A convenient access to αVβ3/αVβ5 integrin ligand conjugates: Regioselective solid-phase functionalisation of an RGD based peptide

Article abstract of DOI:10.1016/S0040-4039(01)00293-3

The cyclopeptide (-RGDfK-) is a potent and selective α_Vβ₃/α_Vβ₅ integrin ligand. A methodology for the conjugation of cyclo(-RGDfK-) through the regioselective derivatisation of lysine side chains, either in solution or directly on the solid support, is described. This provides a rapid and flexible chemical entry to conjugated integrin ligands bearing reporter groups for biological investigations or reactive chemical functions for the preparation of new vector systems.

Full text of DOI:10.1016/S0040-4039(01)00293-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 2787–2790
A convenient access to aVb3/aVb5 integrin ligand conjugates:
regioselective solid-phase functionalisation of an RGD
based peptide
Didier Boturyn and Pascal Dumy*
LEDSS, Inge´nierie Mole´culaire et Chimie des Compose´s Bioorganiques, UMR CNRS 5616, Universite´ Joseph Fourier, BP 53,
38041 Grenoble Cedex 9, France
Received 14 February 2001; revised 19 February 2001; accepted 20 February 2001
Abstract—The cyclopeptide (-RGDfK-) is a potent and selective aVb3/aVb5 integrin ligand. A methodology for the conjugation of
cyclo(-RGDfK-) through the regioselective derivatisation of lysine side chains, either in solution or directly on the solid support,
is described. This provides a rapid and flexible chemical entry to conjugated integrin ligands bearing reporter groups for biological
investigations or reactive chemical functions for the preparation of new vector systems. © 2001 Elsevier Science Ltd. All rights
reserved.
The integrin family of adhesion molecules participate in
important cell–cell and cell–extracellular matrix interac-
tions in a diverse range of biological processes.1 The
aVb3/aVb5 integrins are expressed in various cell types,
such as endothelial cells, melanoma, platelets,
osteoclasts and smooth muscle cells. Furthermore, they
play an important role in angiogenesis as well as in
tumour cell migration by interacting with vitronectin on
the extracellular matrix mainly through the recognition
of the tripeptide sequence RGD.2 The search for highly
selective ligands to target the endothelial cell integrin
aVb3 is currently the focus of many research groups as
they may represent new therapeutics or valuable diag-
nosis in a number of areas such as metastasis, angio-
genesis, arthritis and retinopathy.
specificity of these peptides for the aVb3 integrin is
increasingly exploited to target endothelial cells for
drugs or DNA delivery,7 as well as for the diagnosis of
tumour formation.8
For these reasons, there is a current need for selective
RGD ligands, which may be readily synthesised and
easily derivatised. The cyclopeptide (-DfKRG-), con-
taining a lysine following a
D-phenylalanine residue,
displays an excellent affinity and selectivity toward
aVb3/aVb5 receptors.9 Most importantly, this peptide
represents a very interesting ligand for biological inves-
tigations of integrin because it can be easily function-
alised through the lysine side chain without significant
modification of its binding capacity. Indeed, the lysine
side-chain functionalisation has been used for surface-
mediated osteoblast adhesion or for biological detection
of the binding to the integrin.10 However, there is no
general synthesis described so far for this purpose. This
communication describes a methodology suitable for
the synthesis of cyclic (-DfKRG-) peptides as well as
for the cognate peptide conjugates through the lysine
side chain in solution and on solid phase.
To date, the cyclic (-RGDfX-) peptides3 and related
N-methylated analogues4 developed by Kessler’s group
are among the most active and selective compounds for
the aVb3 and aVb5 integrin subtype receptors. For
instance, the combined use of c(-RGDf(NMe)V-) and
an antibody–cytokine fusion protein led to an impres-
sive in vivo tumour eradication in mice.5 They are also
under investigation in clinical studies; they are used
especially in phase I/II in patients with progressive or
recurrent anaplastic glioma, and in phase I in patients
with HIV related Kaposi’s sarcoma.6 Moreover, the
The linear-protected peptides were assembled using
standard Fmoc solid-phase chemistry with a highly
acid-labile o-chlorotrityl chloride® as well as Sasrin®
resins. The very mild acid conditions for the release of
the linear-protected peptide from this support (1%
TFA/CH2Cl2, 2 min or AcOH/TFE/CH2Cl2, 1/1/3, 2 h)
are compatible with the side-chain protection stability.
* Corresponding author. Fax: 33(0)4.76.51.43.82; e-mail: pascal.
dumy@ujf-grenoble.fr
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00293-3

2788
D. Boturyn, P. Dumy / Tetrahedron Letters 42 (2001) 2787–2790
This is mandatory for the subsequent head-to-tail cycli-
sation performed in solution. Further, the use of glycine
as a C-terminal end ensures that cyclisation between the
lysine a-amino (head) to glycine a-carboxyl (tail)
groups is racemisation free. Our synthetic plans have to
fulfil the following synthetic aspects: (1) the removal of
the protecting group of the lysine side chain has to be
compatible with the stability of aspartic and arginine
side-chain protection. (2) Modifications of the lysine
side chain have to be feasible directly on the solid
support before the release of the linear-protected pep-
tide and its solution cyclisation. For this purpose, the
Pd°-labile Aloc group on the lysine was preferred. In
the first instance (Scheme 1, right side), Fmoc-K-
(Aloc)-OH was introduced during the SPPS assembly in
an orthogonal combination with tBu and Pmc groups
of aspartic and arginine, respectively.
quently by the same Fmoc chemistry following the
Na-Aloc removal. The added advantage of this two
directional approach is that all the steps are carried out
on the resin, thus opening the way to automation of the
whole process, including the cyclisation as well as the
final side-chain deprotection.
The assembly of all RGD containing peptides was
carried out as outlined in Scheme 1. The Aloc cleavage
on the resin or in solution was readily performed by the
well-established Pd(PPh3)4/PhSiH3 procedure.12 In gen-
eral, the linear-protected peptides were obtained in
good yield and high purity. The cyclisation step, per-
formed at high dilution in DMF using PyBOP reagent,
provided the corresponding RGD cyclic peptides in
almost quantitative yields after a simple ether precipita-
tion. It is interesting to note that the presence of
chemical appendage on the lysine side chain does not
seem to interfere with the cyclisation reaction.
This provided the flexibility required to regioselectively
address the lysine side chain of 1 (Scheme 1) in solution
subsequently to the cyclisation step. On the other hand,
the converse combination, e.g. Aloc-K(Fmoc)-OH,11
allowed the regioselective elongation from the side
chain directly on the solid support by Fmoc chemistry
(Scheme 1, left side). Most notably, the normal back-
bone elongation of the peptide is performed subse-
Finally, deprotection using a solution of 95% TFA,
2.5% triisopropylsilane (TIS) and 2.5% water afforded,
after 2 hours, the desired deprotected peptides 1–4. It is
worthy of note that usually only one final purification
was required at the last step for the whole synthesis.
Scheme 1. Regioselective modification of the lysine side chain directly on the solid support (left) before cyclisation or in solution
after cyclisation (right) to afford cyclopeptides 2–4 conjugated with X.

D. Boturyn, P. Dumy / Tetrahedron Letters 42 (2001) 2787–2790
2789
The simplicity and versatility of these two approaches
are exemplified by appending to the RGD cyclopeptide
a biotin containing b-alanine linker (compounds 4a–f),
an aminooxy or a glyoxylaldehyde group through the
lysine side chain (Scheme 1). Compounds 4 bear a
biotin tag for detecting the binding of the ligand to cell
presenting integrins during biological assays. Our
approach permitted to readily assess different linker
lengths (n=0–5) between the biotin and the RGD
ligand by parallel SPPS elongation of the lysine side
chain with Fmoc-bAla-OH. Compounds 2 and 3 repre-
sent important functional groups required for chemose-
lective oxime formation with the corresponding
complementary molecule. They open up the possibility
for further chemical manipulation of RGD cyclic pep-
tide without protecting groups. Such an approach is
widely exploited with success in our laboratory for
multivalent presentation of cyclic RGD peptides as well
as DNA derivatisation.13
National de la Recherche Scientifique (CNRS). The
‘Institut Universitaire de France’ is greatly acknowl-
edged for financial support.
References
1. Hynes, R. O. Cell 1992, 69, 11–25.
2. (a) Brooks, P. C.; Clark, C. F.; Cheresh, D. A. Science
1994, 264, 569–571; (b) Giannis, A.; Ru¨bsam, F. Angew.
Chem., Int. Ed. Engl. 1997, 36, 588–590.
3. Haubner, R.; Finsinger, D.; Kessler, H. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 1374–1389.
4. Dechantsreiter, M. A.; Planker, E.; Matha, B.; Lohof, E.;
Holzeman, G.; Jonczyk, A.; Goodman, S. L.; Kessler, H.
J. Med. Chem. 1999, 42, 3033–3040.
5. Lode, H. N.; Moehler, T.; Xiang, R.; Jonczyk, A.;
Gillies, S. D.; Cheresh, D. A.; Reisfeld, R. A. Proc. Natl.
Acad. Sci. USA 1999, 96, 1591–1596.
6. National Cancer Institute clinical trials database online.
See: http://cancertrials.nci.nih.gov
7. Colin, M.; Maurice, M.; Trugnan, G.; Kornprobst, M.;
Harbottle, R. P.; Knight, A.; Cooper, R. G.; Miller, A.
D.; Capeau, J.; Coutelle, C.; Brahimi-Horn, M. C. Gene
Ther. 2000, 7, 139–152.
8. Haubner, R.; Wester, H.-J.; Reuning, U.; Senekowitsch-
Schmidtke, R.; Diefenbach, B.; Kessler, H.; Sto¨cklin, G.;
Schwaiger, M. J. Nucl. Med. 1999, 40, 1061–1071.
9. Kantlehner, M.; Finsinger, D.; Meyer, J.; Schaffner, P.;
Jonczyk, A.; Diefenbach, B.; Nies, B.; Kessler, H. Angew.
Chem., Int. Ed. 1999, 38, 560–562.
10. Kantlehner, M.; Schaffner, P.; Finsinger, D.; Meyer, J.;
Jonczyk, A.; Diefenbach, B.; Nies, B.; Ho¨lzemann, G.;
Goodman, S. L.; Kessler, H. ChemBioChem 2000, 1,
107–114.
Aminooxy containing peptide 2 was introduced using
the activated N-hydroxysuccinimide (O-carboxy-
methyl)-N-Boc-hydroxylamine derivative (Scheme 1,
J=BocNHOCH2CO). Peptide 3, bearing an aldehyde
function, was generated after serine residue sodium
periodate oxidation14 of unprotected cyclo(-D-f-K(o-S)-
R-G-). For this purpose, Boc-Ser(tBu)-OH was previ-
ously coupled with PyBOP/DIEA to the lysine o-amino
group (Scheme 1, J=BocSer(OtBu)). Following the
same procedure, biotin was also conjugated to the
peptide subsequently to elongate the lysine side chain
with b-alanine residues as spacer molecules. Products
1–4 were purified by preparative RP-HPLC after the
final deprotection and the structure of peptides con-
firmed by ES-MS.15 Compounds obtained either by
solid-phase or solution derivatisation of the lysine were
found to be identical. As anticipated, the solid-phase
approach is much more advantageous than the solution
derivatisation in terms of rapidity, product purity and
quantity (110–190 mg, 60–80% overall yield relative to
the initial resin loading versus 30% in solution).
11. To
a solution containing 2 g (4.2 mmol) of H-
K(Fmoc)OH and 2.2 g of Na2CO3 (20.7 mmol) in 50 mL
of H2O/dioxan (3/2) at 0–4°C, 500 mL (4.7 mmol) of allyl
chloroformate in 10 mL of dioxan was added over a 1 h
period. After 4 h, the reaction mixture was washed three
times with 10 mL of ether. The aqueous solution was
added dropwise to a cold 1 M HCl solution and the
product was subsequently extracted with ethyl acetate
and filtered over Na2SO4 to afford 1.80 g (4 mmol, 95%)
of the desired product as a white powder. NMR (300
In conclusion, the strategy described here provides a
convenient synthetic access to various conjugates of the
bioactive cyclo(-D-f-K-R-G-) peptides. Most impor-
tantly, the derivatisation is achieved directly on the
solid support, the cyclisation and final deprotection
remaining the sole steps performed in solution. These
conditions are flexible and easily prone to automation
for peptide synthesis. This is an important point one
has to consider with respect to the biological impact
these compounds may represent in the near future. The
solid-phase method should also be suitable for large-
scale synthesis given some similarity to those recently
reported.16 Biological exploitations of such conjugates
are currently underway in our laboratory.13b
1
MHz for H, DMSO-d6 solvent, J in Hz): 12.50 (1H, s,
CO2H), 7.88 (2H, d, J=7.3, ArH-Fmoc), 7.68 (2H, d,
J=7.3, ArH-Fmoc), 7.46 (1H, d, J=8.0, NH-Aloc), 7.41
(2H, t, J=7.3, ArH-Fmoc), 7.32 (2H, t, J=7.3, ArH-
Fmoc), 7.27 (1H, t, J=5.9, NH-Fmoc), 5.89 (1H, m,
CHꢀCH2), 5.29 (1H, d, J=15.6, CHꢀCH2), 5.17 (1H, d,
J=10.4, CHꢀCH2), 4.47 (2H, d, J=5.3, CH2-CHꢀCH2),
4.29 (2H, d, J=6.9, CH2-Fmoc), 4.20 (1H, t, J=6.3,
CH-Fmoc), 3.89 (1H, m, CHa), 2.96 (2H, m, CH2d), 1.61
(2H, m, CH2b), 1.35 (4H, m, CH2g, CH2o) ppm.
12. (a) Dangles, O.; Guibe´, F.; Balavoine, G.; Lavielle, S.;
Marquet, A. J. Am. Chem. Soc 1987, 52, 4984–4993; (b)
Thieriet, N.; Alsina, J.; Giralt, E.; Guibe´, F.; Albericio,
F. Tetrahedron Lett. 1997, 38, 7275–7278.
Acknowledgements
13. (a) Forget, D.; Boturyn, D.; Defrancq, E.; Lhomme, J.;
Dumy, P., submitted; (b) Boturyn, D.; Coll, J. L.; Favrot,
M. C.; Dumy, P., work in preparation.
This work was supported by the Association pour la
Recherche sur le Cancer (ARC) and the Centre

2790
D. Boturyn, P. Dumy / Tetrahedron Letters 42 (2001) 2787–2790
14. Geoghegan, K. F.; Stroh, J. G. Bioconjugate Chem. 1992,
676.7); 3, M+H]+=678.4 (C29H43N9O10, m/z calcd
677.7); 4a, M+H]+=830.8 (C37H55N11O9S, m/z calcd
829.9); 4b, M+H]+=902.2 (C40H60N12O10S, m/z calcd
901.1); 4c, M+H]+=973.2 (C43H65N13O11S, m/z calcd
972.1); 4d, M+H]+=1044.3 (C46H70N14O12S, m/z calcd
1043.2); 4e, M+H]+=1114.7 (C49H75N15O13S, m/z calcd
1113.5); 4f, M+H]+=1186.0 (C52H80N16O14S, m/z calcd
1185.4).
3, 138–146.
,
15. HPLC characterisation (Nucleosil 100 A 5 mm C18 parti-
cles, 250×4.6 mm; linear gradient 95:5 A:B to 0:100 A:B
in 30 min; solvent B: 0.09% TFA in 90% acetonitrile and
solvent A 0.09% TFA, 1 mL/min, detection u=214 and
250 nm), 2, tR=13.9 min; 3, tR=15.2 min; 4a, tR=17.7
min; 4b, tR=16.9 min; 4c, tR=16.7 min; 4d, tR=16.4
min; 4e, tR=16.1 min; 4f, tR=16.0 min. ES-MS (positive
mode): 2, M+H]+=677.0 (C29H44N10O9, m/z calcd
16. Dai, X.; Su, Z.; Liu, J. O. Tetrahedron Lett. 2000, 41,
6295–6298.
.