Chemoselective acylation of fully deprotected hydrazino acetyl peptides. Application to the synthesis of lipopeptides

Article abstract of DOI:10.1021/jo0010577

Fully deprotected N-terminal α-hydrazino acetyl peptides were synthesized and chemoselectively acylated on the hydrazine moiety with various fatty acid succinimidyl esters or N-(cholesterylcarbonyloxy) succinimide to give lipopeptides of high purity. The buffer and pH were adjusted in order to minimize the oxidation of the hydrazine moiety and to achieve the best conversion and selectivity. The acylation was performed in a citrate - Phosphate buffer/2-methylpropan-2-ol mixture of pH 5.1. The pK_a of the α-hydrazino acetyl group on our model peptide was found to be 6.45, i.e., about 2 units lower than the pK_a of a glycyl residue. The reaction was subsequently applied to the synthesis of a 38AA peptide derivatized by a palmitoyl group.

Full text of DOI:10.1021/jo0010577

J . Org. Chem. 2001, 66, 443-449
443
Ch em oselective Acyla tion of F u lly Dep r otected Hyd r a zin o Acetyl
P ep tid es. Ap p lica tion to th e Syn th esis of Lip op ep tid es
Dominique Bonnet, Nathalie Ollivier, He´le`ne Gras-Masse, and Oleg Melnyk*
UMR 8525 CNRS-Institut Pasteur de Lille-Universite´ de Lille 2. Institut de Biologie de Lille,
1 rue du Professeur Calmette 59019 Lille, France
Oleg.melnyk@pasteur-lille.fr
Received J uly 13, 2000
Fully deprotected N-terminal R-hydrazino acetyl peptides were synthesized and chemoselectively
acylated on the hydrazine moiety with various fatty acid succinimidyl esters or N-(cholesterylcar-
bonyloxy) succinimide to give lipopeptides of high purity. The buffer and pH were adjusted in order
to minimize the oxidation of the hydrazine moiety and to achieve the best conversion and selectivity.
The acylation was performed in a citrate-phosphate buffer/2-methylpropan-2-ol mixture of pH
5.1. The pKa of the R-hydrazino acetyl group on our model peptide was found to be 6.45, i.e., about
2 units lower than the pKa of a glycyl residue. The reaction was subsequently applied to the synthesis
of a 38AA peptide derivatized by a palmitoyl group.
Sch em e 1: Acyla tion of r-Hyd r a zin op ep tid es
In tr od u ction
Many papers reporting the derivatization of synthetic
peptides by fatty acids, by cholesterol derivatives, and
in general by lipophilic moieties have been published
during the most recent decade. First of all, this modifica-
tion is now widely recognized as a means of enhancing
the transport across biological membranes1 or biological
barriers such as the blood-brain barrier,2 the derm,3 or
the intestinal barrier.4 The modification of a peptide by
two lipid chains allows its stable insertion into mem-
branes.5 The lipidation has also been reported to influ-
ence significantly the conformation of peptides1b,6 or their
stability in vivo.4 Finally, several experiments have
revealed that cytotoxic T lymphocyte responses can be
induced fairly efficiently by simple lipopeptide vaccines
in mice,7 in primates,8 and in humans.9
However, the development of these applications is
hampered by the difficulty in producing lipopeptides,
which are sparingly soluble and whose purification is
often troublesome. The synthesis of large lipopeptides can
be achieved by stepwise solid-phase methods. However,
the final cleavage and deprotection step in highly acid
media do not allow the modification of peptides by acid-
sensitive lipophilic moieties such as unsaturated fatty
acids. Thus, the lipophilic moiety cannot easily be
modulated, while the nature of the fatty acid is known
to have a profound effect upon the interaction with the
membrane including its alteration.10 On the other hand,
the lipidation of a separately synthesized and purified
peptide or protein is an attractive approach provided that
there is a clear discrimination among the different
nucleophiles present in the fully deprotected compound.11
In this context, we have examined the reaction of
R-hydrazino acetyl peptides 1 with activated fatty acids
in buffered water/2-methylpropan-2-ol mixtures (Scheme
1).12 We report in this paper that hydrazino peptides 1
were chemoselectively acylated on the hydrazine moiety
to give conjugates 2 in good yield and purity. The pH of
the solvent mixture and the buffer concentration and
composition were optimized to obtain the greatest selec-
tivity. This novel methodology was applied to the syn-
* Corresponding author: Oleg Melnyk, Tel: 33 (0)3 20 87 12 15, Fax:
33 (0)3 20 87 12 33.
(1) (a) Loing, E.; Delanoye, A.; Sergheraert, C.; Tartar, A.; Gras-
Masse, H. Pept. Res. 1996, 9, 229-232. (b) Thiam, K.; Loing, E.;
Verwaerde, C.; Auriault, C.; Gras-Masse, H. J . Med. Chem. 1999, 42,
3732-3736. (c) Stephens, G.; O’Luanaigh, N.; Reilly, D.; Harriott P.;
Walker, B.; Fitzgerald D.; Moran, N. J . Biol. Chem. 1998, 273, 32,
20317-2032.
(2) Chopineau, J .; Robert, S.; Fe´nart, L.; Cecchelli, R.; Lagoutte, B.;
Paitier, S.; Dehouck, M.-P.; Domurado, D. J . Controlled Release 1998,
56, 231-237.
(3) Foldvari, M.; Attah-Poku, S.; Hu, J .; Li, Q.; Hugues, H.; Babiuk,
L. A.; Kruger, S. J . Pharm. Sci. 1998, 87, 1203-1208.
(4) Yodoya, E.; Uemura, K.; Tenma, T.; Fujita, T.; Murakami, M.;
Yamamoto, A.; Muranishi, S. J . Pharm. Exp. Ther. 1994, 271, 1509-
1513.
(5) (a) Na¨gele, E.; Schelhaas, M.; Kuder, N.; Waldmann J . Am.
Chem. Soc. 1998, 120, 6889-6902. (b) Liu, L.; Dudler, T.; Gelb, M. H.
J . Biol. Chem. 1996, 271, 23269-23276.
(6) Niidome, T.; Kawakami, R.; Okamoto, K.; Ohmori, N.; Mihara,
H.; Aoyagi, H. J . Pept. Res. 1997, 50, 458-464.
(7) Martinon, F.; Gras-Masse, H.; Boutillon, C.; Chirat, F.; Deprez,
B.; Guillet J . G.; Gomard, E.; Tartar, A.; Le´vy, J .-P. J . Immunol. 1992,
149, 3416-3422.
(10) De Vries, J . E.; Vork, M. M.; Roemen, T. H. M.; Yong, Y. F.;
Cleutjens, J . P. M.; Van Der Vusse, G. J .; Bilsen, M. V. J . Lipid Res.
1997, 38, 1384-1394.
(8) Bourgault, I.; Chirat, F.; Tartar, A.; Le´vy, J .-P., Guillet, J .-G.;
Venet, A. J . Immunol. 1994, 152, 2530-2537.
(11) (a) Chopineau, J .; Robert, S.; Fe´nart, L.; Cecchelli, R.; Lagoutte,
B.; Paitier, S.; Dehouck, M.-P.; Domurado, D. J . Controlled Release
1998, 56, 231-237. (b) Ekrami, H. M.; Kennedy, A. R.; Shen, W.-C.
FEBS Lett. 1995, 371, 283-286.
(9) (a) Vitiello, A.; Ishioka, G.; Grey, H. M.; Rose, R.; Farness, P.;
LaFond, R.; Yuan, L.; Chisari, F. V.; Furze, J .; Bartholomeuz, R.;
Chesnut, R. W. J . Clin. Invest. 1995, 95, 341-349. (b) Livingston, B.
D.; Crimi, C.; Grey, H.; Chesnut, R. W.; Sette, A.; J . Immunol. 1997,
159, 1383-1392.
(12) (a) Preliminary communication: Bonnet, D.; Rommens, C.; Gras-
Masse, H.; Melnyk, O. Tetrahedron Lett. 2000, 41, 45-48. (b) Bonnet,
D.; Gras-Masse, H.; Melnyk, O. FR 99 10626 (08/19/1999).
10.1021/jo0010577 CCC: $20.00 © 2001 American Chemical Society
Published on Web 12/22/2000

444 J . Org. Chem., Vol. 66, No. 2, 2001
Bonnet et al.
Sch em e 2: Syn th esis of P ep tid es 4-6
thesis of a 38 amino acid peptides derivatized by a
palmitoyl group.
Resu lts a n d Discu ssion
The discrimination between R- and side-chain groups
in polypeptides has been the subject of intense research
and remains one of the most intractable problems for the
peptide or protein chemist. It is rare for the R-amino
group of the peptide chain to be distinguishable from its
side-chain counterparts, notwithstanding the differences
in pK between side-chain (pK∼10) and terminal groups
(pK∼8). The acetimidylation reaction, unusually selective
for the ꢀ-amino group, when compared with the R-amino
group, is an exception to this rule.13 While discrimination
is not perfect, the R-amino group tends to be left free to
a large and often useful extent, even when ꢀ-amino
groups are fully substituted.14 The acylation of peptides
is much too difficult to control. For example, reaction of
glucagon with acetic anhydride at pH 5.5 gave two
products, 70% acetylated only on the R-amino group of
the N-terminal His and 30% diacetylated at the N-
terminus plus the ꢀ-amino group of Lys12.15 Desacetylthy-
mosin R1, a 28 amino acid peptide, reacted with acetic
anhydride to yield 90% of the N-terminally acetylated
peptide,16 but reaction of trypsinogen with acetic anhy-
dride exhibited little if any selectivity for the N-termi-
nus.17 For small peptides, with only a single competing
lysine group, the reagent iodoacetic anhydride was found
to be selective for the N-terminal R-amine at pH 6.18
However, the selectivity was highly dependent upon both
the N-terminal amino acid and the presence/absence of
His or Tyr residues. These amino acids are suspected of
catalyzing the acylation via the formation of unstable
acylimidazoles or phenyl esters. In addition, a large
excess of anhydride was used to compensate for the
hydrolysis of the reagent.
One way to circumvent these difficulties is to increase
the difference in pKa between the ꢀ-amino groups and
the nucleophile to be acylated. Working at a pH below 6
should improve the selectivity and minimize histidine or
tyrosine residues from acting as acylation catalysts. In
addition, lowering the pH should increase the half-life
of the activated ester. This latter point is particularly
important in the case of activated fatty acids, whose
reduced solubility in partial aqueous media preclude the
use of large excesses. Some fatty acids are also expensive.
The hydrazino moiety of R-hydrazino acetic acid ethyl
ester19 has a pKa of 5.97 at 25 °C, i.e., about 2 pK units
below that of an R-amino group. Interestingly, R-hy-
drazino acids are known to be regioselectively acylated
at the Nâ position, so that only a single isomer is
produced.20 The acylation of R-hydrazino acids should
also be less sensitive to the nature of the side-chain, due
to the greater distance between the â-nitrogen and the
R-carbon when compared with R-amino acids. Finally, the
“R-effect” exhibited by hydrazines was expected to have
a favorable impact upon the selectivity.21 Initially, we
have examined the chemoselective acylation of R-hy-
drazino acetyl peptides 1 (R ) H) using the model peptide
6 derived from the POL 476-484 sequence (Scheme 2).
Peptide 6 contained a Lys in position 7 for selectivity
studies.
F u n ction a liza tion of P ep tid es by a n R-Hyd r a zin o
Acetyl Gr ou p . The Fmoc/t-Bu solid-phase peptide syn-
thesis22 was used throughout this study since the R-hy-
drazino acetyl moiety was found to be stable in the
concentrated trifluoroacetic acid mixtures employed for
the final cleavage and deprotection steps.23 Owing to the
commercial availability of R-hydrazino acetic acid ethyl
ester, we focused our attention on the coupling of Boc-
protected derivatives of R-hydrazino acetic acid. Mono-
protection of R-hydrazino acetic acid on the â-nitrogen
did not suppress the nucleophilic character of the nitro-
gens and the partial polymerization of the activated ester.
Similar observations have been made by Collet et al.24
On the other hand, (Boc)NHN(Boc)CH2CO2H 7 or (Boc)2-
NN(Boc)CH2CO2H 8 were found to be useful derivatives
using either N-[(1H-benzotriazol-1-yl)(dimethylamino)-
methylene]-N-methylmethanaminium hexafluorophos-
(13) Hunter, M. L.; Ludwig, M. J . J . Am. Chem. Soc. 1962, 84, 3491-
3504.
(14) (a) Mahrenholtz, A. M.; Flanders, K. C.; Hoosein, N. M.; Gurd,
F. R. N.; Gurd, R. S. Arch. Biochem. Biophys. 1987, 257, 379-386. (b)
Nureddin, A.; Inagami, T. Biochem. J . 1975, 147, 71-81.
(15) Desbuquois, B. Eur. J . Biochem. 1975, 60, 335-347.
(16) Wetzel, R.; Heyneker, H. L.; Goeddel, D. V.; J hurani, P.;
Shapiro, J .; Crea, R.; Low, T. L. K.; McClure, J . E.; Thurman, G. B.;
Goldstein, A. L. In Cellular Responses to Molecular Modulators; Miami
Winter Symposium; Mozes, L. W., Scott, W. A., Shultz, J ., Werner, R.,
Eds.; Academic: New York, 1981; Vol. 18, pp 251-270.
(17) Magee, A. I.; Grant, D. A. W.; Hermon-Taylor, J .; Offord, R. E.
Biochem. J . 1981, 197, 239-244.
(18) Wetzel, R.; Halualani, R.; Stults, J . T.; Quan, C. Bioconjugate
Chem. 1990, 1, 114-122.
(19) Krueger P. J . In The chemistry of the hydrazo, azo and azoxy
groups; Patai, A., Ed; J ohn Wiley: New York, 1975; Chapter 7, pp 159-
175.
(20) Amour, A.; Collet, A.; Dubar, C.; Reboud-Ravaux, M. Int. J .
Pept. Protein Res. 1994, 43, 297-304.
(21) Isaacs, N. S. In Physical Organic Chemistry; Longman Scientific
& Technical: Harlow Essex, U.K., 1987; p 247.
(22) Fields, G. B.; Noble, R. L. Int. J . Pept. Protein Res. 1990, 35,
161-214.
(23) Bonnet, D.; Samson, F.; Rommens, C.; Gras-Masse, H.; Melnyk,
O. J . Peptide Res. 1999, 54, 270-278.
(24) Guy, L.; Vidal, J .; Collet, A. J . Med. Chem. 1998, 41, 4833-
4843.

Acylation of Hydrazino Acetyl Peptides
J . Org. Chem., Vol. 66, No. 2, 2001 445
phate (HBTU)/N-hydroxybenzotriazole (HOBt)25 or ben-
zotriazole-1-yloxytris(dimethylamino)phosphonium hexa-
fluorophosphate (BOP)26 activation. The chemistry re-
lated to the preparation, reactivity, and coupling of
compounds 7 and 8 will be reported in detail elsewhere.27
Fully protected derivative 8 was used throughout this
study for the preparation of R-hydrazino acetyl peptides.
Thus, peptidyl resin 3 (Scheme 2), elaborated starting
from Fmoc-L-Val-Wang resin, was acylated with 8 using
in situ BOP activation. Model peptide 6 was isolated with
a 35% yield following RP-HPLC purification.
p Ka of th e Hyd r a zin o Acetyl Moiety. The syntheses
of peptides 4 and 5 (Scheme 2) were undertaken to permit
the determination of the pKa of the N-terminal amino and
hydrazino groups using 15N NMR spectroscopy. The 15N-
labeled R-hydrazino acetyl moiety of peptide 5 was
elaborated on the solid phase by N-electrophilic amina-
tion of the N-terminal Gly[15N] residue with N-Boc-3-(4-
cyanophenyl)oxaziridine (BCPO).28 Reaction of BCPO
with a primary amine yields a Boc-protected hydrazine
plus an imine resulting from the trapping of the liberated
p-cyanobenzaldehyde by unreacted amines. The advan-
tage of adopting the solid-phase strategy is based on the
selective hydrolysis the imine. Thus, the unmasked
amino groups can be rereacted with BCPO following
neutralization with diisopropylethylamine. The N-ami-
nation/hydrolysis/neutralization cycles are repeated until
the disappearance of the amino groups, as confirmed by
the use of ninhydrin29 or 2,4,6-trinitrobenzene-1-sulfonic
acid30 tests. Unhindered primary amines such as the
R-amino group of Gly or ꢀ-amino groups usually require
5 to 10 cycles. The process was fully automated on a solid-
phase peptide synthesizer. Peptides 4 and 5 were isolated
with a 65 and 54.5% overall yield, respectively, following
RP-HPLC purification.
F igu r e 1. Chemical shifts of labeled R-amino (peptide 4) or
hydrazino (peptide 5) groups as a function of pH at 20 °C
(15NH3 was used as an external reference).
Sch em e 3: Rea ction of P ep tid e 6 w ith P a lm itic
Acid Su ccin im id yl Ester
The variation in the chemical shift of the 15N nucleus
in peptides 4 and 5 as a function of pH is shown in Figure
1. These data corresponded to a pKa of 6.45 for the
R-hydrazino group, and 8.45 for the R-amino group. Thus,
as expected, the pKa of the R-hydrazino acetyl moiety is
about 2 pKa units below that of the R-amino group of Gly.
Ch em oselective Acylation of P eptide 6 with P alm -
itic Acid Su ccin im id yl Ester (Sch em e 3). The acyla-
tions were performed in a water/2-methylpropan-2-ol: 1/1
mixture, in which both peptides and activated fatty acids
proved soluble. Several studies have demonstrated that
the addition of an organic solvent to water does not alter
significantly the pKa of amines.31 Thus, the difference in
pKa between the R-hydrazino and amino groups found
in water should be conserved upon addition of 2-meth-
ylpropan-2-ol. Preliminary experiments were undertaken
with a 50 mM citrate-phosphate buffer (pH 5.0, 5.5, and
6.0) using 1.2 equiv of palmitic acid succinimidyl ester
on a 1 mg scale. The greatest selectivity and RP-HPLC
yield were obtained at pH 5.0. However and unexpect-
edly, aggregation occurred when the reactions were
performed on a larger scale, but this problem was
resolved by reducing the concentration of the buffer to
25 mM. The optimal pH was determined more precisely
as shown in Figure 2a and 2b using this buffer concen-
tration. Figure 2a corresponds to the RP-HPLC yield for
peptide 9, the lipopeptide resulting from the acylation
of the hydrazino moiety. The best yield was obtained at
pH 5.11 following 75 h of reaction at room temperature.
Figure 2b gives the extent of diacylation, which dramati-
cally increased above pH 5.11. Using these optimal
experimental conditions, lipopeptide 9 was isolated with
a 69% yield following RP-HPLC purification (6 mg scale,
Table, entry 1). Figure 3a shows the RP-HPLC profile of
the reaction mixture after 48 h at room temperature. The
minor, final eluting peak corresponded to peptide 10
acylated both at the hydrazino and ꢀ-amino groups. This
side product was isolated with a 6.9% yield following RP-
HPLC purification. As expected, the control experiment
with the native peptide H-GRGR-POL476-484 (peptide 12)
resulted in a low conversion and a poor selectivity (Figure
(25) Knorr, R.; Trzeciak, A.; Bannwarth, W.; Gillensen, D. Tetra-
hedron Lett. 1989, 30, 1927-1930.
(26) Rivaille, P.; Gautron, J . P.; Milhaud, G. Tetrahedron 1980, 36,
3413-3419.
(27) Bonnet, D.; Gras-Masse, H.; Melnyk, O. Manuscript in prepara-
tion.
(28) (a) Vidal, J .; Drouin, J .; Collet, A. J . Chem. Soc., Chem.
Commun. 1991, 435-437. (b) Vidal, J .; Guy, L.; Ste´rin, S.; Collet, A.
J . Org. Chem. 1993, 58, 4791-4793. (c) Klinguer, C.; Melnyk, O.; Loing,
E.; Gras-Masse, H. Tetrahedron Lett. 1996, 37, 7259-7262. (d) Melnyk,
O.; Rommens, C.; Gras-Masse, H. Tetrahedron Lett. 1999, 40, 1491-
1494. (e) Bonnet, D.; Rommens, C.; Gras-Masse, H.; Melnyk, O.
Tetrahedron Lett. 1999, 40, 7315-7318. (f) Samson, F.; Bonnet, D.;
Rommens, C.; Gras-Masse, H.; Melnyk, O. J . Pept. Sci. 1999, 5, 352-
359.
(29) Sarin, V. K.; Kent, S. B. H.; Tam, J . P.; Merrifield, R. B. Anal.
Biochem. 1981, 117, 147-157.
(30) Cayot, P.; Tainturier, G. Anal. Biochem. 1997, 249, 184-200.
(31) (a) Homandberg, G. A.; Mattis, J . A.; Laskowski, M. Biochem-
istry 1978, 17, 5220-5227. (b) Homandberg, G. A.; Laskowski, M.
Biochemistry 1979, 18, 586-592. (c) Michaelis, L.; Mizutani, M. Z.
Phys. Chem. 1925, 116, 135-160. (d) Mizutani, M. Z. Phys. Chem. 1925
116, 350-358.

446 J . Org. Chem., Vol. 66, No. 2, 2001
Bonnet et al.
F igu r e 3. (a and b) RP-HPLC of the crude reaction mixtures
after 48 h at room temperature, C3 Zorbax column (1 mL/min,
215 nm, linear gradient 0-80% acetonitrile in water in 30 min,
0.05% TFA). (a) peptide 6 H2N-GRGR-POL.476-484 (b) control
peptide 12 H-GRGR-POL.476-484 Peaks a and b corresponded
to peptides Palm-GRGRILKEPVHGV-OH and H-GRGRILK-
(Palm)EPVHGV-OH, respectively, as determined by ES-MS
analyses and Edman degradation.
F igu r e 2. (a) RP-HPLC yield for peptide 9 (monoacylation).
(b) RP-HPLC yield for peptide 10 (diacylation).
Ta ble 1. Isola ted Yield s for Lip op ep tid es 9 a n d 14-17
entry
peptide
R group
isolated yield (%)
1
2
3
4
5
6
9
14
15
16
17
18
palmitoyl
oleyl
cholesterylcarbonyl
cis-9,10-epoxystearyl
linoleyl
69
49
54
31
52
47
stearyl
3b). The product resulting from acylation of the N-
terminal glycine was isolated with a 28.6% yield, whereas
the isomer bearing a palmitoyl group on the Lys7 side-
chain represented 8.7%.
Peptide 9 was extensively characterized to ensure the
absence of contamination by isomer 11. Peptide 9 was
found to be homogeneous by RP-HPLC and CZE. In
addition, the trypsic digest of peptide 9 yielded the
expected fragments CH3(CH2)14CONH-GR-OH, CH3-
(CH2)14CONH-GRGR-OH, and ILKEPVHGV-OH, which
were isolated and characterized by ES-MS. Fragments
corresponding to peptide 11 were not detected.
In flu en ce of th e Bu ffer . Recently, Mukerjee et al.
have reported that pKa and pH values of acetate and
phosphate buffers are affected differently upon addition
of DMSO.32 2-Methylpropan-2-ol may have similar ef-
fects. Thus, the selectivity of the acylation reaction may
depend on the composition of the buffer.
F igu r e 4. Acylation of peptide 6 in citrate/phosphate or
acetate buffers.
citrate-phosphate buffer. This result may be correlated
to the higher apparent pH of the acetate/2-methylpropan-
2-ol mixture (6.36) when compared with the citrate-
phosphate/2-methylpropan-2-ol mixture (5.85).
Besides the products resulting from acylation of pep-
tide 6 by palmitic acid succinimidyl ester, the crude
reaction mixtures contained minor amounts of a new
product, slightly more hydrophobic by RP-HPLC than the
starting hydrazino peptide 6, and which by ES-MS
corresponded to Ac-RGR-POL476-484 (peptide 13). This
side-product could result from the air oxidation of the
R-hydrazino acetyl moiety. The stability of peptide 6 in
pH 5.1 citrate-phosphate or acetate buffers was exam-
ined (Figure 5). Peptide 6 decomposed exclusively into
The selectivity of the reaction of peptide 6 with palmitic
acid succinimidyl ester was carefully examined in pH 5.1
citrate-phosphate or acetate buffers diluted with 2-me-
thylpropan-2-ol (Figure 4). The RP-HPLC yield for pep-
tide 9 and the selectivity was found to be higher in the
(32) Mukerjee, P.; Ostrow, J . D. Tetrahedron Lett. 1998, 39, 423-
426.

Acylation of Hydrazino Acetyl Peptides
J . Org. Chem., Vol. 66, No. 2, 2001 447
Sch em e 5: Ca ta lytic Hyd r ogen a tion of
Lip op ep tid e 14
Sch em e 6: Acyla tion of P ep tid e 19
F igu r e 5. Formation of byproduct 13 in 25 mM acetate or
citrate-phosphate buffers diluted with 2-methylpropan-2-ol.
Sch em e 4: Syn th esis of Lip op ep tid es 14-18
Sch em e 7: Syn th esis of P ep tid e 21 by Solid P h a se
N-Electr op h ilic Am in a tion
13 in both media. The RP-HPLC yield for 13 in citrate-
phosphate buffer reached 9.6% following 72h at room
temperature, but was 3 times as high in acetate buffer.
Clearly, the lower yield for lipopeptide 9 in acetate buffer
when compared with the citrate-phosphate can be
ascribed to both a lower selectivity for the acylation and
to a reduced stability of the starting R-hydrazino acetyl
peptide.
Ch em oselective Acylation of P eptide 6 with Oth er
F a tty Acid Su ccin im id yl Ester s. Acylation of peptide
6 with other activated fatty acids or cholesterol deriva-
tives was then examined (Scheme 4, Table). Oleic,
linoleic, stearic, and cis-9,10-epoxystearic acids were
converted into the corresponding succinimidyl esters
using diisopropylcarbodiimide/N-hydroxysuccinimide in
dry THF/CH2Cl2 at 4 °C. After removal of the solvents
in vacuo, the activated esters were redissolved in 2-meth-
ylpropan-2-ol and added to the peptide solution. N-
(Cholesterylcarbonyloxy)succinimide was obtained as a
solid following reaction of cholesteryl chloroformate with
N-hydroxysuccinimide in the presence of triethylamine.
These lipophilic derivatives behaved as palmitic acid
succinimidyl ester in terms of both reactivity and selec-
tivity. Lipopeptides 14, 15, and 17 were isolated with 49-
54% yield following RP-HPLC purification. Lipopeptide
18 was found to be identical to the product obtained
following catalytic hydrogenation of compound 14 in
aqueous acetic acid (Scheme 5). The acylation involving
cis-9,10-epoxystearic acid succinimidyl ester highlights
the mildness of the process, since the epoxy moiety is
known to be very sensitive to acids. Peptide 16 was
purified by RP-HPLC at pH 7.2, since RP-HPLC at pH 2
resulted in the opening of the epoxide (Table, entry 4).
Ap p lica tion to th e Syn th esis of La r ge Lip op ep -
tid es. The application of the methodology to the synthe-
sis of large lipopeptides was examined with peptide 19,
which contained three lysines in its sequence (Scheme
6). Peptide 19 is derived from the murine interferon-γ
(IFN-γ) 95-132 sequence. A construct resulting from the
modification of the IFN-γ 95-132 sequence by a palmi-
toyl group was found to reproduce the activities of the
recombinant IFN-γ, on both human and mouse cells.3b
The acylation of peptide 19 with palmitic acid succin-
imidyl ester was found to be chemoselective and led to
formation of lipopeptide 20 in 58% yield following RP-
HPLC purification (Scheme 6). Thus, the methodology
could be applied to large R-hydrazino acetyl peptides,
despite the presence of several Lys residues in the
sequence. However, a scrambled analogue of peptide 19
led to poor results (data not shown), due to aggregation
of both the starting material and of the lipopeptide.
Exten sion of th e Meth od ology to Oth er R-Hy-
d r a zin o Acid s. The solid-phase N-electrophilic amina-
tion allows the automated synthesis of peptides func-

448 J . Org. Chem., Vol. 66, No. 2, 2001
Bonnet et al.
triethylamine-phosphate buffer in water; Buffer E: pH 7.2 25
mM triethylamine-phosphate buffer in water/acetonitrile: 1/4
by vol.
Sch em e 8: Acyla tion of P ep tid e 21
Capillary zone electrophoresis (CZE) was performed on an
Applied Biosystems 270A-HT apparatus in a 75 µm × 50 cm
fused silica capillary, with a 40 mA current and a 30 kV field.
The analyses were undertaken in a pH 3.0 sodium citrate
buffer unless otherwise stated.
Electrospray mass spectrometry (ES-MS) studies were
performed on a Micromass Quatro II Electrospray Mass
Spectrometer. The m/z range 200-2100 was scanned using an
ionspray voltage of 4500 V. The orifice plate voltage was
ranged between 40 and 60 V. The temperature in the ioniza-
tion chamber was 60 °C. Time-of-flight plasma desorption mass
spectrometry (TOF-PDMS) was undertaken on a Bio-ion 20
Plasma Desorption Mass Spectrometer.
Amino acid compositions were determined using a Beckman
amino acid analyzer model 7300 with ninhydrin detection,
following hydrolysis in evacuated sealed tubes with 6 N HCl/
phenol: 10/1 (v/v) at 110 °C during 24 h.
tionalized at the N-terminus by an R-hydrazino acid.
Thus, the lipidation of such peptides was examined using
the model hydrazino peptide 21 (Scheme 7), whose
sequence was derived from the 989-995 membrane
proximal portion of the GpIIb cytoplasmic tail.3c Peptide
21 was synthesized as described in Scheme 7 using
standard Fmoc/t-Bu chemistry. Following peptide elonga-
tion, the R-amino group was reacted with BCPO as
described for peptide 5. Unreacted amino groups were
capped with palmitic acid succinimidyl ester. Cleavage
in concentrated TFA followed by RP-HPLC purification
afforded compound 21 in a 56% yield. Acylation of peptide
21 with palmitic acid succinimidyl ester was found to be
very clean, and lipopeptide 22 was isolated with a 60%
yield following RP-HPLC purification (Scheme 8).
Syn th esis of P ep tid es 4 a n d 5. Peptidyl-resin H-R(Pmc)-
GR(Pmc)ILK(Boc)E(t-Bu)PVH(Trt)GV-Wang-PS was elabo-
rated starting from 0.25 mmol of Fmoc-Val-Wang-PS resin
(0.52 mmol/g, Novabiochem). Fmoc-Gly[15N]-OH (Euriso-top,
82 mg, 0.275 mmol) was coupled manually using HBTU/DIEA:
104.2 mg (0.275 mmol)/93.9 µL (0.55 mmol) activation in DMF
(30 min). The resin was washed with DMF (2 × 2 min), CH2-
Cl2 (2 × 2 min), and capped with Ac2O/DIEA/CH2Cl2: 10/5/85
by vol (2 and 20 min). The resin was then rinsed with CH2Cl2
(2 × 2 min) and DMF (3 × 2 min) and treated with piperidine/
DMF: 1/4 by vol (2 and 15 min). After rewashing with DMF (1
min), CH2Cl2 (1 min), and DMF (1 min), half of the resin was
taken up in diethyl ether, dried in vacuo, and treated with
TFA/water/anisole: 95/2.5/2.5 by vol during 2 h. The crude
peptide was precipitated in 100 mL of cold diethyl ether/
heptane: 1/1 by vol, redissolved in water and lyophilized. The
crude peptide (224.6 mg) was purified by RP-HPLC on a C18
Hyperprep column (15 × 300 mm) using A and B buffers (0-
100% B in 100 min, 3 mL/min, rt, detection at 215 nm). Peptide
4, 160.7 mg (65%), white powder, ES-MS [M + H]+ calcd
1419.7, found 1418.95. AA(calcd): found; G(3): 2.96, R(2): 2.0,
I(1): 0.95, L(1): 1.0, K(1): 1.0, E(1): 1.1, V(2): 2.0, H(1): 0.94.
The other half was subjected to the BCPO procedure (10
cycles) as described elsewere.30 An aliquot of the peptidyl resin
was subjected to the Kaiser test (negative). The resin was
washed with CH2Cl2 and diethyl ether and dried in vacuo. The
peptidyl resin was cleaved and deprotected as above to give
228 mg of a white powder. The crude peptide was purified by
RP-HPLC on a C18 Hyperprep column (15 × 300) using buffers
A and C (0-60% C in 60 min, 3 mL/min, rt, detection at 215
nm). Peptide 5, 136 mg (54.3%), white powder, ES-MS [M +
H]+ calcd 1434.7, found 1434.0. AA(calcd): found; G(2): 2.0,
R(2): 2.0, I(1): 0.96, L(1): 1.0, K(1): 1.0, E(1): 1.1, V(2): 2.0, H(1):
0.87.
Con clu sion
The R-hydrazino group of R-hydrazino acids has unique
properties when compared with the functional groups
present naturally in peptides. Owing to its low pKa, the
hydrazino moiety was able to react chemoselectively in
aqueous media at pH 5.1 with fatty acid succinimidyl
esters, thus giving a very mild access to lipopeptides. The
functionalization of peptides by an R-hydrazino acetyl
moiety is very easy using a fully protected derivative of
R-hydrazino acetic acid. The methodology, which permit-
ted the use of near stoichiometric amounts of the acti-
vated ester, was extended to hydrazino lysine. Most of
the purification efforts were concentrated upon the
isolation of the hydrophilic and soluble hydrazino precur-
sor. Thus, thanks to the very late introduction of the
lipophilic tail, lipopeptides of high purity could be isolated
with good yields.
Exp er im en ta l Section
Solid-phase peptide syntheses were performed using stan-
dard Fmoc/t-Bu chemistry on Applied Biosystems 430A or
431A or Perseptive Pioneer peptide synthesizers. The amino
acids were activated using N-[(1H-benzotriazol-1-yl)(dimethyl-
amino)methylene]-N-methylmethanaminium hexafluorophos-
phate N-oxide (HBTU)/N-hydroxybenzotriazole (HOBt)/diiso-
propylethylamine activation (AA/HBTU/HOBt/DIEA: 4 equiv/
4 equiv/ 4 equiv/ 8 equiv) in DMF. Side chain protections were
as follows: His(Trt), Glu(t-Bu), Arg(Pmc), Lys(Boc), Asn(Trt),
Tyr(t-Bu), Ser(t-Bu), Q(Trt). Fmoc amino acids were purchased
from Senn Chemicals. Amino acid composition of peptidyl
resins was checked by amino acid analysis (Beckman amino
acid analyzer model 7300, ninhydrin detection), following
hydrolysis of an aliquot in propionic acid/6 N HCl: 1/1 at 140
°C during 3 h.
Ch a r a cter iza tion of th e P u r ified P ep tid es. The RP-
HPLC analyses were performed at 50 °C on C3 Zorbax (4.6 ×
250 mm) or C18 Hyperprep (4.6 × 250 mm) columns using
water-acetonitrile or water/2-propanol linear gradients. The
following buffers were used: Buffer A: water containing 0.05%
TFA by vol; Buffer B: acetonitrile/water: 4/1 by vol containing
0.05% TFA by vol; Buffer C: isopropanol/water: 3/2 by vol
containing 0.05% TFA by vol; Buffer D: pH 7.2 25 mM
pKa Deter m in a tion . The pKa of the R-amino group of
peptide 4 and of the R-hydrazino group of peptide 5 were
determined using 15N NMR spectroscopy on a Bruker DRX300
NMR spectrometer at 20 °C. Peptides 4 and 5 were dissolved
in D2O at a concentration of 97.7 and 96.9 mg/mL, respectively.
The pH was adjusted with 1 N aqueous NaOH. Chemical shifts
were referenced relative to external 15NH3.
Syn th esis of H2N-GRGR-P OL476-484-OH P ep tid e (p ep -
tid e 6). Peptide 6 was elaborated starting from 0.125 mmol
of Fmoc-L-Val-Wang PS resin (0.73 mmol/g, Applied Biosys-
tems, Foster City, CA). Following peptide elongation, (Boc)2N-
N(Boc)CH2CO2H (58.5 mg, 0.15 mmol) was coupled twice
manually using BOP (66.3 mg, 0.15 mmol)/DIEA (78.4 µL, 0.45
mmol) activation in DMF (30 min). The peptidyl-resin was
washed with DMF, CH2Cl2, and diethyl ether and dried in
vacuo. Cleavage and deprotection were performed using 10 mL
of a TFA/water/anisole: 95/2.5/2.5 by vol mixture during 2 h
at room temperature. The crude peptide was precipitated in
cold diethyl ether/heptane: 1/1 by vol, redissolved in water/
acetic acid: 5/1 by vol, and lyophilized. Purification was
performed on a C18 Hyperprep (15 × 300 mm) column using

Acylation of Hydrazino Acetyl Peptides
J . Org. Chem., Vol. 66, No. 2, 2001 449
buffers A and C (0-50% C in 100 min, 3 mL/min, rt, detection
at 215 nm). The pure fractions were pooled and lyophilized to
give 84.2 mg (34%) of a white powder. AA(calcd): found; G(2):
1.9, R(2): 2.0, I(1): 1.0, L(1): 1.2, K(1): 1.0, E(1): 1.6, V(2): 2.0,
H(1): 0.9. ES-MS: M calcd 1432.7, found 1433.0.
Activa tion of F a tty Acid s, Typ ica l Exp er im en ta l P r o-
ced u r e. Oleic acid (4.96 mg, 17.5 µmoles) was dissolved in
175 µL of THF/CH2Cl2: 1/1 by vol 17.7 µL (17.5 µmol) of
N-hydroxysuccinimide 0.99 M in THF and 15.0 µL (13.5 µmol)
of diisopropylcarbodiimide 0.90 M in CH2Cl2 were added at 4
°C. The reaction mixture was stirred overnight. The solvents
were then removed in vacuo, and the residual oil was redis-
solved in 600 µL of 2-methylpropan-2-ol and immediately used
in the acylation reactions.
P r ep a r a tion of N-(Ch olester ylca r bon yloxy)su ccin i-
m id e. Cholesteryl chloroformiate (J anssen, 500 mg, 1.13
mmol) and N-hydroxysuccinimide (140.9 mg, 1.22 mmol) were
dissolved in 2 mL of CH2Cl2 at room temperature. Triethy-
lamine (170 µL, 1.22 mmol) was added in one portion. The
reaction was stirred 45 min, diluted with 50 mL of CH2Cl2,
and washed with water saturated with KH2PO4. The organic
phase was dried over sodium sulfate. Removal of the solvents
under reduced pressure yielded 451.6 mg (76%) of a white
powder which was used without further purification. 1H NMR
(300 MHz, DMSO-d6, 296 K) δ 5.41 (1H, m), 4.60 (1H, m), 2.82
(4H, s), 2.48 (2H, m), 2.04-0.85 (38H, m), 0.68 (3H, s). 13C
NMR (150 MHz, DMSO-d6, 296 K) δ: 169.8, 151.7, 139.4, 124.8,
82.7, 57.0, 56.5, 50.2, 42.6, 39.9, 39.8, 37.8, 36.9, 36.4, 36.0,
32.1, 28.4, 28.2, 27.6, 25.6, 24.4, 24.0, 22.9, 22.7, 21.2, 19.3,
28.8, 11.9.
Acyla tion of P ep tid e 6. Typ ica l Exp er im en ta l P r oce-
d u r e. Peptide 6 (6.0 mg, 3.0 µmol) was dissolved in 900 µL of
a citrate-phosphate pH 5.2 buffer. The pH of the solution was
adjusted to 5.1 with a 0.2 M Na2HPO4 aqueous solution.
Palmitic acid succinimidyl ester (Sigma, 1.16 mg, 3.3 µmol)
was dissolved in 900 µL of 2-methylpropan-2-ol and added to
the peptide solution. The reaction was monitored by RP-HPLC
on a C3 Zorbax column using buffers A and B (0-100% B in
30 min then 5 min at 100% B, 1 mL/min, 50 °C, detection at
215 nm). Following 72 h at room temperature, the reaction
mixture was diluted with 5 mL of water/acetic acid: 4/1 by vol
and purified on a C3 Zorbax column (0-60% B in 70 min, 3
mL/min, 50 °C, detection at 215 nm). Peptide 9, 4.41 mg (69%),
white powder, MALDI-TOF [M + H]+ calcd 1672.2, found
1672.2.
Peptides 14/15 and 17/18 were synthesized and purified
similarly.
Starting from 5.02 mg of peptide 6, 2.58 mg (49%) of peptide
14 was obtained, white powder, ES-MS [M + H]+ calcd 1697.2,
found 1697.0.
Starting from 5.04 mg of peptide 6, 2.27 mg (54%) of peptide
15 were obtained, white powder, ES-MS [M + H]+ calcd 1845.6,
found 1845.6.
Starting from 5.07 mg of peptide 6, 2.83 mg (52%) of peptide
17 were obtained, white powder, ES-MS [M + H]+ calcd 1695.1,
found 1695.5.
Starting from 5.01 mg of peptide 6, 2.53 mg (47%) of peptide
18 were obtained, white powder, ES-MS [M + H]+ calcd 1699.2,
found 1699.7.
Peptide 16 was purified on a C3 Zorbax column using buffers
D and E (0-60% E in 70 min, 3 mL/min, 50 °C, detection at
215 nm). Starting from 4.96 mg of peptide 6, 1.67 mg (31%) of
peptide 16 were obtained, white powder, ES-MS [M + H]+
calcd 1713.1, found 1713.1.
Syn th esis of P ep tid e 19. Peptide elongation was per-
formed starting from 0.2 mmol of a Fmoc-PAL-PS resin (0.16
mmol/g, Perseptive). Then, an aliquot of the peptidyl resin
(127.3 µmol) was acylated with (Boc)2N-N(Boc)CH2CO2H as
described previously for peptide 6. Cleavage and deprotection
were achieved using TFA/water/phenol/ethanedithiol/thioani-
sole: 10 mL/0.5 mL/0.75 g/0.25 mL/0.5 mL during 2 h at room
temperature. The crude peptide was precipitated in diethyl
ether/heptane: 1/1 by vol, dissolved in water/acetic acid: 5/1
by vol and lyophilized. RP-HPLC purification was performed
on a C3 Zorbax (15 × 300 mm) column using buffers A and C
(25-70% C in 60 min, 3 mL/min, 50 °C, detection at 215 nm).
Following lyophilization, 126 mg (17%) of peptide 19 were
obtained as a white powder. AA(calcd): found; R(6): 6.0, K(3):
2.8, E(7): 7.3, A(2): 1.9, L(4): 4.0, S(3): 2.7, F(2): 1.8. ES-MS:
[M + H]+ calcd 4645.4, found 4644.5.
Synthesis of Lipopeptide 20 by Acylation of 19 with Palmitic
Acid Succinimidyl Ester. 8.15 mg (1.4 µmol) of peptide 19 were
reacted with palmitic acid succinimidyl ester as described
previously. The crude peptide was purified by RP-HPLC on a
C3 Zorbax (15 × 300 mm) column using buffers A and B (0-
60% C in 80 min, 3 mL/min, 50 °C, detection at 215 nm) to
give 4.79 mg (57.8%) of a white powder. ES-MS: [M + H]+ calcd
4883.8, found 4882.5.
Syn th esis of P ep tid e 21. Peptide elongation was per-
formed starting from 0.25 mmol of Rink amide AM-PS resin
(0.70 mmol/g, Senn Chemicals). Then, the peptidyl resin was
subjected to the BCPO procedure (10 cycles), neutralized with
diisospropylethylamine/CH2Cl2: 1/19 by vol (2 × 2 min), and
capped with palmitic acid succinimidyl ester in DMF. The resin
was washed with diethyl ether (2 × 2 min) and dried in vacuo.
The final cleavage and deprotection steps were performed
using 10 mL of TFA/water/thioanisole/TIS: 94/2.5/2.5/1 by vol
during 1 h 30 min. The crude peptide was precipitated in cold
diethyl ether/pentane: 1/1 by vol, dissolved in water/acetic acid:
9/1 by vol, and lyophilized. The RP-HPLC purification was
performed on a C3 Zorbax (15 × 300 mm) column using buffers
A and C (0-70% C in 70 min, 2 mL/min, 30 °C, detection at
215 nm) to give 124 mg (56%) of a white powder. TOF-PDMS:
[M + H]+ calcd 896.1, found 895.9. AA(calcd): found; G(1): 0.94,
V(1): 1.0, F(2): 2.0, K(1): 1.0, R(1): 1.0.
Syn th esis of P ep tid e 22. Peptide 21 (5.06 mg, 3.75 µmol)
was reacted with palmitic acid succinimidyl ester as described
previously. Following 48 h at room temperature, the reaction
mixture was purified by RP-HPLC on a C3 Zorbax column
using A and B buffers (0-100% B in 90 min, 3 mL/min, 30 °C,
detection at 215 nm) to give 3.26 mg (60%) of a white powder.
ES-MS: [M + H]+ calcd 1133.9, found 1133.8. AA(calcd): found;
G(1): 0.99, V(1): 1.03, F(2): 2.0, K(1): 1.02, R(1): 0.97.
Ack n ow led gm en t. We thank Bernadette Coddeville
and Guy Ricart for the ES-MS studies (Universite´ des
Sciences et Technologies de Lille) and Dr. Steve Brooks
for proofreading the manuscript. We gratefully acknowl-
edge financial support from Institut Pasteur de Lille,
Agence Nationale de Recherche sur le SIDA (A.N.R.S.),
and Centre National de la Recherche Scientifique
(C.N.R.S.). Dominique Bonnet holds a C.N.R.S.-Re´gion
Nord-Pas-de-Calais fellowship.
Su p p or tin g In for m a tion Ava ila ble: 1H and 13C NMR of
N-(cholesterylcarbonyloxy)succinimide. RP-HPLC, mass spec-
tra, amino acid analysis, and capillary zone electrophoresis of
peptides 4-6, 9, 14. RP-HPLC, mass spectra, and capillary
zone electrophoresis of peptides 16, 17, 19, and 20. RP-HPLC,
mass spectra and amino acid analysis of peptides 21 and 22.
RP-HPLC and mass spectra of peptides 15 and 18. This
material is available free of charge via the Internet at
http://pubs.acs.org.
J O0010577