Development of Isoxazoline-Containing Peptidomimetics as Dual αvβ3 and α5β1 Integrin Ligands

Article abstract of DOI:10.1002/cmdc.201100372

Isoxazoline-containing peptidomimetics, designed to be effective α_vβ₃ and α₅β₁ integrin ligands, were synthesized through an original procedure involving N,O-bis(trimethylsilyl)hydroxyamine conjugate addition to alkylidene acetoacetates, followed by intramolecular hemiketalization. To mimic the RGD recognition sequence, basic and acidic terminal appendages were introduced, and the final products were tested in cell adhesion inhibition assays. All the synthesized compounds proved to be excellent ligands for both integrin receptors, and a strong influence on intracellular signaling and phosphorylation pathways was demonstrated by evaluation of fibronectin-induced phosphorylation of ERK. The molecular basis of the observed inhibitory activity was suggested on the results of docking experiments. Molecular copycats! Isoxazoline-containing peptidomimetics, designed to be effective α_vβ₃ and α₅β₁ integrin ligands, were synthesized by hydroxyamine conjugate addition to alkylidene acetoacetates followed by intramolecular hemiketalization. Cell adhesion assay results suggest a very effective ligand-receptor interaction, as confirmed by docking studies.

Full text of DOI:10.1002/cmdc.201100372

MED
DOI: 10.1002/cmdc.201100372
Development of Isoxazoline-Containing Peptidomimetics
as Dual avb3 and a5b1 Integrin Ligands
Alessandra Tolomelli,*[a] Luca Gentilucci,[a] Elisa Mosconi,[a] Angelo Viola,[a]
Samantha Deianira Dattoli,[b] Monica Baiula,[b] Santi Spampinato,[b] Laura Belvisi,[c] and
Monica Civera[c]
Isoxazoline-containing peptidomimetics, designed to be effec-
tive avb3 and a5b1 integrin ligands, were synthesized through
an original procedure involving N,O-bis(trimethylsilyl)hydroxya-
mine conjugate addition to alkylidene acetoacetates, followed
by intramolecular hemiketalization. To mimic the RGD recogni-
tion sequence, basic and acidic terminal appendages were in-
troduced, and the final products were tested in cell adhesion
inhibition assays. All the synthesized compounds proved to be
excellent ligands for both integrin receptors, and a strong in-
fluence on intracellular signaling and phosphorylation path-
ways was demonstrated by evaluation of fibronectin-induced
phosphorylation of ERK. The molecular basis of the observed
inhibitory activity was suggested on the results of docking ex-
periments.
Introduction
Development of novel tools for the early detection,[1] diagno-
sis,[2] and therapy[3] of cancer is one of the most important
goals in medicinal chemistry. The complex mechanism of cell–
cell and cell–matrix adhesion is crucial for developmental pro-
cesses, as such connections lead to cellular communication
and signaling. Among the broad spectrum of cell-adhesion
molecules, integrin receptors,[4] a large family of heterodimeric
transmembrane glycoproteins, play a fundamental role, and
thus they are involved in the pathogenesis of several diseases
such as atherosclerosis, osteoporosis, cancer, and a variety of
inflammatory disorders. The avb3 integrin is present at low
levels on healthy tissues, whereas it is overexpressed in certain
pathologies such as metastatic melanoma, late-stage glioblas-
toma, and breast and prostate tumors.[5] It is also preferentially
expressed on cancer blood vessels, mediating angiogenesis ini-
tiated by basic fibroblast growth factor (bFGF) or tumor ne-
crosis factor-a (TNF-a).[6] Another member of this class of re-
ceptors, a5b1 integrin, was first identified as an attachment
point for cell penetration by certain viruses and bacteria,[7] but
it has also recently been reported to play a facilitating role in
cancer cell invasion during migration through connective tis-
sues, through the generation of enhanced contractile forces.[8]
The a5b1 integrin has also been unambiguously identified as a
pro-angiogenic receptor, as antagonists to its extracellular
matrix partner fibronectin are able to block growth-factor- and
tumor-induced angiogenesis.[9] Simultaneous blockade of avb3
and a5b1 integrins was reported to inhibit bFGF-induced angio-
genesis,[10] and evidence of cross-talk between the two recep-
tors has been provided.[11] Therefore, the identification of syn-
thetic ligands that are able to engage with these cell-surface
receptors could aid in the design of imaging biomarkers for
early detection of cancer and assessment of therapy re-
sponse,[12] as well as in the engineering of cell-targeted anti-
cancer agents.[13]
Processes such as angiogenesis, which involve several mech-
anisms, can be addressed by targeting multiple pathogenic
pathways with a single compound.[14] This approach may pro-
vide enhanced efficacy because pharmacokinetic problems re-
lated to the administration of more than one drug can be
minimized. Antagonists to both avb3 and a5b1 integrins may
have a synergistic effect and could suppress tumor angiogene-
sis. Some selected examples of dual avb3 and a5b1 integrin in-
hibitors[15] are shown in Figure 1.
The design and synthesis of dual non-peptidic integrin in-
hibitors has generally been planned on the basis of the Arg-
Gly-Asp (RGD) motif, which is present in a wide range of extra-
cellular matrix (ECM) proteins.[16] X-ray crystallographic analysis
of the complex between avb3 integrin and the privileged syn-
thetic ligand cilengitide[17] shows that the recognition se-
quence binds to the receptor mainly through electrostatic in-
teractions with regions in the protein having opposite charges:
arginine interacts with two aspartic acid residues situated in
the a unit of the protein, while the carboxylate group of the
ligand binds to the metal-ion-dependent adhesion site
(MIDAS). No high-resolution structural information regarding
a5b1 integrin is currently available to understand the molecular
[a] Dr. A. Tolomelli, Prof. L. Gentilucci, Dr. E. Mosconi, Dr. A. Viola
Department of Chemistry “G. Ciamician”, University of Bologna
Via Selmi 2, 40126 Bologna (Italy)
E-mail: alessandra.tolomelli@unibo.it
[b] Dr. S. D. Dattoli, Dr. M. Baiula, Prof. S. Spampinato
Department of Pharmacology, University of Bologna
Via Irnerio 48, 40126 Bologna (Italy)
[c] Dr. L. Belvisi, Dr. M. Civera
Department of Organic and Industrial Chemistry, and the Interdisciplinary
Center for Biomolecular Studies and Industrial Applications (CISI)
University of Milan, Via Venezian 21, 20133 Milan (Italy)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201100372.
2264
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 2264 – 2272

Isoxazoline-Containing Peptidomimetics
five-membered rings as scaffolds with the goal of obtaining a
small library of functionalized isoxazoline-based integrin li-
gands.
Results and Discussion
Chemistry
The conjugate addition of nitrogen-containing nucleophiles to
electron-deficient olefins is one of the most frequently used
and versatile methods for CÀN bond construction in organic
chemistry.[29] Alkylidene malonates have been extensively em-
ployed as Michael acceptors.[30] In particular, a chiral Lewis-
acid-catalyzed Michael addition of hydroxyamino derivatives to
alkylidene malonates has been reported by our research
group.[31] The use of acetoacetates in place of malonates in
this field is rather unusual and has the advantage of introduc-
ing a reactive keto functionality that may be further elaborate-
d.[28a]
Figure 1. Selected examples of dual avb3 and a5b1 integrin inhibitors report-
ed in the literature.
basis of ligand–receptor binding, but homology models have
been proposed on the basis of the avb3 integrin–cilengitide
complex structure[18] and through binding pocket mapping
using a gain-of-function approach.[19]
According to these considerations, the isoxazolidine 1 was
prepared by treatment of tert-butylalkylidene acetoacetates
with N,O-bis(trimethylsilyl)hydroxyamine in the presence of a
catalytic amount of zinc trifluoromethanesulfonate (Scheme 1).
Reaction conditions were optimized on the basis of our previ-
ous work in order to avoid the formation of an oxime byprod-
uct resulting from the undesired 1,2-addition process.[32] Direct
addition of silica gel in the reaction workup induced rapid con-
version of the intermediate adduct to trans-5-hydroxyisoxazoli-
dine-4-carboxylate 1a–d via intramolecular hemiketalization.
The procedure, previously reported for ethyl and methyl
esters, was successfully applied to tert-butylalkylidene acetoa-
cetates.[32]
Several research groups have been involved in the develop-
ment of small constrained peptidomimetics of the RGD motif,
the enhanced bioavailability of which could result in more
promising drugs.[20] Typically, a rigid heterocyclic scaffold is in-
corporated as the central core of a small non-peptidic mole-
cule, having acidic and basic moieties properly oriented and
distanced for binding. Over the last few years, extensive stud-
ies have been undertaken regarding the synthesis of sequen-
ces that contain unusual linear or cyclic amino acids.[21] Thus,
the design and synthesis of new enantiopure non-proteino-
genic amino acids represent a central issue for chemists work-
ing in this broad research area. In a similar way, the prepara-
tion of small heterocyclic rings through simple and efficient
routes is receiving increased attention for their potential appli-
cation as scaffolds in the development of bioactive compound
libraries.[22]
In this context, isoxazolidines have been extensively ex-
plored both as synthetic intermediates[23] and as conformation-
al constraints.[24] These five-membered heterocycles may be en-
visaged as masked amino acids, as 1,3-amino alcohol equiva-
lents, or as furanose ring mimics in the preparation of bioac-
tive compounds. Indeed, isoxazolines have been incorporated
as conformational constraint elements in several transcriptional
activators.[25] In a similar way, analogues of the antimitotic
combretastatin containing an isoxazoline or isoxazole ring
have been reported,[26] and the synthesis of nucleoside ana-
logues containing an isoxazolidinic and isoxazolinic ring is of
current significant interest in the development of antiviral and
anticancer agents.[27]
Scheme 1. Synthesis of 5-hydroxyisoxazolidine-4-carboxylates 1a–d via con-
jugate addition of N,O-bis(trimethylsilyl)hydroxyamine to alkylidene aceto-
acetates: a) Zn(OTf)2 (5%), CH2Cl2.
Introduction of the heterocyclic core into a peptidomimetic
structure that possesses the fundamental features for receptor
affinity required selection of the proper appendages to mimic
aspartate and arginine side chains. The basicity and length of
the guanidine-mimicking group was found to play a central
role, while the presence of a carboxylate function to mimic as-
partate is necessary to create an electrostatic interaction with
the metal cation in the binding pocket. We recently synthe-
sized a small library of RGD mimetics containing 5,6-dihydro-
pyridin-2-ones and evaluated the effect of various side chains
on the inhibition of adhesion.[15c] This study suggested that
linking 4-aminobenzylamine and malonic acid to a rigid scaf-
fold, as basic and acidic termini, respectively, may provide
access to efficient avb3 and a5b1 integrin ligands.
We recently described the synthesis of ethyl 5-hydroxyisoxa-
zolidine-4-carboxylate and related functionalized isoxazolines
through a Lewis-acid-induced Michael addition of hydroxya-
mine derivatives to alkylideneacetoacetates, followed by intra-
molecular hemiketal formation.[28] On the basis of our experi-
ence in the introduction of heterocyclic scaffolds as central
cores of RGD peptidomimetics,[15c] we decided to test these
ChemMedChem 2011, 6, 2264 – 2272
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
2265

A. Tolomelli et al.
MED
Therefore, racemic trans-5-hydroxyisoxazolidine-4-carboxy-
lates 1a–d were acylated at the nitrogen atom by treatment
with benzyl malonyl chloride, generated in situ from the corre-
sponding acid and thionyl chloride, in the presence of pyridine
(Scheme 2). The reaction afforded compounds 2a–d in yields
ranging from 60 to 65%. Dehydration of the isoxazolidines
2a–d to the corresponding unsaturated racemic isoxazolines
3a–d was accomplished by transformation of the hydroxy
moiety into a mesylate group; subsequent base-induced elimi-
nation led to loss of the stereocenter at position 4 of the het-
erocycle (85–90% yield). The tert-butyl and benzyl esters were
chosen as terminal groups to effect their selective removal.
Indeed, treatment of 3a–d with excess trifluoroacetic acid
(TFA) in dichloromethane afforded acids 4a–d in nearly quanti-
tative yield. Coupling with 4-nitrobenzylamine by following the
protocol usually applied for peptide synthesis (EDCI/HOBt/
Et3N) allowed introduction of the arginine mimetic chain, bear-
ing a terminal nitro group as a masked form of the aromatic
amino moiety. Compounds 5a–d were isolated after purifica-
tion by flash chromatography on silica gel in yields ranging
from 60 to 70%. Finally, hydrogenation of 5a–d with palladi-
um/carbon as catalyst at atmospheric pressure led to simulta-
neous removal of the benzyl ester moiety and reduction of the
nitro group to amine, affording compounds 6a–d in excellent
yields (70–85%). Purification of the final isoxazoline com-
pounds with Dowex 50WX2-200 ion-exchange resin was
always performed before evaluation of bioactivity.
with the aspartate groups present in the integrin binding site,
we designed the synthesis of 9 starting from 4b (Scheme 3).
By direct reaction of 4b with 4-aminobenzylamine in the pres-
ence of HBTU and triethylamine, regioisomer 7 was exclusively
obtained in 80% yield, thus confirming that the aromatic
amine is less reactive than the aliphatic amine. Elongation was
carried out by coupling 7 with N-Cbz-glycine chloride, generat-
ed in situ from the corresponding acid and thionyl chloride, in
the presence of triethylamine to obtain 8 in 58% yield. Finally,
simultaneous removal of the benzyl ester and the benzyloxy-
carbonyl protecting group by hydrogenation on Pd/C followed
by treatment with Dowex 50WX2-200 ion-exchange resin af-
forded 9 in nearly quantitative yield.
With the aim of determining whether elongated molecules
are better suited for fitting into the receptor pocket and if a
more basic aliphatic amine could create stronger interactions
Scheme 3. Synthesis of RGD mimetic 9b: a) 4-aminobenzylamine, HBTU,
Et3N, CH2Cl2, RT, 80%; b) N-Cbz-Gly, SOCl2, Et3N, CH2Cl2, RT, 58%; c) H2/Pd/C,
MeOH, RT, >95%.
Pharmacology
Inhibition of avb3 and a5b1 integrin-mediated cell adhesion
The ability of racemic compounds 6a–d and 9 to inhibit the
adhesion of K562 (human erythroleukemia expressing a5b1 in-
tegrin) or SK-MEL-24 (human malignant melanoma expressing
avb3 integrin) cells to immobilized fibronectin was evaluated.
These cell models are widely used to investigate potential an-
tagonists of avb3 integrin-mediated SK-MEL-24 cell adhesion.[33]
The bioassays were not performed on intermediates 1a–d,
2a–d, 3a–d, 4a–d and 5a–d, which do not possess the func-
tional groups and spacing required for receptor affinity. In
these experiments, cells were seeded onto plates coated with
different substrata and allowed to adhere before quantitation
of the number of adherent cells in the presence of increasing
concentrations of test compounds. Under these conditions, no
significant cell adhesion was observed for bovine serum albu-
min (BSA)-coated plates (negative control) or nonspecific sub-
strate-coated plates (i.e., collagen I for SK-MEL-24 expressing
Scheme 2. Synthesis of RGD mimetics 6a–d: a) monobenzyl malonate,
SOCl2, pyridine, CH2Cl2, RT, 60–65%; b) CH3SO2Cl, Et3N, CH2Cl2, RT, 85–90%;
c) TFA, RT, >95%; d) 4-nitrobenzylamine, EDCI, HOBt, Et3N, CH2Cl2/DMF, RT,
60–70%; e) H2/Pd/C, MeOH, RT, 70–85%.
2266
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 2264 – 2272

Isoxazoline-Containing Peptidomimetics
avb3 integrin and poly-l-lysine
for K562 expressing a5b1 integ-
rin; data not shown). The results
Table 1. Inhibition of avb3 and a5b1 integrin-mediated cell adhesion to fibronectin (FN) in the presence of com-
pounds 6a–d, 9, or reference standards.
Entry
Compd
IC50[a] [nm]
are summarized in Table 1.
avb3
a5b1
The ability of the new com-
pounds to inhibit the adhesion
of SK-MEL-24 and K562 cells to
fibronectin or vitronectin was
compared with that of the stan-
dard compounds, Ac-Asp-Arg-
Leu-Asp-Ser-OH and H-Gly-Arg-
Gly-Asp-Asn-Pro-OH, known to
be potent inhibitors of cell adhe-
sion mediated by avb3 integ-
rin.[34] Among the five new RGD
mimetics tested, compound 6b
exhibited the highest potency as
an inhibitor of cell adhesion
mediated by avb3 and a5b1 in-
tegrins. The IC50 values against
both integrin-mediated cell ad-
hesion events were in the nano-
molar range (Entry 2, Table 1).
The results listed in Table 1 show
that the isoxazoline ring is an ef-
fective scaffold for inducing the
proper orientation of the Asp-
and Arg-mimicking chain, as
smaller members of the library
displayed excellent inhibitory ac-
tivities toward adhesion mediat-
ed by both receptors, and can
thus be regarded as dual inhibitors.
1
2
3
6a
6b
6c
32Æ3
12Æ4
8.8Æ0.6
360Æ70
1.05Æ0.3
1320Æ80
1030Æ50
4
6d
20Æ6
5
9
15500Æ900
>100000
6[b,c]
7[b]
Ac-Asp-Arg-Leu-Asp-Ser-OH
H-Gly-Arg-Gly-Asp-Asn-Pro-OH
25Æ3
>100000
>100000
234Æ32
[a] Values represent the mean ÆSD of three experiments. [b] Reference compounds. [c] See reference [15c].
Compound 6a, which has a smaller methyl substituent on
the isoxazoline ring, maintained notable efficacy, showing IC50
values only tenfold less potent than compound 6b (IC50:
32 nm versus avb3 integrin and 12 nm versus a5b1 integrin,
Entry 1). The sec-butyl-substituted derivative 6c showed de-
creased inhibitory activity (Entry 3), but an increase in selectivi-
ty toward avb3 integrin. A strong preference of compound 6d
for avb3 integrin was observed, as it has an IC50 value of 20 nm
as an avb3 integrin antagonist, whereas its IC50 value toward
a5b1 is 1 mm (Entry 4). Finally, compound 9, which was de-
signed to verify the effect of elongation and a terminal aliphat-
ic amine on adhesion inhibition, exhibited poor IC50 values, in
the millimolar to sub-millimolar range (Entry 5).
nectin, we tested its ability to block the adhesion of SK-MEL-24
cells to vitronectin relative to the standard peptide Ac-Asp-
Arg-Leu-Asp-Ser-OH. Compound 6b showed an IC50 value of
3.14 nm, while the reference peptide showed an IC50 value in
the sub-micromolar range (IC50: 0.33 mm) versus vitronectin-
mediated cell adhesion.
Effect of an integrin antagonist on fibronectin-induced ERK
phosphorylation
The mechanism by which components of the ECM generate in-
tracellular signaling through integrins requires increased phos-
phorylation of cytoplasmic second messengers. Phosphoryla-
tion of ERK plays a central role in fibronectin-mediated survival
signaling through integrins: in fact, cell adhesion activates ERK
by binding of a5b1 integrins at the cell surface to ECM proteins
such as fibronectin.[35] Therefore, we investigated the most ef-
fective compound 6b on fibronectin-induced phosphorylation
of ERK in K562 cells, which express a5b1 integrin.
In agreement with previous studies, Ac-Asp-Arg-Leu-Asp-
Ser-OH displayed nanomolar inhibitory activity toward the cell-
adhesive capacity driven selectively by avb3 integrin, whereas
the peptide H-Gly-Arg-Gly-Asp-Asn-Pro-OH had IC50 values in
the sub-micromolar range against avb3 integrin. Interestingly,
compounds 6a and 6b are more potent than the reference
compounds Ac-Asp-Arg-Leu-Asp-Ser-OH and H-Gly-Arg-Gly-
Asp-Asn-Pro-OH, as they inhibit cell adhesion mediated by
avb3 and a5b1 integrin with IC50 values lower than those of the
reference compounds. Moreover, as compound 6b exhibited
the highest potency as an inhibitor of cell adhesion to fibro-
K562 cells were serum-starved in RPMI-1640 containing 1%
fetal bovine serum (FBS) for 16 h; thereafter, they were pre-in-
cubated with compound 6b for 30 min in suspension and
plated for 30 min on fibronectin or poly-l-lysine (used as non-
specific substrate). In agreement with other studies,[35] the ERK
ChemMedChem 2011, 6, 2264 – 2272
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
2267

A. Tolomelli et al.
MED
pathway is activated in K562 cells exposed to phorbol-12-myr-
istate-13-acetate (PMA, 65 nm; Figure 2). In cells exposed to fi-
bronectin, a significant increase in phosphorylated ERK is ob-
served after 60 min, whereas no change is detected for shorter
exposure times, as previously described. This late ERK phos-
phorylation induced by fibronectin has been previously report-
ed.[35]
Molecular docking
To interpret the inhibitory activity observed for compound 6b
in the avb3-mediated cell adhesion assays at the molecular
level, a computational model for the binding of 6b to avb3 in-
tegrin was built by using a previously established and validat-
ed docking approach.[20c] Starting from the X-ray crystal struc-
ture of the extracellular segment of integrin avb3 in complex
with the cyclic pentapeptide ligand cilengitide (PDB ID:
1L5G),[17] automated docking calculations were carried out with
the Glide software package[36] according to the procedure de-
scribed in the Experimental Section below. The docking proto-
col was successful in reproducing the crystallographic binding
mode of cilengitide at the interface of the a and b subunits,
forming all the specific electrostatic and hydrogen bond inter-
actions. The docking protocol was then applied to small libra-
ries of cyclic and linear RGD peptidomimetics to evaluate their
ability to properly fit into the receptor site. Good agreement
between docking results and experimental biological data was
observed.[20c,37] In all the calculations, the experimentally ob-
served binding mode of cilengitide with avb3 integrin was
taken as a reference model for the analysis of docking results
in terms of ligand–protein interactions.
In this work, the Glide docking protocol was applied to com-
pound 6b (both the R and S enantiomers) to generate a struc-
tural model for the interaction with the ligand binding site of
the avb3 integrin receptor. Automated docking calculations of
both enantiomers produced top-ranked poses conserving the
key ionic interactions. As shown in Figure 3, the acidic and
basic pharmacophore groups of compound (R)-6b act like an
electrostatic clamp, interacting with charged regions of the re-
ceptor binding site. In particular, the ligand carboxylate group
is coordinated to the metal cation in the MIDAS region of the
b3 subunit, and the ligand aromatic amine moiety interacts
with the negatively charged side chains of Asp218 and
Asp150 in the av subunit. Further stabilizing interactions in-
volve the formation of hydrogen bonds between the ligand
carboxylate group and the backbone amide hydrogen atoms
Figure 2. Compound 6b prevents fibronectin-induced phosphorylation of
ERK. K562 cells were serum-starved in RPMI-1640 containing 1% FBS for
16 h; cells were then pre-incubated with compound 6b for 30 min in sus-
pension. The cells were kept in suspension (Ctrl) or plated on fibronectin
(FN) or on poly-l-lysine (poly-l-Lys). After 30 min, cells were lysed, and ly-
sates were analyzed by Western blot using an antibody directed against
phosphorylated ERK (pERK) or total ERK (tERK). a) Western blot shows that
control cells exposed to PMA (65 nm) or plated on FN had a much stronger
signal for phosphorylated ERK than those plated on poly-l-lysine (used as
nonspecific substrate). Pre-incubation with compound 6b before plating on
fibronectin caused a significant decrease in the amount of phosphorylated
ERK in K562 cells. b) Densitometric analysis of the bands (mean ÆSEM;
n=6): p42 (&), p44 (&); the amount of phosphorylated ERK is normalized
to the total amount of ERK. Densitometric quantification confirmed that
compound 6b specifically decreases fibronectin-induced phosphorylation of
ERK. In control cells, as expected, PMA induced a significant increase in ERK
phosphorylation; *p<0.01 versus Ctrl, **p<0.001 versus FN60’ (Newman–
Keuls test after ANOVA).
Western blots showed that for K562 cells not pre-incubated
with compound 6b, exposure to fibronectin for 60 min elicited
a much stronger signal for phosphorylated ERK than those
plated on poly-l-lysine (Figure 2a). Pre-incubation with com-
pound 6b (1 mm) caused a significant decrease in the amount
of fibronectin-induced ERK phosphorylation in K562 cells.
Figure 3. Docking best pose of compound (R)-6b (thick grey lines) into the
crystal structure of the extracellular domain of avb3 integrin. Selected integ-
rin residues involved in interactions with the ligand are shown in black. The
Mn2+ ion at MIDAS is shown as a dark CPK sphere. Nonpolar hydrogen
atoms have been removed for clarity.
2268
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 2264 – 2272

Isoxazoline-Containing Peptidomimetics
(5 mmol, 5 equiv) and isoxazolidine 1a–d (1 mmol) were added.
The solution was left to stir overnight, and then the reaction mix-
ture was diluted with CH2Cl2 (10 mL) and washed with H2O (2ꢁ
10 mL). The organic layer was dried (Na2SO4) and filtered, and the
solvent was removed in vacuo. Purification of the crude residue by
flash chromatography on silica gel, (eluent cyclohexane/EtOAc 9:1)
afforded compounds 2a–d as racemic mixtures of trans isomers.
of Asn215 and Tyr122 in the b unit. A ring-stacking effect be-
tween the ligand 4-aminobenzylamine and the Tyr178 side
chain of the av subunit is also observed.
Conclusions
A novel class of isoxazoline-containing peptidomimetics de-
signed as effective avb3 and a5b1 integrin ligands was devel-
oped on the basis of the well-known interactions of these
transmembrane receptors with bioactive ligands. For this pur-
pose, the isoxazolidine ring was synthesized by following an
original procedure involving N,O-bis(trimethylsilyl)hydroxya-
mine conjugate addition to alkylidene acetoacetates, followed
by intramolecular hemiketalization. Appendages that mimic ar-
ginine and aspartate side chains were introduced, and the final
products were tested in cell adhesion inhibition assays. Most
of the synthesized compounds proved to be excellent integrin
ligands, showing IC50 values in the nanomolar range for both
receptors. The effect of the most active compound 6b on fi-
bronectin-induced phosphorylation of ERK also allowed verifi-
cation that ligand–receptor binding has a strong influence on
intracellular signaling and phosphorylation pathways. More-
over, docking experiments yielded insight into the possible in-
teractions between 6b and avb3 integrin, thus providing infor-
mation on the molecular basis of the observed inhibitory activ-
ity. Because avb3 and a5b1 integrins play such an important
role in tumor angiogenesis, the new compounds reported
herein will likely serve as crucial leads for the development of
therapeutic and diagnostic tools.
tert-Butyl 2-[3-(benzyloxy)-3-oxopropanoyl]-5-hydroxy-3-isopro-
pyl-5-methylisoxazolidine-4-carboxylate (2b): Yellow oil (292 mg,
66%): 1H NMR (200 MHz, CDCl3): d=0.71 (d, J=7.0 Hz, 3H), 0.88
(d, J=7.0 Hz, 3H), 1.40 (s, 9H), 1.54 (s, 3H), 1.88 (m, 1H), 2.87 (d,
J=6.6 Hz, 1H), 3.43 (d, J=15.6 Hz, 1H), 3.60 (d, J=15.6 Hz, 1H),
4.51 (m, 1H), 4.61 (bs, 1H), 5.08 (s, 2H),7.29 ppm (s, 5H); 13C NMR
(50 MHz, CDCl3): d=17.7, 18.6, 23.0, 27.9, 32.2, 41.2, 58.8, 65.0,
67.1, 82.8, 105.7, 128.2, 128.5, 128.6, 135.4, 168.0, 168.7, 169.2 ppm;
LC–MS (ESI): tR =10.4 min, m/z: 444 [M+Na], 865 [2M+Na].
General procedure for the synthesis of dehydration of 2a–d to
3a–d: CH3SO2Cl (116 mL, 1.5 mmol, 1.5 equiv) and Et3N (348 mL,
2.5 mmol, 2.5 equiv) were added in a single portion to a stirred so-
lution of 2a–d (1 mmol) in dry CH2Cl2 (5 mL) at room temperature.
The solution was stirred at room temperature and stopped upon
disappearance of the starting material spot (by TLC), typically after
1 h. The reaction mixture was then diluted with CH2Cl2 (10 mL) and
washed with H2O (2ꢁ10 mL). The organic layer was dried (Na2SO4)
and filtered, and the solvent was removed in vacuo. Purification of
the crude residue by flash chromatography on silica gel (eluent cy-
clohexane/EtOAc 9:1) afforded compounds 3a–d.
tert-Butyl
2-[3-(benzyloxy)-3-oxopropanoyl]-3-isopropyl-5-
methyl-2,3-dihydroisoxazole-4-carboxylate (3b): Yellow oil
(359 mg, 89%): 1H NMR (200 MHz, CDCl3): d=0.71 (d, J=6.6 Hz,
3H), 0.89 (d, J=6.6 Hz, 3H), 1.41 (s, 9H), 1.55 (m, 1H), 2.01 (s, 3H),
3.44 (d, J=15.8 Hz, 1H), 3.60 (d, J=15.8 Hz, 1H), 5.08 (s, 2H), 5.12
(bs, 1H), 7.36 ppm (m, 5H); 13C NMR (50 MHz, CDCl3): d=10.9,
15.6, 19.7, 29.0, 31.3, 40.9, 66.9, 68.1, 80.7, 105.3, 128.1, 128.2,
128.4, 135.0, 161.1, 162.2, 166.3, 169.4 ppm; LC–MS (ESI): tR =
12.4 min, m/z: 404 [M+1], 829 [2M+Na].
Experimental Section
Chemistry
General: All chemicals were purchased from commercial suppliers
and were used without further purification. Anhydrous solvents
were purchased in sure-seal bottles over molecular sieves and
were used without further drying. Flash chromatography was per-
formed on silica gel (230–400 mesh). Dowex 50WX2-200(H) ion-ex-
change resin was used for purification of free amino acids. NMR
Spectra were recorded with Varian Gemini 200, Mercury Plus 400,
or Unity Inova 600 MHz spectrometers. Chemical shifts (d) are re-
ported in ppm relative to the solvent peak of CDCl3 set at d=
7.27 ppm (1H NMR) or d=77.0 ppm (13C NMR), CD3OD set at d=
3.31 ppm (1H NMR) or d=49.0 ppm (13C NMR), or D2O set at d=
4.79 ppm (1H NMR). Coupling constants (J) are given in Hz. LC–MS
analyses were performed on an HP1100 liquid chromatograph cou-
pled with an electrospray ionization mass spectrometer (LC–
ESIMS), using H2O/CH3CN as solvent at 258C (positive scan 100–
500 m/z, fragmentor 70 V). Compounds 1a–d were prepared by
following reported procedures.[28b] Complete characterization of in-
termediates bearing the isopropyl side chain and of final com-
pounds are reported below. Data for all other intermediates are re-
ported in the Supporting Information.
General procedure for tert-butyl ester hydrolysis (compounds
4a–d): TFA (15 equiv, 15 mmol, 1.15 mL) was added to a stirred so-
lution of 3a–d (1 mmol) in CH2Cl2 (2 mL) at 08C. After 2 h, the mix-
ture was washed with H2O (2ꢁ5 mL), the organic layer was dried
(Na2SO4), filtered, and the solvent was removed in vacuo. The acids
4a–d were used in the following step without further purification.
2-[3-(Benzyloxy)-3-oxopropanoyl]-3-isopropyl-5-methyl-2,3-dihy-
droisoxazole-4-carboxylic acid (4b): Yellow oil (340 mg, 98%):
1H NMR (200 MHz, CDCl3): d=0.80 (d, J=6.8 Hz, 3H), 0.98 (d, J=
6.8 Hz, 3H), 2.12 (s, 3H), 2.20 (m, 1H), 3.53 (d, J=16.0 Hz, 1H), 3.74
(d, J=16.0 Hz, 1H), 5.16 (s, 2H), 5.26 (bs, 1H), 7.34 ppm (m, 5H);
13C NMR (50 MHz, CDCl3): d=11.4, 15.4, 19.8, 23.9, 31.2, 40.9, 67.3,
103.7, 128.3, 128.4, 128.5, 134.9, 164.8, 166.4, 168.6, 169.9 ppm;
LC–MS (ESI): tR =9.2 min, m/z: 348 [M+1], 370 [M+Na], 717 [2M+
Na].
General procedure for coupling of 4a–d with p-nitrobenzyla-
mine (compounds 5a–d): EDCI (1.1 equiv, 0.55 mmol, 105 mg) and
Et3N (2.1 equiv, 1.05 mmol, 146 mL) were added to a stirred solution
of acid 4 (0.5 mmol) in dry CH2Cl2 (2 mL) under N2 atmosphere.
After 30 min, HOBt (1.1 equiv, 0.55 mmol, 74 mg) and p-nitrobenzy-
lamine·HCl (1.1 equiv, 0.55 mmol, 103 mg) were added. The solu-
tion was stirred overnight and then the mixture was diluted with
CH2Cl2 (8 mL) and washed with acidic H2O (2ꢁ5 mL, pH 3) and
basic H2O (2ꢁ5 mL, pH 8). The two phases were separated, the or-
General procedure for the synthesis of N-malonamido deriva-
tives 2a–d: SOCl2 (1.5 mmol, 1.5 equiv) was added dropwise to a
stirred solution of benzyl malonic acid (1.5 mmol, 1.5 equiv) in dry
CH2Cl2 (5 mL) at 08C. After removing the ice bath, the solution was
stirred for 30 min at room temperature, and then pyridine
ChemMedChem 2011, 6, 2264 – 2272
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
2269

A. Tolomelli et al.
MED
ganic layer was dried (Na2SO4) and filtered, and the solvent was re-
moved in vacuo. Compounds 5a–d were isolated by flash chroma-
tography on silica gel (cyclohexane/EtOAc 8:2 as eluent).
13C NMR (50 MHz, CD3OD): d=11.3, 16.5, 19.8, 30.7, 33.4, 42.2,
43.5, 69.0, 107.5, 120.9, 129.2, 136.4, 138.1, 159.7, 165.3, 165.5,
167.9, 170.4 ppm; LC–MS (ESI): tR =0.83 min, m/z: 419 [M+1].
Benzyl 3-[3-isopropyl-5-methyl-4-(4-nitrobenzylcarbamoyl) iso-
xazol-2(3H)-yl]-3-oxopropanoate (5b): Yellow oil (166 mg, 69%):
1H NMR (200 MHz, CDCl3): d=0.75 (d, J=6.6 Hz, 3H), 0.84 (d, J=
6.6 Hz, 3H), 1.24 (m, 1H), 2.02 (s, 3H), 3.39 (d, J=15.8 Hz, 1H), 3.61
(d, J=15.8 Hz, 1H), 4.51 (d, J=5.8 Hz, 2H), 5.07 (s,2H), 5.30 (bs,
1H), 6.67 (bt, 1H), 7.19 (m, 5H), 7.34 (d, J=8.4 Hz, 2H), 8.11 ppm
(d, J=8.4 Hz, 2H); 13C NMR (50 MHz, CDCl3): d=10.9, 16.4, 19.1,
31.2, 43.5, 66.1, 67.6, 105.8, 122.9, 127.3, 127.5, 128.3, 128.7, 135.8,
140.9, 145.3, 156.9, 162.5, 166.0, 168.2 ppm; LC–MS (ESI): tR =
10.0 min, m/z: 482 [M+1], 423 [M+Na].
Biology
Cell culture: SK-MEL-24 cells (American Type Culture Collection,
ATCC, Rockville, MD, USA) were routinely grown in minimum es-
sential medium (MEM, Cambrex, Walkersville, MD, USA) supple-
mented with 10% fetal bovine serum (FBS; Invitrogen, Carlsbad,
CA, USA), nonessential amino acids, and sodium pyruvate. K562
cells (ATCC) were maintained as a stationary suspension culture in
RPMI-1640 and l-glutamine with 10% FBS. Cells were kept at 378C
under a 5% CO2 humidified atmosphere; 40 h prior to experiments,
K562 cells were treated with 25 ngmLÀ1 PMA (Sigma–Aldrich SRL,
Milan, Italy) to induce differentiation and to increase expression of
cell surface antigens.[38]
General procedure for hydrogenation of 5a–d: To a solution of
5a–d (0.5 mmol) in MeOH, Pd/C (10 mg) was added in one portion.
The reaction mixture was stirred vigorously at room temperature
under H2 atmosphere overnight. The solution was filtered to
remove catalyst, and the solvent was evaporated in vacuo. The
crude compound was treated with Dowex 50WX2-200 ion-ex-
change resin, eluting with NH4OH (0.5m). Compounds 6a–d were
isolated after removal of the aqueous solvent in vacuo.
Cell adhesion assays:[33] 96-well plates (Corning Costar, Celbio,
Milan, Italy) were coated by passive adsorption with fibronectin
(10 mgmLÀ1) or poly-l-lysine (0.002%; Sigma–Aldrich SRL) overnight
at 48C. Cells were counted with a hemocytometer and pre-incubat-
ed with various concentrations of each compound for 30 min at
room temperature to reach a ligand–receptor equilibrium. Stock
solutions (10–2m) of the assayed compounds were prepared in
phosphate-buffered saline (PBS). At the end of the incubation
time, the cells were plated (50000 cells per well) and incubated at
room temperature for 1 h. All the wells were then washed with
PBS to remove nonadherent cells, and 50 mL hexosaminidase [4-ni-
trophenyl-N-acetyl-b-d-glucosaminide dissolved at a concentration
of 7.5 mm in 0.09m citrate buffer (pH 5) and mixed with an equal
volume of 0.5% Triton X-100 in H2O] was added. This product is a
chromogenic substrate for b-N-acetylglucosaminidase, whereby it
is transformed into 4-nitrophenol, the absorbance of which can be
measured at l 405 nm. As previously described,[39] there is a linear
correlation between absorbance and enzymatic activity. Therefore,
it is possible to identify the number of adherent cells among treat-
ed wells, interpolating the absorbance values of unknowns in a cal-
ibration curve. The reaction was blocked by the addition of 100 mL
of a stopping solution [50 mm glycine and 5 mm EDTA (pH 10.4)],
and the plates were read in a Victor2 Multilabel Counter (Perki-
nElmer, Waltham, MA, USA). Experiments were carried out in quad-
ruplicate and repeated at least three times. Data analysis and IC50
values were calculated using GraphPad Prism 3.0 (GraphPad Soft-
ware, San Diego, CA, USA).
Benzyl 3-[3-isopropyl-5-methyl-4-(4-aminobenzylcarbamoyl) iso-
xazol-2(3H)-yl]-3-oxopropanoate (6b): Yellow oil (143 mg, 79%):
1H NMR (200 MHz, CD3OD): d=0.73 (d, J=6.6 Hz, 3H), 0.86 (d, J=
6.6 Hz, 3H), 1.89 (m, 1H), 2.10 (s, 3H), 4.23 (d, J=15.0 Hz, 1H), 4.35
(d, J=15.0 Hz, 1H), 4.81 (s, 2H), 5.31 (bs, 1H), 7.00 (d, J=8.0 Hz,
2H), 7.20 (d, J=8.0 Hz, 2H), 8.21 ppm (bt, 1H); 13C NMR (50 MHz,
CDCl3): d=11.4, 17.0, 20.1, 27.8, 30.7, 45.5, 69.8, 117.0, 120.7, 129.5,
129.7, 136.3, 146.3, 161.6, 163.4, 167.1 ppm; LC–MS (ESI): tR =
1.46 min, m/z: 362 [M+1], 745 [2M+Na].
Coupling of 4b with p-aminobenzylamine (compound 7): HBTU
(1.3 equiv, 0.65 mmol, 246 mg) and Et3N (2.3 equiv, 1.15 mmol,
160 mL) were added to a stirred solution of acid 4b (0.5 mmol) in a
1:1 mixture of dry CH2Cl2 (2 mL) and CH3CN (2 mL), under N2 at-
mosphere. After 30 min, p-aminobenzylamine (1.2 equiv, 0.6 mmol,
68 mL) was added, and the solution was stirred overnight. The mix-
ture was diluted with CH2Cl2 (8 mL) and washed with acidic H2O
(2ꢁ5 mL, pH 3) and basic H2O (2ꢁ5 mL, pH 8). The two phases
were separated, the organic layer was dried (Na2SO4) and filtered,
and the solvent was removed in vacuo. Compound 7 was used in
the following step without further purification.
Coupling with N-Cbz-glycine: SOCl2 (1.2 mmol, 2.4 equiv) was
added dropwise to a stirred solution of N-Cbz-glycine (1.2 equiv,
0.6 mmol, 125 mg) in dry CH2Cl2 (5 mL) at 08C. After removing the
ice bath, the solution was stirred for 30 min at room temperature,
and then Et3N (1.2 mmol, 2.4 equiv, 167 mL) and compound 7
(0.5 mmol) were added. The solution was left to stir overnight, and
then the reaction mixture was diluted with CH2Cl2 (10 mL) and
washed with H2O (2ꢁ10 mL). The organic layer was dried (Na2SO4)
and filtered, and the solvent was removed in vacuo. Purification of
the crude residue by flash chromatography on silica gel (eluent cy-
clohexane/EtOAc 9:1) afforded compound 8. Following hydrogena-
tion and purification procedures reported above for 5, compound
9 was isolated in nearly quantitative yield.
Western blot analysis: K562 cells were incubated in RPMI-1640 with
1% FBS for 16 h. Plates were coated with fibronectin (10 mgmLÀ1
)
and blocked with 1% BSA (Sigma–Aldrich SRL). Subsequently, 4ꢁ
106 cells were pre-incubated with the various compounds for
30 min. Cells were allowed to adhere for 1 h on fibronectin in
RPMI-1640 with 1% FBS. The cells were then lysed in mammalian
protein extraction reagent (M-PER; Pierce, Rockford, IL, USA) sup-
plemented with phosphatase inhibitor (Sigma–Aldrich SRL) for
10 min at 48C by gentle shaking. Cell debris was removed by cen-
trifugation (14000 g for 15 min at 48C), and protein concentrations
were estimated by BCA assay (Pierce, Rockford, IL, USA). Protein ex-
tracts (100 mg) were denatured at 958C for 3 min before loading
and separation by 12% SDS-PAGE. The membranes were blocked
in 1% BSA and incubated for 2 h with anti-phospho-ERK 1/2 (extra-
cellular signal-regulated kinase 1/2) (1:2500) or anti-total ERK 1/2
antibodies (1:5000) (Promega, Madison, WI, USA), and thereafter
with anti-rabbit peroxidase-conjugated secondary antibody. Digital
3-{4-[4-(2-Aminoacetamido)benzylcarbamoyl]-3-isopropyl-5-
methylisoxazol-2(3H)-yl}-3-oxopropanoic acid (9): Yellow oil
1
(410 mg, 98%): H NMR (200 MHz, CD3OD): d=0.78 (d, J=6.6 Hz,
3H), 0.89 (d, J=6.6 Hz, 3H), 1.96 (m, 1H), 2.07 (s, 3H), 3.31 (m, 2H),
3.52 (m, 2H), 4.23 (d, J=15.6 Hz, 1H), 4.44 (d, J=15.6 Hz, 1H), 5.40
(bs, 1H), 7.22 (d, J=6.2 Hz, 2H), 7.43 ppm (d, J=6.2 Hz, 2H);
2270
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 2264 – 2272

Isoxazoline-Containing Peptidomimetics
[4] a) M. Barczyk, S. Carracedo, D. Gullberg, Cell Tissue Res. 2010, 339, 269–
280; b) N. J. Anthis, I. D. Campbell, Trends Biochem. Sci. 2011, 36, 191–
198.
images were acquired and analyzed by following previously report-
ed methods.[40]
[5] a) C. C. Kumar, M. Malkowski, Z. Yin, E. Tanghetti, B. Yaremko, T. Nechuta,
J. Varner, M. Liu, E. M. Smith, B. Neustadt, M. Presta, L. Armstrong,
Cancer Res. 2001, 61, 2232–2238; b) B. Felding-Habermann, T. E.
O’Toole, J. W. Smith, E. Fransvea, Z. M. Ruggeri, M. H. Ginsberg, P. E.
Hughes, N. Pampori, S. J. Shattil, A. Saven, B. M. Mueller, Proc. Natl.
Acad. Sci. USA 2001, 98, 1853–1858.
Molecular docking
All calculations were run using the Schrçdinger suite of programs
(http://www.schrodinger.com) through the Maestro graphic inter-
face.
[6] C. J. Avraamides, B. Garmy-Susini, J. A. Varner, Nature Rev. Cancer 2008,
8, 604–617.
Protein setup: The recently determined crystal structure of the ex-
tracellular domain of the integrin avb3 receptor in complex with cil-
engitide and in the presence of the pro-adhesive ion Mn2+ (PDB
ID: 1L5G)[17] was used for docking studies. Docking was performed
only on the globular head of the integrin because the head group
of integrin has been identified in the X-ray structure as the ligand
binding region. The protein structure was set up for docking as fol-
lows: the protein was truncated to residues 41–342 for chain a
and residues 114–347 for chain b. Due to a lack of parameters, the
Mn2+ ions in the experimental protein structure were modeled via
replacement with Ca2+ ions. The resulting structure was prepared
with the Protein Preparation Wizard of the graphical user interface
Maestro and the OPLSAA force field.
[7] a) K. L. Schornberg, C. J. Shoemaker, D. Dube, M. Y. Abshire, S. E. Delos,
A. H. Bouton, J. M. White, Proc. Natl. Acad. Sci. USA 2009, 106, 8003–
8008; b) G. Tran Van Nhieu, R. R. Isberg, EMBO J. 1993, 12, 1887–1895.
[8] C. T. Mierke, B. Frey, M. Feliner, M. Herrmann, B. Fabry, J. Cell. Science
2011, 124, 369–383.
[9] S. Kim, K. Bell, S. A. Mousa, J. A. Varner, Am. J. Pathol. 2000, 156, 1345–
62.
[10] S. Kim, M. Harris, J. A. Varner, J. Biol. Chem. 2000, 275, 33920–33928.
[11] D. P. Ly, S. A. Corbett, J. Surg. Res. 2005, 123, 200–205.
[12] a) J. K. Willmann, N. van Bruggen, L. M. Dinkelborg, S. S. Gambhir,
Nature Rev. Drug Disc. 2008, 7, 591–607; b) M. Schottelius, B. Laufer, H.
Kessler, H. J. Wester, Acc. Chem. Res. 2009, 42, 969–980; c) A. J. Beer, M.
Schwaiger, J. Nucl. Med. 2011, 52, 335–336; d) H. Xie, P. Diagaradjane,
A. A. Deorukhkar, B. Goins, A. Bao, W. T. Phillips, Z. Wang, J. Schwartz, S.
Krishnan, Int. J. Nanomed. 2011, 259–269.
Docking: Automated docking calculations were performed with
Glide (Grid-based Ligand Docking with Energetics)[36] within the
framework of Impact version 4.5 in a standard precision (SP) mode.
The grid-generation step started from the extracellular fragment of
the X-ray structure of the avb3 complex with cilengitide, prepared
as described above in Protein setup. The center of the grid-enclos-
ing box was defined by the center of the bound ligand. The en-
closing box dimensions, which are automatically deduced from the
ligand size, fit the entire active site. For the docking step, the size
of the bounding box for placing the ligand center was set at 12 ꢂ.
No further modifications were applied to the default settings. The
GlideScore function was used to select 20 poses for each ligand.
Glide was initially tested for its ability to reproduce the crystallized
binding geometry of cilengitide. The program was successful in re-
producing the experimentally determined binding mode of this
compound, as it corresponds to the best-scored pose.
[13] a) I. S. Shin, B.-S. Jang, S. N. Danthi, J. Xie, S. Yu, N. Le, J.-S. Maeng, I. S.
Hwang, K. C. P. Li, J. A. Carrasquillo, C. H. Paik, Bioconjugate Chem. 2007,
18, 821–828; b) R. Schmidmaier, P. Baumann, Curr. Med. Chem. 2008,
15, 978–990; c) J. M. Cantor, M. H. Ginsberg, D. M. Rose, Immunol. Rev.
2008, 223, 236–251; d) S. A. Mousa, Mol. Biotechnol. 2008, 38, 33–40;
e) V. Hofmeister, D. Schrama, J. C. Becker, Cancer Immunol. Immunother.
2008, 57, 1–17; f) J. A. Nemeth, M. T. Nakada, M. Trikha, Z. Lang, M. S.
Gordon, G. C. Jayson, R. Corringham, U. Prabhakar, H. M. Davis, R. A.
Beckman, Cancer Invest. 2007, 25, 632–646.
[14] a) A. Cavalli, M. L. Bolognesi, A. Minarini, M. Rosini, V. Tumiatti, M. Reca-
natini, C. Melchiorre, J. Med. Chem. 2008, 51, 347–372; b) R. Morphy, Z.
Rankovic, Drug Discovery Today 2007, 12, 156–160; c) K. Nadrah, M. S.
Dolenc, Curr. Med. Chem. 2005, 12, 1449–1466; d) A. Perdih, M. S.
Dolenc, Curr. Med. Chem. 2010, 17, 2371–2392.
[15] a) J. M. Smallheer, C. A. Weigelt, F. J. Woerner, J. S. Wells, W. F. Daneker,
S. A. Mousa, R. R. Wexler, P. K. Jadhav, Bioorg. Med. Chem. Lett. 2004, 14,
383–387; b) K. Meerovitch, F. Bergeron, L. Leblond, B. Grouix, C. Poirier,
M. Bubenik, L. Chan, H. Gourdeau, T. Bowlin, G. Attardo, Vasc. Pharma-
col. 2003, 40, 77–89; c) F. Benfatti, G. Cardillo, S. Fabbroni, P. Galzerano,
L. Gentilucci, R. Juris, A. Tolomelli, M. Baiula, A. Spartꢅ, S. Spampinato,
Bioorg. Med. Chem. 2007, 15, 7380–7390.
Acknowledgements
[16] E. Ruoslahti, Annu. Rev. Cell. Dev. Biol. 1996, 12, 697–715.
[17] J. P. Xiong, T. Stehle, R. Zhang, A. Joachimiak, M. French, S. L. Goodman,
M. A. Arnaout, Science 2002, 296, 151–155.
[18] a) D. Heckmann, A. Meyer, L. Marinelli, G. Zahn, R. Stragies, H. Kessler,
Angew. Chem. 2007, 119, 3641–3644; Angew. Chem. Int. Ed. 2007, 46,
3571–3574; b) L. Marinelli, A. Meyer, D. Heckmann, A. Lavecchia, E. Nov-
ellino, H. Kessler, J. Med. Chem. 2005, 48, 4204–4207.
This study was carried out with fundamental contributions of the
“Fondazione del Monte di Bologna e Ravenna”, MIUR (PRIN 2008)
and the University of Bologna (Italy). We also thank the Italian
Minister for Foreign Affairs (MAE) (General Direction for the Cul-
tural Promotion and Cooperation) for financial support of a bilat-
eral project between Italy and Mexico. Mr. Andrea Garelli is grate-
fully acknowledged for the LC–ESI-MS analysis. We also thank the
Consorzio Interuniversitario Nazionale “Metodologie e Processi In-
novativi di Sintesi” (CINMPIS) for a grant to A.V.
[19] A. P. Mould, E. J. Koper, A. Byron, G. Zahn, M. J. Humpries, Biochem. J.
2009, 424, 179–189.
[20] a) B. Cacciari, G. Spalluto, Curr. Med. Chem. 2005, 12, 51–70; b) D. Heck-
mann, A. Meyer, B. Laufer, G. Zhan, R. Stragies, H. Kessler, ChemBioChem
2008, 9, 1397–1407; c) L. Manzoni, L. Belvisi, D. Arosio, M. Civera, M. Pil-
kington-Miska, D. Potenza, A. Caprini, E. M. V. Araldi, E. Monferini, M.
Mancino, F. Podestꢅ, C. Scolastico, ChemMedChem 2009, 4, 615–632.
[21] a) A. Muller, F. Schumann, M. Koksch, N. Sewald, Lett. Pept. Sci. 1997, 4,
275–281; b) P. J. Coleman, K. M. Brashear, C. A. Hunt, W. F. Hoffmann,
J. H. Hutchinson, M. J. Breslin, C. A. McVean, B. C. Askew, G. D. Hartman,
S. B. Rodan, G. A. Rodan, C. T. Leu, T. Prueksaritanont, C. Fernandez-Met-
zler, B. Ma, L. A. Libby, K. M. Merkle, G. L. Stump, A. A. Wallace, J. J.
Lynch, R. Lynch, M. E. Duggan, Bioorg. Med. Chem. Lett. 2002, 12, 31–
34; c) S. Urman, K. Gaus, Y. Yang, U. Strijowski, N. Sewald, S. De Pol, O.
Reiser, Angew. Chem. 2007, 119, 4050–4053; Angew. Chem. Int. Ed.
2007, 46, 3976–3978; d) A. Trabocchi, G. Menchi, N. Cini, F. Bianchini, S.
Keywords: cancer
·
cell adhesion
·
drug discovery
·
heterocycles · peptidomimetics
[1] L. U. Hunter, A. Eisbruch, Cancer J. 2011, 17, 151–154.
[2] M. Alonso-Basanta, C. Chapman, J. F. Dorsey, R. Rengan, S. M. Hahn,
Cancer J. 2011, 17, 190–194.
[3] a) M. Lꢃpez-Lꢄzaro, Mol. Med. 2010, 16, 144–153; b) V. Pavet, M. M.
Portal, J. C. Moulin, R. Herbrecht, H. Gronemeyer, Oncogene 2011, 30,
1–20.
ChemMedChem 2011, 6, 2264 – 2272
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
2271

A. Tolomelli et al.
MED
Raspanti, A. Bottoncetti, A. Pupi, L. Calorini, A. Guarna, J. Med. Chem.
2010, 53, 7119–7128.
[22] a) A. D. Abell, Lett. Pept. Sci. 2002, 8, 267–272; b) A. Schuffenhauer, P.
Ertl, S. Roggo, S. Wetzel, M. A. Koch, H. Waldmann, J. Chem. Inf. Model.
2007, 47, 47–58.
[23] a) T. Shibue, T. Hirai, I. Okamoto, N. Morita, H. Masu, I. Azumaya, O.
Tamura, Chem. Eur. J. 2010, 16, 11678–11688; b) A. R. Minter, A. A.
Fuller, A. K. Mapp, J. Am. Chem. Soc. 2003, 125, 6846–6847.
[24] S. S. Bhella, A. P. Singh Pannu, M. Elango, A. Kapoor, M. S. Hundal,
M. P. S. Ishar, Tetrahedron 2009, 65, 5928–5935.
[25] a) S. P. Rowe, R. J. Casey, B. B. Brennan, S. J. Buhrlage, A. K. Mapp, J. Am.
Chem. Soc. 2007, 129, 10654–10655; b) R. J. Casey, J. P. Desaulniers,
J. W. Hojfeldt, A. K. Mapp, Bioorg. Med. Chem. 2009, 17, 1034–1043.
[26] D. Simoni, G. Grisolia, G. Giannini, M. Roberti, R. Rondanin, L. Piccagli, R.
Barucchello, M. Rossi, R. Romagnoli, F. P. Invidiata, S. Grimaudo, M. K.
jung, E. Hamel, N. Gebbia, L. Crosta, V. Abbadessa, A. Di Cristina, L. Du-
sonchet, M. Meli, M. Tolomeo, J. Med. Chem. 2005, 48, 723–736.
[27] a) P. Merino, S. Franco, F. L. Merchan, T. Tejero, J. Org. Chem. 2000, 65,
5575–5589; b) Y. Xiang, H. J. Gi, D. Niu, R. F. Schinazi, K. Zhao, J. Org.
[30] a) A. Alexakis, J. Vastra, P. Mangeney, Tetrahedron: Asymmetry 1993, 4,
2427–2430; b) D. A. Evans, T. Rovis, M. C. Kozlowski, C. W. Downey, J. S.
Tedrow, J. Am. Chem. Soc. 2000, 122, 9134–9142; c) W. Zhuang, T.
Hansen, K. A. Jørgensen, Chem. Commun. 2001, 347–348; d) J. Schup-
pan, A. J. Minnaard, B. L. Feringa, Chem. Commun. 2004, 792–793;
e) C. L. Cao, X. L. Sun, J. L. Zhou, Y. Tang, J. Org. Chem. 2007, 72, 4073–
4076.
[31] G. Cardillo, S. Fabbroni, L. Gentilucci, M. Gianotti, R. Percacciante, A. Tol-
omelli, Tetrahedron: Asymmetry 2002, 13, 1407–1410.
[32] F. Benfatti, G. Cardillo, L. Gentilucci, E. Mosconi, A. Tolomelli, Synlett
2008, 2605–2608.
[33] S. Caltabiano, W. T. Hum, G. J. Attwell, D. N. Gralnick, L. J. Budman, A. M.
Cannistraci, F. J. Bex, Biochem. Pharmacol. 1999, 58, 1567–1578.
[34] H. Fujii, H. Komazawa, H. Mori, M. Kojima, I. Itoh, J. Murata, I. Azuma, I.
Saiki, Biol. Pharm. Bull. 1995, 18, 1681–1688.
[35] a) B. D. Cox, M. Natarajan, M. R. Gladson, L. Candece, J. Cell. Biochem.
2006, 99, 35–52; b) D. D. Schlaepfer, C. R. Hauck, D. J. Sieg, Progr. Bio-
phys. Mol. Biol. 1999, 71, 435–478.
[36] Glide version 4.5, 2007, Schrçdinger LLC, New York, NY (USA).
[37] L. Manzoni, M. Bassanini, L. Belvisi, I. Motto, C. Scolastico, M. Castorina,
C. Pisano, Eur. J. Org. Chem. 2007, 1309–1317.
[38] L. Hunꢄkovꢄ, J. Sedlꢄk, M. Klobusickꢄ, M. Sulikovꢄ, B. Chorvꢄth, Neoplas-
ma 1995, 42, 249–253.
´
ˇ
´
ˇ
Chem. 1997, 62, 7430–7434; c) E. Hyrosovꢄ, M. Medvecky, L. Fisera, C.
Hametner, J. Frçhlich, M. Marchetti, G. Allmaier, Tetrahedron 2008, 64,
3111–3118; d) B. Loh, L. Vozzolo, B. J. Mok, C. C. Lee, R. J. Fitzmaurice, S.
Caddick, A. Fassati, Chem. Biol. Drug Des. 2010, 75, 461–474.
[28] a) F. Benfatti, G. Cardillo, S. Contaldi, L. Gentilucci, E. Mosconi, A. Tolo-
melli, E. Juaristi, G. Reyes-Rangel, Tetrahedron 2009, 65, 2478–2483;
b) F. Benfatti, A. Bottoni, G. Cardillo, L. Gentilucci, M. Monari, E. Mosconi,
M. Stenta, A. Tolomelli, Eur. J. Org. Chem. 2008, 6119–6127.
[39] H. Shibata, T. Yagi, Clin. Chim. Acta 1996, 251, 53–64.
[40] A. Bedini, M. Baiula, S. Spampinato, J. Neurochem. 2008, 105, 2166–
2178.
[29] a) P. Perlmutter, Conjugate Addition Reaction in Organic Synthesis, Perga-
mon Press, Oxford, 1992; b) M. P. Sibi, S. Manyem, Tetrahedron 2000, 56,
8033–8061.
Received: July 30, 2011
Revised: August 30, 2011
Published online on September 26, 2011
2272
www.chemmedchem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2011, 6, 2264 – 2272