
and commercial ALN and CLD were used as reference compounds.
Bisphosphonates 1aef and commercial ALN and CLD were added to
a suspension of PE/CA-PJ15 cells in HBSS (Hanks' Balanced Salt
Solution, 1 mM CaCl2) buffer previously lysed with 1% Triton X-100.
The variation of the Ca2þ/FURA complex fluorescence was moni-
tored over time (a representative example for compound 1a is re-
ported in Supplementary Information, Fig. SI1). The fluorescence
value after cells lysis was set at 100%. A fluorescence decrease was
observed upon the addition of indole 1a (10e30 mM), with a con-
centration effect: the higher the concentration, the more consistent
fluorescence decrease.
Each BP exhibited a different chelating ability assessed by the
fluorescence drop upon addition of the compound at 30 mM con-
centration (Fig. 3). The unsubstituted compound 1a proved to be
the most active chelating compound with a fluorescence drop of
73 2.4%, followed by fluorinated compound 1b (34 2.6% cut).
Other BPs (1c, 1e, and CLD) reduced the fluorescence of Ca2þ/FURA
complex of about 10% whilst ALN, 1d, and 1f were the less effective
chelating species causing fluorescence decrease of only 5% (Fig. 3).
Thus, slight modifications to the indole moiety substituents
seem to affect the chelating properties substantially. Indeed,
although being considered bioisosteric, the replacement of one
hydrogen with a fluorine in position 5 or 6, afforded two less
effective compounds (1b and 1e). Similarly, the substituent tri-
fluoromethyl (products 1ced) caused a further loss in activity.
Moreover, substitution of the indole scaffold with a pyrrolepyridine
ring to give compound 1f did not dramatically change the
bisphosphonate activity with respect to ALN.
Fig. 4. BPs chelating ability of [Ca2þ
]
in the PE/CA-PJ15 cells in Ca2þ free medium is
c
expressed as decrease or increase of [Ca2þ
average of 10 experiments SEM.
] (D
[Ca2þ]c, nM). Data are expressed as the
c

C.A. Palmerini et al. / European Journal of Medicinal Chemistry 102 (2015) 403e412
407
[Ca2þ]c induced by 1a was greater than that induced by 1b, and by
CLD > 1d > 1e in the 10 and 30
mM BPs concentration range (Fig. 5).
far superior to those induced by 1ce1e and commercial ALN and
CLD. On the contrary, azaindole 1f did not exhibit any chelating
The diagrams in Figs. 4 and 5 clearly suggest that BPs (especially 1a
and 1b) were able to chelate [Ca2þ]c, even at lower concentrations of
activity, but induced a significant enhancement of the [Ca2þ
]
ligand (10 m
M) and in the presence of Ca2þ in the medium (see
c
levels. Thus, azacompound 1f has a unique agonistic behavior, with
respect to other IBPs, and it seems to favor the release of the ion
from internal stores. Further studies are currently ongoing to
examine in depth this surprising result, which to our knowledge
has never been reported before.
Table 3 for a summary).
Even in these experimental conditions, the aza-analog 1f shows
a peculiar behavior. Indeed, it did not show any chelating activity,
and on the contrary it raised the [Ca2þ]c levels in a dose dependent
way. The effect was even more evident than that observed in the
absence of Ca2þ (Figs. 4 and 5). Therefore, 1f acted as an agonist,
probably by favoring the release of the ion from internal stores, also
in absence of Ca2þ in the incubation medium. This finding might be
correlated with the different nature of the indole scaffold, whereas
the presence of a basic aza-group (pKa ¼ 7.96, predicted by MoKa
[46]) provides a protonated species under physiological condition
which could mimic the second messenger cADPR, by acting on the
ryanodine receptors and behave as a pro-apoptotic agent.
Changes of the total [Ca2þ]c induced by consecutive additions of
IBPs 1ae1f and commercial BPs to reach a final concentration of 10
and then 30
mM were not linear. The non-linear fluorescence
response at those concentration may be attributable to the different
dissociation constants of the two competing complexes Fura-Ca2þ
and BP-Ca2þ. In addition, a saturating effect of BPs at the nanomolar
levels of [Ca2þ
concentrations less than 10
measurable by this method.
] may be present. However, bisphosphonates, at
c
m ] not
M, lead to changes in [Ca2þ
c
2.3. Preliminary ADME studies
2.2.3. BPs chelating ability in the PE/CA-PJ15 cells in Ca2þ
containing medium
-
Finally, preliminary studies on the ADME properties of the BPs
were performed. In particular, we evaluated the intestinal perme-
ability in Caco-2 cells for all the tested compounds in silico, using
the VolSurfþ software [47]. Fig. 6 shows the projection of the BPs
compounds on the VolSurfþ Caco-2 model [50] for permeability
prediction, generated using 750 experimental values of perme-
ability in Caco-2 cells. The points that represent the experimental
permeability values in the VolSurfþ model are colored in a blue-to-
red scale, where blue and red correspond to high and low perme-
ability, respectively. The projected BPs compounds are reported as
yellow points.
In order to assess whether the presence of extracellular Ca2þ (as
in physiological condition) affects the activity of intracellular BPs,
all IBPs were further tested in presence of 1.0 mM Ca2þ in the in-
cubation medium. Results are summarized in Fig. 5 where D ]
values (nM) reported for all tested compounds were measured after
50 s from the addition.
[Ca2þ
c
Once again, indole 1a was the most active compound in seizing
[Ca2þ]c, the effectiveness following the order: 1a > 1b > ALN > 1c >
The BPs compounds show a good predicted cell permeability,
compared to the commercial reference compounds. In addition,
substituents in the indole nucleus seem to have only a slight effect
on permeability. Among the synthesized compounds, only azain-
dole 1f seems to demonstrate a slightly lower permeability. This
effect is probably due to the presence of a basic center in the 1H-
pyrrolo[3,2-c]pyridine ring (pKa ¼ 7.96 according to MoKa [46]),
which is positively charged at physiological pH. This in silico pre-
diction not only makes BPs promising compounds for further
studies, but also suggests that the significant chelating ability
detected in the cytosol might be related to a better permeability of
the BPs, although the cell lines used for in silico and experimental
studies are different.
Compounds 1a and 1b resulted to be the most promising
compounds for their chelating ability, thus their metabolic stability
was determined upon incubation in human liver microsomes for
30 min (a description of the assay is reported in the Experimental
Section). Compounds 1a and 1b were rapidly metabolized, with the
parent compounds being reduced to about 20% at the end of the
incubation (See Figs. SI3 and SI4 in Supporting Information). Me-
tabolites were identified by LC/MS analysis and a similar metabolic
pathway was observed for both compounds. Indeed, compounds 1a
and 1b underwent mono-aliphatic hydroxylation as proved by the
presence of fragment 156.0808 for metabolite of compound 1a
(1aþ16) and the presence of fragment 174.0714 for metabolite of
compound 1b (1bþ16), (Fig. 7). Traces of the di-hydroxylated
(1aþ32) form were also detected for compound 1a, suggesting
that the attack occurs always at the aliphatic portion of the com-
pound for the presence of fragment 130.0561 (Fig. 7).
3. Conclusions
Ca2þ is an important second messenger participating in many
cellular activity. Several studies have shown that signaling
Fig. 5.
of 1 mM [Ca2þ] in the medium. Data are the average of 10 experiments SEM.
D ] (nM) induced by different BPs in the PE/CA-PJ15 cells in the presence
[Ca2þ
c

408
C.A. Palmerini et al. / European Journal of Medicinal Chemistry 102 (2015) 403e412
Table 3
Summary table of BPs effects on cellular Ca2þ in different conditions.
Comp.
% Fluorescence in Extracellular Ca2þ-assaya
D
[Ca2þ
]
c (Ca2þ-free medium), nMb
D ]
[Ca2þ c (1 mM Ca2þ in the medium), nMb
1a
1b
1c
1d
1e
1f
ALN
CLD
ꢀ73.0 2.4
ꢀ34.0 2.6
ꢀ16.5 1.5
ꢀ5.0 0.5
ꢀ18.0 1.2
ꢀ6.0 1.0
ꢀ7.0 0.6
ꢀ12.0 1.2
ꢀ27.0 2.8
ꢀ14.0 2.8
ꢀ8.6 0.9
ꢀ8.0 1.0
ꢀ8.0 1.9
þ36.0 3.2
ꢀ7.5 1.2
ꢀ7.6 1.4
ꢀ44.0 5.0
ꢀ33.0 4.0
ꢀ11.0 1.4
ꢀ7.3 0.5
ꢀ4.6 0.5
þ67.0 3.0
ꢀ20.0 3.2
ꢀ7.0 0.9
a
Values obtained after the addition of 30 mM BPs.
Values obtained after the total addition of 60 mM BPs. Data are the average of 10 experiments SEM.
b
homeostasis [51,52]. When physiochemical events deregulate Ca2þ
homeostasis, Ca2þ acts an intrinsic stressor that may trigger cell
damage [53]. The rationale of this study was to investigate the ef-
fects of commercial and synthetic bisphosphonates on extracellular
and cytosolic calcium levels in human oral squamous cancer cells
(PE/CA PJ-15) used as a model. Both commercial (ALN, CLD) and
synthetic bisphosphonates 1aef tested in this work are able to bind
extracellular Ca2þ but to a different extent.
The presence of the unsubstituted indole nucleus tied down to a
gem-diphosphonate carbon (1a), significantly enhances the Ca2þ
effect compared to the other tested compounds (ALN, CLD, and IBPs
1be1e). Chelating ability was also confirmed in the presence of
extracellular Ca2þ (as in physiological conditions), where BPs
showed greater activity, although not in a dose dependent manner.
Moreover the presence of fluorine in different positions on the
indole showed a detrimental effect on the chelating property.
Preliminary ADME studies of IBPs suggested that the cell mem-
brane permeability has a pivotal role in cytosolic calcium binding
ability with respect to commercial compounds. In addition, the
presence of the indole scaffold makes these IBPs substrates for P450
cytochromes. The synthesis of a new class of chelating agents
having the indole scaffold linked to the alkylbisphosphonate moi-
ety may find a use in therapeutic doses lower than those of com-
mercial BPs, in all those pathologies caused by a Ca2þ deregulated
homeostasis. The long term objective of the research will be to
assess the role of a new class of bisphosphonates able to modulate
[Ca2þ]c altered during oxidative stress.
Fig. 6. Projection of the BPs in the PLS model for permeability prediction available in
the VolSurfþ model (t1/t2 plot).
molecules such as Ca2þ are deregulated in cancer; moreover pro-
apoptotic agents (as thapsigargin, staurosporine, Ca2þ ionophores,
stress oxidative) induce cell death by interfering with Ca2þ
4. Experimental section
4.1. Chemistry
1a+16
1a+32
1H NMR, 13C NMR {1H} (acquired using J-modulated (J-MOD)
spineecho experiments), 19F NMR {1H}, and 31P NMR {1H} were
recorded on Bruker Advance II 400 MHz spectrometer at room
temperature with tetramethylsilane, 85% H3PO4 and trichloro-
N
N
H
H
fluoromethane as internal standard. Chemical shifts (d) are re-
Chemical Formula: C11H10N•
Exact Mass: 156,0813
Chemical Formula: C9H8N•
Exact Mass: 130,0657
ported in parts per million (ppm), and peak multiplicity are
reported as s (singlet), d (doublet), t (triplet), q (quartet), p (pentet),
hept (heptet), m (multiplet), or br s (broad singlet). HRMS spectra
were registered on Agilent Technologies 6540 UHD Accurate Mass
Q-TOF LC/MS, HPLC 1290 Infinity as an external standard. IR spectra
were registered in CHCl3 solution in the 4000e625 cmꢀ1 range.
Tetraethyl ethane-1,1-bis(phosphonate) (TEBP) was prepared ac-
cording to Degenhardt's protocol [41] consisting in the condensa-
tion of tetraethyl methanebis(phosphonate) with formaldehyde in
the presence of diethylamine in refluxing methanol. TsOH-
catalyzed elimination of methanol in refluxing toluene gave the
expected product in 82% of yield.
1b+16
F
N
H
Chemical Formula: C11H9FN•
Exact Mass: 174,0719
The general procedures to prepare N-Boc-3-methyleneindolines
6aee was previously reported [34]. All the new precursors of the
Fig. 7. Main fragments for the hydroxylated metabolites.

C.A. Palmerini et al. / European Journal of Medicinal Chemistry 102 (2015) 403e412
409
indole bisphosphonates 1aef, were prepared as previously re-
ported [34] and characterized as follows:
4.1.5.2. 1-(tert-Butoxycarbonyl)-3-[3,3-bis(diethoxyphosphoryl)pro-
pyl]-5-fluoroindole (6b) (66%). Viscous oil; 1H NMR (400 MHz,
CDCl3)
d
8.06 (s, 1H), 7.44 (s, 1H), 7.22 (dd, J ¼ 8.9 and 2.4 Hz, 1H),
7.01 (td, J ¼ 9.1 and 2.5 Hz,1H), 4.40e2.19 (m, 8H), 2.95 (t, J ¼ 7.5 Hz,
4.1.1. tert-Butyl pyridin-4-ylcarbamate (2f) (74%)
2H), 2.45e2.20 (m, 3H), 1.65 (s, 9H), 1.33 (t, J ¼ 7.0 Hz, 6H), 1.30 (t,
tert-butyl pyridin-4-ylcarbamate (2f) (74%) was prepared from
J ¼ 7.0 Hz, 6H); 13C NMR (101 MHz, CDCl3)
d
159.1 (d, J ¼ 238 Hz),
4-aminopyridine (4.00 g, 21.2 mmol) [34]. White solid. 1H NMR
149.4, 131.6, 131.3 (d, J ¼ 9.3 Hz), 126.1, 119.3 (d, J ¼ 3.9 Hz), 116.1 (d,
J ¼ 9.0 Hz), 112.0 (d, J ¼ 25 Hz), 104.7 (d, J ¼ 24 Hz), 83.6, 62.6 (d,
J ¼ 6.6 Hz, 2C), 62.4 (d, J ¼ 6.7 Hz, 2C), 35.9 (t, J ¼ 133 Hz), 28.1 (3C),
25.2 (t, J ¼ 4.8 Hz), 24.0 (t, J ¼ 6.7 Hz), 16.3 (d, J ¼ 4.9 Hz, 4C); 19F
(200 MHz, CDCl3)
d 8.45 (m, 2H), 7.30 (m, 2H), 1.52 (s, 9H).
4.1.2. tert-Butyl (3-bromopyridin-4-yl)carbamate (3f) (52%)
tert-Butyl (3-bromopyridin-4-yl)carbamate (3f) (52%) was pre-
pared from pyridine 2f (2.70 g, 13.9 mmol) [34]. White solid; mp
NMR (376 MHz, CDCl3)
d
ꢀ121.4 (s); 31P NMR (126 MHz, CDCl3)
d
24.8 (s); IR ymax 2988, 1728, 1599, 1251, 1158, 1027 cmꢀ1. Elem.
91e92 ꢁC. 1H NMR (400 MHz, CDCl3)
d 8.58 (s, 1H), 8.37 (d,
Anal. Calcd for C24H38FNO8P2: C, 52.46; H, 6.97; N, 2.55. Found C,
52.41; H, 7.10; N, 2.47.
J ¼ 5.6 Hz, 1H), 8.15 (d, J ¼ 5.6 Hz, 1H), 7.18 (s, 1H), 1.55 (s, 9H); 13
C
NMR (101 MHz, CDCl3)
28.1 (3C).
d 151.6, 151.4, 149.4, 143.1, 113.1, 109.6, 82.4,
4.1.5.3. 1-(tert-Butoxycarbonyl)-3-[3,3-bis(diethoxyphosphoryl)pro-
pyl]-5-trifluormethylindole (6c) (75%). Viscous oil; 1H NMR
4.1.3. tert-Butyl (3-bromopyridin-4-yl)(prop-2-yn-1-yl)carbamate
(4f) (69%)
(400 MHz, CDCl3)
d
8.16 (broad d, J ¼ 7.6 Hz,1H), 7.78 (s,1H), 7.48 (d,
J ¼ 8.7 Hz, 1H), 7.44 (s, 1H), 4.25e4.00 (m, 8H), 2.96 (t, J ¼ 7.2 Hz,
2H), 2.34e2.20 (m, 3H), 1.61 (s, 9H), 1.27 (t, J ¼ 7.1 Hz, 6H), 1.23 (t,
tert-Butyl
(3-bromopyridin-4-yl)(prop-2-yn-1-yl)carbamate
(4f) (69%) was prepared by alkylation of carbamate 3f (1.90 g,
6.95 mmol) with propargyl bromide [34]. It was obtained as a
brown oil after chromatography on SiO2 (eluent, 1:1 petroleum
J ¼ 7.1 Hz, 6H); 13C NMR (101 MHz, CDCl3)
d 149.3, 137.2, 130.1, 124.8
(q, J ¼ 270 Hz), 124.7, 124.7 (q, J ¼ 32 Hz), 121.1 (q, J ¼ 3.1 Hz), 119.7,
116.4 (q, J ¼ 3.1 Hz), 115.5, 84.2, 62.7 (d, J ¼ 6.7 Hz, 2C), 62.5 (d,
J ¼ 6.6 Hz, 2C), 35.9 (t, J ¼ 133 Hz), 28.1 (3C), 25.4, 23.8 (t, J ¼ 6.6 Hz),
ether/diethyl ether). 1H NMR (400 MHz, CDCl3)
d
8.80 (s, 1H), 8.55
(d, J ¼ 5.1 Hz, 1H), 7.36 (bs, 1H), 4.71 (bs, 1H), 4.08 (bs, 1H), 2.25 (s,
1H), 1.43 (s, 9H); 13C NMR (101 MHz, CDCl3)
153.2, 152.6, 149.5,
16.3 (d, J ¼ 5.8 Hz, 4C); 19F NMR (376 MHz, CDCl3)
d
ꢀ61.4 (s); 31P
d
NMR (126 MHz, CDCl3) d 24.8 (s); IR ymax 2988, 2932, 1732, 1622,
147.8, 125.2, 121.7, 82.3, 78.5, 73.1, 37.8, 28.2 (3C).
1255, 1159, 1027 cmꢀ1. Elem. Anal. Calcd for C25H38F3NO8P2: C,
50.09; H, 6.39; N, 2.34. Found C, 50.16; H, 6.31; N, 2.28.
4.1.4. 1-(tert-Butoxycarbonyl)-3-methylene(pyrrolo[3,2-c]pyridine)
(5f) (10%)
4.1.5.4. 1-(tert-Butoxycarbonyl)-3-[3,3-bis(diethoxyphosphoryl)pro-
pyl]-6-trifluormethylindole (6d) (64%). Viscous oil; 1H NMR
1-(tert-Butoxycarbonyl)-3-methylene(pyrrolo[3,2-c]pyridine)
(5f) (10%) was prepared by radical cyclization of propargyl pyridine
4f (1.45 g, 4.7 mmol) [34]. After chromatography of the crude on
silica gel (eluent 8:2 petroleum ether/diethyl ether) a white solid
(10% yield) was collected exhibiting the following characteristics:
(400 MHz, CDCl3)
d
7.75 (d, J ¼ 7.7 Hz, 1H), 7.62 (d, J ¼ 7.6 Hz, 1H),
7.39 (s, 1H), 7.33 (t, J ¼ 7.7 Hz, 1H), 4.22 (m, 8H), 2.99 (t, J ¼ 7.5 Hz,
2H), 2.32e2.21 (m, 3H), 1.63 (s, 9H), 1.34 (t, J ¼ 7.1 Hz, 6H), 1.31 (t,
J ¼ 7.1 Hz, 6H); 13C NMR (101 MHz, CDCl3)
d 149.9,132.9,131.0,126.7
mp 90 ꢁC (dec). 1H NMR (400 MHz, CDCl3)
d
8.61 (s, 1H), 8.36 (d,
J ¼ 5.2 Hz, 1H), 7.75 (bs, 1H), 5.59 (t, J ¼ 2.9 Hz, 1H), 5.15 (s, 1H), 4.59
(s, 2H), 1.58 (s, 9H); 13C NMR (101 MHz, CDCl3)
151.1, 150.7, 142.3,
(q, J ¼ 270 Hz), 126.6, 123.5 (d, J ¼ 5.0 Hz), 122.7, 122.1, 118.9, 112.8,
84.5, 63.1 (d, J ¼ 6.6 Hz, 2C), 62.7 (d, J ¼ 6.6 Hz, 2C), 35.9 (t,
J ¼ 132 Hz), 27.8 (3C), 25.3 (t, J ¼ 6.7 Hz), 23.8 (t, J ¼ 6.6 Hz), 16.3 (d,
d
J ¼ 6.2 Hz, 4C); 19F NMR (376 MHz, CDCl3)
d
ꢀ61.0 (s); 31P NMR
138.5, 125.2, 109.9, 103.5, 82.0, 53.5 (2C), 28.2 (3C).
(126 MHz, CDCl3)
d 25.0 (s); IR ymax 2989, 2937, 1756, 1603, 1248,
1155, 1026 cmꢀ1. Elem. Anal. Calcd for C25H38F3NO8P2: C, 50.09; H,
6.39; N, 2.34. Found C, 50.18; H, 6.48; N, 2.21.
4.1.5. General procedure for the synthesis of tetraethyl [3-(1-tert-
butoxycarbonylindol-3-yl)propane-1,1-diyl]bis(phosphonates)
(6aef)
4.1.5.5. 1-(tert-Butoxycarbonyl)-3-[3,3-bis(diethoxyphosphoryl)pro-
pyl]-6-fluoroindole (6e) (75%). Viscous oil; 1H NMR (400 MHz,
A
mixture of the suitable N-tert-butoxy carbonyl-3-
methyleneindoline 5aef (1.0 mmol) and VBP (1.1 mmol) was
heated at 100 ꢁC for 2e4 h in a sealed tube without solvent. The
crude reaction product was washed with water (5 mL) under
vigorous stirring in order to eliminate the unreacted VBP. Chro-
matography of the viscous residue on neutral aluminum oxide
(eluent, 99:1 petroleum ether/ethanol) allowed to collect the pure
product that was characterized as follows.
CDCl3)
d
7.77 (broad s, 1H), 7.42 (dd, J ¼ 8.6 and 5.4 Hz, 1H), 7.32 (s,
1H), 6.91 (td, J ¼ 8.7 and 2.4 Hz, 1H), 4.1 (m, 8H), 2.91 (t, J ¼ 7.0 Hz,
2H), 2.43e2.24 (m, 3H), 1.60 (s, 9H), 1.26 (t, J ¼ 7.1 Hz, 6H), 1.23 (t,
J ¼ 7.1 Hz, 6H); 13C NMR (101 MHz, CDCl3)
d
161.0 (d, J ¼ 238 Hz),
158.3, 149.4, 126.7, 123.1, 119.6 (d, J ¼ 9.7 Hz), 119.4, 110.5 (d,
J ¼ 24 Hz), 102.6 (d, J ¼ 28 Hz), 83.8, 62.6 (d, J ¼ 6.7 Hz, 2C), 62.4 (d,
J ¼ 6.6 Hz, 2C), 35.9 (t, J ¼ 133 Hz), 28.1 (3C), 25.3 (t, J ¼ 6.9 Hz), 24.0
(t, J ¼ 6.8 Hz), 16.4 (d, J ¼ 6.1 Hz, 2C),16.3 (d, J ¼ 6.2 Hz, 2C); 19F NMR
4.1.5.1. 1-(tert-Butoxycarbonyl)-3-[3,3-bis(diethoxyphosphoryl)pro-
(376 MHz, CDCl3)
d
ꢀ118.1 (s); 31P NMR (126 MHz, CDCl3)
d 24.9 (s);
pyl]indole (6a) (78%). Viscous oil; 1H NMR (400 MHz, CDCl3)
d
8.12
IR ymax 2993, 2932, 1727, 1382, 1250, 1159, 1027 cmel; Elem. Anal.
Calcd for C24H38FNO8P2: C, 52.46; H, 6.97; N, 2.55. Found C, 52.49;
H, 7.15; N, 2.42.
(bd, J ¼ 7.2 Hz, 1H), 7.58 (d, J ¼ 7.3 Hz, 1H), 7.41 (s, 1H), 7.30 (td,
J ¼ 7.2 and 1.1 Hz,1H), 7.23 (td, J ¼ 7.2 and 1.1 Hz,1H), 4.45e2.25 (m,
8H), 3.00 (t, J ¼ 7.4 Hz, 2H), 2.46e2.27 (m, 3H), 1.67 (s, 9H), 1.33 (t,
J ¼ 7.0 Hz, 6H), 1.30 (t, J ¼ 7.0 Hz, 6H); 13C NMR (101 MHz, CDCl3)
4.1.5.6. 1-(tert-Butoxycarbonyl)-3-[3,3-bis(diethoxyphosphoryl)pro-
d
149.7, 135.6, 130.4, 124.3, 123.0, 122.3, 119.5, 119.0, 115.2, 83.4, 62.6
pyl]-pyrrolo[3,2-c]pyridine (6f) (52%). Viscous oil. 1H NMR
(d, J ¼ 6.7 Hz, 2C), 62.4 (d, J ¼ 6.7 Hz, 2C), 35.9 (t, J ¼ 133 Hz), 28.2
(400 MHz, CDCl3)
d
8.87 (s, 1H), 8.45 (d, J ¼ 5.7 Hz, 1H), 7.95 (broad
(3C), 25.3 (t, J ¼ 4.8 Hz), 24.1 (d, J ¼ 6.8 Hz), 16.3 (d, J ¼ 5.4 Hz, 4C);
s, 1H), 7.42 (s, 1H), 4.2 (m, 8H), 3.05 (t, J ¼ 7.5 Hz, 1H), 2.3 (m, 3H),
31P NMR (126 MHz, CDCl3)
d 25.0 (s); IR ymax 2988, 2933,1726, 1605,
1.67 (s, 9H), 1.33 (t, J ¼ 7.1 Hz, 6H), 1.28 (t, J ¼ 7.1, 6H); 13C NMR
1253, 1026 cmꢀ1. Elem. Anal. Calcd for C24H39NO8P2: C, 54.23; H,
7.40; N, 2.64. Found C, 54.27; H, 7.45; N, 2.67.
(101 MHz, CDCl3) d 149.1, 144.2, 142.1, 139.9, 126.6, 123.7, 118.8,
110.0, 84.6, 62.8 (2C), 62.7 (2C), 35.9 (t, J ¼ 133 Hz), 28.1 (3C), 25.5 (t,

410
C.A. Palmerini et al. / European Journal of Medicinal Chemistry 102 (2015) 403e412
J ¼ 6.8 Hz), 23.9 (t, J ¼ 6.8 Hz), 16.4 (2C), 16.3 (2C); 31P NMR
7.14e7.01 (m, 1H), 4.24e4.11 (m, 8H), 3.08 (t, J ¼ 7.3 Hz, 2H), 2.41e2.28
(m, 3H), 1.34e1.22 (m, 12H); 13C NMR (101 MHz, CDCl3)
137.9, 126.6,
(126 MHz, CDCl3)
d
24.71 (s). Elem. Anal. Calcd for C23H38N2O8P2: C,
d
51.88; H, 7.19; N, 5.26. Found C, 51.91; H, 7.29; N, 5.30.
124.1, 121.2 (d, J ¼ 32 Hz), 118.2 (d, J ¼ 3.0 Hz), 116.3 (d, J ¼ 4.3 Hz),
115.1, 111.6, 62.7 (d, J ¼ 7.5 Hz, 2C), 62.5 (d, J ¼ 7.9 Hz, 2C), 35.6 (t,
J ¼ 133 Hz), 26.7 (t, J ¼ 5.2 Hz), 24.0 (t, J ¼ 6.9 Hz), 16.2 (d, J ¼ 4.0 Hz,
4.1.5.7. General procedure for the synthesis of 3-[3,3-bis(diethoxy-
phosphoryl)propyl]indole 1aef. The suitable carbamate 6aef (50 mg)
was cautiously heated at 150e190 ꢁC in a sealed tube with an air
heater until gas evolvement ceased (~5e10 min). After cooling, a
viscous brown oil was obtained whose 1H NMR spectrum was
compatible with the structure of the indole bisphosphonates 1aef.
High purity products were obtained in practically quantitative yield
by flash chromatography of the crude on SiO2 (dichloromethane/
ethanol 99:1 as eluent). They were characterized as follows.
2C), 16.1 (d, J ¼ 4.5 Hz, 2C); 19F NMR (376 MHz, CDCl3)
d
ꢀ60.5 (s); 31
P
NMR (126 MHz, CDCl3)
d 25.07 (s); HRMS: calcd for C19H30FNO6P2
450.1611 (MþHþ), found 450.1610 (MþHþ).
4.1.5.13. Tetraethyl (3-(3-(pyrrolo[3,2-c]pyridin-3-yl)propane-1,1-
diyl)bis(phosphonate) (1f). Colorless oil 1H NMR (400 MHz, CDCl3)
d
8.94 (bs,1H), 8.23 (d, J ¼ 5.2 Hz,1H), 7.34 (d, J ¼ 5.9 Hz, 1H), 7.13 (s,
1H), 4.2 (m, 8H), 3.1 (m, 2H), 2.3 (m, 3H), 1.3 (m, 12H). 13C NMR
(101 MHz, CDCl3) 142.2, 140.7, 140.2, 127.3, 124.8, 114.8, 105.0, 62.6
(d, J ¼ 7.9 Hz, 2C), 62.5 (d, J ¼ 6.8 Hz, 2C), 36.2 (t, J ¼ 133 Hz), 26.3 (t,
J ¼ 4.9 Hz), 24.2 (t, J ¼ 6.9 Hz), 16.4 (d, J ¼ 6.3 Hz, 2C), 16.2 (d,
4.1.5.8. Tetraethyl (3-(indol-3-yl)propane-1,1-diyl)bis(phosphonate)
(1a). Colorless oil. 1H NMR (400 MHz, CDCl3)
d 8.50 (s, 1H), 7.64 (d,
J ¼ 7.7 Hz, 1H), 7.37 (d, J ¼ 8.0 Hz, 1H), 7.22e7.08 (m, 2H), 7.06 (s,
J ¼ 6.0 Hz, 2C); 31P NMR (126 MHz, CDCl3)
d 23.64 (s); HRMS: calcd
1H), 4.26e4.02 (m, 8H), 3.08 (t, J ¼ 6.8 Hz, 2H), 2.55e2.26 (m, 3H),
for C18H30FN2O6P2 433.3961 (MþHþ), found 433.3961 (MþHþ).
1.41e1.21 (m, 12H); 13C NMR (101 MHz, CDCl3)
d 136.5, 127.4, 122.1,
121.9, 119.1, 118.7, 114.6, 111.2, 62.5 (d, J ¼ 6.7 Hz, 2C), 62.4 (d,
J ¼ 6.7 Hz, 2C), 35.7 (t, J ¼ 133 Hz), 26.0 (t, J ¼ 4.8 Hz), 24.4 (t,
J ¼ 6.8 Hz), 16.1 (d, J ¼ 3.3 Hz, 2C), 16.3 (d, J ¼ 3.3 Hz, 2C); 31P NMR
4.2. Biological tests
4.2.1. Materials and methods
(126 MHz, CDCl3)
d
24.1 (s, 1H); HRMS: calcd for C19H31NO6P2
Materials Fura 2-AM (Fura-2-pentakis(acetoxymethyl)ester),
Triton X-100 (t-octylphenoxypolyethoxyethanol), EGTA (ethylene
432.4080 (M þ Hþ), found 432.4081 (M þ Hþ).
glycol-bis(b
-aminoethyl ether)-N,N,N0,N0-tetracetic acid), Clodro-
4.1.5.9. Tetraethyl (3-(5-fluoroindol-3-yl)propane-1,1-diyl)bis(phosph-
nate (CLD), Alendronate (ALN), Trypan blue, and Iscove's modified
Dulbecco Medium were purchased from SigmaeAldrich corpora-
tion (St. Louis, Missouri, USA). Other reagents (reagent grade) were
obtained from common commercial sources. Synthetic compounds
1aef were ꢂ98% pure as determined by UHPLC: column, Phe-
onate) (1b). Colorless oil 1H NMR (400 MHz, CDCl3)
d 9.30 (s, 1H),
7.20 (m, 2H), 7.02 (s, 1H), 6.83 (td, J ¼ 9.1 and 2.2 Hz, 1H), 4.21e4.09
(m, 8H), 2.97 (t, J ¼ 7.4, 2H), 2.3 (m, 3H), 1.28 (t, J ¼ 7.0 Hz, 6H),1.24 (t,
J ¼ 7.0 Hz, 6H); 13C NMR (101 MHz, CDCl3) 157.4 (d, J ¼ 234 Hz),133.0,
127.5 (d, J ¼ 9.5 Hz), 124.0, 114.2, 111.8 (d, J ¼ 9.5 Hz), 109.7 (d,
J ¼ 26 Hz), 103.4 (d, J ¼ 23 Hz) 62.5 (d, J ¼ 6.7 Hz, 2C), 62.4 (d,
J ¼ 6.7 Hz, 2C), 35.5 (t, J ¼ 133 Hz), 25.8 (t, J ¼ 4.7 Hz) 24.2 (t,
J ¼ 6.5 Hz), 16.2 (d, J ¼ 3.8 Hz, 2C), 16.1 (d, J ¼ 3.8 Hz, 2C). 19F NMR
nomenex AERIS Peptide 1.2 mm ꢃ 1000 mm (1.7
mm); flow rate,
0.8 mL/min; acquisition time, 20 min; DAD 190e650 nm; oven
temperature, 45 ꢁC; gradient of acetonitrile in water containing
0.1% of formic acid (0e100% in 20 min).
(376 MHz, CDCl3)
d
ꢀ126.0 (s). 31P NMR (126 MHz, CDCl3)
d 25.17 (s);
HRMS: calcd for C19H30FNO6P2 450.1611 (MþHþ), found 450.1609
4.2.2. Cell preparations
(MþHþ).
PE/CA PJ-15 cells were cultivated (37 ꢁC, 5% CO2, 72 h) in Iscove-
Dulbecco medium complemented (IMDM) with 1% L-glutamine,
4.1.5.10. Tetraethyl (3-(5-trifluoromethyl-3-yl)propane-1,1-diyl)bis(ph-
penicillin/streptomycin (100 U/mL each) and 10% inactivated fetal
calf serum. The medium was replaced every 72 h and cells rinsed
before use with phosphate buffered saline (PBS) to remove excess
serum. The incubation medium was then discarded and the cells
washed again with phosphate buffered saline (PBS). Trypsin (0.05%
in 0.02% EDTA, Euroclone) was added to each flasks to detach cells.
The action of trypsin was stopped after 3 min by the addition of fetal
calf serum (10%). Cells were harvested by centrifugation at
400ꢃ g ꢃ 5 min and suspended in a volume of medium to obtain cell
concentration of 1 ꢃ 106 cells/mL. Cell proliferation was monitored
by counting the viable cells by Trypan blue exclusion method.
osphonate) (1c). Colorless oil. 1H NMR (400 MHz, CDCl3)
d 9.76 (bs,
1H), 7.88 (s, 1H), 7.42 (d, J ¼ 8.6 Hz, 1H), 7.33 (d, J ¼ 8.4 Hz, 1H), 7.08
(d, J ¼ 2.1 Hz, 1H), 4.19e4.05 (m, 8H), 3.06 (t, J ¼ 7.3, 2H), 2.55e2.10
(m, 3H), 1.35e1.10 (m, 12H); 13C NMR (101 MHz, CDCl3) 138.05,
126.67, 125.56 (q, J ¼ 271 Hz), 124.16, 121.20 (q, J ¼ 32 Hz), 118.28 (q,
J ¼ 3.1 Hz),116.34 (q, J ¼ 4.4 Hz),115,12,111.67, 62.67 (t, J ¼ 7.4 Hz, 4C),
35.70 (t, J ¼ 134 Hz), 26.21(t, J ¼ 4.8 Hz), 24.16 (t, J ¼ 7.0 Hz), 16.31 (d,
J ¼ 4.5 Hz, 2C), 16.25 (d, J ¼ 4.5 Hz, 2C); 19F NMR (376 MHz, CDCl3)
d
ꢀ60.5 (s). 31P NMR (126 MHz, CDCl3)
d 25.07 (s); HRMS: calcd for
C
20H30F3NO6P2 500.1579 (MþHþ), found 500.1577 (MþHþ).
4.1.5.11. Tetraethyl (3-(6-trifluoromethyl-3-yl)propane-1,1-diyl)bis(ph-
osphonate) (1d). 1H NMR (400 MHz, CDCl3)
8.55 (s, 1H), 7.84 (d,
4.2.3. Test of Trypan blue exclusion
d
Aliquots (20 mL) of cell suspensions were diluted 1:10 with
J ¼ 7.8 Hz, 1H), 7.46 (d, J ¼ 7.5 Hz,1H), 7.19 (s, 2H), 4.24e4.01 (m, 8H),
Trypan blue (2.22 mg/mL of PBS) and counted using a Burker
chamber after standing at room temperature for 5 min. The blue-
stained cells were considered as non-viable.
3.11 (t, J ¼ 7.1 Hz, 2H), 2.51e2.45 (m, 3H), 1.29 (t, J ¼ 7.0 Hz, 6H), 1.26
(t, J ¼ 7.0 Hz, 6H); 13C NMR (101 MHz, CDCl3)
d 131.9, 129.0, 124.7 (q,
J ¼ 272 Hz), 123.2, 123.0, 119.5 (q, J ¼ 4.4 Hz), 118.6, 115.4, 113.2 (q,
J ¼ 32 Hz), 62.6 (d, J ¼ 6.5 Hz, 2C), 62.4 (d, J ¼ 6.5 Hz, 2C), 35.7 (t,
J ¼ 133 Hz), 26.0 (t, J ¼ 4.6 Hz), 23.9 (t, J ¼ 6.7 Hz), 16.4 (d, J ¼ 5.6 Hz,
4.2.4. Ca2þ extracellular determination
The determination of the extracellular calcium was performed
on an aliquot of the cell suspension (1 mL) incubated in 3 mL of
HBSS (þ1 mM CaCl2). The cells were lysed with triton X-100. The
variation of fluorescence of the sample (Ca2þ/FURA-2) was deter-
2C), 16.3 (d, J ¼ 5.6 Hz, 2C); 19F NMR (376 MHz, CDCl3)
d
ꢀ60.87; 31
P
NMR (126 MHz, CDCl3)
d 25.07; HRMS: calcd for C20H30F3NO6P2
500.1579 (MþHþ), found 500.1582 (MþHþ).
mined after the addition of (10e30 mM) BPs.
4.1.5.12. Tetraethyl (3-(6-fluoroindol-3-yl)propane-1,1-diyl)bis(phosph-
onate) (1e). Colorless oil 1H NMR (400 MHz, CDCl3)
(s, 1H), 7.45 (d, J ¼ 8.4 Hz, 1H), 7.36 (d, J ¼ 8.4 and 1.2 Hz, 1H),
d
9.75 (s, 1H), 7.93
4.2.5. Cytosolic calcium determination
FURA-2 AM (2 L of a 2 mM solution in DMSO) was added to
m

C.A. Palmerini et al. / European Journal of Medicinal Chemistry 102 (2015) 403e412
411
1 mL of cell suspension (about 10 ꢃ 106 cells) and incubated for
Acknowledgments
60 min at 37 ꢁC, in the dark. Cells were then harvested by centri-
fugation at 800ꢃ g ꢃ 10 min and finally suspended in HBSS buffer
(140 mM NaCl, 5.3 mM KCl, 1 mM MgCl2, 5 mM glucose, 25 mM
Hepes, adjusted to pH 7.4) to a final cell concentration of 1 ꢃ 106/
mL.
The authors would like to thank Prof. Renzo Ruzziconi for his
scientific support and Molecular Discovery Ltd. for financial sup-
port. Dr. Cristiano Zuccaccia is kindly acknowledged for his
contribution in the elucidation of NMR spectra.
An aliquot of this suspension (1 mL) was centrifuged at 800 g for
5 min, cells harvested and suspended in 3 mL of calcium-free HBSS
(0.1 mM EGTA). Fluorescence was measured with a PerkineElmer
LS 50 B spectrophotofluorometer equipped with a double excita-
tion system (ex. 340 and 380 nm, em. 510 nm). Slit widths were set
at 10 nm for excitation and 7.5 nm for emission. Cytosolic calcium
concentrations ([Ca2þ]c) were calculated as reported elsewhere
[49]. When necessary, bisphosphonates and Ca2þ were added, as
reported in the “Results” section.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2015.08.019.
References
[1] R.E. Coleman, Metastatic bone disease: clinical features, pathophysiology and
treatment strategies, Cancer Treat. Rev. 27 (2001) 165e176.
[2] H. Fleisch, Bisphosphonates in osteoporosis, Eur. Spine J. Off. Publ. Eur. Spine
Soc. Eur. Spinal Deform. Soc. Eur. Sect. Cerv. Spine Res. Soc. 12 (Suppl. 2)
(2003) S142eS146.
4.3. Computational method
[3] H. Fleisch, J. Maerki, R.G. Russell, Effect of pyrophosphate on dissolution of
hydroxyapatite and its possible importance in calcium homeostasis, pro-
ceedings of the society for experimental biology and medicine, Soc. Exp. Biol.
Med. 122 (1966) 317e320.
4.3.1. Permeability prediction
The VolSurfþ software (www.moldiscovery.com) was used to
predict cell-permeability. The method has been extensively
described elsewhere [47]. To describe the BPs compounds, the GRID
[47] probes for water (OH2), hydrophobic (DRY) and H-bonding
carbonyl (O) interactions were used to generate the 3D interaction
energies and a grid space of 0.5 Å. The most abundant protonation
state for each compound was used.
[4] H.A. Fleisch, R.G. Russell, S. Bisaz, R.C. Muhlbauer, D.A. Williams, The inhibi-
tory effect of phosphonates on the formation of calcium phosphate crystals
in vitro and on aortic and kidney calcification in vivo, Eur. J. Clin. Invest. 1
(1970) 12e18.
[5] J. Meldolesi, T. Pozzan, The endoplasmic reticulum Ca2þ store: a view from the
lumen, Trends Biochem. Sci. 23 (1998) 10e14.
[6] D.M. Bautista, R.S. Lewis, Modulation of plasma membrane calcium-ATPase
activity by local calcium microdomains near CRAC channels in human T
cells, J. Physiol. 556 (2004) 805e817.
[7] M.J. Berridge, M.D. Bootman, P. Lipp, Calciumea life and death signal, Nature
395 (1998) 645e648.
4.4. Statistical analysis
[8] D.E. Clapham, Intracellular calcium - replenishing the stores, Nature 375
(1995) 634e635.
Results are expressed as means SEM.
[9] M.R. Webster, M. Zhao, M.A. Rudek, C.L. Hann, C.L. Freel Meyers, Bisphos-
phonamidate clodronate prodrug exhibits potent anticancer activity in non-
small-cell lung cancer cells, J. Med. Chem. 54 (2011) 6647e6656.
[10] M.J. Rogers, New insights into the molecular mechanisms of action of
bisphosphonates, Curr. Pharm. Des. 9 (2003) 2643e2658.
[11] R.G. Russell, Bisphosphonates: mode of action and pharmacology, Pediatrics
119 (Suppl. 2) (2007) S150eS162.
[12] J.E. Dunford, M.J. Rogers, F.H. Ebetino, R.J. Phipps, F.P. Coxon, Inhibition of
protein prenylation by bisphosphonates causes sustained activation of Rac,
Cdc42, and Rho GTPases, J. Bone Miner. Res. Off. J. Am. Soc. Bone Miner. Res.
21 (2006) 684e694.
4.5. Metabolism assay
Substrates were incubated with human liver microsomes (HLM,
0.5 mg protein/ml) (BD Biosciences) according to manufacturer's
recommendations with minor modification. Briefly, substrates at
5
m
M final concentration were preincubated in a shaking water bath
for 5 min at 37 ꢁC in 100 mM potassium phosphate buffer (pH 7.4)
in a total volume of 250 L. The reactions were initiated by addic-
tion of 1 mM NADPH. After incubation for 0 and 30 min, 250 L of
[13] P. Clezardin, F.H. Ebetino, P.G. Fournier, Bisphosphonates and cancer-induced
bone disease: beyond their antiresorptive activity, Cancer Res. 65 (2005)
4971e4974.
m
m
[14] J.R. Green, Bisphosphonates: preclinical review, Oncologist. 9 (Suppl. 4) (2004)
3e13.
cold acetonitrile (containing 0.6 mM labetalol as internal standard)
were added to the mix to terminate the reaction. Proteins were
precipitated by centrifugation at 12,000 g for 5 min at 4 ꢁC, and
aliquots of supernatants were analyzed by HPLC-MS/MS. The LC/MS
analysis were run on a Agilent 6540 UHD accurate mass Q-TOF LC/
MSMS system (Agilent Technologies, Palo Alto, CA) governed by
Agilent MassHunter software (B.05.00 version). The system consists
of a binary pump, autosampler, thermostated column compart-
ment, DAD detector, source, and Q-TOF spectrometer. Chromato-
graphic separation of the metabolites (Agilent 1290 UHPLC system)
was performed with Aeris Peptide 1.7 J XB-C18, 100 ꢃ 2.1 mm
(Phenomenex USA) at a constant temperature of 40 ꢁC. The mobile
phases consisted of A: H2O/0.1% formic acid and B: acetonitrile/0.1%
formic acid at the flow of 0.3 mL/min with the following gradient:
Time 0 min, B 0%; Time 20 min, B 100%. The DAD Detector stored all
the acquired spectra in the 10e640 nm range (2 nm spectrum step).
The ion source was an Agilent Dual JetStream operating under
positive ionization mode (4000 V), with as nitrogen the desolvating
gas (320 ꢁC, 10 L/min, 35 psig). The fragmentor was set to 110 V, the
skimmer to 65 V, and the octrapole RF to 750 V. The spectrometric
data were collected in All Ion mode in the 100$1000 mass range,
with 3 scans/sec at Collision Energy of 0, 30, 40 V. The TOF operated
at 2 GHz.
[15] M. Caraglia, M. Marra, C. Viscomi, A.M. D'Alessandro, A. Budillon, G. Meo,
C. Arra, A. Barbieri, U.R. Rapp, A. Baldi, P. Tassone, S. Venuta, A. Abbruzzese,
P. Tagliaferri, The farnesyltransferase inhibitor R115777 (ZARNESTRA) en-
hances the pro-apoptotic activity of interferon-alpha through the inhibition of
multiple survival pathways, Int. J. Cancer. Journal International du Cancer 121
(2007) 2317e2330.
[16] M. Bezzi, M. Hasmim, G. Bieler, O. Dormond, C. Ruegg, Zoledronate sensitizes
endothelial cells to tumor necrosis factor-induced programmed cell death:
evidence for the suppression of sustained activation of focal adhesion kinase
and protein kinase B/Akt, J. Biol. Chem. 278 (2003) 43603e43614.
[17] B. Ory, M.F. Heymann, A. Kamijo, F. Gouin, D. Heymann, F. Redini, Zoledronic
acid suppresses lung metastases and prolongs overall survival of
osteosarcoma-bearing mice, Cancer 104 (2005) 2522e2529.
[18] J. Liu, W. Liu, H. Ge, J. Gao, Q. He, L. Su, J. Xu, L.-q Gu, Z.-s Huang, D. Li, Syn-
theses and characterization of non-bisphosphonate quinoline derivatives as
new FPPS inhibitors, Biochim. Biophys. Acta (BBA)-Gen. Subj. 1840 (2014)
1051e1062.
[19] D. Ledoux, Y. Hamma-Kourbali, M. Di Benedetto, A. Foucault-Bertaud,
O. Oudar, O. Sainte-Catherine, M. Lecouvey, M. Kraemer, A new dimethyl ester
bisphosphonate inhibits angiogenesis and growth of human epidermoid
carcinoma xenograft in nude mice, Anti-Cancer Drugs 17 (2006) 479e485.
[20] G. De Rosa, G. Misso, G. Salzano, M. Caraglia, Bisphosphonates and cancer:
what opportunities from nanotechnology? J. Drug Deliv. 2013 (2013) 637976.
[21] X.L. Xu, W.L. Gou, A.Y. Wang, Y. Wang, Q.Y. Guo, Q. Lu, S.B. Lu, J. Peng, Basic
research and clinical applications of bisphosphonates in bone disease: what
have we learned over the last 40 years? J. Transl. Med. 11 (2013) 303.
[22] R. Bergner, I.J. Diel, D. Henrich, M. Hoffmann, M. Uppenkamp, Differences in
nephrotoxicity of intravenous bisphosphonates for the treatment of

412
C.A. Palmerini et al. / European Journal of Medicinal Chemistry 102 (2015) 403e412
ꢀ
malignancy-related bone disease, Onkologie 29 (2006) 534e540.
[23] I.J. Diel, R. Bergner, K.A. Grotz, Adverse effects of bisphosphonates: current
issues, J. Support. Oncol. 5 (2007) 475e482.
[39] H. Couthon-Gourves, G. Simon, J.-P. Haelters, B. Corbel, Synthesis of novel
diethyl indolylphosphonates and tetraethyl indolyl-1, 1-bisphosphonates by
Michael addition, Synthesis 2006 (2006) 81e88.
[24] http://pusbchem.ncbi.nlm.nih.gov, in.
[25] J.J. Vepsalainen, Bisphosphonate prodrugs, Curr. Med. Chem.
[40] A. Chiminazzo, L. Sperni, M. Damuzzo, G. Strukul, A. Scarso, Copper-mediated
1,4-conjugate addition of boronic acids and indoles to vinylidenebi-
sphosphonate leading to gem-bisphosphonates as potential antiresorption
bone drugs, Chemcatchem 6 (2014) 2712e2718.
[41] C.R. Degenhardt, D.C. Burdsall, Synthesis of ethenylidenebis (phosphonic acid)
and its tetraalkyl esters, J. Org. Chem. 51 (1986) 3488e3490.
[42] G.F. Smith, A.E. Walters, 194. Indoles. Part V. 3-alkylindole dimers, J. Chem.
Soc. (Resumed) (1961) 940e943.
[43] E. Ascic, S.L. Buchwald, Highly diastereo- and enantioselective CuH-catalyzed
synthesis of 2,3-disubstituted lndolines, J. Am. Chem. Soc. 137 (2015)
4666e4669.
[44] J. Bergman, E. Koch, B. Pelcman, Coupling reactions of indole-3-acetic acid
derivatives. Synthesis of arcyriaflavin A, J. Chem. Soc. Perkin Trans. 1 (2000)
2609e2614.
9 (2002)
1201e1208.
[26] A. Townshend, D.T. Burns, R. Lobinski, E.J. Newman, G. Guilbault,
Z. Marczenko, H. Onishi, Dictionary of Analytical Reagents, Taylor & Francis,
1993.
[27] G. Hagele, Z. Szakacs, J. Ollig, S. Hermens, C. Pfaff, NMR-controlled titrations:
characterizing aminophosphonates and related structures, Heteroat. Chem. 11
(2000) 562e582.
[28] http://www.sanofi.ca, in.
[29] W.K. Hagmann, The many roles for fluorine in medicinal chemistry, J. Med.
Chem. 51 (2008) 4359e4369.
[30] S. Purser, P.R. Moore, S. Swallow, V. Gouverneur, Fluorine in medicinal
chemistry, Chem. Soc. Rev. 37 (2008) 320e330.
[31] P.V. Ramachandran, Welcome to 'fluorine in medicinal chemistry', Future
Med. Chem. 1 (2009) 771e772.
[45] C. CharletFagnere, J. Laronze, J.Y. Laronze, L. Toupet, R. Vistelle, D. Lamiable,
C. Mouchard, P. Renard, G. Adam, Studies on the dimerization of melatonin (5-
methoxy-N-acetyltryptamine) and related compounds in acid medium.
Oxidation of some 2-arylindolines into 2-arylindoles, B Soc. Chim. Fr. 133
(1996) 39e50.
[46] F. Milletti, L. Storchi, G. Sforna, G. Cruciani, New and original pKa prediction
method using grid molecular interaction fields, J. Chem. Inf. Model. 47 (2007)
2172e2181.
[32] M. Franz, T. Hansen, P. Richter, L. Borsi, F.D. Bohmer, P. Hyckel, P. Schleier,
D. Katenkamp, L. Zardi, H. Kosmehl, A. Berndt, Complex formation of the
laminin-5 gamma2 chain and large unspliced tenascin-C in oral squamous cell
carcinoma in vitro and in situ: implications for sequential modulation of
extracellular matrix in the invasive tumor front, Histochem. Cell Biol. 126
(2006) 125e131.
[33] U.S. Singh, R. Shankar, A. Kumar, R. Trivedi, N. Chattopadhyay, N. Shakya,
S. Palne, S. Gupta, K. Hajela, Synthesis and biological evaluation of indolyl
bisphosphonates as anti-bone resorptive and anti-leishmanial agents, Bioorg.
Med. Chem. 16 (2008) 8482e8491.
[34] F. Bellezza, A. Cipiciani, R. Ruzziconi, S. Spizzichino, Nucleus- and side-chain
fluorinated 3-substituted indoles by a suitable combination of organome-
tallic and radical chemistry, J. Fluor. Chem. 129 (2008) 97e107.
[35] S. Lepri, F. Buonerba, P. Maccaroni, L. Goracci, R. Ruzziconi, Are carboxylic
esters really refractory to DAST? On the fluorination of alpha-hydroxyesters
with DAST, J. Fluor. Chem. 171 (2015) 82e91.
[47] G. Cruciani, P. Crivori, P.-A. Carrupt, B. Testa, Molecular fields in quantitative
structureepermeation relationships: the VolSurf approach, J. Mol. Struct.
THEOCHEM 503 (2000) 17e30.
[48] K.K. Kulasekara, O.M. Lukandu, E. Neppelberg, O.K. Vintermyr,
A.C. Johannessen, D.E. Costea, Cancer progression is associated with increased
expression of basement membrane proteins in three-dimensional in vitro
models of human oral cancer, Arch. Oral Biol. 54 (2009) 924e931.
[49] G. Grynkiewicz, M. Poenie, R.Y. Tsien, A new generation of Ca2þ indicators
with greatly improved fluorescence properties, J. Biol. Chem. 260 (1985)
3440e3450.
[36] R. Ruzziconi, G. Ricci, A. Gioiello, H. Couthon-Gourves, J.P. Gourves, First
general approach to cyclohex-3-ene-1,1-bis(phosphonates) by Diels-Alder
cycloaddition of tetraethyl vinylidenebis(phosphonate) to 1,3-dienes, J. Org.
Chem. 68 (2003) 736e742.
[37] H. Krawczyk, Indoles as nucleophiles in a self-catalytic michael reaction,
Synthesis 2002 (2002) 1351e1354.
[50] F. Broccatelli, G. Cruciani, L.Z. Benet, T.I. Oprea, BDDCS class prediction for new
molecular entities, Mol. Pharm. 9 (2012) 570e580.
[51] S.R. Denmeade, J.T. Isaacs, The SERCA pump as a therapeutic target: making a
“smart bomb” for prostate cancer, Cancer Biol. Ther. 4 (2005) 14e22.
[52] S. Weisthal, N. Keinan, D. Ben-Hail, T. Arif, V. Shoshan-Barmatz, Ca(2þ)-
mediated regulation of VDAC1 expression levels is associated with cell death
induction, Biochim. Biophys. Acta 1843 (2014) 2270e2281.
[53] C. Cerella, M. Diederich, L. Ghibelli, The dual role of calcium as messenger and
stressor in cell damage, death, and survival, Int. J. Cell Biol. 2010 (2010)
546163.
[38] M.V. Shevchuk, J.-M. Sotiropoulos, K. Miqueu, V.D. Romanenko, V.P. Kukhar,
Tetrakis (trimethylsilyl) ethenylidene-1, 1-bisphosphonate:
a mild and
convenient Michael acceptor for the synthesis of 2-aminoethylidene-1, 1-
bisphosphonic acids and their potassium salts, Synlett 2011 (2011)
1370e1374.