Synthetic studies towards peptidyl nucleoside antibiotics: First synthesis of a polyoxamic acid derivative enabling direct coupling with α-amino acid esters

Article abstract of DOI:10.1039/a700815e

The reaction of the Mukaiyama's aldehyde-derived N-benzyl imine with an hydroxyketene equivalent followed by exposure of the resulting α-hydroxy β-lactam to NaOCl and 2,2,6,6-tetramethylpiperidinyl-1-oxyl (TEMPO) provides a novel α-amino acid N-carboxy an

Full text of DOI:10.1039/a700815e

View Article Online / Journal Homepage / Table of Contents for this issue
Synthetic studies towards peptidyl nucleoside antibiotics: first synthesis of a
polyoxamic acid derivative enabling direct coupling with a-amino acid esters
Claudio Palomo,* Mikel Oiarbide and Aitor Esnal
Departamento de Qu´ımica Orga´nica, Facultad de Qu´ımica, Universidad del Pa´ıs Vasco, Apdo. 1072, 20080 San Sebastia´n, Spain
The reaction of the Mukaiyama’s aldehyde-derived N-
benzyl imine with an hydroxyketene equivalent followed by
exposure of the resulting a-hydroxy b-lactam to NaOCl and
2,2,6,6-tetramethylpiperidinyl-1-oxyl (TEMPO) provides a
novel a-amino acid N-carboxy anhydride formally derived
from polyoxamic acid.
triethylamine, with the imine 3 gave 5 [mp 74–76 °C, [a]2D5 +1.6
(CH2Cl2, c 1.0)] in 85% yield. In each case a single
diastereoisomer was produced as judged by 1H NMR analysis of
the crude reaction products. The relative cis-disposition of the
substituents at the C3 and the C4 positions of the b-lactam ring
was determined by the value of the proton NMR coupling
constant (J3,4 = 5 Hz) and the absolute configuration of both
b-lactams 4 and 5 was primarily established by the assumption
of an uniform stereochemical outcome with regard to that
observed in closely related reactions.7,8 Further, chemical
correlation of both 4 and 5 with the known 4-formyl derivative
68 confirmed the assignment.
To complete the synthesis of the polyoxamic acid skeleton
the 3-hydroxy b-lactam 4 was converted into the NCA 7 by
using a solution of commercial bleach and a catalytic amount of
2,2,6,6-tetramethylpiperidinyl-1-oxyl (TEMPO). The trans-
formation occurred almost instantaneously (1–2 min)‡ via an
intermediate a-keto b-lactam which underwent chemoselective
Baeyer–Villiger rearrangement to produce 7 in almost quantita-
tive yield in a single pot operation. That is, from the readily
available imine 3 a simultaneously amino-protected and
carboxy-activated form of polyoxamic acid is obtained in 90%
overall yield. From this approach two key elements are
especially noteworthy. First, the creation of the a-amino
stereogenic centre of the polyoxamic acid with virtually
complete diastereoselectivity as compared with the reported
procedures employing Mukaiyama’s aldehyde as the starting
The polyoxin 1 family of nucleoside antibiotics, Fig. 1, has as a
common structural feature a dipeptide comprised of an unusual
hydroxylated a-amino acid, commonly named polyoxamic acid
2, linked to one of the related nucleoside a-amino acids.1 As a
consequence of the potent antifungal activity associated with
this type of peptidyl nucleosides1,2 many methods for the
preparation of 2 have been proposed over the last few years.3
However, most of them involve a large number of steps and,
therefore, a practical short route to this amino acid still remains
of interest. On the other hand, and most important, all of these
methods deal with the synthesis of the amino acid in its free
form rather than with the generation of a simultaneously amino-
protected and carboxy-activated species ready for a subsequent
acylation reaction. Consequently, it would be more desirable
and also a conceptually new approach to peptidyl nucleoside
antibiotics if the synthesis of the polyoxamic moiety could be
combined with a peptide coupling step.4 Here we disclose our
initial results on the first synthesis of polyoxamic acid that
fulfils all of these requirements.
As Scheme 1 illustrates, the approach takes advantage of the
facile construction of an a-hydroxy b-lactam via standard [2 +
2] cycloaddition of hydroxyketene equivalents with imines,5
followed by its quantitative conversion into an a-amino acid
N-carboxy anhydride (NCA) according to our recently reported
method.6 For the first step of the synthesis we elected to use the
imine 3 derived from Mukaiyama’s aldehyde, already incorpo-
rating the trihydroxylated subunit of the desired amino acid and,
at the same time, providing chirality to the corresponding NCA
precursor.† Thus, the cycloaddition reaction of acetoxyketene,
generated from acetoxyacetyl chloride and triethylamine, with
the imine 3 and subsequent saponification of the acetoxy group
of the resulting crude product furnished the crystalline a-hy-
droxy b-lactam 4 [mp 98–102 °C, [a]2D5 212.9 (CH2Cl2, c 1.0)]
in 90% yield over the two steps. Likewise, the reaction of
benzyloxyketene, generated from benzyloxyacetyl chloride and
Me
Me
Me
Me
R2
O
O
O
R4O
vi
CO2R3
OBn
N
H
NHR1
O
N
OH
Bn
8 R1 = Bn, R3 = Bn, R4 = Bn
9 R1 = H, R3 = Boc, R4 = H
3
v
i
Me
O
O
Me
H
iv
OBn
Me
Me
O
O
H
H
NBn
OH
RO
OBn
O
7
NBn
OH
R2
O
H
H
BnO
CHO
O
NH
CO2R1
iii
O
OH
O
4 R = H
5 R = Bn
ii
NBn
O
N
H2N
O
N
H
O
a R2 = CH2CHMe2
b R2 = Bn
6
OH NH2
O
HO
c R2 = CHMe2
OH
Me
1
Scheme 1 Reagents and conditions: i, AcOCH2COCl, Et3N, CH2Cl2,
0 °C ? room temp., 16 h then LiOH (6 equiv.), H2O2 (18 equiv.), THF–
H2O, 0 °C, 3 h, 90% or BnOCH2COCl, Et3N, CH2Cl2, 0 °C ? room temp.,
16 h, 85%; ii, BnBr, NaH, DMF, 220 °C; iii, 5 with HO4Cl, wet THF, room
temp., 2–4 h, then NaIO4, H2O–Me2CO, 0 °C, 2 h; iv, 4 with NaOCl,
CH2Cl2, TEMPO (cat.), NaHCO3, KBr, KH2PO4–K2HPO4 (pH 7.0), 0 °C,
1 min; v, (S)-H2NCH(R2)CO2Bn, CH2Cl2, room temp., 24 h; vi, H2, Pd–C
(10% w/w), (Boc)2O, EtOH, room temp., 24 h
OH
R1 OH,
Me
R2 CO2H, CH2OH, Me, H
N
CO2H
HO
CO2H
OH NH2
2
Fig. 1
Chem. Commun., 1997
691

View Article Online
material3c–g and, second, the generation, for the first time, of the
polyoxamic acid 2 as an active species allowing further
coupling reactions. On pursuing this latter aspect, however, the
coupling of 7 with an a-amino acid ester was essentially a
question of whether epimerization at the a-centre of the NCA 7
could take place; if not, the approach would open up a new and
short route to peptidyl nucleoside antibiotics. We were gratified
to observe that the reaction of freshly prepared 7 with
(S)-leucine benzyl ester in dichloromethane as solvent at room
temperature for 24 h proceeded with no appreciable epimeri-
zation§ to give the desired peptide 8 in 85% yield. The reaction
of 7 with both (S)-phenylalanine and (S)-valine benzyl ester also
led to the formation of the corresponding coupling peptides in
86 and 90% yield, respectively. The peptide products were then
transformed into their Boc-derivatives 9 for characterization
purposes.
The reaction time for the coupling process could be shortened
by using a more polar solvent such as DMF, but under these
conditions isomerization at the Ca of the NCA occurred¶ giving
rise to an equimolar mixture of epimers. This observation
prompted us to examine further the influence of the nature of the
solvent upon the degree of isomerization. To this end we carried
out parallel experiments using an array of common solvents
with different polarities. As the results in the Table 1 show, an
increase of m, the dipolar moment of the solvent, leads to
increased isomerization, the largest isomerization being when
the coupling was performed in HMPA. It thus appears that
dipeptide products with the unnatural configuration at the Ca
position of the polyoxamic acid might also be obtained from this
approach, albeit some further improvement would be required.
Finally, as Scheme 2 illustrates, the method can also be applied
to the synthesis of tripeptides containing the polyoxamic
framework as part of the backbone. Therefore, it is expected that
more complex amino acid derived peptides could be made
accessible from this approach without the necessity to pre-
viously prepare each individual non-proteinogenic a-amino
acid.
This work has been supported by Ministerio de Educacio´n y
Ciencia (MEC) of the Spanish Government (Project SAF:
95/0749) and in part by the Basque Government (Project: PI95/
93). A grant from the Basque Government to A. E. is gratefully
acknowledged.
Footnotes
† For a conceptually different b-lactam route to the unnatural enantiomer of
polyoxamic acid, see ref. 3(k).
‡ Prolonged exposure of 4 to the reaction conditions causes a reduction in
chemical yield.
§ Under these reaction conditions the degree of isomerization was
determined to be less than 0.5% by HPLC analysis of the crude reaction
nature and comparison with the chromatogram of a sample containing the
mixture of epimers, the latter prepared by carrying out the coupling reaction
in DMF as solvent, see text.
¶ The possibility of the isomerization of the other labile Ca stereocentre of
the amino ester residue under these reaction conditions was ruled out by
comparison of the HPLC retention times and NMR spectra of the crude
product obtained by coupling of the NCA 7 with both optically pure and
racemic amino esters.
References
1 For reviews on this subject, see K. Isono, J. Antibiot., 1988, 41, 1711;
K. Isono, Pharm. Ther., 1991, 52, 269.
2 A. B. Cooper, J. Desai, R. G. Lovey, A. K. Saksena, V. M.
Girijavallabhan, A. K. Ganguly, D. Loebenberg, R. Parmegiani and
A. Cacciapuoti, Bioorg. Med. Chem. Lett., 1993, 3, 1079 and references
cited therein.
3 For reviews, see (a) P. Garner, in Studies in Natural Products Chemistry,
ed. Atta-Ur Rahman, Elsevier, Amsterdam, 1988, vol. 1, part A, p. 397;
(b) G. Casiraghi, F. Zanardi, G. Rassu and P. Spanu, Chem. Rev., 1995,
95, 1677. For recent methods, see (c) A. K. Saksena, R. G. Lovey,
V. M. Girigavallabhan and A. K. Ganguly, J. Org. Chem., 1986, 51,
5024; (d) I. Savage and E. J. Thomas, J. Chem. Soc., Chem. Commun.,
1989, 717; (e) T. Mukaiyama, K. Suzuki, T. Yamada and F. Tabuga,
Tetrahedron, 1990, 46, 265; (f) A. Dondoni, S. Franco, F. L. Mercha´n,
P. Merino and T. Tejero, Tetrahedron Lett., 1993, 34, 5479; (g)
R. F. W. Jackson, N. J. Palmer, M. J. Wythes, W. Clegg and
M. R. J. Elsegood, J. Org. Chem., 1995, 60, 6431; (h) P. Garner and
J. M. Park, J. Org. Chem., 1988, 53, 2979; (i) A. Dureault, F. Carreaux
and J. C. Depezay, Synthesis, 1991, 150; (j) M. Irama, H. Hioki and S. Ito,
Tetrahedron Lett., 1988, 29, 3125; (k) B. K. Banik, M. S. Manhas and
A. K. Bose, J. Org. Chem., 1993, 58, 307; (l) M. M. Paz and F. J. Sardina,
J. Org. Chem., 1993, 58, 6990; (m) F. Matsuura, Y. Hamada and
T. Shiori, Tetrahedron Lett., 1994, 35, 733; (n) B. M. Trost, A. C.
Krueger, R. C. Bunt and J. Zambrano, J. Am. Chem. Soc., 1996, 118,
6520; (o) S. H. Kang and H.-W. Choi, Chem. Commun., 1996, 1521.
4 For a recent review on the synthesis of complex nucleoside antibiotics,
see S. Knapp, Chem. Rev., 1995, 95, 1859. For recent total synthesis of
(+)-polyoxin J, see N. Chida, K. Koizumi, Y. Kitida, C. Yokoyama and
S. Ogawa, J. Chem. Soc., Chem. Commun., 1994, 111; A. Dondoni,
F. Junquera, F. L. Mercha´n, P. Merino and T. Tejero, J. Chem. Soc.,
Chem. Commun., 1995, 2127.
Table 1 Solvent effect on the coupling of 7 with (S)-LeuOBna
b
Entry
Solvent
m
8:epi-8c
1
2
3
4
5
6
7
Et2O
4.34
100:0d
100:0d
85:15
92:8
50:50
51:49
28:72
CH2Cl2
MeCN
MeNO2
DMF
Me2SO
HMPA
5.17
11.48
11.88
12.88
13.00
18.48
a Reactions conducted on a 0.25 mmol scale at room temp. for 24 h in the
corresponding solvent (2.5 ml); 7:(S)-LeuOBn, 1:2. b From C. Reichardt in
Solvent Effects in Organic Chemistry, ed. H. F. Ebel, Verlag Chemie Gmbh,
5 For reviews, see G. I. Georg and V. T. Ravikumar, in The Organic
Chemistry of b-Lactams, ed. G. I. Georg, VCH, New York, 1993, p. 295;
J. Backes, in Houben-Weyl Methoden der Organischen Chemie, ed. E.
Muller and O. Bayer, Band E16B, Theime, Stuttgart, 1991, p. 31.
6 C. Palomo, J. M. Aizpurua, C. Cuevas, R. Urchegui and A. Linden,
J. Org. Chem., 1996, 61, 4400.
7 For cycloadditions of glyceraldehyde imines, see C. Hubschwerben and
G. Schmid, Helv. Chim. Acta, 1983, 66, 2206; D. R. Wagle, G. Garai,
J. Chiang, M. S. Manhas and A. K. Bose, J. Org. Chem, 1988, 29, 5065;
S. Saito, T. Ishikawa and T. Moriwake, Synlett., 1993, 139.
c
1
Weinheim, 1979, p. 270. Determined by H NMR analysis of the crude
products. d Corroborated by HPLC analysis.
Me
Me
Me
Me
O
O
Me
Me
H2N
H
N
N
H
CO2H
Me
HO
i, ii
CO2Bn
Me
N
H
8 C. Palomo, F. P. Coss´ıo, C. Cuevas, B. Lecea, A. Mielgo, P. Roma´n,
A. Luque and M. J. Mart´ınez-Ripoll, J. Am. Chem. Soc., 1992, 114, 9360;
M. Jayaraman, A. R. A. S. Deshmukh and B. M. Bahawal, J. Org. Chem.,
1994, 59, 932.
O
O
NHBoc
O
60%
Me
Me
Received in Cambridge, UK, 4th February 1997; Com.
7/00815E
Scheme 2. Reagents and conditions: i, 7, CH2Cl2, room temp., 24 h, 80%;
ii, H2, Pd–C (10% w/w), Boc2O, EtOH, room temp., 24 h, 75%
692
Chem. Commun., 1997