Efficient and direct solid phase synthesis of ketomethylenimino and ketomethylenamino peptides

Article abstract of DOI:10.1016/j.tetlet.2004.09.164

The reaction between the free amino terminus of a solid-supported peptide and a glyoxal leads to two families of pseudopeptides, the ketomethylenimino and the ketomethylenamino peptides. The aim of this study is the synthesis of pseudopeptides on solid supports, in order to quickly obtain modified peptides. We report a convenient step-by-step synthesis of ketomethylenimino ψ[CO-CH=N] and ketomethylenamino ψ[CO-CH₂-NH] peptides. The key is the reaction between the free amino terminus of the supported peptide and a glyoxal-modified amino acid, leading to a ketomethylenimino bond, which can be reduced to a ketomethylenamino bond.

Full text of DOI:10.1016/j.tetlet.2004.09.164

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 8603–8606
Efficient and direct solid phase synthesis of ketomethylenimino and
ketomethylenamino peptides
*
´
Elise Bernard and Regis Vanderesse
´
Laboratoire de Chimie Physique Macromoleculaire, UMR CNRS-INPL 7568, Groupe ENSIC, 1 rue Grandville,
BP 451, 54001 NANCY Cedex, France
Received 14 September 2004; revised 22 September 2004; accepted 28 September 2004
Abstract—The aim ofthis study is the synthesis ofpseudopeptides on solid supports, in order to quickly obtain modified peptides.
We report a convenient step-by-step synthesis ofketomethylenimino w[CO–CH@N] and ketomethylenamino w[CO–CH2–NH] pep-
tides. The key is the reaction between the free amino terminus of the supported peptide and a glyoxal-modified amino acid, leading
to a ketomethylenimino bond, which can be reduced to a ketomethylenamino bond.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Several diseases involve a protease imbalance and it is of
great importance to find new protease inhibitors.
Human Leucocyte Elastase (HLE) is a serine protease
that hydrolyzes a wide variety ofproteins, including
collagen and elastin. One ofits natural inhibitors is
a1-proteinase inhibitor (a1-PI) and its imbalance with
HLE may induce various chronic inflammatory diseases
such as acute respiratory distress syndrome (ARDS) or
pulmonary emphysema.1,2 This explains the increasing
need ofefficient inhibitors, resistant to hydrolysis caused
by the enzyme. A possible class ofenzyme inhibitors is
the pseudopeptides. They can be very interesting tools
for scientific and mechanistic investigation and have
been widely used for this purpose for a long time. Our
group has extensively studied, on model pseudopeptides,
the structural influence ofdifferent amide bond surro-
gates3 and ofthe nitrogen substitution ofr the C aH
group.4 These peptide surrogates present the advantage
ofallowing structural modulation ofthe peptide back-
bone with retention ofthe side chains generally required
for biological activity. Furthermore, it is possible to
obtain (i) a control ofthe structural flexibility, (ii) an
increased selectivity for the binding site, (iii) increased
resistance to hydrolysis, and (iv) improved permeability
through various biological barriers.
The replacement ofa natural peptide bond with a
ketomethylenamino bond w[CO–CH2–NH] is ofgreat
interest due to its increased flexibility, and therefore a
potentially better adaptability to the binding site ofan
enzyme, and to a putative better resistance to hydrolysis.
The ketomethylenamino link has been introduced by
Meyer et al.5 for the design of ACE inhibitors. It has
later been used as such or in various N-alkylated or
reduced forms to give ACE,6 furin,7 and HIV8 protease
inhibitors.
Herein, we present the incorporation ofketomethyleni-
mino and ketomethylenamino moieties in a hexapeptide,
derived from the target sequence of a1-PI, H-Ala1-Ala2-
Pro3-Val4-Ala5-Ala6-OH.7 HLE cleaves this peptide at
the Val4-Ala5 position.9 Therefore, we developed a con-
cise and general method for the direct incorporation of
ketomethylenimino and ketomethylenamino moieties
into Wang-resin-supported peptides.
Until now8,10 ketomethylenamino peptides have been
synthesized in homogeneous phase via the incorporation
ofthe modified peptide building blocks that require
laborious purification steps with concomitant racemiza-
tion at the stereogenic center. One method (Pathway A,
Scheme 1) involves the formation of a chloromethyl-
ketone.8 A second method uses the formation of the gly-
oxal ofan N-protected amino acid obtained via the
oxidation ofits corresponding diazo derivative by di-
methyldioxirane (DMD) (Pathway B, Scheme 1). The
imino bond formed is reduced by SiCl3H.10
Keywords: Solid phase peptide synthesis; Pseudopeptide; Glyoxal;
Ketomethylenimino; Ketomethylenamino.
*
Corresponding author. Tel.: +33 3 83 17 52 04; fax: +33 3 83 37 99
77; e-mail: regis.vanderesse@ensic.inpl-nancy.fr
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.09.164

8604
E. Bernard, R. Vanderesse / Tetrahedron Letters 45 (2004) 8603–8606
R1
R1
IBCF, NMM
CH2N2
Z
OH
Z
OH
N
H
N
H
O
O
pathway A
pathway B
HCl
DMD
Z
R1
R1
H
O
Z
N
H
Cl
N
H
O
O
α-aminoester
R1 R2
KI, DMF
α-aminoester
R1
R2
reduction
Z
Z
N
H
N
H
CO2R3
N
N
CO2R3
H
O
O
Scheme 1. General synthesis ofa ketomethylenamino dipeptide.
Table 1. Yields for the preparation of the modified amino acids
These two methods are aimed at the preparation of N-Z-
protected pseudodipeptides, which are incompatible
with direct solid phase synthesis (SPPS). These
approaches entail: (i) the preparation and purification
ofthe Z-pseudodipeptide ester; (ii) the substitution of
the Z group for a Boc or an Fmoc group; (iii) the sapon-
ification ofthe C-terminus ester, and (iv) the coupling of
the building block on the growing chain. To minimize
the number ofthe steps, we developed a direct method
in only one step via SPPS. An Fmoc aminoglyoxal is
coupled to the free N-terminus of a peptide linked to
the resin to form the ketomethylenimino bond, which
can be reduced, in situ, to the ketomethylenamino bond.
The ketomethylenimino link has been introduced at the
Val4-Ala5 position and the ketomethylenamino at the
Ala2-Pro3, Pro3-Val4, and Val4-Ala5 positions.
Fmoc-amino acid
Diazo (%)
Glyoxal (%)
Fmoc-Ala-OH
Fmoc-Pro-OH
Fmoc-Val-OH
92
93
98
>99
>99
>99
ing first to the ketomethylenimino link. This short step
synthesis ofketomethylenimino derivatives is advanta-
geous in view ofthe fragility ofthis class ofcompounds.
To form the ketomethylenamino bond, a few drops of
acetic acid are added prior to 3equiv ofNaBH 3CN,
added portionwise for 1h, in order to reduce the imino
bond. After stirring overnight, a Kaiser test clearly indi-
cates the complete coupling. The following peptide cou-
plings, as well as the final cleavage ofthe pseudopeptide
from the resin, are achieved according to the methods
known in the literature (Scheme 3).14 The overall yield
ofthe final crude pseudopeptides ranges rfom 59% to
90%. They were characterized by nuclear magnetic reso-
nance (13C, 1H, COSY, TOCSY NMR) and mass spectr-
oscopy. COSY and TOCSY NMR experiments
confirmed the inclusion ofeither the ketomethylenimino
link (dH = 7.99 for CO–CH@N) or the ketomethyl-
enamino link (dH = 3.68 for CO–CH2–NH). Purity of
hexapseudopeptides tested is >99% after HPLC.
Initially, we synthesized the Fmoc aminoglyoxals Fmoc-
Ala-CHO, Fmoc-Pro-CHO, and Fmoc-Val-CHO by the
Groarke10 method with 90% overall yields (Scheme 2).
The methylene group ofthe pseudopeptidic link w[CO–
CH2–NH] was then included via homologization on the
12
C-terminus part by action ofdiazomethane
on the
mixed anhydride ofthe N-Fmoc protected original ami-
no acid. The diazo compounds were synthesized in good
to excellent yields (Table 1). Purification was achieved
via silica gel column flash chromatography. The diazo
compounds were oxidized into the corresponding gly-
oxal immediately prior to use. The oxiꢁdizing agent
DMD is prepared via the action ofOxone on acetone
in the presence ofNaHCO 3.13 The DMD solution quan-
titatively oxidizes the diazo compound in 10min at 0ꢂC.
After removal of the solvent, the glyoxal must be used
without further purification, due to its low stability.
In conclusion, we propose an efficient method for the
synthesis ofketomethylenimino
w[CO–CH@N] and
ketomethylenamino w[CO–CH2–NH] peptides on a
solid support. This method uses N-protected amino-
glyoxals as reagents, readily obtained by a diazota-
tion–oxidation pathway. In this way, we avoid the
purification, transprotection, and nonfree racemization
saponification steps needed for the pseudodipeptide
building-block method previously described. The hexa-
These Fmoc aminoglyoxals are then directly reacted
with the growing peptide linked to the Wang-resin, lead-
pseudopeptide
H-Ala-Ala-Pro-Valw[CO–CH2–NH]-
R
R
R
H
IBCF, CH2N2
OH
DMD
FmocHN
FmocHN
FmocHN
O
N2
CH2Cl2
O
O
O
Scheme 2. Synthesis ofFmoc- a-aminoglyoxals.11

E. Bernard, R. Vanderesse / Tetrahedron Letters 45 (2004) 8603–8606
8605
O
O
H
FmocHN
O
N
+
H
N
H
H
O
DMF
O
O
N
FmocHN
N
H
O
H
a) NaBH3CN, AcOH, DMF
b) Piperidine , DMF
piperidine , DMF
O
O
O
O
H
H2N
N
H2N
N
N
H
N
H
H
H
H
O
O
a) Fmoc-Xaa-OH, TBTU, HOBt, DIEA
b) piperidine , DMF
c) TFA, EDT, H2O
O
O
O
O
O
O
H
N
H
H
H2N
N
N
OH
H2N
N
N
N
OH
N
H
N
H
N
N
H
H
H H
O
O
O
O
O
O
ketomethylenimino
Scheme 3. Solid phase synthesis ofthe ketomethylenimino and ketomethylenamino peptides (example given for the Val 4-Ala5 position).
ketomethylenamino
W. Biochem. Biophys. Res. Commun. 1984, 124, 141–147;
(b) Gordon, E. M.; Natarajan, S.; Pluscec, J.; Weller, H.
N.; Godfrey, J. D.; Rom, M. B.; Sabo, E. F.; Engebrecht,
J.; Cushman, D. W. Biochem. Biophys. Res. Commun.
1984, 124, 148–155; (c) Gordon, E. M.; Godfrey, J. D.;
Pluscec, J.; Von Langen, D.; Natarajan, S. Biochem.
Biophys. Res. Commun. 1985, 126, 419–426; (d) Natarajan,
S.; Gordon, E. M.; Sabo, E. F.; Godfrey, J. D.; Weller, H.
N.; Pluscec, J.; Rom, M. B.; Engebrecht, J.; Cushman, D.
W.; Deforrest, J. M.; Powell, J. J. Enzyme Inhib. 1988, 2,
91–97.
Ala-Ala-OH exhibits inhibition toward HLE (IC50:
1.90 · 10À4 M) and is currently being studied by NMR
in order to determine its structure when bound to its
receptor.
Acknowledgements
`
Pr. Michele Reboud-Ravaux (Institut Jacques Monod,
University ofParis VI/VII) is grateuflly acknowledged
for the enzymatic tests on HLE. We would also like
to thank Mathilde Achard and Olivier Fabre for
their technical assistance and Dr. Matthew Wilkinson
(University ofAmsterdam) ofr the revision ofthe
English text.
7. Angliker, H. J. Med. Chem. 1995, 38, 4014–4018.
8. (a) Rich, D. H.; Green, J.; Toth, M. V.; Marshall, G. R.;
Kent, S. B. H. J. Med. Chem. 1990, 33, 1285–1288; (b)
Rich, D. H.; Vara Prasad, J. V. N.; Sun, C.-Q.; Green, J.;
Mueller, R.; Houseman, K.; MacKenzie, D.; Malkovsky,
M. J. Med. Chem. 1992, 35, 3803–3812.
9. Guy, L.; Vidal, J.; Collet, A.; Amour, A.; Reboud-
Ravaux, M. J. Med. Chem. 1998, 41, 4833–4843.
References and notes
10. Groarke, M.; Hartzoulakis, B.; McKervey, M. A.;
Walker, B.; Williams, C. H. Bioorg. Med. Chem. Lett.
2000, 10, 153–155.
1. Pochet, L.; Doucet, C.; Dive, G.; Wouters, J.; Masereel,
B.; Reboud-Ravaux, M.; Pirotte, B. Bioorg. Med. Chem.
2000, 8, 1489–1501.
2. For more information about emphysema, see Snider, L.
G. Am. Rev. Respir. Dis. 1992, 146, 1615–1622; Parfrey,
H.; Mahadeva, R.; Lomas, D. A. Int. J. Biochem. Cell
Biol. 2003, 35, 1009–1014.
3. (a) Vanderesse, R.; David, L.; Grand, V.; Marraud, M.;
Mangeot, J.-P.; Aubry, A. Tetrahedron Lett. 1997, 38,
2669–2672; (b) Lecouvey, M.; Frochot, C.; Miclo, L.;
Orlewski, P.; Marraud, M.; Gaillard, J. L.; Cung, M. T.;
Vanderesse, R. Lett. Pept. Sci. 1997, 4, 359–364; (c)
Vanderesse, R.; Grand, V.; Limal, D.; Vicherat, A.;
Marraud, M.; Didierjean, C.; Aubry, A. J. Am. Chem.
Soc. 1998, 120, 9444–9451.
11. Typical procedure Fmoc-Val-OH (1.63g, 4.79mmol) was
dissolved in anhydrous THF (10mL) under nitrogen at
À25ꢂC. N-Methylmorpholine (0.55mL, 5mmol) and
isobutylchloroformate were then added to the mixture
and after 5min of stirring, the mixture was cooled down to
À78ꢂC. Anhydrous ether (10mL) was added and the fine
suspension was filtered under N2. Diazomethane (294mg,
7mmol) was added dropwise and the reaction was allowed
to stir for 2h at room temperature. After evaporation of
the solvent, the residue was dissolved in CH2Cl2, and
washed with water (3 · 10mL) before drying over MgSO4
and concentration. Fmoc-Val-CHN2 were purified via
flash chromatography using EtOAc/hexane (1/1, v/v).
(98%) 1H NMR (300MHz, CDCl3): d = 0,90 (d,
J = 6.7Hz, 3H); 0.98 (d, J = 6.7Hz, 3H); 2.10 (qd,
J = 6.7Hz, 1H); 4.12 (m, 1H); 4.22 (t, J = 6.6Hz, 1H);
4.44 (d, J = 6.6Hz, 2H); 5.30 (s, 1H); 5.36 (d, J = 8.1Hz,
1H); 7.30 (m, 2H); 7.40 (dd, J = 7.3Hz, 2H); 7.71 (m, 2H);
7.87 (d, J = 7.3Hz, 2H). 13C NMR (CDCl3): d = 18.09
(CH3c); 20.07 (CH3c); 31.77 (CHb); 47.96 (CH Fmoc);
´
4. Andre, F.; Boussard, G.; Marraud, M.; Didierjean, C.;
Aubry, A. Lett. Pept. Sci. 1995, 2, 239–242.
5. Meyer, R. F.; Essenburg, A. D.; Smith, R. D.; Kaplan, H.
R. J. Med. Chem. 1982, 25, 996–999.
6. (a) Natarajan, S.; Gordon, E. M.; Sabo, E. F.; Godfrey, J.
D.; Weller, H. N.; Pluscec, J.; Rom, M. B.; Cushman, D.

8606
E. Bernard, R. Vanderesse / Tetrahedron Letters 45 (2004) 8603–8606
63.47 (CH@N2, CHa); 67.53 (CH2 Fmoc); 120.66, 125.73,
imino or a ketomethylenamino bond. The simple conden-
sation ofthe glyoxal gives rise to the ketomethylenimino
bond whereas the addition ofNaBH 3CN (3equiv) por-
tionwise over 1h as reducing agent with a few drops of
acetic acid leads to the ketomethylenamino bond. The
reaction mixtures were stirred overnight. The following
amino acids are classically introduced. Finally, the last
Fmoc group was deprotected with 20% piperidine in DMF
and the resin was washed several times with CH2Cl2 before
being dried. For the cleavage step, the peptide–resin was
stirred over 2h in a solution ofethanedithiol (250 lL) and
water (500lL) in TFA (10mL). The peptide was then
precipitated by addition ofcool ether. Atfer removal of
the solvent and lyophilization ofthe crude material, the
pseudohexapeptide was purified by HPLC with a linear
gradient ofA: 0.1% TFA in water and B: 0.08% TFA and
20% water in acetonitrile, from 80% A to 50% A over
30min. H-Ala-Ala-Prow[CO–CH@N]-Val-Ala-Ala-OH
Mass: m/z [ES] calcd 510.28, found 511.39 for [M + H]+;
H-Ala-Alaw[CO-CH2-N]-Pro-Val-Ala-Ala-OH Mass: m/z
[ES] calcd 512.30, found 513.44 for [M + H]+ H-Ala-Ala-
Prow[CO–CH2–NH]-Val-Ala-Ala-OH Mass: m/z [ES]
calcd 512.30, found 513.51 for [M + H]+ H-Ala-Ala-Pro-
Valw[CO–CH2–NH]-Ala-Ala-OH Mass: m/z [ES] calcd
512.30, found 513.47 for [M + H]+.
127.77, 128.39, 142.03, 144.47 (C arom); 156.99 (C
urethane); 193.88 (C@O). Diazo compound Fmoc-Val-
CHN2 (544mg, 1.5mol) was dissolved in the solution of
DMD (50mL, 4.5mmol). After 10min of stirring, the
solvent was evaporated and the residue taken up in 15mL
ofCH 2Cl2 in order to remove the traces ofwater. The
solvent was removed in vacuo and the glyoxals obtained in
quantitative yields. The products are relatively unstable
and should be used without delay.
12. Fales, H. M.; Jaouni, T. M.; Babashak, J. F. Anal. Chem.
1973, 42, 2302–2303.
13. Adam, W.; Bialas, J.; Hadjiarapoglou, L. Chem. Ber.
1991, 124, 2377.
14. Typical procedure To a swollen Wang-resin in DMF, the
first amino acid (3equiv) was added with TBTU (3equiv),
HOBt (3equiv) and DIEA (9equiv). The reaction mixture
was stirred for 20 or 40min. The resin was washed with
CH2Cl2 (3 · 5mL), MeOH (3 · 5mL), and DMF
(3 · 5mL). A Kaiser test indicated a complete coupling.
After deprotection of Fmoc group with 20% piperidine in
DMF (5mL) for 2, 5, and 8min, the next amino acids were
introduced as previously described. Then, glyoxal (3equiv)
dissolved in DMF (5mL) was added to the peptide linked
to the resin. It is then possible to obtain a ketomethylen-