Cyclic RGD peptide by ring-closing metathesis

Article abstract of DOI:10.1016/S0040-4039(02)00642-1

A cyclic pseudopeptide containing the RGD sequence has been synthesized in high yield by ring-closing metathesis using a Grubbs' catalyst in tetrahydrofuran. Conformational analysis has been correlated to structure modeling and preliminary biological acti

Full text of DOI:10.1016/S0040-4039(02)00642-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 3765–3767
Cyclic RGD peptide by ring-closing metathesis
Christian Bijani,a Ste´phane Varray,a Rene´ Lazaro,a Jean Martinez,a Fre´de´ric Lamatya,* and
Nelly Kiefferb
aLaboratoire des Aminoacides, Peptides et Prote´ines (LAPP), CNRS-Universite´s Montpellier 1 et 2, Place Euge`ne Bataillon,
34095 Montpellier Cedex 5, France
bLaboratoire Franco-Luxembourgeois de recherche biome´dicale (CNRS/CRP-Sante´), Luxembourg
Received 1 March 2002; accepted 14 March 2002
Abstract—A cyclic pseudopeptide containing the RGD sequence has been synthesized in high yield by ring-closing metathesis
using a Grubbs’ catalyst in tetrahydrofuran. Conformational analysis has been correlated to structure modeling and preliminary
biological activity as an integrin binding inhibitor has been evaluated. © 2002 Elsevier Science Ltd. All rights reserved.
Conformationally restricted structures obtained by
cyclizing aminoacid sequences often present relatively
to their linear precursors several advantages such as
increased metabolic stability, bioavailability and recep-
tor selectivity.1,2 True cyclic peptides resulting from
head-to-tail cyclization have been used as inhibitors of
tion have been applied to improve receptor selectivity in
RGD-dependent adhesion proteins,7 enkephalin,8
cholecystokinin,9 tachykinin.10
Since the recent discovery of efficient and robust cata-
lysts,11 cyclic pseudopeptides obtained by ring-closing
metathesis (RCM) start to become more popular due to
the smoothness of the catalysis conditions and the
tolerance for functional groups.12 In that way, cyclic
dithiopeptides can be modified and stabilized by
replacement of the S–S link for a CꢀC bond.12 More-
over, the olefin metathesis reaction should be consid-
ered as a new way for introducing the dehydrocarba
surrogate of the peptide bond.13a This enzymatically-
resisting lipophilic isosteric mimic increases the whole
biodisponibility of the target molecule.13b
integrin avb3 or of pancreatic trypsin,4 synthetic
3
immunogens,5 antigens for Herpes Simplex Virus.6
Those obtained by side chain-to-side chain lactamiza-
G
f
R
D
i
Boc NH
Chx
ii H NH
Boc
Boc
OH
Chx
NH
i
Chx
NH
H
OH
ii
Boc
Boc
Membrane adhesion proteins belonging to the integrin
superfamily are involved in different major diseases
such as: thrombosis, osteoporosis, diabetic retinopathy,
angiogenesis (reviewed in Refs. 14 and 15). Since many
integrins bind to their natural putative ligands of the
extracellular matrix mainly through the same RGD
sequence thus, their binding specificity must rely on the
spatial arrangement of the lateral chain charges of this
tripeptide. Increasing the selectivity of their inhibitors is
currently a challenging objective.
Chx
Chx
NH
NH
i
Ts
Boc
Boc
OHii H
Chx
Ts
NH
i
O
Chx
Chx
Ts
Ts
H
NH
NH
OH
ii
O
Experimental conditions: i CH2Cl2/TFA, ii BOP/DIEA, DMF
Scheme 1. Linear peptide precursor synthesis.
Starting from the c(RGDfV) peptide identified by
Kessler and Coll.3 as a potent cyclopeptide highly spe-
cific of the avb3 versus aIIBb3 integrin binding inhibi-
tion, we designed a dehydrocarba analog of the parent
peptide prepared by ring-closing metathesis of the cor-
responding diolefinic linear precursor containing the
* Corresponding author. Tel.: +33-467-143-847; fax: +33-467-144-866;
e-mail: frederic@univ-montp2.fr
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00642-1

3766
C. Bijani et al. / Tetrahedron Letters 43 (2002) 3765–3767
1
3-enoic acid residue)] as judged by a H NMR temper-
ature effect study (Table 2). The chemical shifts of the
amide proton showed a linear upfield shift as the tem-
perature increased, indicating that no major conforma-
tion changes occurred over this temperature range in
DMSO-d6. Temperature coefficients of >[5×10−3] ppm/
K are indicative of freely exchanging amide protons,
whereas values of <[3×10−3] ppm/K are generally con-
sidered to be involved in an intramolecular hydrogen
bond.
same RGDf sequence and bearing an allylic substitu-
tion at the N- and C-terminal positions. The linear
peptide was synthesized using a step by step coupling
strategy in solution as depicted in Scheme 1.
The Boc group was used as temporary protection and
HF-cleavable protection chosen for the side chain of
arginine (Ts) and aspartic acid (Chx). The linear pro-
tected peptide was obtained in nine steps from allyl-
amine to the butenoyl grafting with a global yield of
7.5%. Coupling reactions were carried out in the pres-
ence of BOP and DIEA in DMF as solvent and Boc
deprotection obtained with a CH2Cl2/TFA mixture.
A modeling study using the Genmol program20 gave as
the most stable conformation, a structure including the
two intramolecular NH bonds indicated above but
presenting a shorter distance between Ca Arg and Ca
Asp than that obtained with c(RGDfV) (0.49 nm versus
1.16 nm). Preliminary biological studies of this peptide
were performed in a cell adhesion assay using CHO cell
clones expressing recombinant human avb3 integrins.21
The dehydrocarba cyclopeptide was shown to inhibit
avb3-dependent cell adhesion to immobilized fibrinogen
with an IC100% at 0.25 mM.
The protected precursor has been submitted to RCM in
conventional conditions as it may be seen in Scheme 2.
Reactions were carried out at a concentration of 2 mM
and cyclization was monitored by LC/MS. Poor results
were firstly obtained using the first Grubbs’ catalyst I in
CH2Cl2 (entry 1, Table 1) until the solvent and catalyst
were changed towards more polar media and the sec-
ond Grubbs’ catalyst II. In THF, rarely used until
now16 for RCM, a quantitative cyclization yield (entry
6, Table 1) was hopefully obtained. This good result is
probably related to a better solubility of the peptide
and may also be due to a higher conformational free-
dom allowing a favorable position of the two olefinic
groups. One cannot also exclude a positive interaction
of THF oxygens with the catalyst, releasing it from
possible unproductive chelates due to the presence of
complexing functional groups of the peptidic
structure.17,18
In summary, a cyclic pseudopeptide containing an
olefinic moiety was successfully prepared by a catalytic
RCM performed in THF. This methodology would be
easily applied to other dehydrocarba analogs contain-
Table 2. Chemical shift assignments and temperature
coefficients of NH of cyclo-(R-G-
D
-f-Apa) in DMSO-d6
Entry
NH proton
lNH
DlNH (ppb)/DT (K)
1
2
3
4
5
Asp
Phe
Arg
Apa
Gly
8.73
8.50
8.35
7.47
7.42
−4.9
−5.4
−4.7
−1.8
−1.7
After HF deprotection, the cyclopseudopeptide was
purified and analyzed.19 1H NMR analysis19 showed a
high amount (more than 90%) of the cis-olefin
stereomer and two out of five possible intramolecular
hydrogen bonds [NH-Gly and NH-Apa (5-aminopent-
H
N
PCy3
Cl
HN
Ts
NH
Ru
L = PCy3 I
NH
O
Cl
Ph
HN
COOChx
O O
NH
L = IMes II
L
O
O
O
O
H
N
H
N
Solvent,
Temperature
HN
N
H
N
H
NH
HN
HN
NH
OChx
O
O
HN
O
O
Ts
Scheme 2. Peptide cyclization by olefin metathesis.
Table 1. Reaction conditions for the ring-closing metathesis
Entry
Catalyst (% mol)
Solvent
Temperature (°C)
Time (h)
Conversion (%)
1
2
3
4
5
6
I (10)
II (20)
II (20)
II (20)
II (20)
II (20)
CH2Cl2
CH2Cl2
CH2Cl2/THF (2/1)
CH2Cl2/THF (1/2)
THF
Reflux
Reflux
Reflux
Reflux
Room temperature
Reflux
24
3
3
3
3
0
59
74
89
82
THF
3
100

C. Bijani et al. / Tetrahedron Letters 43 (2002) 3765–3767
3767
ing the RGD sequence designed to increase the avb3
on a Waters Millennium with a photodiode array detec-
tor 996, wavelength 214 nm, using as a reversed phase
Nucleosil C18 column, 5m, (250×10 mm) with a flow rate
of 1 ml/min, eluants (solvent A: 0.1% TFA in H2O,
solvent B: 0.1% TFA in CH3CN).
binding inhibitor activity.
Acknowledgements
Butenoyl-L-Arg(Ts)-Gly-L-Asp(Cy)-D-PheNH-Allyl: Rf 0.5
(EtOAc 100%). 1H NMR (CD3OD) l 1.25–1.90 (m,
16H), 2.40 (s, 3H), 2.60–2.80 (m, 2H), 2.85–3.10 (m, 3H),
3.10–3.35 (m, 8H), 3.80 (d, J=5.0 Hz, 2H), 3.85 (s, 3H),
4.25 (t, 1H), 4.50–4.60 (m, 1H), 4.70 (t, J=46.0 Hz, 3H),
5.0–5.30 (m, 5H), 5.65–4.05 (m, 2H), 7.15–7.40 (m, 5H),
7.70–7.80 (d, J=9.0 Hz, 2H). 13C NMR (CD3OD) l
20.41, 23.20, 23.70, 25.40, 28.60, 29.67, 31.45, 31.50,
36.21, 37.60, 40.48, 41.76, 43.00, 50.14, 54.24, 55.80,
73.65, 115.25, 117.91, 126.13, 126.77, 128.51, 129.30,
129.32, 131.70, 134.12, 137.65, 141.16, 142.55, 157.65,
170.61 (2C), 171.61, 172.12, 173.47, 174.31. ESI-MS
837.4 (M+H+). HRMS calcd 837.3970, found 837.3969.
HPLC 12.07 min. IR cm−1 1130.7, 1253.2, 1450.4, 1542.0,
1638.6, 1720.2, 2931.8, 3290.8.
We thank the MENRT and the CNRS for financial
support.
References
1. Scott, C. P.; Abel-Santos, E.; Wall, M.; Wahnon, D. C.;
Benkovic, S. J. Proc. Natl. Acad. Sci. USA 1999, 96,
13638.
2. Gudmundsson, O. S.; Nimkar, K.; Gangwar, S.; Siahaan,
T.; Borchardt, R. T. Pharm. Res. 1999, 16, 16.
3. (a) Kessler, H. Angew. Chem., Int. Ed. Engl. 1982, 21,
512; (b) Haubner, R.; Finsinger, D.; Kessler, H. Angew.
Chem., Int. Ed. Engl. 1997, 36, 1374.
cyclo-(L-Arg(Ts)-Gly-L-Asp(Cy)-D-Phe-cis-Apa): Linear
4. Kasher, R.; Oren, D. A.; Barda, Y.; Gilon, C. J. J. Mol.
peptide (270 mg, 0.32 mmol) was dissolved in THF (160
ml) then Grubbs’ catalyst II (55 mg, 6.4.10−2 mmol) was
added to the solution. The mixture was stirred at reflux
overnight. THF was evaporated and residue was purified
over preparative HPLC to yield the title compound
Biol. 1999, 292, 421.
5. Brugghe, H. F.; Timmermans, H. A.; Van Unen, L. M.;
Ten Hove, G. J.; Van De Verken, G.; Poolman, J. T.;
Hoogerhout, P. Int. J. Pept. Protein Res. 1994, 43, 166.
6. Mezo, G.; Majer, Z.; Valero, M. L.; Andreu, D.; Hudecz,
F. J. Pept. Sci. 1999, 5, 272.
7. Peishoff, C. E.; Ali, F. E.; Bean, J. W.; Calvo, R.;
D’Ambrosio, C. A.; Eggleston, D. S.; Hwang, S. M.;
Kline, T. P.; Koster, P. F.; Nichols, A.; Powers, D.;
Romoff, T.; Samanen, J. M.; Stadel, J.; Vasko, J. A.;
Kopple, K. D. J. Med. Chem. 1992, 35, 3962.
8. Schiller, P. W.; Nguyen, T. M.-D.; Lemieux, C.; Maziak,
L. A. J. Med. Chem. 1985, 28, 1766.
9. Charpentier, B.; Dor, A.; Roy, P.; England, P.; Pham,
H.; Durieux, C.; Roques, B. P. J. Med. Chem. 1989, 32,
1184.
10. Holzemann, G.; Jonczyk, A.; Eiermann, V.; Pachler, K.
G. R.; Barnickel, G.; Regoli, D. Biopolymers 1991, 31,
691.
11. Trnka, T. N.; Grubbs, R. H. Acc. Chem. Res. 2001, 34,
18.
12. Miller, S. J.; Blakwell, H. E.; Grubbs, R. H. J. Am.
Chem. Soc. 1996, 118, 9606.
13. (a) Hann, M. M.; Sammes, P. G.; Kennewell, P. D.;
Taylor, J. B. J. Chem. Soc., Perkin 1 1982, 307; (b) Hann,
M. M.; Sammes, P. G.; Kennewell, P. D.; Taylor, J. B. J.
Chem. Soc., Chem. Commun. 1980, 234.
1
(60%). H NMR (CD3OD) l 1.25–1.95 (m, 15H), 2.40 (s,
3H), 2.50–2.55 (m, 1H), 2.80–2.95 (m, 2H), 3.10 (q,
J=6.5 Hz, 1H), 3.15–3.25 (sl, 2H), 3.35 (dd, J=4.5 Hz,
J=14.0 Hz, 1H), 3.80–3.95 (m, 4H), 4.35 (q, J=4.5 Hz,
1H), 4.45 (t, J=6.5 Hz, 1H), 4.60 (dd, J=4.5 Hz, J=10.5
Hz, 1H), 4.65–4.75 (m, 1H), 5.65–5.85 (m, 2H), 7.20–7.35
(m, 7H), 7.75 (d, J=8.5 Hz, 2H). 13C NMR (CD3OD) l
20.42, 23.66, 25.40, 27.96, 31.45, 35.29, 36.74, 39.75,
41.34, 41.99, 51.49. 53.43, 55.00, 73.60, 125.74, 126.16,
126.75, 128.49, 129.11, 129.26, 129.34, 137.93, 169.96,
170.14, 171.93, 172.15, 172.69, 173.15. HPLC 11.175 min.
HRMS calcd 809.3665, found 809.3665.
cyclo-(L-Arg-Gly-L-Asp-D-Phe-cis-Apa): Cyclic peptide
(0.040 g, 0.049 mmol) was added to a solution of anhy-
dric fluorhydric acid at 0°C for 1 h in the presence of
anisole. The mixture was heated to 20°C and HF was
evaporated. The residue was precipitated in ether then
purified over preparative HPLC to yield 0.020 g (60%) of
1
the title compound. H NMR (DMSO-d6) l 1.15 (sl, 1H),
1.30–1.50 (m, 4H), 1.75 (sl, 1H), 2.20–2.30 (m, 2H),
2.60–2.75 (m, 2H), 2.90–3.10 (m, 3H), 3.60–3.80 (m, 4H),
4.00–4.10 (m, 1H), 4.10–4.20 (m, 1H), 4.25–4.35 (m, 1H),
5.50–5.70 (m, 2H), 7.05–7.20 (m, 5H), 7.35 (sl, 1H), 7.40
(sl, 1H), 8.25 (d, J=8.5 Hz, 1H), 8.40 (d, J=9.0 Hz, 1H),
8.65 (d, J=5.0 Hz, 1H), 12.35 (sl, 1H). 13C NMR
(DMSO-d6) l 26.16, 28.61, 35.85, 37.03, 41.73, 42.34,
51.87, 52.99, 54.75, 126.87, 127.03, 128.94, 129.53, 139.33,
157.51, 170.02, 171.07, 171.29, 171.79. ESI-MS m/z 573.2
(M+H+). HRMS calcd for C26H37O7N8 573.2785, found
573.2778. HPLC 7.312 min.
14. Hynes, R. O. Cell 1992, 69, 11.
15. Arnaout, M. A. Immunol. Rev. 1990, 114.
16. Marsella, M. J.; Maynard, H. D.; Grubbs, R. H. Angew.
Chem., Int. Ed. Engl. 1997, 36, 1101.
17. Conde-Frieboes, K.; Andersen, S.; Breinholt, J. Tetra-
hedron Lett. 2000, 41, 9153.
18. Varray, S.; Gauzy, C.; Lamaty, F.; Lazaro, R.; Martinez,
J. J. Org. Chem. 2000, 65, 6787 and references cited
therein.
20. Pe`pe, G.; Siri, D. Stud. Phys. Theor. Chem. 1990, 71, 93.
21. Morel-Kopp, M. C.; Melchior, C.; Chen, P.; Ammerlaan,
W.; Lecompte, T.; Kaplan, C.; Kieffer, N. Thromb.
Haemost. 2001, 86, 1425.
1
19. All new compounds gave satisfactory analytical data: H
and 13C NMR analyses were performed with a 400 MHz
NMR spectrometer. The HPLC analyses were carried out