Radiosynthesis and micro-SPECT analysis of triazole-based RGD integrin ligands as non-peptide molecular imaging probes for angiogenesis

Article abstract of DOI:10.1016/j.bmc.2014.12.065

Taking advantage of click chemistry, we synthesized triazole-containing RGD peptidomimetics capable of binding to α_vβ₃ integrin with diverse potency, and selected ¹²⁵I-labeled compounds proved to interact in vitro and in v

Full text of DOI:10.1016/j.bmc.2014.12.065

Bioorganic & Medicinal Chemistry 23 (2015) 1112–1122
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Radiosynthesis and micro-SPECT analysis of triazole-based RGD
integrin ligands as non-peptide molecular imaging probes for
angiogenesis
Francesca Bianchini a, , Pierangelo Fabbrizzi b, , Gloria Menchi b,c, Silvia Raspanti d, Anna Bottoncetti e,
Alessandro Passeri c,d, Elena Andreucci a, Antonio Guarna b,c, Lido Calorini a, , Alberto Pupi c,d,e,
,
⇑
⇑
Andrea Trabocchi b,c,
⇑
a Department of Clinical and Experimental Biomedical Science ‘Mario Serio’, University of Florence, Viale Morgagni 50, 50134 Florence, Italy
b Department of Chemistry ‘Ugo Schiff’, University of Florence, Via della Lastruccia 13, 50019 Sesto Fiorentino, Florence, Italy
c Interdepartmental Center for Preclinical Development of Molecular Imaging (CISPIM), University of Florence, Viale Morgagni 85, 50134 Florence, Italy
d Department of Clinical and Experimental Biomedical Science ‘Mario Serio’, University of Florence, Largo Brambilla 3, 50134 Florence, Italy
e Azienda Ospedaliera Universitaria Careggi (AOUC), Largo Brambilla 3, 50134 Florence, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Taking advantage of click chemistry, we synthesized triazole-containing RGD peptidomimetics capable of
binding to
vb3 integrin with diverse potency, and selected 125I-labeled compounds proved to interact
in vitro and in vivo with
vb3 integrin expressed by melanoma cells. Two 125I-compounds containing
either 2-aminobenzimidazole or 2-aminopyridine groups as the arginine bioisostere with the capacity
Received 11 October 2014
Revised 22 December 2014
Accepted 28 December 2014
Available online 13 January 2015
a
a
to selectively bind cells of highly expressing
imaging studies.
a
vb3 melanoma xenografts were found using micro-SPECT
Keywords:
Integrins
Ó 2015 Elsevier Ltd. All rights reserved.
Molecular imaging
Cancer
Angiogenesis
Amino acids
1. Introduction
action at the molecular level and in obtaining information such
as receptor density and occupancy.5,6 Promising anti-angiogenic
treatments for cancer therapy have become well-established in
recent years. Angiogenesis, consisting of the generation of new
capillaries from existing blood vessels, is a common process asso-
ciated to several pathological processes, including the growth of
tumor tissues. Angiogenesis is a multistep process in which the
proteolytic degradation of the basal membrane allows for loosen-
ing the endothelial cells adhesion and their proliferation and
migration. Final re-adhesion of newly-formed endothelial cells
gives rise to new capillary vessels. Anti-angiogenic therapy is fun-
damental in blocking endothelial cells growth and inducing their
necrosis, resulting in the reduced endothelial network to starve
the tumor mass in term of nutrient and oxygen supply.
During the last decade medical imaging technologies experi-
enced a remarkable growth, and nowadays occupy a central func-
tion in clinical oncology.1 In this view, the use of biomarkers for
molecular imaging in pre-clinical studies has been proven useful
in accelerating the development of targeted cancer therapies.
Among all molecular imaging techniques, positron emission
tomography (PET) and single-photon emission computed tomogra-
phy (SPECT) played an increasing role in pre-clinical investigations,
taking advantage of the high sensitivity and the specificity of these
approaches.2 Indeed, molecular imaging studies using radiophar-
maceuticals are oriented towards receptor-based studies,3,4
because this approach is helpful in confirming the mechanism of
Integrins are transmembrane heterodimeric glycoprotein
receptors consisting of various combinations of
a and b subunits.
⇑
These receptors regulate many important biological processes such
as cell adhesion, migration and invasion, cell growth and differen-
tiation, and apoptosis. The role of integrins in the regulation of
angiogenesis is an essential feature in tumor progression and met-
astatic dissemination.7–10 As biomarkers for molecular imaging of
Corresponding authors. Tel.: +39 055 4598207; fax: +39 055 4598900 (L.C.);
tel.: +39 055 7947527; fax: +39 055 4224392 (A.P.); tel.: +39 055 4573507; fax: +39
055 4574913 (A.T.).
E-mail addresses: lido.calorini@unifi.it (L. Calorini), a.pupi@dfc.unifi.it (A. Pupi),
andrea.trabocchi@unifi.it (A. Trabocchi).
These authors contributed equally to the work.
http://dx.doi.org/10.1016/j.bmc.2014.12.065
0968-0896/Ó 2015 Elsevier Ltd. All rights reserved.

F. Bianchini et al. / Bioorg. Med. Chem. 23 (2015) 1112–1122
1113
angiogenesis are needed to monitor the progress of the therapeutic
the radionuclide therapy using 131I (emitting b- and
easily assessed.
c-rays) can be
treatment at the molecular level, efforts have been dedicated to
integrins as selective targets in molecular imaging approaches to
set up tailored anti-angiogenic treatments. Accordingly, several
in vivo imaging probes for PET and SPECT, for optical imaging,
and for magnetic resonance imaging (MRI) have been developed.11
Cilengitide is a RGD (arginine–glycine–aspartic acid) mimetic inte-
grin inhibitor representing a relevant entry in the panorama of
molecular imaging agents for defining new tumor therapies.
Although a number of RGD peptidomimetics have been reported
in the literature,12–15 much effort has been focused on the applica-
tion of Cilengitide as a chemical probe for the molecular imaging of
angiogenesis.
Our previous contribution in this field consisted of developing a
Cilengitide-like RGD cyclopentapeptide containing the cyclic amino
acid (R)-morpholine-3-carboxylic acid,16,17 and the successful eval-
uation of this molecule as a 125I-labeled SPECT imaging agent for
angiogenesis in a melanoma xenograft murine model.18 In view of
advancing the in vivo study of RGD radioligands, we became inter-
ested in small molecule radiotracers possessing a non-peptide char-
acter. Accordingly, we recently reported the synthesis and integrin
binding evaluation of triazole peptidomimetics of the RGD sequence
by means of copper-catalyzed azide–alkyne cycloaddition
(CuAAC).19 One compound of this class showed high binding capac-
In this work, we report the modification of compound 1 (Fig. 1) by
replacing the phenyl ring with a phenol group or by alkylating the
guanidine moiety with a tyramine appendage, in order to introduce
an iodine atom on the aromatic ring. This project was undertaken
with the aim of assessing whether these RGD peptidomimetics,
appropriately labeled, could be suitable for SPECT/PET molecular
imaging using iodine radionuclides and in order to explore the struc-
ture–activity relationship of the compounds in vivo. The study was
conceived by modulating the arginine bioisostere and the position
of the radioisotope, namely on the aromatic ring of the azide frag-
ment or using a tyramine moiety installed on the guanidine group.
The modulation of the guanidine bioisostere was studied to under-
stand the role of such chemical moiety in some in vitro biological
characteristics of tumor cells directed by integrins and in the phar-
macokinetics and pharmacodynamics profile of the radioligand.
2. Results and discussion
2.1. Synthesis
Triazole-based RGD ligands were achieved by combining azides
and alkynes (see Supplementary data for their preparation) in the
Cu-catalyzed azide–alkyne cycloaddition, followed by acid-mediated
hydrolysis of the protecting groups of side chain isosteres
(Schemes 1 and 2). CuP(OEt)3I proved to work smoothly in the
Cu-catalyzed reaction, as compared to previous application of
other copper salts,19 resulting in the preparation of side-chain pro-
tected (Pg) RGD mimetics Pg-6–Pg-9 in 55–89% yields. Subsequent
deprotection in 3 M HCl furnished the corresponding RGD mimet-
ics 6–9 in pure form. The synthesis of compound 7 (Scheme 1)
required a different hydrolysis procedure, as the treatment with
3 M HCl resulted in product degradation. Specifically, the removal
of the protecting groups was achieved using a 1:1 mixture of
ity for avb3/avb5 integrins, thus we envisaged the application of this
molecule as a novel molecular imaging agent by exploring diverse
positions for labeling with 125I radioisotope. The use of 125I for
pre-clinical molecular imaging studies is of special interest, because
it is a readily available radionuclide commonly used in molecular
biology, and many 125I-labeled radiopharmaceuticals are commer-
cially available. Also, iodine-labeled compounds represent a valu-
able tool for translating pre-clinical results up to humans because
the substitution of 125I with 123I provides a probe suitable for SPECT
imaging in humans. Furthermore, molecular imaging by PET can be
accomplished with the positron emitter 124I, and the translation to
H
N
CO2H
H2N
HN
N
N
O
N
N
H
1
R3
R1 Boc
N
N
+
O
X
N3
CO2Me
R2
N
H
alkyne containing Arg mimetic
azide containing Asp mimetic
R1-R2=(CH)3, Ph
R3=OH
X= N-Boc,CH
R1= (CH2)2(4'-OH)Ph
R2=H, R3=H, X= N-Boc
1.click chemistry
2.radiolabeling
HO
125I
NH
H
N
N
CO2H
N
N
O
Arg
N
H
NH
mimetic
125I
H
N
N
CO2H
N
N
O
Arg
mimetic
N
H
Figure 1. Design of RGD peptidomimetic ligands as SPECT imaging probes containing 125I.

1114
F. Bianchini et al. / Bioorg. Med. Chem. 23 (2015) 1112–1122
OH
HO
Boc
N
R
HN
HN
11
O
Boc
O
N
N
N
H
N
HN
H2N
CO2H
N3
CO2Me
N
H
N
H
1. CuP(OEt)3I
2. 3M HCl
10
6: R = H, 89%
2: R = I
CHL-T, NaI
CHL-T, Na125
Boc
N
Boc
I
Boc
N
: R = 125I, 49%
125I-2
N
N
N
12
13
1. CuP(OEt)3I
2. 1:1 DCM-TFA
HO
1. CuP(OEt)3I
2. 3M HCl
R
HO
R
O
N
N
H
N
H
N
CO2H
N
N
H
O
N
N
N
H
N
N
N
CO2H
N
H
7: R = H, 57%
3: R = I
CHL-T, NaI
CHL-T, Na125I
125I-3
: R = 125I, 41%
8: R = H, 55%
4: R = I
CHL-T, NaI
CHL-T, Na125I
125I-4
: R = 125I, 38%
Scheme 1. Synthetic approach for RGD mimetic radioligands 2–4 containing either guanidine or its bioisosteres.
Boc
N
1. CuP(OEt)3I
2. 3M HCl
N
+
O
HO
N3
CO2Me
HN
82%
N
H
Boc
15
14
O
O
N
N
N
N
N
Chloramine-T
Na125I or NaI
H
N
H
N
HN
HN
HN
CO2H
CO2H
N
N
H
N
H
HN
R5
5
: R5=I
9
OH
OH
125I-5: R5=125I, 48%
Scheme 2. Synthetic approach for the RGD mimetic radioligand 5 containing the tyramine linkage.
TFA–dichloromethane instead of an aqueous system, resulting in a
clean hydrolysis to give 7.
(Table 1).19 IC50 values showed all iodinated compounds possessing
a higher binding affinity toward
corresponding precursors. The
a
vb3 integrin as compared to their
The iodination procedure consisted of the direct labeling with
NaI (1:1 molar ratio) using chloramine-T (CHL-T) as the oxidizing
reagent in a phosphate buffered saline solution (PBS). The reaction
was quenched with sodium metabisulfite, and the crude product
was purified by semipreparative HPLC to give the corresponding
iodinated compounds 2–5 in pure form. The same approach was
avb3 inhibition curves of all RGD
mimetics but 6, 2 and 3 were better described by a two-sites model
(Table 1), as showed by Hill slope values different from unity, as a
consequence of binding of the peptidomimetic compound with
two conformational states of integrins, in agreement with other
ligands already reported in the literature.20–24 Compound 6 and
its iodinated analogue 2 showed a one-site binding profile and bind-
then applied for radiolabeling using Na125 125
I in 1:18 molar ratio
I
with respect to RGD mimetic precursors 6–9 to favor the monoio-
dination. The crude reaction mixture was purified by HPLC, and the
fractions were collected through radioactivity monitoring with a
ing affinity in the low micromolar range (2.1 and 0.66
lM, respec-
tively). Compounds 3 and 4, proved to bind to vb3 integrin with
a
higher affinity in the nanomolar range and according to a one-site
and two-sites models, respectively, suggesting an important role
of the guanidine isostere towards binding activity. Compounds 9
and 5 showed divergent ligand capacity, as the former demon-
c-counter, resulting in a clean iodination reaction with a radio-
chemical yield of 38–49% for compounds 125I-2–125I-5.
2.2. Integrin binding assays
strated poor binding to
to the high-affinity conformation but quite poor binding to the
low-affinity conformational state of vb3 integrin. These results sug-
gested that alkylation of the guanidine group produced a significant
modification of the binding affinity towards vb3 integrin as
avb3 integrin, and the latter good binding
Solid-phase assays were applied to test the ability of RGD pepti-
a
domimetics 6–9 and their corresponding iodinated compounds 2–5
to compete with 125I-echistatin for binding to pure
avb3 integrin
a

F. Bianchini et al. / Bioorg. Med. Chem. 23 (2015) 1112–1122
1115
Table 1
Inhibition of 125I-echistatin specific binding to purified human integrin proteins
avb3 by triazole-containing RGD peptidomimetics 6–9 (right) and their corresponding iodinated
derivatives (2–5)
Iodinated Compd
IC50
Hill coeff.
Compd
—
IC50
—
Hill coeff.
—
1
High: 16.4 5.3 nM
Low: 3.4 1.6
2.1 1.3
37 11 nM
N/A
ꢀ0.38
lM
6
7
8
lM
ꢀ0.79
ꢀ0.75
—
2
3
4
660 110 nM
64 11 nM
High: 0.21 0.16 nM
Low: 630 110 nM
ꢀ1.56
ꢀ0.96
ꢀ0.57
(100% inhib. at 10
86% inhib. at 1.0
lM
M)
l
9
High: 310 220 nM
Low: no convergence
ꢀ0.25
5
High: 41 12 nM
Low: 55 12 lM
ꢀ0.16
Binding data of compound 1 are added as a referencea.
a
High and low IC50 values refer to two-sites binding model.
compared to the introduction of the other moieties as found in com-
pounds 2–4, resulting in a switch of binding affinity from nano- to
micromolar range for tyramine-containing RGD mimetics. This
binding affinity profile was further confirmed by cell adhesion
assays.
substrates. Interestingly, the presence of the iodine atom in 3 and 4
significantly augmented the inhibitory ability of the two com-
pounds on living cells adhesion, in agreement with
binding assays.
avb3 integrin
2.4. Molecular modeling
2.3. Cell biology assays
In order to understand the differences in the binding mode of
selected iodinated-RGD peptidomimetics 2, 3 and 4 to vb3 inte-
grin, a docking simulation was carried out. Compound 5 was not
The candidate RGD peptidomimetic ligands were tested for
their integrin binding affinity towards M21 human melanoma
cells. Cytofluorimetric analysis revealed that M21 human mela-
a
considered for this study having ascertained its poor affinity
noma cells express high levels of
95% positive cells) and low levels of
(25–28% positive cells) (data not shown). We demonstrated the
capacity of RGD peptidomimetics 6–9 and their corresponding
iodinated compounds 2–5, used at a final concentration of 10,
a
vb3 integrin heterodimer (92–
towards
by c[RGDf(Me)V], also known as Cilengitide (Fig. 3, red structure),
and the extracellular fragment of vb3 integrin (PDB code: 1L5G)
avb3 integrin. The crystal structure of the complex formed
avb5 integrin heterodimer
a
provides a general mode of interaction between the integrin and
its ligands.25
1.0, 0.1 or 0.01
l
M, to inhibit the binding of M21 melanoma cells
In particular, Asp carboxylate and Arg guanidinium moieties of
RGD mimetic ligands are key structural elements for receptor rec-
ognition, which display their binding mode through the carboxyl-
ate group interacting with the metal ion-dependent adhesion site
(MIDAS) of the Mn2+ ion and Ser121/Ser123, and the Arg guanidi-
nium group being responsible for salt bridge interactions with
Asp218 and Asp150 side chains. Additional ligand–receptor con-
to various RGD-containing substrates, such as vitronectin, fibro-
nectin or osteopontin.
The analyses were performed in the presence of 2 mM MnCl2, in
order to switch integrins of tumor cells into an activated form. The
results shown in Figure 2, panel A, indicated that compound 9 and
the iodinated compound 5 did not inhibit melanoma binding to
RGD-containing substrates, demonstrating that the alkylation of
the guanidine group with tyramine play a detrimental effect
towards binding affinity. Compound 6 did not inhibit the binding
of M21 melanoma cells to vitronectin and fibronectin, while weakly
inhibiting the adhesion to osteopontin when used at a higher dose.
The iodinated compound 2, at the higher dose, weakly inhibited
M21 adhesion to fibronectin and osteopontin but not to vitronectin,
and the inhibition significantly differed from the non-iodinated
compound only in the case adhesion to fibronectin (Fig. 2, panel
B). Compounds 7 and 3 inhibited M21 adhesion to vitronectin at
the higher doses and the inhibition was completely abolished at
lower doses. Also, they inhibited the M21 adhesion to fibronectin
and osteopontin at high levels with the higher doses and this inhi-
bition was maintained even when used at lower doses (Fig. 2, panel
C). Compounds 8 and 4 expressed an inhibitory ability profile on
tumor cell adhesion that is comparable to that of compound 7. In
particular, the iodinated compound 4 showed a significant inhibi-
tory activity on fibronectin and osteopontin adhesion at the lower
concentration, as compared to the corresponding non-iodinated
compound 8 (Fig. 2, panel D). These results suggest that compounds
6 and 9 and their iodinated analogues 2 and 5, which do not display
a sufficient RGD binding inhibition of tumor cell adhesion, in agree-
ment with competition assay results as reported in Table 1, could
not be useful for in vivo analysis. On the contrary, compounds 7
tacts occur between Tyr122 in hydrophobic
p-stacking and
Asn215. Autodock 4.026 was used for molecular docking calcula-
tions to evaluate the binding energies of selected conformations
of RGD mimetic ligands 2–4. Docked conformations were analyzed
by taking into account the binding interactions as observed in the
crystal structure of the bound ligand–protein complex. Docking
calculations of all RGD peptidomimetics 2–4 resulted in a cluster
of conformations characterized by key interactions with Asp218
and the MIDAS site of the integrin, thus confirming these interac-
tions being necessary for the molecular recognition of Arg–Gly–
Asp-containing ligands (Fig. 3). Specifically, the COOH group was
found in the MIDAS site and interacting with Ser121, and the guan-
idinium group, or its isosteres for 3 and 4, undergoing a salt-bridge
interaction with Asp218 and Asp150.
The main conformation of 2 showed two
p-stacking interac-
tions experienced by the -Tyr aromatic ring facing Tyr-122, and
D
the triazole ring interacting with Tyr178 (Fig. 3, top). This binding
profile was in agreement with the ligand properties of Cilengitide
with the exception of the characteristic
p-stacking interactions
between 2 and Tyr178. A different binding mode geometry was
assessed for compound 3 as a result of two cation–
p interac-
tions27,28 experienced by Arg214 and the 30-iodophenol moiety of
3, and between Tyr178 and the benzimidazole group. The latter
interaction replaced the triazole ring in the
p-interaction with
and 8, which display a quite comparable affinity for the
avb3 inte-
Tyr178 (Fig. 3, middle). Compound 4 showed a binding mode sim-
grin receptor, could be used successfully for in vivo imaging studies,
ilar to 2, consisting of a preferred salt bridge interaction of the
due to their inhibitory activity on M21 tumor cell adhesion on RGD
2-aminopyridyl moiety with Asp150 and a compact
p-stacking

1116
F. Bianchini et al. / Bioorg. Med. Chem. 23 (2015) 1112–1122
Figure 2. Percentage of inhibition of integrin-mediated M21 cell adhesion to vitronectin, fibronectin and osteopontin, by compounds 9 and 5 (panel A, black and grey,
respectively), by compounds 6, and 2 (panel B, black and grey, respectively), by compounds 7 or 3 (panel C, black and grey, respectively) and by compounds 8 or 4 (panel D,
black and grey, respectively). Experiments were conducted in triplicate and repeated at least three times. Data are presented as means SEM from three independent
experiments, statistical analysis was conducted between the values of the same dose and class of compound, The asterisks indicate a significant difference at p <0.05 as
compared to the corresponding non-iodinated compounds.
interaction between the triazole ring and Tyr178 (Fig. 3, bottom).
Thus, the modulation of the guanidine isostere proved to affect
lowed by HPLC purification. Fractions containing the chemically
pure 125I-compounds, as determined with a
-counter, were pooled
c
the binding mode towards
a
vb3 integrin at the molecular level.
together (38–49% radiochemical yield) and used for in vivo experi-
ments after solvent removal and solubilization in a physiological
saline solution. Biodistribution studies of bioactive 125I-2–125I-4
compounds were performed by micro SPECT-CT imaging. Figure 4,
upper panel, shows transaxial slices of tumor bearing mice iv
injected with 125I-2–125I-4 compounds, taken at the tumor level.
Compound 125I-5 did not show any activity in vivo either at
30 min post-injection or at later acquisitions, as expected (data
not shown). Normalized activity concentration (nACt) in different
organs has been quantified in the liver and kidneys as most
Also, the binding mode of 3 and 4 with respect to Cilengitide sug-
gested the interaction with Tyr122 being not crucial for ligand
affinity, and showing a different binding asset of the heterocyclic
guanidine bioisosteres with respect to Asp218 and Asp150.
2.5. Imaging of melanoma xenografts
Radiolabeling of compounds 6–9 was achieved in a solution of
PBS containing Na125I, and chloramine-T as the oxidizing agent, fol-

F. Bianchini et al. / Bioorg. Med. Chem. 23 (2015) 1112–1122
1117
centrated in the kidneys at 30 min post injection (Fig. 4), while at
later acquisitions (60 and 120 min) the activity was almost com-
pletely lost. The kinetics of this elimination pathway was found to
be extremely rapid. A weak activity was found at 30 min p.i. associ-
ated with the tumor, when normalized to muscular tissue was rap-
idly cleared, as demonstrated by the absence of any activity in the
later acquisitions.
Interestingly, the insertion of a benzimidazole as arginine
mimetic portion in compound 125I-3 (Fig. 4), resulted in an increase
of activity found in the liver and a significant decrease of activity
found in the kidneys, suggesting reduced renal elimination kinetics
as compared to compound 125I-2, and accounting for a longer per-
sistence in the circulatory system. Finally, the activity associated
with the tumor mass was found to increase as compared to com-
pound 125I-2 during the observation time. Compound 125I-4
showed a strong involvement of hepatobiliary and renal elimina-
tion pathways since these kinetics were similar but more intense
compared to those of compound 125I-3. Interestingly, compound
125I-4 showed a faster time-dependent increase of the activity in
the tumor mass, which was higher as compared to compound
125I-3 (Fig. 4). These data were found to be in agreement with their
receptor affinity (Table 1) and their behavior in the inhibition of
melanoma cells adhesion (Fig. 2). Differences in the metabolic pat-
terns of excretion between compound 125I-2 and the other two
ligands, 125I-3 and 125I-4, were likely related to the arginine
mimetic portion, as the presence of the benzimidazole and of the
aminopyridine moiety contributed to more lipophilic character
and less basic behavior of the molecule, ultimately resulting in a
longer circulation time. However, 24 h post-injection animals were
analyzed and no significant residual signal was found for any of the
radioligands. Compounds 125I-3 and 125I-4 reached a significant
uptake within the tumor mass when data were normalized to
the liver (Table 2), as a key organ of the reticuloendothelial system,
due to the low tumor-to-background contrast in the liver.
Displacement experiments were conducted to ascertain the
binding specificity of the radiolabeled compounds (Fig. 5). These
studies were chosen instead of the simple use of a low expressing
a
vb3 cell line, since integrins are known to be expressed not only
by tumor cells and endothelial cells,29 but also by other stromal
cells such cancer-associated fibroblasts.30 Hence, saturation with
the cold compound would interfere with
avb3 expressing cells of
a different origin, thus resulting in a lower background signal
and a more reliable evaluation of the binding specificity.
As shown in Figure 5, tumor associated activity in mice injected
with compound 125I-3 was reduced after cold compound pre-treat-
ment. Interestingly, the displacement of compound 125I-4 by the
corresponding cold compound was more significant, suggesting a
better in vivo ligand binding specificity.
3. Conclusions
a
vb3 integrin-targeted radiotracers are putatively able to reveal
the presence of a tumor mass and to monitor the growth and dif-
fusion of tumor cells, and to anticipate the efficacy of an antiangio-
genic treatment. Thus, they are of great interest as molecular
probes for SPECT and PET molecular imaging. In this view, we
developed RGD-peptidomimetics using click chemistry, and a first
selection of iodinated molecules was evaluated for their in vitro
and in vivo capacity to inhibit the adhesion of human melanoma
cells to RGD substrates, such as vitronectin and fibronectin. Molec-
ular modeling calculations suggested a diverse binding modality
depending upon the guanidine bioisostere present in the RGD pep-
tidomimetic. Indeed, the optimization of the arginine mimetic por-
tion of the molecule was found crucial in modulating the in vivo
profile, and SPECT images clearly showed that three of the four
selected 125I-compounds have the capacity to monitor the growth
Figure 3. RGD mimetic ligands (top: 2 in yellow, middle: 3 in cyan, bottom: 4 in
green) docked into the binding region of vb3 integrin, highlighting protein residues
a
(magenta) that form key interactions: Asp218/Asp150 versus guanidine or isostere,
Mn2+ versus COOH group, Tyr122 (compound 2) or Arg214 (compound 3) versus 3-
I-phenol moiety, Tyr178 and triazole (compounds
2 and 4) or benzimidazole
(compound 3). Coordinates of reference compound Cilengitide, c[RGDf(Me)V] (data
from PDB: 1L5G) are shown in red for comparison. Non-polar hydrogen atoms are
omitted for clarity.
representative organs of the hepatobiliary and urinary excretion
pathways, and in tumor tissue (Fig. 4, lower panel). Bladder was
omitted since the organ has a rapid rate of filling and emptying.
Compound 125I-2 showed a sudden distribution which mostly con-

1118
F. Bianchini et al. / Bioorg. Med. Chem. 23 (2015) 1112–1122
Figure 4. Biodistribution studies of compounds 125I-2, 125I-3, 125I-4 in human melanoma xenografts. Upper panel: SPECT/CT transaxial slices taken at the tumor level (arrow)
at different time points of observation, representative images of one mouse (n = 3). Co-registered CT images were acquired at 10 min and SPECT images were acquired at 30,
60 and 120 min after tail vein injection of compounds 125I-2, 125I-3, 125I-4, (9.25 MBq in 0.2 mL of saline). Lower panel: distribution analysis of 125I-2, 125I-3, 125I-4 in mice at
different time points; data are expressed as mean SE of different normalized activity concentration (nACt) of target tissue from three different animals. The arrows show the
area where tumor is present.
Table 2
Uptake efficiency normalized to the liver for compounds 125I-2, 125I-3 and 125I-4
nAC values normalyzed to liver
30 min
60 min
120 min
125I-2
125I-3
125I-4
n.d.
n.d.
n.d.
n.d.
0.46 0.03
0.37 0.05
n.d.
0.54⁄ 0.02
0.58⁄⁄ 0.03
Significantly different (⁄p <0.05 and ⁄⁄p <0.005).
of a melanoma xenograft. Specifically, we ascertained that benz-
imidazole and aminopyridine moieties as guanidine bioisosteres
produced a rapid uptake in the tumor mass, as evaluated in com-
parison to background activity in the muscle and in the liver.
Moreover, such compounds resulted in longer permanence of
the radiotracer in the tumor tissue. The uptake kinetics of these
two compounds in the tumor mass showed an increasing trend,
as compared to other radio-halogenated cyclic-RGD peptidomi-
metic showing a rapid uptake and excretion.31 The new com-
pounds presented in this study demonstrated a site specific
receptor interaction, according to the results obtained in displace-
ment experiments. Nevertheless, compounds 125I-3 and 125I-4 will
need further investigation and structure refinement in a hit-to-
lead process in order to achieve promising RGD peptidomimetic
markers for molecular imaging in early detection of cancer lesion
and for monitoring the antiangiogenic therapeutic response.
Figure 5. Displacement experiments in tumor bearing mice (n = 3) were performed
three days after saturation experiments in the same mice. Displacements were
achieved with an excess of the ‘cold’ ligand (18 mg/kg) injected 10 min before the
corresponding radiolabeled tracer. Data are expressed as mean SE of different
normalized activity concentration (nACt) of the target tissue from three different
animals at different time points p.i. Discrepancies between nACt of different
compounds at each time point were significant ⁄p <0.05. Differences between nACt
of saturation versus displacement were significant at each time point with ⁄p <0.05
for compound 125I-3 and ⁄⁄p <0.001 for compound 125I-4.
4. Experimental section
4.1. General
purifications were performed using Merck silica gel (40–63 lm),
or via an automated Biotage Isolera system (SNAP silica cartridges).
TLC analyses were performed on Merck silica gel 60 F254 plates.
Elemental analyses were recorded on a Perkin Elmer 240 CHN
1H NMR and 13C NMR spectra were recorded on a Varian Gemini
200 or a Varian Gemini 300. Flash column chromatography (FCC)

F. Bianchini et al. / Bioorg. Med. Chem. 23 (2015) 1112–1122
1119
Analyzer. ESI mass spectra were recorded on a Thermo LCQ-Fleet. All
commercially available reagents and solvents were purchased from
Sigma–Aldrich and used as received, unless otherwise specified.
Analytical HPLC purity tests and semipreparative HPLC purifications
of radioligands were performed on a Beckman-Coulter System
allowed us to obtain the final product 8 as a white solid in quantita-
tive yield. 1H NMR (300 MHz, D2O) d 7.62 (s, 1H); 7.60–7.56 (m, 1H),
7.52–7.48 (m, 1H), 7.07 (d, J = 8.4 Hz, 2H), 6.71–6.66 (m, 3H), 6.65–
6.61 (m, 1H), 5.09–4.95 (m, 3H), 3.16 (t, J = 6.9 Hz, 2H), 2.77–2.73
(m, 2H), 2.67 (t, J = 6.9 Hz, 2H), 1.91–1.81 (m, 2H). 13C NMR
(50 MHz, D2O) d 173.6, 165.5, 154.1, 151.5, 145.6, 142.3, 133.6,
130.9, 127.9, 126.9, 124.1, 114.6, 112.1, 111.3, 51.5, 49.3, 44.7,
39.7, 39.0, 25.8, 20.6. ESI-MS m/z 425.2 [(M+H)+, 100]. Anal. Calcd
for C21H24N6O4: C, 59.42; H, 5.70; N, 19.80. Found: C, 59.55; H,
5.76; N, 19.66. HPLC: tR = 12.80 min, 99% purity.
GOLD (column Grace Discovery Sciences Alltima C18 10 lm;
250 ꢁ 10 mm), using H2O/CH3CN gradient eluant buffered with
0.1% TFA (flow rate 2.5 mL/min, k = 254 nm, gradient eluant:
CH3CN 10%/5 min, CH3CN 10–90%/25 min). HPLC analysis was used
to determine purity and all tested compounds possessed >95%
purity.
4.6. (S)-3-(2-(4-(3-(3-(4-hydroxyphenethyl)guanidino)propyl)-
1H-1,2,3-triazol-1-yl)acetamido)-3-phenylpropanoic acid
hydrochloride (9)
4.2. General procedure (A) for the synthesis of RGD mimetic
ligands 6–9
Azide (1.1 equiv) was dissolved in THF (0.15 M solution), then
alkyne (1 equiv) and CuP(OEt)3I (0.11 equiv) were added. The reac-
tion was performed under microwave irradiation at 120 °C for 1 h,
then THF was evaporated, the crude residue dissolved in EtOAc or
DCM, and washed with concentrated ammonia solution and brine.
The organic layer was dried over Na2SO4, evaporated under vac-
uum and purified via flash column chromatography. The protected
adduct was suspended in 3 M HCl (10 mL/mmol) and stirred at
30 °C overnight. The solution was evaporated to dryness in vacuo
to afford the corresponding product in pure form.
Following the general procedure (A), Pg-9 was obtained as a
colorless oil in 82% yield. The hydrolysis of Pg-9 in 3 M HCl allowed
us to obtain the final product 9 as a white solid in quantitative
yield. 1H NMR (300 MHz, D2O) d 7.47 (s, 1H), 7.22–7.14 (m, 5H),
6.90 (d, J = 8.4 Hz, 2H), 6.58 (d, J = 8.4 Hz, 2H), 5.15–5.05 (m, 1H),
5.01 (d, J = 3.6 Hz, 2H), 3.19 (t, J = 6.6 Hz, 2H), 2.87 (t, J = 6.6 Hz,
2H), 2.77 (d, J = 6.6 Hz, 2H), 2.55 (t, J = 6.6 Hz, 2H), 2.45 (t,
J = 7.2 Hz; 2H), 1.60–1.52 (m, 2H). 13C NMR (50 MHz, CDCl3) d
174.4, 166.5, 155.5, 154.1, 139.9, 130.2, 129.4, 128.9, 128.1,
127.0, 126.3, 115.4, 52.4, 50.7, 42.3, 40.1, 39.8, 33.5, 27.1, 21.3.
ESI-MS m/z 494.4 [(M+H+, 100]. Anal. Calcd for C25H32ClN7O4: C,
56.65; H, 6.09; N, 18.50. Found: C, 56.73; H, 6.18; N, 18.24. HPLC:
tR = 23.34 min, 98% purity.
4.3. (S)-3-(2-(4-(3-Guanidinopropyl)-1H-1,2,3-triazol-1-yl)aceta
mido)-3-(4-hydroxyphenyl)propanoic acid hydrochloride (6)
Following the general procedure (A), the side chain-protected
(Pg) Pg-6 was obtained as a pale yellow oil in 89% yield. The
hydrolysis of Pg-6 allowed us to obtain the final product 6 as a
white solid in quantitative yield. 1H NMR (300 MHz, D2O) d
7.63 (s, 1H), 7.10 (d, J = 7.8 Hz, 2H), 6.72 (d, J = 7.8 Hz, 2H), 5.2–
4.9 (m+s, 3H), 3.02 (t, J = 6.6 Hz, 2H), 2.78–2.75 (m, 2H), 2.63 (t,
J = 6.6 Hz, 2H), 1.78 (t, J = 6.6 Hz, 2H). 13C NMR (50 MHz, D2O) d
173.6, 165.2, 155.8, 154.3, 130.9, 127, 114.7, 51.9, 49.4, 39.4,
39.1, 26.3, 20.4. ESI-MS m/z 412.3 [(M+Na)+, 66], 390.33
[(M+H)+, 100]. Anal. Calcd for C17H24ClN7O4: C, 47.94; H, 5.68;
N, 23.02. Found: C, 48.11; H, 5.73; N, 22.90. HPLC: tR = 10.22 min,
98% purity.
4.7. General procedure (B) for the synthesis of 2–4
Deprotected RGD mimetic ligand was dissolved in 1:1 H2O–
MeOH mixture (0.05 M), and NaI (1 equiv), dissolved in the mini-
mum quantity of water, was added to the mixture. Then a solution
of chloramine-T, dissolved in the minimum quantity of water, was
added and the resulting mixture was stirred for 20 min at rt. After-
wards, an aqueous solution of sodium bisulfite (mixture with meta-
bisulfite, 1 equiv) was added, followed by water addition (starting
volume ꢁ 4) and the solution was washed with DCM. The aqueous
layer was evaporated and aliquots of the mixture were purified
via semipreparative HPLC giving suitable quantities of iodinated
product for 1H NMR and biological assays.
4.4. (S)-2-((3-(1-(2-((2-Carboxy-1-(4-hydroxyphenyl)ethyl)
amino)-2-oxoethyl)-1H-1,2,3-triazol-4-yl)propyl)amino)-1H-
benzo[d]imidazol-3-ium trifluoroacetate (7)
4.8. (S)-Amino((3-(1-(2-((2-carboxy-1-(4-hydroxy-3-iodophenyl)
ethyl)amino)-2-oxoethyl)-1H-1,2,3-triazol-4-yl)propyl)amino)
methaniminium trifluoroacetate (2)
Following the general procedure (A), Pg-7 was obtained as a
colorless oil in 57% yield. Following a different hydrolysis proce-
dure, Pg-7 (45 mg, 0.06 mmol) was dissolved in 2 mL of a 1:1 mix-
ture of DCM–TFA and stirred at room temperature 2 h. The solvent
was evaporated to yield the deprotected product 7 in quantitative
yield. 1H NMR (300 MHz, D2O) d 7.62 (s, 1H), 7.08–7.03 (s+d,
4+2H), 6.64 (dd, J = 6.6, 2.1 Hz, 2H), 5.08–5.00 (m, 3H), 3.19 (t,
J = 6.9 Hz, 2H), 2.76–2.66 (m, 4H), 1.90 (t, J = 6.9 Hz, 2H). 13C
NMR (50 MHz, D2O) d 173.7, 165.7, 154.1, 149.5, 146.0, 130.9,
128.1, 126.8, 123.9, 122.4, 114.5, 110, 51.3, 49.3, 44.6, 41.0, 39.1,
26.4, 20.7. ESI-MS m/z 464.3 [(M+H)+, 100]. Anal. Calcd for
Iodination of compound 6 was achieved according to general
procedure (B) as above to give pure 2 after HPLC purification. 1H
NMR (300 MHz, D2O) d 7.66 (s, 1H), 7.59 (s, 1H), 7.11 (d,
J = 8.4 Hz, 2H), 6.79 (d, J = 8.4 Hz, 2H), 5.10–5.03 (m, 3H), 3.04 (t,
J = 6.9 Hz, 2H), 2.78 (bd, J = 6.9 Hz, 2H), 2.66 (t, J = 7.2 Hz, 2H),
1.81 (t, J = 7.2 Hz, 2H). ESI-MS m/z 537.5 [(M+Na)+, 100], 516.1
[(M+H)+, 18]. HPLC: tR = 13.22 min, 97% purity.
4.9. (S)-2-((3-(1-(2-((2-Carboxy-1-(4-hydroxy-3-iodophenyl)
ethyl)amino)-2-oxoethyl)-1H-1,2,3-triazol-4-yl)propyl)amino)-
1H-benzo[d]imidazol-3-ium trifluoroacetate (3)
C25H26F3N7O6: C, 51.99; H, 4.54; N, 16.98. Found: C, 52.25; H,
4.64; N, 16.79. HPLC: tR = 15.91 min, 98% purity.
Iodination of compound 7 was achieved according to general
procedure (B) as above to give pure 3 after HPLC purification.
1H NMR (300 MHz, D2O) d 7.63 (s, 1H), 7.45 (d, J = 2.1 Hz, 1H),
7.05–6.99 (m, 5H), 6.62 (d, J = 8.7 Hz; 1H), 5.10–4.93 (m, 3H),
3.23–3.10 (m, 2H), 2.71 (t, J = 6.9 Hz, 4H), 1.92 (t, J = 6.9 Hz,
2H). ESI-MS m/z 612.14 [(M+Na)+, 24], 590.16 [(M+H)+, 100]. HPLC:
tR = 17.96 min, 97% purity.
4.5. (S)-3-(4-Hydroxyphenyl)-3-(2-(4-(3-(pyridin-2-ylamino)
propyl)-1H-1,2,3-triazol-1-yl)acetamido)propanoic acid
hydrochloride (8)
Following the general procedure (A), Pg-8 was obtained as a col-
orless yellow oil in 55% yield. The hydrolysis of Pg-8 in 3 M HCl

1120
F. Bianchini et al. / Bioorg. Med. Chem. 23 (2015) 1112–1122
4.10. (S)-2-((3-(1-(2-((2-carboxy-1-(4-hydroxy-3-iodophenyl)
ethyl)amino)-2-oxoethyl)-1H-1,2,3-triazol-4-yl)propyl)
amino)pyridin-1-ium trifluoroacetate (4)
defined as 125I-echistatin bound in the presence of an excess
(1 M) of unlabeled echistatin. After incubation at room tempera-
ture for 3 h, plates were washed three times with blocking/binding
l
buffer and counted in a Top-Count NXT microplate scintillation
Iodination of compound 8 was achieved according to general
procedure (B) as above to give pure 4 after HPLC purification. 1H
NMR (300 MHz, D2O) d 7.62 (s, 1H), 7.59–7.53 (m, 3H), 7.06 (dd,
J = 8.1, 2.1 Hz, 1H), 6.70 (d, J = 8.1 Hz, 1H), 6.66–6.58 (m, 3H),
5.05–4.97 (m, 3H), 3.12 (t, J = 6.9 Hz, 2H), 2.76–2.64 (m, 4H),
1.91–1.83 (m, 2H). ESI-MS m/z 551.2 [(M+1)+, 38], 439.3 (100).
HPLC: tR = 14.84 min, 96% purity.
counter (PerkinElmer Life Sciences, Boston, MA) using 200 lL/well
of MicroScint-40 liquid scintillation (PerkinElmer Life Sciences,
Boston, MA). Data are shown as means SD from three indepen-
dent experiments. IC50 values were determined by fitting binding
inhibition data by nonlinear regression using GraphPad Prism 4.0
Software Package (GraphPad Prism, San Diego, CA). Any possible
binding of displaced radioactive ligand to BSA was evaluated in
every assay by comparing the total bound value to the total added
value as blank controls, resulting in a deviation due to radioligand
binding to BSA of about 5%, which may be comprised within the
experimental errors.
4.11. (S)-((3-(1-(2-((2-Carboxy-1-phenylethyl)amino)-2-
oxoethyl)-1H-1,2,3-triazol-4-yl)propyl)amino)((4-hydroxy-3-
iodophenethyl)amino)methaniminium trifluoroacetate (5)
Iodination of compound 9 was achieved according to general
procedure (B) as above to give pure 5 after HPLC purification. 1H
NMR (300 MHz, D2O) d 7.50 (s, 1H), 7.41 (s, 1H), 7.41 (s, 1H),
7.29–7.11 (m, 5H), 6.92 (d, J = 8.4 Hz, 1H), 6.64 (d, J = 8.4 Hz, 1H),
5.15–5.00 (m, 3H), 3.25–3.15 (m, 2H), 2.87 (t, J = 6.9 Hz, 2H), 2.80
(d, J = 6.6 Hz, 2H), 2.60–2.51 (m, 2H), 2.44 (t, J = 6.9 Hz, 2H),
1.58–1.51 (m, 2H). ESI-MS m/z 620.2 [(M+H)+, 100]. HPLC:
tR = 25.34 min, 97% purity.
4.14. Cell lines and culture conditions
The M21 human melanoma cell line was obtained from the
American Type Culture Collection (ATCC, Rockville, MD). Mela-
noma cells were grown in Dulbecco’s modified Eagle medium, con-
taining 4500 mg/L glucose (DMEM 4500, GIBCO) supplemented
with 10% foetal calf serum (FCS) at 37 °C in a humidified incubator
containing 10% CO2. 5.0 ꢁ 105 melanoma cells were seeded in
100 mm Sarstedt dishes and propagated every 3 days by incuba-
tion with a trypsin–EDTA solution. Cultures were periodically
monitored for mycoplasma contamination. For the use in the
experiments, cells (passages 2–4) were grown to confluence in
plates coated with 1% bovine gelatin (Sigma, St. Louis).
4.12. Radiolabeling with Na125
I
A solution of PBS (100 lL,
l
L, 0.05 M) containing Na125I (50
5 mCi, 2.29 nmol), either compound 6, 7, 8, or 9 (10 L of a
l
0.05 M PBS solution, 41.2 nmol), and chloramine-T (1.88 L of a
l
freshly prepared 0.05 M PBS solution, 8.24 nmol) was prepared.
The reaction mixture was stirred at room temperature for 20 min
and then quenched by adding a solution of sodium metabisulfite
4.15. Cytofluorimetric assay
Cells were detached by gentle treatment with Accutase (Lonza),
(2.18
lL of a freshly prepared 0.05 M PBS solution, 8.24 nmol).
a 0.5 mM EDTA solution, washed, and incubated for 1 h at 4 °C in
The mixture was extracted with DCM (2 ꢁ 0.2 mL) and purified
the presence of anti-
integrin vb3, clone LM609, Millipore) and anti-
antibody (1 g/50 L, anti-integrin vb5, Santa Cruz 13588). Cells
were then washed and incubated for 1 h at 4 °C with a specific
secondary antibody, 5 g/mL goat antimouse IgG conjugated with
FITC (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Integrin-
Positive cells were analyzed at 488 nm on the flow cytometer
FACScan system (BD-FACS Canto).
a
vb3 monoclonal antibody (1
lg/50 lL, anti-
by HPLC, and the radioactivity of fractions collected every 30 s
a
avb5 monoclonal
was determined with a
c
-counter. Fractions containing the chem-
l
l
a
ically pure 125I-compound were pooled together and used for
in vivo experiments after solvent removal and solubilization in
physiological saline solution (38–49% radiochemical yield). 125I-
2: 49% radiochemical yield, HPLC: tR = 13.39 min, 99% radiochemi-
cal purity; 125I-3: 41% radiochemical yield, HPLC: tR = 18.17 min,
99% radiochemical purity; 125I-4: 38% radiochemical yield, HPLC:
tR = 15.01 min, 99% radiochemical purity; 125I-5: 48% radiochemi-
cal yield, HPLC: tR = 25.49 min, 99%.
l
4.16. Cell adhesion assay
Plates (96 wells) were coated with vitronectin (10
lg/mL), oste-
4.13. Solid-phase integrin binding assay
opontin (0.5 g/mL) or fibronectin (1 g/mL), by overnight incuba-
l
l
tion at 4 °C. Plates were washed with PBS and then incubated at
37 °C for 1 h with PBS–1% BSA. After being washed tumor cells
were counted and resuspended in serum free medium, and
exposed to compound (final concentration was 10 nM, 0.1, 1, or
125I-echistatin32 with a specific activity of 2000 Ci/mmol, was
purchased from Perkin Elmer, and integrin
avb3 from human pla-
centa was purchased from Millipore. Purified
a
vb3 vb3 integrin
a
was diluted in coating buffer [20 mM Tris (pH 7.4), 150 mM NaCl,
10 lM) at 37 °C for 30 min to allow the ligand–receptor equilib-
2 mM CaCl2, 1 mM MgCl2, 1 mM MnCl2] at concentrations of 500 or
1000 ng/mL. An aliquot of diluted integrin (100 lL/well) was
rium to be reached. Assays were performed in the presence of
2 mmol/L MnCl2. Cells were then plated (4–5 ꢁ 104 cells/well)
and incubated at 37 °C for 1 h. All the wells were washed with
PBS to remove the non adherent cells, and 0.5% crystal violet solu-
tion in 20% methanol was added. After 2 h of incubation at 4 °C,
plates were examined at 540 nm in a counter ELX800 (Bio TEK
Instruments). Experiments were conducted in triplicate and were
repeated at least three times. The values are expressed as% inhibi-
tion SEM of cell adhesion relative to untreated cells.
added to a 96-well microtiter plate (Optiplate-96 HB, PerkinElmer
Life Sciences, Boston, MA), and plates were incubated overnight at
4 °C, followed by washings with blocking/binding buffer [20 mM
Tris (pH 7.4), 150 mM NaCl, 2 mM CaCl2, 1 mM MgCl2, 1 mM
MnCl2, 1% BSA] and incubated at room temperature for additional
2 h. After rinsing twice with the same buffer, competition binding
assays were performed with a constant concentration of 125I-
echistatin (0.05 nM) and with concentrations of tested compound
ranging from 0.01 to 100 nM. All assays were performed in tripli-
cate in a final volume of 0.2 mL, each containing the following spe-
cies: 0.05 mL of 125I-echistatin, 0.04 mL of tested compound, and
0.11 mL of blocking/binding buffer. Nonspecific binding was
4.17. Small-animal SPECT and CT studies
The in vivo studies were conducted using a FLEX™ Triumph™
Pre-Clinical Imaging System (Trifoil-Imaging, Northridge, CA). The

F. Bianchini et al. / Bioorg. Med. Chem. 23 (2015) 1112–1122
1121
engrafted tumor tracer uptake was studied at different time-points
(at 30, 60 and 120 min p.i.) after tail vein injection. In vivo exper-
iments were conducted in accordance with national guidelines and
approved by the ethical committee of Animal Welfare Office of Ital-
ian Work Ministry and in agreement with the legal mandates and
Italian guidelines for the care and maintenance of laboratory
animals. Eight-ten week-old male SCID bg/bg mice (Charles River
Laboratories International) were kept under anesthesia and
injected subcutaneously with a suspension mixture of 1ꢁ106
time point. Comparison between two different radiotracers for DE
was also made using the multiple-comparison Student–Newman–
Keuls test, after demonstration of significant differences among
medians by non-parametric variance analysis, according to the
Kruskal–Wallis test. The level of significance was set at p <0.05.
4.19. Docking calculations
Automated docking studies were carried out using the Lamarck-
ian Genetic Algorithm (LGA) as a search engine implemented in the
Autodock 4.0.1 program.26 The AutoDockTools 1.4.5 (ADT) graphi-
cal interface34 was used to prepare integrin and ligands PDBQT
files. Coordinates of RGD mimetic ligands 2–4 were generated
using Spartan (version 5.147), and then energy-minimized through
the AM1 semi-empirical method to calculate the equilibrium
M21 melanoma cells resuspended in 100 lL of PBS diluted 1:1
with a Matrigel solution (BD Biosciences). Animals were kept
under pathogen-free conditions and allowed to feed freely. Tumor
growth was monitored every other day. Saturation experiments
were performed in order to visualize grafted tumors.
125I-compounds (each radiolabeled with 0.2–0.3 mCi in 200
lL of
physiological saline solution) were injected in the tail vein of
tumor bearing mice kept under isoflurane anesthesia. CT scan
was performed followed by the SPECT tomography at 30 min p.i.
with animals being kept under anesthesia on a warmed bed and
monitoring for breath frequency. Animals were ip injected with
geometry. The coordinates of
avb3 receptor were retrieved from
the Protein Data Bank (PDB code: 1L5G), and ligand–protein com-
plex was unmerged for achieving free receptor structure. Water
molecules were removed. For protein receptor and RGD mimetic
ligands 2–4, all hydrogen atoms were added, Gasteiger charges
were computed, and nonpolar hydrogen atoms were merged. A
charge value of +2.0 to each Mn atom of protein receptor was suc-
cessively added. Three-dimensional energy scoring grids of 0.375 Å
resolution and 40 ꢁ 40 ꢁ 40 Å dimensions were computed. The
center of the grid was set to be coincident with mass center of
ligands preliminary fitted on the X-ray structure of c[RGDf(Me)V]
150 lL of a solution of KI (4 mg/L) 1 h before injection of the differ-
ent compounds to avoid the accumulation in the thyroid gland of
the 125I atom possibly released from the molecule. The CT scan
was acquired with 50 kV and 320
tion factor was set to 2 resulting in an axial field of view of
59.2 mm. 512 projections, 500 s each, were acquired on a circular
orbit in 256 s with an acquisition matrix of 1184 ꢁ 1120, 115.6
lA tube settings. The magnifica-
l
lm
in the
avb3 complex (1L5G). A total of 50 runs with a maximum
pixel size. Having the animal bed in the same position, 64 SPECT
projections (20 s each) were acquired on a circular orbit with
radius of rotation of 40 mm and a 5-pinhole high-resolution
collimator (1 mm hole diameter, 75 mm focal distance) for a total
acquisition time of about 22 min. The acquisition matrix was
80 ꢁ 80, 1.5 mm pixel size. The energy window was set on the
125I photopeak ranging from 20 keV and 36 keV. Once the scanning
session was complete, animals were recovered in a housing unit
and monitored until awake.
of 2,500,000 energy evaluations were carried out for each ligand,
using the default parameters for LGA. Cluster analysis was per-
formed on docked results using a root–mean–square (rms) toler-
ance of 1.5 Å. Analysis of the binding mode, calculation of the
binding energy, and prediction of the binding activity of docked
conformations were carried out using Autodock plugin within
PyMol software.35
Acknowledgments
Giannetto Comis is acknowledged for technical assistantship on
the use of MicroPET/SPECT/CT instrumentation. Financial support
from University of Florence, Istituto Toscano Tumori, and MIUR
PRIN2010-2011, cod. 2010NRREPL, and Regione Toscana Regional
Council through POR CRO FSE 2007–2013 and UNIFI_FSE2012
funding actions is acknowledged. Ente Cassa di Risparmio is
acknowledged for financing the MicroPET/SPECT/CT instrument
and the Ente CRF grant (2012.0302) entitled ‘Melanoma a cattiva
prognosi: basi biologiche e possibili ricadute in termini di salute
pubblica e progettazione di biofarmaci’.
4.18. Data analysis of small-animal SPECT and CT studies
The CT image data set was reconstructed using FBP (Filtered
Back Projection) algorithm implementing an apodization low pass
filter devoted to statistical noise minimization. The reconstruction
matrix was 512 ꢁ 512 ꢁ 512, 180
lm pixel size. SPECT images
were reconstructed with OSEM (Ordered Subset Expectation Max-
imization) algorithm (8 subsets and 5 iterations) and a reconstruc-
tion matrix of 60 ꢁ 60 ꢁ 60, 1.5 mm pixel size. The fused images
were visually inspected to detect regional SPECT activities and
measured using the analysis software Vivid (Gamma Medica-Ideas,
Northridge, CA). Elliptical ROIs (Regions Of Interest) were
delineated by hand corresponding to a volume of approximately
0.3–0.5 mL for each organ. Absolute activity concentration in the
target (the tumor or the organ) (AACt) and in the reference region
(AACr) (muscular tissue) were measured.33 Data were also
decay-corrected to the time of injection and residual activity in
the syringe. Finally, AAC data for each organ were normalized to
activity (nACt) in muscular tissue and corrected for background
activity according to the formula:
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2014.12.065.
References and notes
1. Weissleder, R. Science 2006, 312, 1168.
2. Franc, B. L.; Acton, P. D.; Mari, C.; Hasegawa, B. H. J. Nucl. Med. 2008, 49, 1651.
3. Gomes, C. M.; Abrunhosa, A. J.; Ramos, P.; Pauwels, E. K. Adv. Drug Delivery Rev.
2011, 63, 547.
4. Willmann, J. K.; van Bruggen, N.; Dinkelborg, L. M.; Gambhir, S. S. Nat. Rev. Drug
Disc. 2008, 7, 591.
5. Wong, D. F.; Tauscher, J.; Gründer, G. Neuropsychopharmacology 2009, 34, 187.
6. Van Dort, M. E.; Rehemtulla, A.; Ross, B. D. Curr. Comput. Aided Drug Des. 2008,
4, 46.
7. Cabodi, S.; del Pilar Camacho-Leal, M.; Di Stefano, P.; Defilippi, P. Nat. Rev.
Cancer 2010, 10, 858.
AACt ꢀ AACr
nACt ¼
AACr
where nACt is normalized activity concentration in target tissue,
AACt and ACCr are the absolute activity concentrations of the target
tissue and of the reference tissue respectively.
The biodistribution data expressed as nACt ratios are reported
as mean plus SEM based on the results from three animals at each
8. Shattil, S. J.; Kim, C.; Ginsberg, M. H. Nat. Rev. Mol. Cell. Biol. 2010, 11, 288.
9. Desgrosellier, J. S.; Cheresh, D. A. Nat. Rev. Cancer 2010, 10, 9.

1122
F. Bianchini et al. / Bioorg. Med. Chem. 23 (2015) 1112–1122
10. Arnaout, M. A.; Mahalingam, B.; Xiong, J. P. Annu. Rev. Cell Dev. Biol. 2005, 21,
381.
11. Schottelius, M.; Laufer, B.; Kessler, H.; Wester, H. J. Acc. Chem. Res. 2009, 42,
969.
12. Onthank, D. C.; Liu, S.; Silva, P. J.; Barrett, J. A.; Harris, T. D.; Robinson, S. P.;
Scott Edwards, D. Bioconjugate Chem. 2004, 15, 235.
22. Takagi, J.; Petre, B. M.; Walz, T.; Springer, T. Cell 2002, 110, 599.
23. Humphries, M. J.; McEwan, P. A.; Barton, S. J.; Buckley, P. A.; Bella, J.; Mould, A.
P. Trends Biochem. Sci. 2003, 28, 313.
24. Shimaoka, M.; Springer, T. A. Nat. Rev. Drug Disc. 2003, 2, 703.
25. Xiong, J. P.; Stehle, T.; Zhang, R.; Joachimiak, A.; Frech, M.; Goodman, S. L.;
Arnaout, M. A. Science 2002, 296, 151.
13. Jang, B. S.; Lim, E.; Hee Park, S.; Shin, I. S.; Danthi, S. N.; Hwang, I. S.; Le, N.; Yu,
S.; Xie, J.; Li, K. C.; Carrasquillo, J. A.; Paik, C. H. Nucl. Med. Biol. 2007, 34, 363.
14. Neubauer, S.; Rechenmacher, F.; Beer, A. J.; Curnis, F.; Pohle, K.; D’Alessandria,
C.; Wester, H.-J.; Reuning, U.; Corti, A.; Schwaiger, M.; Kessler, H. Angew. Chem.,
Int. Ed. 2013, 52, 11656.
15. Monaco, A.; Michelin, O.; Prior, J.; Rüegg, C.; Scapozza, L.; Seimbille, Y. J. Label
Compd. Radiopharm. 2014, 57, 365.
16. Cini, N.; Trabocchi, A.; Menchi, G.; Bottoncetti, A.; Raspanti, S.; Pupi, A.; Guarna,
A. Bioorg. Med. Chem. 2009, 17, 1542.
26. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W. E.; Belew, R. K.;
Olson, A. J. J. Comput. Chem. 1998, 19, 1639.
27. Zacharias, N.; Dougherty, D. A. Trends Pharmacol. Sci. 2002, 23, 281.
28. Gallivan, J. P.; Dougherty, D. A. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 9459.
29. Weis, S. M.; Cheresh, D. A. Cold Spring Harb. Perspect. Med. 2011, 1, a006478.
30. Calorini, L.; Bianchini, F. Cell Commun. Signal. 2010, 8, 24.
31. Haubner, R.; Wester, H. J.; Weber, W. A.; Mang, C.; Ziegler, S. I.; Goodman, S. L.;
Senekowitsch-Schmidtke, R.; Kessler, H.; Schwaiger, M. Cancer Res. 2001, 61,
1781.
17. Sladojevich, F.; Trabocchi, A.; Guarna, A. J. Org. Chem. 2007, 72, 4254.
18. Bianchini, F.; Cini, N.; Trabocchi, A.; Bottoncetti, A.; Raspanti, S.; Vanzi, E.;
Menchi, G.; Guarna, A.; Pupi, A.; Calorini, L. J. Med. Chem. 2012, 55, 5024.
19. Trabocchi, A.; Menchi, G.; Cini, N.; Bianchini, F.; Raspanti, S.; Bottoncetti, A.;
Pupi, A.; Calorini, L.; Guarna, A. J. Med. Chem. 2010, 53, 7119.
20. Dijkgraaf, I.; Kruijtzer, J. A.; Frielink, C.; Soede, A. C.; Hilbers, H. W.; Oyen, W. J.;
Corstens, F. H.; Liskamp, R. M.; Boerman, O. C. Nucl. Med. Biol. 2006, 33, 953.
21. Zanardi, F.; Burreddu, P.; Rassu, G.; Auzzas, L.; Battistini, L.; Curti, C.; Sartori, A.;
Nicastro, G.; Menchi, G.; Cini, N.; Bottoncetti, A.; Raspanti, S.; Casiraghi, G. J.
Med. Chem. 2008, 51, 1771.
32. Kumar, C. C.; Nie, H.; Armstrong, L.; Zhang, R.; Vijay-Kumar, S.; Tsarbopoulos,
A. FEBS Lett. 1998, 429, 239.
33. Israel, O.; Front, D.; Hardoff, R.; Ish-Shalom, S.; Jerushalmi, J.; Kolodny, G. M. J.
Nucl. Med. 1991, 32, 1157.
34. Gillet, A.; Sanner, M.; Stoffler, D.; Olson, A. Structure 2005, 13, 483.
35. DeLano, W. L. The PyMOL Molecular Graphics System, Schrödinger, LLC (http://
www.pymol.org).