Straightforward synthesis of N-hydroxy peptides

Article abstract of DOI:10.1055/s-2003-38347

N-Hydroxy dipeptides are readily synthesized by reduction of the corresponding oximes. The two diastereomers obtained are easily separated by flash chromatography. They can be coupled with a third amino acid moiety - without protection of the hydroxyl group - to give N-hydroxy tripeptides.

Full text of DOI:10.1055/s-2003-38347

LETTER
671
Straightforward Synthesis of N-Hydroxy Peptides
S
traightforwar
a
d Synthesis
s
of
N
-
Hyd
c
roxy Peptid
a
es l Maire, Véronique Blandin, Monique Lopez, Yannick Vallée*
LEDSS, UMR CNRS-Université Joseph Fourier, BP 53, 38041 Grenoble Cedex 9, France
Fax +33(4)76635983; E-mail: Syrca.Ledss@ujf-grenoble.fr
Received 30 January 2003
tides – involving or not a transient protection of the hy-
droxyl group during the coupling with another peptidic
fragment.
Abstract: N-Hydroxy dipeptides are readily synthesized by reduc-
tion of the corresponding oximes. The two diastereomers obtained
are easily separated by flash chromatography. They can be coupled
with a third amino acid moiety – without protection of the hydroxyl
group – to give N-hydroxy tripeptides.
Routes to non-racemic N-hydroxy α-amino acids are
known8 but require multi-step procedures from optically
active starting materials. We preferred to investigate the
reduction of peptidic oximes 2 as a direct synthesis of ter-
minal N-hydroxy dipeptides 3, in order to have rapid ac-
cess to a small library of molecules. Compounds 2 were
obtained in two steps from the corresponding α-amino es-
ters via the α-keto amides 1 (Scheme 2). The results are
summarized in Table 1.
Key words: N-hydroxy peptides, oximes, reduction, hydroxyl-
amines, N-/O-acylation
There has been a growing interest in N-hydroxy peptide
type compounds1 in the last decade. This modified peptid-
ic structure – including one or more hydroxamate motifs
– has been observed in some natural products such as the
apoptosis inducers polyoxypeptins A and B.2 The anti-
HIV activity of some other N-hydroxy pseudopeptides has
been tested.3 Moreover the additional OH group induces
conformational changes in the peptide.4 Furthermore the
strong iron complexation properties of the hydroxamate
moiety was used for the design of good siderophores.5
Acid chlorides were generated by reaction of the corre-
sponding α-keto acids with α,α-dichloromethyl-
methylether9 and were either isolated (Entries 1–4 and 8)
or prepared in situ (Entries 5–7). They were then reacted
with various α-amino esters – HLeuOEt (Entries 1, 5),
HValOEt (Entries 2, 6, 8), HPheOEt (Entries 3, 7) and
HProOEt (Entry 4) – to give α-keto amides 1 in good
yields. Condensation of 1 with hydroxylamine hydrochlo-
ride in ethanol followed by addition of triethylamine led
to oxime 2 as a single isomer (R2 = Me) or as a mixture of
E and Z-isomers (R2 = i-Pr, Ph).10 Reduction of 2 was car-
The obvious method for the synthesis of these compounds
would be the direct coupling of a N-hydroxy amino acid
(or a terminal N-hydroxy peptide, Scheme 1) with an acti-
vated peptide residue. However, two coupling products
can be obtained: the O-acylated compound A and the de-
sired N-acylated compound B. Protection of the OH group
as a benzyl ether has thus been used to overcome this
problem.6 This requires an additional deprotection step
but above all N-benzyloxy amino acids show a decreased
reactivity in coupling reactions when bulky residues are at
stake.7
O
Cl
R2
R
O
NH
R
O
N
CO2Et
R2
R2
R2
R1
CO2Et
CH2Cl2
O
R1
1a-h
In this communication we report (i) the reduction of pep-
tidic oximes as a way to terminal N-hydroxy dipeptides
and (ii) two methods for obtaining inner N-hydroxy pep-
NH2OH
EtOH
a : R = H, R1 = iBu, R2 = Me
b : R = H, R1 = iPr, R2 = Me
c : R = H, R1 = Bn, R2 = Me
d : R-R1 = -(CH2)3-, R2 = Me
e : R = H, R1 = iBu, R2 = iPr
f : R = H, R1 = iPr, R2 = iPr
g : R = H, R1 = Bn, R2 = iPr
h : R = H, R1 = iPr, R2 = Ph
NOH
O
R
R
R'
R
N
CO2Et
OH
OH
O
N
H
N
N
H
H
R1
2a-h
A
and/or
B
O
O
O
R
OH
N
O
R'
BH3·NMe3
HCl / EtOH
N
N
H
H
O
R'
O
NHOH
O
R
N
CO2Et
R1
Scheme 1
Synlett 2003, No. 5, Print: 10 04 2003.
3a-h
Art Id.1437-2096,E;2003,0,05,0671,0674,ftx,en;D02503ST.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0936-5214
Scheme 2

672
P. Maire et al.
LETTER
Table 1 Synthesis of Compounds 1, 2 and 313
O
H
FmocHN
N
CO2Et
N
Entry
Compound Yield (%)
OH
O
1a
2a
3b
(S, S, S)-4
1
2
3
4
5
6
7
8
a
b
c
d
e
f
87
86
90
81
95
86
70
73
84
80
80
93
H
O
N
H
N
CO2Et
86
80
FmocHN
O
O
93
40
65c
76c
55c
95
88
(S, S, S)-5
Figure 1
91
g
h
83
no reaction
TMSCI
(2 equiv)
a Isolated yields.
HO
NH
TMSO
pyridine
(4 equiv)
b Total yield of the two diastereomers.
NH
H
H
c Starting from keto acid; the acyl chloride was not isolated.
N
CO2Et
N
CO2Et
CH2Cl2
O
O
ried out with 2 equivalents of trimethylamine/borane
complex in 7 N ethanolic hydrochloric acid.11 α-Oximino
amides 2 bearing an alkyl R2 substituent (Entries 1 to 7)
were generally reduced in high yields whereas an aromat-
ic R2 substituent seems detrimental to the reactivity (Entry
8).12
(S, S)-3b
1) FmocAlaCl (1 equiv)
CH2Cl2
(S,S,S)-4
2) H3O+
Scheme 3
The diastereoselectivity of the reduction was poor in all
cases, leading roughly to a 1:1 mixture of (R,S) and (S,S)
N-hydroxy dipeptides 3 as estimated from the H NMR
This one-pot procedure was next extended to the prepara-
tion of other N-hydroxy peptides (Table 2, Method A).
The coupling between two Alanine residues succeeded in
fair to good yields and both diastereomers of 3 show the
same reactivity (compare Entries 2/3, 6/7, 8/9). However
increasing the bulkiness of one of the partners gave poorer
results (Entries 2/4, 7/9/10, 14). Running the reaction at
40 °C improved the yield to some extent (Entry 11).
1
spectra of the crude materials and/or by isolated yield of
each diastereomer after separation by liquid chromatogra-
phy.14 The absolute configuration of each diastereomer
was assigned by hydrogenation of the N-hydroxylamine
in the presence of Raney nickel and comparison of the
dipeptide obtained with an authentic (S,S) sample pre-
pared using standard peptide coupling methods.
We then reinvestigated the direct reaction with an acyl
chloride. Reaction of N-Fmoc-Alanine chloride (1 equiv)
with (S,S)-3b in the presence of pyridine (1 equiv) led to
a 65:35 mixture of 4/5 in 50% isolated yield. Changing the
base to sodium hydrogenocarbonate (3 equiv) resulted in
a dramatic improvement, the N-hydroxy tripeptide 4 was
produced in 82% yield. With these new conditions the
coupling reactions occured in high yields, regardless of
steric hindrance (Table 2, Method B). Even the reaction
involving two Valine residues can be performed success-
fully (Entry 15).
Scarce examples of coupling methods for the preparation
of inner N-hydroxy peptides can be found in the litera-
ture.15 In our hands, coupling between FmocAlaOH and
1
(S,S)-3b led to two isomers showing different H NMR
spectra. The products resulting from N-acylation (4)16 and
O-acylation (5)17 (Figure 1) were identified by com-
parison with an authentic sample of Fmoc-
AlaΨ[CO(NOH)]AlaVal-OEt (prepared via the N-benz-
yloxy dipeptide). All the coupling conditions that we used
(activation with DCC/HOBt, EDCI, BOP, DMTMM)18,19
gave 5 as the major product.
Finally, the N-hydroxy tripeptide 4 was deprotected quan-
titatively with piperidine. Coupling with N-Fmoc-Valine
using a standard method produced the N-hydroxy tet-
rapeptide Fmoc-ValAlaΨ[CO(NOH)]AlaVal-OEt in 76%
yield (Scheme 4).
To avoid this O-acylation we envisaged a temporary O-
protection of the hydroxylamine as a trimethylsilyl ether20
as shown in Scheme 3. In our tuned conditions, the O-pro-
tected compound was obtained in situ.21 Addition of N-
Fmoc-Alanine chloride followed by an aqueous acidic
work-up removing the TMS protecting group gave access
to the N-hydroxy tripeptide 4 in 70% yield after recrystal-
lization.
In conclusion, we have prepared a series of N-hydroxy
peptides using a straightforward sequence. Once the inner
N-hydroxy tripeptide is obtained, it can be further elongat-
ed via classical peptide synthesis without protection of the
Synlett 2003, No. 5, 671–674 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Straightforward Synthesis of N-Hydroxy Peptides
673
hydroxyl group. We are currently broadening the scope of
this methodology to incorporate other N-hydroxy amino
acid residues with (R) or (S) configurations.
References
(1) For reviews see: (a) Ottenheijm, H. C. J.; Herscheid, J. D.
M. Chem. Rev. 1986, 86, 697. (b)Marraud, M.; Vanderesse,
R. In Houben-Weyl, Methods of Organic Chemistry, 4th ed.,
Vol. E22c; Goodman, M., Ed.; Thieme: Stuttgart, 2002, .
(2) (a) Isolation and structure determination: Umezawa, K.;
Nakazawa, K.; Ikeda, Y.; Naganawa, H.; Kondo, S. J. Org.
Chem. 1999, 64, 3034. (b) Biosynthesis of polyoxypeptin
A: Umezawa, K.; Ikeda, Y.; Kawase, O.; Naganawa, H.;
Kondo, S. J. Chem. Soc., Perkin Trans. 1 2001, 1550.
(3) (a) Garrouste, P.; Pawlowski, M.; Tonnaire, T.; Sicsic, S.;
Dumy, P.; de Rosny, E.; Reboud-Ravaux, M.; Fulcrand, P.;
Martinez, J. Eur. J. Med. Chem. 1998, 33, 423.
Fmoc-Val-OH (1 equiv)
EDCI (1.1 equiv)
HOBt (1 equiv)
O
H
H2N
N
CO2Et
N
CH2Cl2
OH
O
O
N
H
H
N
N
CO2Et
FmocHN
(b) Marastoni, M.; Bazzaro, M.; Salvadori, S.; Bortolotti, F.;
Tomatis, R. Bioorg. Med. Chem. 2001, 9, 939.
O
OH
O
(4) (a) Dupont, V.; Lecoq, A.; Mangeot, J.-P.; Aubry, A.;
Boussard, G.; Marraud, M. J. Am. Chem. Soc. 1993, 115,
8898. (b) Takeuchi, Y.; Marshall, G. R. J. Am. Chem. Soc.
1998, 120, 5363.
Scheme 4
(5) Selected examples: (a) Akiyama, M.; Katoh, A.; Mutsui, Y.;
Watanabe, Y.; Umemoto, K. Chem. Lett. 1996, 915.
(b) Hara, Y.; Akiyama, M. J. Am. Chem. Soc. 2001, 123,
7247.
(6) Kolosa, T.; Chimiak, A. Tetrahedron 1977, 33, 3285.
(7) Formation of the N-benzyloxy amide link is very sensitive to
steric hindrance and generally requires strong acylating
systems such as HATU {N-[(dimethylamino)-1H-1,2,3-
triazolo[4,5-b]pyridin-1-ylmethylene]-N-methylmethan-
aminium hexafluorophosphate} see: (a) Akiyama, M.;
Iesaki, K.; Katoh, A.; Shimizu, K. J. Chem. Soc., Perkin
Trans. 1 1986, 851. (b) Bianco, A.; Zabel, C.; Walden, P.;
Jung, G. J. Peptide Sci. 1998, 4, 471.
(8) (a) Nucleophilic substitution on α-bromo acid: Kolasa, T.;
Chimiak, A. Tetrahedron 1974, 30, 3591. (b) Use of α-
hydroxy acid via the triflate derivative: Feenstra, R. W.;
Stokkingreef, E. H. M.; Nivard, R. J. F.; Ottenheijm, H. C. J.
Tetrahedron 1988, 44, 5583. (c) Under Mitsunobu
conditions: Hanessian, S.; Yang, R.-Y. Synlett 1995, 633.
(d) Oxyamination of N-acylsultam enolate: Oppolzer, W.;
Tamura, O.; Deerberg, J. Helv. Chim. Acta 1992, 75, 1965.
(e) Oxidation of α-amino acid derivatives: Grundke, G.;
Keese, W.; Rimpler, M. Synthesis 1987, 1115. (f) See also:
Feenstra, R. W.; Stokkingreef, E. H. M.; Reichwein, A. M.;
Lousberg, W. B. H.; Ottenheijm, H. C. J. Tetrahedron 1990,
46, 1745. (g) See also: Detomaso, A.; Curci, R. Tetrahedron
Lett. 2001, 42, 755.
Table 2 Synthesis of N-Hydroxy Peptides
Entry Fmoc-
AA1-Cl
NHOH-AA2-AA3-OEt 3 Method Aa Method Bb
Yieldc (%) Yieldc (%)
1
2
Gly
Ala
Ala
Phe
Phe
Ala
Ala
Phe
Phe
Val
Val
Ala
Ala
Ala
Val
Gly
AlaLeu
AlaLeu
AlaLeu
AlaLeu
AlaLeu
AlaVal
AlaVal
AlaVal
AlaVal
AlaVal
AlaVal
AlaPhe
AlaPhe
ValVal
ValVal
AlaLeu
(R,S)-a
(S,S)-a
(R,S)-a
(S,S)-a
(R,S)-a
(S,S)-b
(R,S)-b
(S,S)-b
(R,S)-b
(R,S)-b
(R,S)-b
(S,S)-c
(R,S)-c
(S,S)-f
(S,S)-f
(R,S)-a
66
69
64
51
33
70
67
57
58
25
40d
–
85
–
3
84
76
77
–
4
5
6
7
82
68
–
8
9
10
11
12
13
14
15
16
78
–
85
–
(9) Ottenheijm, H. C. J.; de Man, J. H. M. Synthesis 1975, 163.
(10) The two oxime isomers are separated by liquid
chromatography. No difference in diastereoselectivity was
observed in their reduction so that they were used as a
mixture.
63
26
–
77
73
85
(11) Tijhuis, M. W.; Herscheid, J. D. M.; Ottenheijm, H. C. J.
Synthesis 1980, 890.
66
(12) Reduction of 2h with sodium cyanoborohydride was also
inefficient.
a Method A: 1) 3, TMSCl (2 equiv), pyridine (4 equiv), CH2Cl2, r.t.,
30 min. 2) Fmoc-AA1-Cl (1 equiv), CH2Cl2, O °C, 10 min, then r.t.,
2 h. 3) H3O+.
(13) All new compounds gave spectroscopic and analytical data
in agreement with the assigned structures. Selected example:
N-hydroxy dipeptide (S,S)-3b: [α]D25 –3.8 (c 2.26, CHCl3).
1H NMR (300 MHz, CDCl3): δ = 7.08 (d, 3J = 9.6 Hz, 1 H,
CONH), 5.46 (br, 2 H, NHOH), 4.61 (dd, 3J = 9.6 and 4.8
Hz, H, CHα Val), 4.21 (m, 2 H, OCH2CH3), 3.66 (q, 3J = 7.1
Hz, 1 H, CHα Ala), 2.23 (m, 1 H, CH-i-Pr), 1.29 (t, 3J = 7.0
Hz, 3 H, OCH2CH3), 1.26 (d, 3J = 7.1 Hz, 3 H, CH3 Ala),
0.97 (d, 3J = 6.9 Hz, 3 H, CH3-i-Pr), 0.91 (d, 3J = 6.9 Hz, 3
H, CH3-i-Pr). 13C NMR (75.5 MHz, CDCl3): δ = 174.0,
172.8, 62.2, 61.6, 56.7, 31.2, 19.2, 17.7, 15.6, 14.3. MS (CI):
b Method B: 3, Fmoc-AA1-Cl (1.03 equiv), NaHCO3 (3 equiv),
CH2Cl2, r.t., 2 h.
c Isolated yields. A hyphen indicates the reaction was not performed
under the current method.
d Reaction carried out at 40 °C.
Synlett 2003, No. 5, 671–674 ISSN 0936-5214 © Thieme Stuttgart · New York

674
P. Maire et al.
LETTER
m/z = 233 (100, M + H+), 187 (60), 159 (53). IR (KBr,
cm–1): 3323 (m), 3254 (w), 2977 (m), 1738 (s), 1663 (s),
1541 (s).
19.0, 18.2, 17.8, 14.5, 14.3. MS (CI): m/z = 543 (27, M +
NH4 ) 304 (100), 179 (69). IR (KBr, cm–1): 3316 (s br), 2964
+
(m), 1734 (s), 1701 (s), 1641 (s), 1533 (s), 1450 (s).
(17) Selected 1H NMR data (250 MHz, CDCl3) for compound 5:
δ = 7.42 (br d, 3J = 4.5 Hz, 1 H, -NH-O-CO-), 3.73 [qd,
3J = 6.9 and 4.5 Hz, 1 H, -CH(CH3)-NH-O(CO)-].
(18) Abbreviations: DCC, dicyclohexylcarbodiimide; HOBt, 1-
hydroxybenzotriazole; EDCI, 1-ethyl-3-(3′-dimethyl-
aminopropyl)carbodiimide; BOP, benzotriazol-1-
(14) Typical example: reduction of 9.33 mmol of 2b gave after
flash chromatography (silica gel; CH2Cl2–MeOH, 97:3) 3.82
mmol of the first diastereomer, 0.53 mmol of a 1:1 mixture
(estimated from 1H NMR) and 3.69 mmol of the second
diastereomer.
(15) See ref. 1 and references therein.
yloxytris(dimethylamino)phosphonium hexafluoro-
phosphate; DMTMM, 4-(4, 6-dimethoxy[1,3,5]triazin-2-
yl)-4-methyl-morpholinium chloride.
(16) Compound (S,S,S)-4 {Fmoc-Ala1Ψ[CO(NOH)]Ala2Val3-
OEt}: [α]D25 –12.7 (c 1.28, CHCl3). 1H NMR (300 MHz,
CDCl3): δ = 8.94 (s, 1 H, OH), 7.73 (d, 3J = 7.4 Hz, 2 H,
CHar Fmoc), 7.58 (d, 3J = 7.4 Hz, 2 H, CHar Fmoc), 7.37
(dd, 3J = 7.4 and 7.4Hz, 2 H, CHar Fmoc), 7.27 (dd, 3J = 7.4
and 7.4 Hz, 1 H, CHar Fmoc), 7.27 (dd, 3J = 7.4 and 7.4 Hz,
1 H, CHar Fmoc), 7.16 (br d, 3J = 8.7 Hz, 1 H, NH Val3),
5.96 (d, 3J = 7.0 Hz, 1 H, NH Ala1), 5.29 (q, 3J = 7.0 Hz, 1
H, CHα Ala2), 4.97 (dq, 3J = 7.0 and 6.8 Hz, 1 H, CHα Ala1),
4.50 (dd, 3J = 8.7 and 4.9 Hz, 1 H, CHα Val3), 4.38–4.27 (m,
2 H, CHH and CH Fmoc), 4.23–4.16 (m, 1 H, CHH Fmoc),
4.17 (q, 3J = 7.1 Hz, 2 H, OCH2CH3), 2.21–2.09 (m, 1 H,
CH-i-Pr Val3), 1.49 (d, 3J = 7.0 Hz, 3 H, CH3 Ala2), 1.42 (d,
3J = 6.8 Hz, 3 H, CH3 Ala1), 1.23 (t, 3J = 7.1 Hz, 3 H,
OCH2CH3), 0.91 (d, 3J = 6.9 Hz, 3 H, CH3-i-Pr Val3), 0.88
(d, 3J = 6.9 Hz, 3 H, CH3-i-Pr Val3). 13C NMR (50 MHz,
CDCl3): δ = 172.8, 172.3, 171.7, 156.4, 143.9, 141.4, 127.8,
127.2, 125.3, 120.1, 67.4, 61.6, 57.3, 54.3, 47.2, 47.2, 31.5,
(19) (a) Preparation of DMTMM: Kunishima, M.; Kawachi, C.;
Morita, J.; Terao, K.; Iwasaki, F.; Tani, S. Tetrahedron
1999, 55, 13159. (b) Use in the synthesis of hydroxamates:
De Luca, L.; Giacomelli, G.; Taddei, M. J. Org. Chem. 2001,
66, 2534.
(20) (a) Reaction of N-alkyl hydroxylamine with mixed
anhydrides of amino acids: Nakonieczna, L.; Chimiak, A.
Synthesis 1987, 418. (b) Acylation of optically active N-
hydroxy Leucine methyl ester with simple acyl chlorides:
Jin, Y.; Kim, D. H. Tetrahedron: Asymmetry 1997, 8, 3699.
(21) Running the reaction in CDCl3 in a NMR tube with 3 equiv
of pyridine indicates that with 0.7 equiv of TMSCl two
exchanging products are first visible. After the addition of an
excess TMSCl (up to 2.2 equiv), only one product remains.
This is in agreement with previous findings (see ref. 20a).
Synlett 2003, No. 5, 671–674 ISSN 0936-5214 © Thieme Stuttgart · New York