Synthesis of cyclic peptides via O-N-acyl migration

Article abstract of DOI:10.1016/j.tetlet.2008.05.049

We describe here a novel and convenient synthesis of head-to-tail cyclic peptide avoiding racemization. Linear depsipeptides including a serine residue as the key element for ester bond formation and acyl transfer were synthesized on 2-chlorotrityl chloride resin. After cleavage from the resin, intramolecular head-to-tail cyclization was performed in solution by C-terminal activation of urethane protected O-acyl serine residue. After removal of the N^α-serine protecting group, the final step consisted in O-N-acyl migration reaction on the 'switch' or 'click' element to restore native cyclic peptides.

Full text of DOI:10.1016/j.tetlet.2008.05.049

Tetrahedron Letters 49 (2008) 4674–4676
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis of cyclic peptides via O–N-acyl migration
*
Jennifer Lécaillon, Pierre Gilles, Gilles Subra, Jean Martinez, Muriel Amblard
Institut des Biomolécules Max Mousseron, UMR 5247 CNRS, Université Montpellier 1, Université Montpellier 2, Faculté de pharmacie, 15 Av. C. Flahault, BP 14 491, 34093
Montpellier cedex 5, France
a r t i c l e i n f o
a b s t r a c t
Article history:
We describe here a novel and convenient synthesis of head-to-tail cyclic peptide avoiding racemization.
Linear depsipeptides including a serine residue as the key element for ester bond formation and acyl
transfer were synthesized on 2-chlorotrityl chloride resin. After cleavage from the resin, intramolecular
head-to-tail cyclization was performed in solution by C-terminal activation of urethane protected O-acyl
serine residue. After removal of the Na-serine protecting group, the final step consisted in O–N-acyl
migration reaction on the ‘switch’ or ‘click’ element to restore native cyclic peptides.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 26 March 2008
Revised 29 April 2008
Accepted 7 May 2008
Available online 14 May 2008
Keywords:
O–N-Acyl migration
Cyclic peptides
Depsipeptides
Epimerization-free
1. Introduction
zation-free’ segment condensation.8–13 The presence of the
hydroxylated side chain of the serine was used to build up the pep-
Despite their involvement in several physiological and physio-
pathological processes, few peptides have been successfully devel-
oped as therapeutic agents, mainly due to their poor bioavailability
and their lack of stability against enzymatic degradation. However
cyclic peptides are of great pharmacological interest in view of
their improved chemical and biochemical stability. Moreover, the
conformational restrictions induced by cyclization significantly
improve their activity and/or selectivity towards biological targets.
Due to the interest in cyclization to overcome the limitations of lin-
ear peptides, many efforts have been devoted to develop efficient
methods for the synthesis of cyclic peptides.1–3 From a strategic
point of view, in most cases head-to-tail cyclization requires acti-
vation of a non-urethane-protected amino acid, leading to epimer-
ization. To minimize this problem the choice of coupling reagent
needs to be carefully examined.1 On-resin cyclization strategies
were developed using as examples, Kaiser oxime,4 thioester5 res-
ins, only compatible with Boc-chemistry, and ‘safety catch’ lin-
ker.6,7 Some of them are described to be free from racemization.
We propose here an alternative route for epimerization-free
cyclization of serine-containing peptides. Our study focused on
the serine as a key element to generate a depsipeptide and to pro-
mote racemization-free head-to-tail cyclization. The O–N-acyl
migration was used to restore the native peptide. This strategy
was the subject of extensive studies and applications, mainly for
the synthesis of peptides containing difficult sequence such as
the amyloid beta peptide and prodrug design but also for ‘racemi-
tide sequence via an O-acyl isopeptide bond. The depsipeptide with
a C-terminal urethane-protected serine residue was then em-
ployed for the cyclization. In order to restore native cyclic peptides,
the last step of the synthesis consisted in an O–N-acyl intramole-
cular transfer on the serine, called ‘switch’ or ‘click’ element
(Scheme 1).8,10
In this study, four model depsipeptides (Table 1) were synthe-
sized on solid support, and the cyclization step was performed in
solution followed by the O–N-acyl transfer. The influence of the
cycle size for the intramolecular migration was investigated. The
synthesis of difficult and non-favorable cyclic tetrapeptide by
using the acyl-transfer reaction strategy as an induction element
was included in this study. Effectively, cyclic depsitetrapeptides
with less severe conformational constraints could be templates
for the very difficult formation of cyclic tetrapeptides.
2. Results
The linear depsipeptides 6–9 were synthesized on solid support
(Scheme 1). Boc-Ser-OH was anchored on a 2-chlorotrityl chloride
resin.
The second amino acid, Fmoc-Ala-OH was added by acylation
on the serine side chain using DIC/DMAP as coupling reagent.
Both reactions were performed at room temperature and under
microwave irradiation, and the latter condition allowed a drastic
reaction time decrease (from 24 h to 40 min). The extent of epi-
merization in the synthesis of Ser-containing depsipeptides is re-
ported to be very low.14 However as it could be a limitation to
develop a racemization-free cyclization strategy, this point was
* Corresponding author. Tel.: +33 4 67 54 86 37; fax: +33 4 67 54 86 54.
E-mail address: muriel.amblard@univ-montp1.fr (M. Amblard).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.05.049

J. Lécaillon et al. / Tetrahedron Letters 49 (2008) 4674–4676
4675
O
O
O
BocNH
BocNH
O
Fmoc SPPS
FmocHN
acylation*
O
BocNH
HO
O
O
Rn
O
O
H
N
FmocNH
N
O
O
H
TFE/AcOH/DCM
Rn+1
O
R1
n
R1
Depsipeptide
4 and 5
6-9
O
O
BocHN
O
BocNH
NH
1- Pip/DMF
OH
Rn+1
O
TFA
Cyclization
2- TFE/AcOH/DCM
H2N
O
Rn
O
H
NH
Rn
N
O
N
O
R1
n
H
HN
Rn+1
O
R1
n
Depsipeptide
O
10-13
14-17
O
O
O
+
NH3
NH2
O
NH
NH
Rn+1
Rn+1
HO
NH
O-N intramolecular
acyl migration
Rn+1
O
O
O
O
NH
O
basic condition
O
NH
Rn
NH
Rn
O
O
NH
R1
N
R1
R1
n
n
n
HN
HN
H
Rn
O
O
18-21
1-3
Scheme 1. General conditions for cyclic peptide synthesis. Resin: 2-chlorotrityl chloride; loading: Boc-Ser-OH (5 equiv), DIEA (10 equiv), DCM 30 min microwave 32 °C;
acylation: Fmoc-AA1-OH (6 equiv/ DIC 3 equiv), DMAP (0.3 equiv), DMF, 2 Â 5 min microwave 70 °C; SPPS: deprotection: Pip/DMF (20/80); coupling: Fmoc-AA-OH (3 equiv),
DIC (3 equiv)/HOBt (3 equiv), DMF; cleavage: TFE/AcOH/DCM (20/10/70); cyclization: NaHCO3 (5 equiv), BOP (1 equiv), DMF; O–N-Acyl shift: aqueous phosphate buffer pH
*
7.4 or DMF/Pip (80/20) solution. : Acylation with Fmoc-Ala-OH and Fmoc-D-Ala-OH gives, respectively, after TFE/AcOH/DCM cleavage compounds 4 and 5.
Table 1
Depsi- and peptide sequences
Peptide sequence
O
O
O
O
O
O
Ser Peptide
+H3N
Peptide
BocNH
Fmoc-Peptide
BocNH
H-Peptide
BocHN
Peptide
O
OH
O
O
Leu-Phe-Ala
Phe-Leu-Phe-Ala
Ala-Phe-Leu-Phe-Ala
Leu-Ala-Phe-Leu-Phe-Ala
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
No transfer
1
2
3
Boc-Ser(Fmoc-Ala)-OH
10.75
Boc-Ser(Fmoc-DAla)-OH
8.15
analyzed by synthesizing D-alanine-containing peptides. After TFE/
0.7
0.6
0.5
AcOH/DCM treatment of a small sample of each batch of resin, the
resulting compounds 4 and 5 were analyzed on chiral RP-HPLC for
determining the loss of configuration. Examination of the spectra
showed that the acylation at room temperature and under micro-
wave occurs with less than 1% epimerization (Fig. 1).
0.40
0.35
0.30
0.25
0.20
0.15
Subsequent amino acids were incorporated by using DIC/HOBt
coupling strategy. After Pip/DMF treatment, the linear depsipep-
tides 6–9 were cleaved from the solid support with a mixture of
TFE/AcOH/DCM to keep the N-terminal Boc protection. The depsi-
peptides 10–13 were cyclized in solution by activation of the car-
boxylic acid function of the N-protected serine residue using BOP,
NaHCO3, DMF, and diluted conditions to favor intramolecular cou-
pling (compounds 14–17). After removing the Boc-protecting
group, the cyclic depsipeptides (18–21) were subjected to the O–
N-acyl migration reaction. They were dissolved and stirred in aque-
ous phosphate buffer at pH 7.4, and the O–N intramolecular acyl
migration for each compound was followed by RP-HPLC and LC/
MS analyses. Under these conditions after several days in the phos-
phate buffer (pH 7.4) at room temperature or at 80 °C or under
microwave irradiation, no transfer could be observed on the depsi-
tetrapeptide 18, certainly due to the important cycle constraint of
the final structure. Only a side reaction corresponding to ester
bond hydrolysis occurred after 72 h.
0.4
0.3
0.2
0.10
0.08
0.00
-0.05
Boc-Ser(Fmoc-DAla)-OH
0.0
-0.1
0
5
10
15
0
5
10
15
Min
Min
Figure 1. HPLC chromatogram (chiralcel OD-RH column; detection at 254 nm; el-
ution under isocratic conditions: 55% B during 30 min at a flow rate of 1 mL/min.
Solvent A = H2O/0.1% TFA. Solvent B = CH3CN/0.1% TFA) of the crude Boc(L)Ser(F-
moc-Ala)-OH (4) and the crude Boc(L)Ser(Fmoc-D-Ala)-OH (5).

4676
J. Lécaillon et al. / Tetrahedron Letters 49 (2008) 4674–4676
It is worth noting that no hydrolysis was observed when O–N-
acyl migration switch was applied to the linear depsipentapeptide
(data not shown), clearly suggesting that hydrolysis was mainly
promoted by a transition state cycle constraint.
For cyclic depsipeptides with higher cycle sizes (depsihexapep-
tide 20 and depsiheptapeptide 21) the intramolecular transposi-
tion reaction was quantitative in phosphate buffer at room
temperature after 12 h affording compounds 2 and 3.
In conclusion, the O–N intramolecular acyl migration was suc-
cessfully applied to the non-racemizing synthesis of cyclic peptides
containing serine residue. We demonstrated the influence of the
cycle size on O–N-acyl intramolecular transfer and the existence
of a hydrolysis side reaction when the cycle constraints become
significant. However, this side reaction can be totally suppressed
under organic conditions. The strategy using more conformational-
ly favorable cyclic depsitetrapeptide intermediates did not allow
the difficult formation of cyclic tetrapeptides. Finally, the efficiency
of this methodology will be applied to threonine-containing pep-
tides and to the design of a new strategy for the synthesis of cyclic
peptides on solid support.
Supplementary data
Experimental procedures, HPLC and mass spectrometry analy-
ses. 1H and 13C NMR data of compounds 19 and 1. Supplementary
data associated with this article can be found, in the online version,
at doi:10.1016/j.tetlet.2008.05.049.
References and notes
1. Davies, J. S. review J. Peptide Sci. 2003, 9, 471–501.
2. Li, P.; Roller, P. P.; Xu, J. Curr. Org. Chem. 2002, 6, 411–440.
3. Lambert, J. N.; Mitchell, J. P.; Roberts, K. D. J. Chem. Soc., Perkin Trans. 1 2001, 5,
471–484.
4. Mihara, H.; Yamabe, S.; Nildome, T.; Aoyagi, H.; Kumagai, H. Tetrahedron Lett.
1995, 36, 4837–4840.
5. Richter, L. S.; Tom, J. Y. K.; Brunier, J. P. Tetrahedron Lett. 1994, 35, 5547–
5550.
6. Yang, L.; Moriello, G. Tetrahedron Lett. 1999, 40, 8197–8200.
7. Bourne, G. T.; Meutermans, W. D. F.; Alewood, P. F.; McGeary, R. P.; Scanlon, M.;
Watson, A. A.; Smythe, M. L. J. Org. Chem. 1999, 64(9), 3095–3101.
8. Mutter, M.; Chandravarkar, A.; Boyat, C.; Lopez, J.; Dos Santos, S.; Mandal, B.;
Mimna, R.; Murat, K.; Patiny, L.; Saucede, L.; Tuchscherer, G. Angew. Chem., Int.
Ed. 2004, 43, 4172–4178.
9. Carpino, L. A.; Krause, E.; Sferdean, C. D.; Schümann, M.; Fabian, H.; Bienert, M.;
Beyermann, M. Tetrahedron Lett. 2004, 45, 7519–7523.
10. Taniguchi, A.; Sohma, Y.; Kimura, M.; Okada, T.; Ikeda, K.; Hayashi, Y.;
Kimura, T.; Hirota, S.; Matsuzaki, K.; Kiso, Y. J. Am. Chem. Soc. 2006, 128, 696–
697.
11. Sohma, Y.; Kiso, Y. ChemBioChem 2006, 7, 1549–1557.
12. Sohma, Y.; Yoshiya, T.; Taniguchi, A.; Kimura, T.; Hayashi, Y.; Kiso, Y.
Biopolymers 2007, 88, 253–262.
13. Mimna, R.; Camus, M. S.; Schmid, A.; Tuchscherer, G.; Lashuel, H. A.; Mutter, M.
Angew. Chem., Int. Ed. 2007, 46, 2681–2684.
Figure 2. RP-HPLC chromatogram at 214 nm (a) of the cyclic depsipentapeptide;
(b) after 72 h in phosphate buffer at room temperature; (c) after 3 h under organic
conditions (DMF/piperidine).
After 72 h in phosphate buffer at room temperature, the depsi-
pentapeptide 19 was completely consumed. However, both O–N
acyl migration and undesired hydrolysis of the ester bond occurred
(Fig. 2b). When using organic conditions DMF/Pip (80/20), no side
reaction was observed and we also avoided the problem of low
solubility of the starting material in phosphate buffer. In such
conditions, the cyclic pentapeptide 1 was quantitatively obtained
after 3 h (Fig. 2c).
14. Yoshiya, T.; Sohma, Y.; Kimuraa, T.; Hayashia, Y.; Kiso, Y. Tetrahedron Lett. 2006,
47, 2523–2526.