Functionalizing αvβ3- or α5β1-selective integrin antagonists for surface coating: A method to discriminate integrin subtypes in vitro

Article abstract of DOI:10.1002/anie.201206370

Stuck with the right choice: αvβ3- or α5β1-selective integrin ligands were functionalized for surface coating without losing activity and selectivity. The coating of nanostructured gold surfaces with these compounds stimulated subtype-selective cell adhesion of genetically modified αvβ3- or α5β1-expressing fibroblasts in vitro. Copyright

Full text of DOI:10.1002/anie.201206370

.
Angewandte
Communications
DOI: 10.1002/anie.201206370
Selective Cell Adhesion
Functionalizing avb3- or a5b1-Selective Integrin Antagonists for
Surface Coating: A Method To Discriminate Integrin Subtypes
In Vitro**
Florian Rechenmacher, Stefanie Neubauer, Julien Polleux, Carlos Mas-Moruno,
Mariarosaria De Simone, Elisabetta Ada Cavalcanti-Adam, Joachim P. Spatz, Reinhard Fꢀssler,
and Horst Kessler*
Investigating the different functions of distinct surface
receptors is essential to understand the complex interactions
between cells and their extracellular environment. Cells use
specific transmembrane receptors of the integrin family to
anchor and respond to extracellular matrix (ECM) proteins.
In doing so, integrins are capable of regulating cell migration,
survival, cell cycle progression, and differentiation, which are
essential tasks for the development of all multicellular
organisms. Integrins are often classified according to their
binding specificity for extracellular ligands. For instance,
arginine–glycine–aspartate (RGD)-containing proteins,[1]
such as fibronectin (Fn), exhibit high binding affinity to 8 of
the 24 integrins expressed in mammals. Among the fibronec-
tin-binding integrins, avb3 and a5b1 play crucial roles during
embryogenesis, angiogenesis, and in pathology.[2–4] Despite
the wealth of information on integrin biology, it is unclear
how cell functions and responses are regulated by a single
integrin subtype. To address this relevant issue, the design of
new active molecules that are able to selectively recognize
distinct integrin subtypes is essential for in vitro studies and
the development of selective drugs for disease therapy
(personalized medicine). Moreover, the conjugation of func-
tional groups to such ligands without affecting their affinity
and selectivity for cell surface receptors remains a difficult
task to achieve.
adhesion of either avb3- or a5b1-expressing cells was clearly
mediated or blocked through coating of gold nanoarrays with
these molecules, thus giving striking evidence for their
selectivity. These compounds are a powerful tool to elucidate
the difference between avb3 and a5b1 integrin-mediated cell
adhesion.
The interaction of integrins with ECM proteins leads to
the clustering of integrins and the recruitment of intracellular
proteins to the integrin cytoplasmic domains.[6] The recruited
proteins, collectively called the adhesome,[7] play crucial roles
in transducing integrin-mediated processes and are thus also
involved in biological processes, such as angiogenesis and
tumor development. They are therefore interesting targets for
pharmacological research.[8,9] Although the functions of
integrins in cell adhesion and formation of focal contacts
have been studied with unselective cyclic RGD peptides and
other ECM mimetics,[10] the exact role of avb3 and a5b1
integrin subtypes played in these processes remains to be
elucidated. Natural integrin ligands are not well suited for
those studies owing to the lack of integrin subtype selectiv-
ity.[11]
Coating of surfaces with peptidic integrin ligands for
biophysical cell adhesion studies, to develop implant materi-
als, or to identify binding motifs was realized in various
cases.[10,12] However, it turned out that peptides, in contrast to
peptidomimetics, were not able to achieve high activity and
sufficient selectivity. Nowadays, it is still challenging to
functionalize selective peptidomimetics for this purpose
without losing activity.
Herein, we report a strategy to synthesize avb3- or a5b1-
specific ligands for the functionalization of nanostructured
gold surfaces[5] and demonstrate that cell adhesion can be
selectively mediated by a single integrin subtype. The
[*] F. Rechenmacher,[+] S. Neubauer,[+] Dr. C. Mas-Moruno,
Dr. M. De Simone
Dr. E. A. Cavalcanti-Adam, Prof. J. P. Spatz
Max Planck-Institute for Intelligent Systems, Department of New
Materials and Biosystems, Stuttgart (Germany)
and
Institute for Advanced Study at the Department of Chemistry,
Technische Universitꢀt Mꢁnchen (Germany)
University of Heidelberg, Institute for Physical Chemistry, Depart-
ment of Biophysical Chemistry
Heidelberg (Germany)
Prof. Dr. H. Kessler
Institute for Advanced Study at the Department of Chemistry
Technische Universitꢀt Mꢁnchen
Lichtenbergstrasse 4, 85748 Garching (Germany)
and
Chemistry Department, Faculty of Science
King Abdulaziz University
P.O. Box 80203, Jeddah 21589 (Saudi Arabia)
E-mail: Kessler@tum.de
Homepage: http://www.org.chemie.tu-muenchen.de
[+] These authors contributed equally to this work.
[**] We thank the IGSSE (International Graduate School of Science and
Engineering), the Bund der Freunde der TU Mꢁnchen e.V., CompInt
(Materials Science of Complex Interfaces) of the Elite Network of
Bavaria for funding, IAS (Institute for Advanced Study) of Techni-
sche Universitꢀt Mꢁnchen, CIPSM (Center for Integrated Protein
Science Munich), and the Max Planck Society for financial support.
Dr. J. Polleux, Prof. R. Fꢀssler
Max Planck Institut fꢁr Biochemie
Department of Molecular Medicine, Martinsried (Germany)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201206370.
1572
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1572 –1575

Angewandte
Chemie
The design of integrin antagonists with the
ability to visualize, block, and inhibit distinct
integrin subtypes during cell adhesion has been
a major goal of research over the last two
decades. Since the discovery of the integrin
recognition motif RGD,[1] we and others have
focused on the design of integrin antagonists
specifically targeting either a5b1 or avb3, while
maintaining selectivity against the platelet
integrin aIIbb3.[13–15] This specificity is partic-
ularly important for possible future clinical use,
as the inhibition of the RGD-binding aIIbb3
integrins in platelet adhesion would have
deleterious side effects in patients. By design-
ing RGD-based peptidomimetics,[16] we
Scheme 1. a5b1-Selective (1) and avb3-selective (2) integrin antagonists and their thiol-
recently obtained the selective antagonist functionalized analogues (3, 4) for binding to gold nanoparticles.
1 that specifically binds and inhibits a5b1
(Scheme 1).[15] However, the synthesis of
a highly active avb3 antagonist that has low affinity for
a5b1 remained an unmet challenge. By designing a library of
b-tyrosine-based scaffolds we have now succeeded for the first
time in synthesizing a subnanomolar active avb3 antagonist 2
(Scheme 1) with selectivity against a5b1 by two orders of
magnitude (Table 1).
Table 1: Selectivity profiles of the free antagonists and their thiol-
functionalized derivatives in comparison to cilengitide.[13]
Compound
IC50(avb3) [nm]
IC50(a5b1) [nm]
1
2
3
4
3001Æ205
0.55Æ0.07
229Æ23
2.3Æ0.02
120Æ27
1.5Æ0.09
130Æ19
11Æ1.2
1.8Æ0.7
Cilengitide
0.20Æ0.07
Docking of peptidomimetics into the X-ray structure of
the avb3 integrin[17] and into the homology model of a5b1
integrin[18] has been performed to elucidate the optimal
positions for functionalizing the integrin antagonists without
affecting their biological activity.[14,15] We found that the
isopropoxy group of the aromatic residue in compound
1 points out of the binding pocket of both a5b1 and avb3,
and it is thus a suitable position for functionalization. We
applied this strategy also to compound 2, as we assumed,
based on the homology model for a5b1,[18] a similar binding
mode as for 1.
For the functionalization of the a5b1-selective antagonist,
molecule 10 was synthesized in a six-step synthesis performed
in solution (Scheme 2). This molecule is a crucial function-
alization building block and offers a convenient way for
a straightforward attachment of any desired moiety to the
selective integrin antagonist 1 by Fmoc-based[19] (Fmoc = 9-
fluorenylmethoxycarbonyl) solid-phase peptide synthesis
(SPPS).[20] It should be mentioned that several previous
attempts for functionalization of 1 with other building blocks
failed both in solution and on solid phase. With compound 10,
the synthesis of the thiol-functionalized a5b1-antagonist 3
could be easily performed on a solid support. After loading
Scheme 2. Synthesis of building block 10 and functionalization of the
a5b1-selective integrin antagonist on a solid support, yielding thiol 3.
a) NaH, 08C, DMF; BnBr, RT; b) nBuLi, À788C, THF; Boc2O, RT;
c) H2, [Pd/C], MeOH; d) 1,1’-(azodicarbonyl)dipiperidine, PBu3, THF;
e) H2, [Pd/C], MeOH; Fmoc-OSu, NaHCO3, THF/H2O; f) TFA/triiso-
propylsilane/H2O 95/2.5/2.5. Alloc=allyloxycarbonyl, Bn=benzyl,
Boc=tert-butoxycarbonyl, Cbz=carboxybenzyl, TFA=trifluoroacetic
acid, Trt=trityl.
Angew. Chem. Int. Ed. 2013, 52, 1572 –1575
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1573

.
Angewandte
Communications
the resin with N-a-Fmoc-N-b-alloc-l-diaminopropionic acid
and Fmoc deprotection, 10 is coupled to the free amine. At
this point, the functionalization was performed to introduce
any group of interest for further biological investigations that
is compatible with Fmoc-based SPPS. Finally, the peptidomi-
metic backbone was built up leading to the thiolated a5b1
ligand 3 (for the detailed synthesis, see the Supporting
Information). Functionalization of the avb3-selective antag-
onist was performed in a similar way.
To prove our concept of functionalization regarding
biological activity, we tested the IC50 values of the synthesized
compounds in a recently reported[21] competitive solid-phase
integrin binding assay to evaluate their activity towards the
integrins avb3 and a5b1 (see the Supporting Information for
the detailed procedure). The activity of the thiolated com-
pound 3 for a5b1 was fully retained (1.5 nm) compared to the
free antagonist 1 (2.3 nm) as well as the selectivity against
avb3. The thiol-functionalized compound 4 retained activity
for avb3 (1.8 nm) and showed remarkable selectivity against
a5b1 (130 nm). Furthermore, all compounds showed no
activity for the platelet integrin aIIbb3, which is of high
importance for further use. Thus, the obtained IC50 values are
a substantial proof to validate our concept of functionaliza-
tion.
cellular adhesion, while it prevents uneven distribution of
immobilized ligands and thus makes integrin-mediated cell
adhesion occur more homogeneously all over the culture
substrate. This approach allows the study of cell adhesion to
defined binding sites for integrins, while avoiding at the same
time unspecific adhesion and integrin binding. Here, gold
nanoparticles with a lateral interdistance of 30 nm served as
anchoring points for the terminal thiol group of the selective
peptidomimetics.[5,22] After passivating the glass substrate
with polyethylene glycol (PEG) and functionalization of the
gold surface with the integrin-selective compounds 3 and 4,
avb3 or a5b1 integrin-expressing fibroblasts were seeded in
low serum-containing (0.5% FBS) medium to reduce non-
specific protein adsorption onto the substrate, and, thus
allowing the cells to exclusively interact with the immobilized
peptidomimetics (Figure 1). The difference in the ability of
the two cell lines to adhere onto surfaces functionalized with
compound 3 or 4 was striking. As demonstrated in Figure 1,
a5b1-expressing fibroblasts adhered after one hour exclu-
sively onto gold surfaces coated with the a5b1-selective
compound 3. In contrast, avb3-expressing cells adhered and
spread only on the immobilized avb3-specific compound 4.
Moreover, both cell lines were not able to adhere onto the
other surface even after six hours. To prove that the cell
adhesion is triggered specifically by the respective peptido-
mimetics, cell adhesion assays were performed on surfaces
coated with a non-integrin binding peptide, the thiolated
c(RADfK), as a negative control. No adhesion of cells could
be identified in this case for both cell lines (data not shown).
Additionally, we quantified the spreading area and the
percentage of spread a5b1- and avb3-expressing cells on the
selective-functionalized nanoarrays after 1 h and 6 h of
incubation (Supporting Information, Figures S1 and S2).
To investigate the selectivity profile of compounds 3 and
4, we performed adhesion assays on functionalized gold
nanoarrays with two different genetically modified fibroblast
lines that express either avb3 or a5b1 integrins. In previous
studies, we used gold nanoarrays presenting cyclic RGD
peptides to determine the spatial requirements for integrin-
mediated cell adhesion and spreading.[5] Moreover, this
approach is very powerful, as quasi-hexagonally organized
nanoparticles with a diameter of about 7 nm enable testing
Figure 1. Cell adhesion assay. Low-magnification phase-contrast microscope images of avb3 or a5b1 integrin-expressing fibroblasts seeded onto
gold nanoarrays and functionalized with an a5b1-selective compound (left) or an avb3-selective compound (right). The 100 mm scale bar applies
for all of the images.
1574
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1572 –1575

Angewandte
Chemie
Micoulet, H. Kessler, B. Geiger, J. P. Spatz, Biophys. J. 2007,
The analysis of cell spreading for each surface clearly
demonstrates a selective cell adhesion mediated by the
compounds 3 and 4 although the spreading of avb3-express-
ing cells occurred with a delay of about one hour when
compared to the a5b1-expressing cells. The similar time
periods required for a5b1- and avb3-expressing fibroblasts to
spread on gold particle surfaces functionalized with com-
pound 3 or 4, respectively, and on fibronectin-coated plastic
surfaces indicates that integrin subtype-mediated cell adhe-
sion was selectively triggered by the two different integrin-
binding peptidomimetics (data not shown).
92, 2964 – 2974; d) M. Arnold, V. C. Hirschfeld-Warneken, T.
Lohmꢀller, P. Heil, J. Blꢀmmel, E. A. Cavalcanti-Adam, M.
Lopꢁz-Garcꢂa, P. Walther, H. Kessler, B. Geiger, J. P. Spatz,
Nano Lett. 2008, 8, 2063 – 2069; e) J. H. Huang, S. V. Graeter, F.
Corbellini, S. Rinck, E. Bock, R. Kemkemer, H. Kessler, J. D.
Ding, J. P. Spatz, Nano Lett. 2009, 9, 1111 – 1116; f) S. H. Shabbir,
J. L. Eisenberg, M. Mrksich, Angew. Chem. 2010, 122, 7872 –
7875; Angew. Chem. Int. Ed. 2010, 49, 7706 – 7709; g) K. A.
Kilian, M. Mrksich, Angew. Chem. 2012, 124, 4975 – 4979;
Angew. Chem. Int. Ed. 2012, 51, 4891 – 4895.
[11] E. F. Plow, T. K. Haas, L. Zhang, J. Loftus, J. W. Smith, J. Biol.
Chem. 2000, 275, 21785 – 21788.
In conclusion, compounds 3 and 4 are the first a5b1- or
avb3-selective integrin antagonists functionalized with thiol
groups for the surface modification of gold-based substrates.
They show a remarkable difference in mediating cell adhesion
of avb3- or a5b1-expressing fibroblasts. Furthermore, we
were able to functionalize antagonists 1 and 2 in an efficient
way without losing activity as well as selectivity, as demon-
strated by the measured IC50 values. Additionally, this study
gives a strong hint that such antagonists could be similarly
functionalized in that position for any purpose (for example
for molecular imaging) while maintaining their biological
profiles. The development of these molecules provides a new
method to investigate differences between a5b1- or avb3-
mediated cell adhesive processes.
[12] a) S. P. Massia, J. A. Hubbell, J. Cell Biol. 1991, 114, 1089 – 1100;
b) D. J. Irvine, A. V. G. Ruzette, A. M. Mayes, L. G. Griffith,
Biomacromolecules 2001, 2, 545 – 556; c) U. Hersel, C. Dahmen,
H. Kessler, Biomaterials 2003, 24, 4385 – 4415; d) W. Kuhlman, I.
Taniguchi, L. G. Griffith, A. M. Mayes, Biomacromolecules
2007, 8, 3206 – 3213; e) Y. Ohmuro-Matsuyama, Y. Tatsu,
Angew. Chem. 2008, 120, 7637 – 7639; Angew. Chem. Int. Ed.
2008, 47, 7527 – 7529; f) J. Sꢃnchez-Cortꢁs, M. Mrksich, Chem.
Biol. 2009, 16, 990 – 1000; g) J. Sꢃnchez-Cortꢁs, M. Mrksich, ACS
Chem. Biol. 2011, 6, 1078 – 1086; h) B. F. Bell, M. Schuler, S.
Tosatti, M. Textor, Z. Schwartz, B. D. Boyan, Clin. Oral Impl.
Res. 2011, 22, 865 – 872.
[13] a) M. A. Dechantsreiter, E. Planker, B. Mathꢄ, E. Lohof, G.
Hçlzemann, A. Jonczyk, S. L. Goodman, H. Kessler, J. Med.
Chem. 1999, 42, 3033 – 3040; b) C. Mas-Moruno, F. Rechen-
macher, H. Kessler, Anti-Cancer Agents Med. Chem. 2010, 10,
753 – 768; c) S. L. Goodman, G. Hçlzemann, G. A. G. Sulyok, H.
Kessler, J. Med. Chem. 2002, 45, 1045 – 1051; d) R. Stragies, F.
Osterkamp, G. Zischinsky, D. Vossmeyer, H. Kalkhof, U.
Reimer, G. Zahn, J. Med. Chem. 2007, 50, 3786 – 3794.
Received: August 8, 2012
Revised: October 3, 2012
Published online: December 12, 2012
[14] 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) D. Heckmann, B. Laufer, L.
Marinelli, V. Limongelli, E. Novellino, G. Zahn, R. Stragies, H.
Kessler, Angew. Chem. 2009, 121, 4501 – 4506; Angew. Chem.
Int. Ed. 2009, 48, 4436 – 4440.
[15] D. Heckmann, A. Meyer, B. Laufer, G. Zahn, R. Stragies, H.
Kessler, ChemBioChem 2008, 9, 1397 – 1407.
[16] C. Dahmen, J. Auernheimer, A. Meyer, A. Enderle, S. L.
Goodman, H. Kessler, Angew. Chem. 2004, 116, 6818 – 6821;
Angew. Chem. Int. Ed. 2004, 43, 6649 – 6652.
[17] J.-P. Xiong, T. Stehle, R. Zhang, A. Joachimiak, M. Frech, S. L.
Goodman, M. A. Arnaout, Science 2002, 296, 151 – 155.
[18] L. Marinelli, A. Meyer, D. Heckmann, A. Lavecchia, E.
Novellino, H. Kessler, J. Med. Chem. 2005, 48, 4204 – 4207.
[19] L. A. Carpino, G. Y. Han, J. Org. Chem. 1972, 37, 3404 – 3409.
[20] R. B. Merrifield, J. Am. Chem. Soc. 1963, 85, 2149 – 2154.
[21] A. O. Frank, E. Otto, C. Mas-Moruno, H. B. Schiller, L. Marine-
lli, S. Cosconati, A. Bochen, D. Vossmeyer, G. Zahn, R. Stragies,
E. Novellino, H. Kessler, Angew. Chem. 2010, 122, 9465 – 9468;
Angew. Chem. Int. Ed. 2010, 49, 9278 – 9281.
Keywords: cell adhesion · gold nanoarrays ·
integrin antagonists · peptide mimetics · surface coatings
.
[1] M. D. Pierschbacher, E. Ruoslahti, Nature 1984, 309, 30 – 33.
[2] C. J. Avraamides, B. Garmy-Susini, J. A. Varner, Nat. Rev.
Cancer 2008, 8, 604 – 617.
[3] R. O. Hynes, Cell 2002, 110, 673 – 687.
[4] R. O. Hynes, Nat. Med. 2002, 8, 918 – 921.
[5] M. Arnold, E. A. Cavalcanti-Adam, R. Glass, J. Blꢀmmel, W.
Eck, M. Kantlehner, H. Kessler, J. P. Spatz, ChemPhysChem
2004, 5, 383 – 388.
[6] B. Geiger, J. P. Spatz, A. D. Bershadsky, Nat. Rev. Mol. Cell Biol.
2009, 10, 21 – 33.
[7] R. Zaidel-Bar, S. Itzkovitz, A. Ma’ayan, R. Iyengar, B. Geiger,
Nat. Cell Biol. 2007, 9, 858 – 868.
[8] S. L. Goodman, M. Picard, Trends Pharmacol. Sci. 2012, 33, 405 –
412.
[9] E. Ruoslahti, Nat. Rev. Cancer 2002, 2, 83 – 90.
[10] a) B. G. Keselowsky, D. M. Collard, A. J. Garcia, J. Biomed.
Mater. Res. Part A 2003, 66A, 247 – 259; b) B. G. Keselowsky,
D. M. Collard, A. J. Garcia, Proc. Natl. Acad. Sci. USA 2005,
102, 5953 – 5957; c) E. A. Cavalcanti-Adam, T. Volberg, A.
[22] J. C. Love, L. A. Estroff, J. K. Kriebel, R. G. Nuzzo, G. M.
Whitesides, Chem. Rev. 2005, 105, 1103 – 1169.
Angew. Chem. Int. Ed. 2013, 52, 1572 –1575
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1575