A highly efficient, asymmetric synthesis of blastidic acid: The β-amino acid component of the antibiotic, (+)-blasticidin S

Article abstract of DOI:10.1016/j.tetlet.2005.08.096

A new approach for the asymmetric synthesis of blastidic acid, the β-amino acid component of the antibiotic, (+)-blasticidin S has been achieved in 11 steps and 30% overall yield. The key transformations involve a one-pot Swern/Wittig reaction sequence to

Full text of DOI:10.1016/j.tetlet.2005.08.096

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 7147–7149
A highly efficient, asymmetric synthesis of blastidic acid: the
b-amino acid component of the antibiotic, (+)-blasticidin S
Roland Bischoff, Niall McDonald and Andrew Sutherland*
WestChem, Department of Chemistry, The Joseph Black Building, University of Glasgow, Glasgow G128QQ, UK
Received 4 August 2005; accepted 18 August 2005
Available online 6 September 2005
Abstract—A new approach for the asymmetric synthesis of blastidic acid, the b-amino acid component of the antibiotic, (+)-blas-
ticidin S has been achieved in 11 steps and 30% overall yield. The key transformations involve a one-pot Swern/Wittig reaction
sequence to prepare the carbon backbone followed by a stereoselective conjugate addition to introduce the b-amino functionality.
Ó 2005 Elsevier Ltd. All rights reserved.
(+)-Blasticidin S 1, a peptidyl-nucleoside antibiotic first
isolated from Streptomyces griseochromogenes has been
used extensively as a fungicide against Pyricularia ory-
zae, the cause for the serious rice blast disease in
Asia.1–3 Structural characterisation of (+)-blasticidin S 1
using both chemical degradation (Scheme 1) and X-ray
analysis revealed two main components, the b-amino
acid, blastidic acid 2 and the hexopyranosyl nucleo-
side, cytosinine 3.4,5 (+)-Blasticidin S 1 has also been
the subject of intense biosynthetic studies with cytosine,
D-glucose, L-arginine and L-methionine shown to be the
primary building blocks.6 More recently, Gould and
co-workers have shown, using labelled precursors,
that the b-amino acid component of 1, blastidic acid 2
is biosynthesised from L-arginine via an intramolecular
migration of the a-nitrogen to the b-position.7
While the biosynthesis of 1 and its components have
been extensively studied, reports on the chemical synthe-
sis of these compounds are much more limited. Goto
and co-workers outlined the first synthesis of cytosinine
3 in 1972, 8 but it was not until 2001 that two routes for
the synthesis of blastidic acid 2 by the groups of Nom-
oto and Ichikawa were first reported.9,10 Further work
by Ichikawa and co-workers has led to the first total
synthesis of (+)-blasticidin S 1.11 Although two routes
have already been developed for the synthesis of blasti-
dic acid 2, we were interested in developing a novel, effi-
cient approach, which would also provide a suitably
protected derivative for the total synthesis of (+)-blastic-
idin S 1. We now report our 11 step synthesis of (+)-
blastidic acid 2 from readily available b-alanine using
a one-pot Swern oxidation/Wittig olefination reaction
sequence to prepare the key E-allylic ester followed
by a stereoselective conjugate addition to introduce the
b-amino functionality.
HO2C
O
N
NH2
NH
O
O
N
H2N
N
N
H
NH2
CH3
1
H+
HO2C
H2N
O
NH2
NH
O
O
N
+
N
NH2
H2N
N
OH
CH3
3
2
Our retrosynthesis of blastidic acid 2 is outlined in
Scheme 2. It was proposed that 2 could be prepared
by coupling commercially available N,N-bis(tert-but-
oxycarbonyl)-1H-pyrazole-1-carboxamidine 4 and a
suitably protected b-ornithine derivative 5. We intended
to synthesise 5 from b-alanine 9 by methylation at an
early stage followed by a one-pot oxidation and Wittig
Scheme 1.
Keywords: Blastidic acid; b-Amino acid; Antibiotic; Asymmetric
synthesis; Conjugate addition.
*
Corresponding author. Tel.: +44 141 330 5936; fax: +44 141 330
4888; e-mail: andrews@chem.gla.ac.uk
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.08.096

7148
R. Bischoff et al. / Tetrahedron Letters 46 (2005) 7147–7149
functionality.13 The b-nitrogen was then methylated
NH
NH2 O
using sodium hydride and methyl iodide to give deriva-
tive 10 in 93% yield. Attempts at reducing 10 to the cor-
responding aldehyde using 1.2 equiv of DIBAL-H
returned a mixture of the starting material, the desired
aldehyde and alcohol 8. Moreover, purification of the
aldehyde by column chromatography led to significant
decomposition. To overcome these problems, methyl
ester 10 was reduced to alcohol 8 in 79% yield using
2.2 equiv of DIBAL-H. Alcohol 8 was then converted
directly to E-allylic ester 6 using a one-pot Swern oxida-
tion/Wittig olefination reaction process thereby, avoid-
ing any problems associated with isolation of the
aldehyde intermediate.14 Having prepared a suitably
derived E-allylic ester, the next stage of the synthesis
required the introduction of the b-amino functional
groupin an asymmetric fashion. This was carried out
using the approach established by Davies and co-work-
ers involving deprotonation of (S)-benzyl-a-methylbenz-
ylamine 7 with n-butyl lithium at ꢀ78 °C followed by
reaction with E-allylic ester 6.15 After work-upof this
H2N
N
OH
CH3
2
CH3
NBn
Ph
NBoc
+
CO2tBu
BocN
CH3
BocHN
N
N
4
5
CH3
CO2tBu
+
BocN
CH3
Ph
N
H
Ph
O
6
7
BocN
CH3
OH
H2N
OH
1
reaction, the H NMR spectrum of the crude material
9
8
showed the presence of the addition product in a dia-
stereomeric ratio of 24:1. Subsequent purification by
column chromatography gave diastereomer 5 in 85%
yield.
Scheme 2.
reaction to give the desired carbon skeleton. Asymmet-
ric conjugate addition with chiral amine 7 using the
Davies protocol would complete the synthesis of 5.12
As well as developing an efficient route to blastidic acid,
we were also interested in preparing an orthogonally
protected intermediate 13, which could be easily coupled
to a cytosinine derivative for the total synthesis of (+)-
blastidicin S 1. Thus, the b-amino functional groupof
5 was deprotected using transfer hydrogenation condi-
tions16 and then subsequently reprotected with benzyl
chloroformate to give b-ornithine derivative 11 in 74%
yield over two steps (Scheme 4). The next stage of the
The first stage of the synthesis involved the preparation
of b-ornithine derivative 5 (Scheme 3). b-Alanine 9 was
converted in a one-pot reaction to N-Boc-b-alanine
methyl ester in essentially quantitative yield using
chlorotrimethylsilane and methanol to form the ester
followed by treatment with triethylamine and di-tert-
butyl dicarbonate resulting in protection of the amino
CH3
O
O
a, b
Ph
NBn
CbzHN
O
BocN
CH3
OMe
OH
a
H2N
OH
t
CO2 Bu
BocN
CH3
OtBu
BocN
CH3
9
10
11
5
c
b
t
d
CO2 Bu
BocN
CH3
BocN
CH3
BocN
BocHN
CbzHN
O
CbzHN
O
8
c
6
N
OH
HN
CH3
OH
CH3
e
12
13
CH3
NBn
d
Ph
t
NH
NH2 O
CO2 Bu
BocN
CH3
H2N
N
OH
CH3
5
2
Scheme 3. Reagents and conditions: (a) (i) TMSCl, MeOH, (ii) NEt3,
Boc2O, 99% over two steps; (b) MeI, NaH, THF, 93%; (c) DIBAL-H
(2.2 equiv), Et2O, ꢀ78 °C to rt, 79%; (d) (i) (COCl)2, DMSO, NEt3,
Scheme 4. Reagents and conditions: (a) (i) 5% Pd/C, ammonium
formate, BuOH, D, (ii) Cbz–Cl, NaHCO3, acetone, H2O, 74% over
two steps; (b) TFA, CH2Cl2; (c) 4, DIPEA, MeOH, 82% over two
steps; (d) TMSI, CHCl3 then HCl, MeOH, 82%.
t
t
n
ꢀ78 °C to rt, (ii) Ph3P@CHCO2 Bu, 97% over two steps; (e) 7, BuLi,
THF, ꢀ78 °C, 85%.

R. Bischoff et al. / Tetrahedron Letters 46 (2005) 7147–7149
7149
2. Isono, K. J. Antibiot. 1988, 41, 1711–1739.
3. Yamaguchi, H.; Yamamoto, C.; Tanaka, N. J. Biochem.
1965, 57, 667–677.
4. Yonehara, H.; Takeuchi, S.; Otake, N.; Endo, T.; Saka-
gami, Y.; Sumiki, Y. J. Antibiot. 1963, 16A, 195–222;
Otake, N.; Takeuchi, S.; Endo, T.; Yonehara, H. Tetra-
hedron Lett. 1965, 6, 1405–1409.
5. Onuma, S.; Nawata, Y.; Saito, Y. Bull. Chem. Soc. Jpn.
1966, 39, 1091.
6. Seto, H.; Yamaguchi, I.; Otake, N.; Yonehara, H.
Tetrahedron Lett. 1966, 7, 3793–3799; Seto, H.; Yama-
guchi, I.; Otake, N.; Yonehara, H. Agric. Biol. Chem.
1968, 32, 1292–1298.
7. Prabhakaran, P. C.; Woo, N.-T.; Yorgey, P.; Gould, S. J.
Tetrahedron Lett. 1986, 27, 3815–3818; Prabhakaran, P.
C.; Woo, N.-T.; Yorgey, P. S.; Gould, S. J. J. Am. Chem.
Soc. 1988, 110, 5785–5791.
8. Kondo, T.; Nakai, H.; Goto, T. Tetrahedron Lett. 1972,
13, 1881–1884; Kondo, T.; Nakai, H.; Goto, T. Tetra-
hedron 1973, 29, 1801–1806.
9. Nomoto, S.; Shimoyama, A. Tetrahedron Lett. 2001, 42,
1753–1755.
synthesis required deprotection of the side-chain amino
functional groupand addition of the guanidine unit.
This was carried out by treatment of 11 with TFA
resulting in the deprotection of both the Boc-group
and the tert-butyl ester. Reaction of 12 with HunigÕs
¨
base (2 equiv) and N,N-bis(tert-butoxycarbonyl)-1H-
pyrazole-1-carboxamidine 417 then gave orthogonally
protected 13 in 82% yield for the two steps. This stage
in the Ichikawa synthesis proved problematic due to
easy lactamisation during deprotection of the side-chain
amino functional group.10,11 In our own synthesis of
selectively labelled L-arginine, cyclic amide products
were also isolated during a similar reaction sequence.18
Hence, in designing our route to blastidic acid, the
tert-butyl ester was used not only to maximise 1,4-addi-
tion during the conjugate addition stepbut also, to
allow the use of reaction conditions to deprotect both
the d-amino groupand the ester, thus preventing forma-
tion of the six-membered lactam. Finally, 13 was
converted to blastidic acid 2 by reaction with trimethyl-
silyl iodide and acidic methanol.19 This method for
deprotection of carbamates results in protonation of
the amino groups as they are formed, again preventing
any untoward cyclisation reactions.11 Purification by
ion exchange chromatography gave blastidic acid in
82% yield.20
10. Ichikawa, Y.; Ohbayashi, M.; Hirata, K.; Nishizawa, R.;
Isobe, M. Synlett 2001, 1763–1766.
11. Ichikawa, Y.; Hirata, K.; Ohbayashi, M.; Isobe, M. Chem.
Eur. J. 2004, 10, 3241–3251.
12. Davies, S. G.; Ichihara, O. Tetrahedron: Asymmetry 1991,
2, 183–186; Costello, J. F.; Davies, S. G.; Ichihara, O.
Tetrahedron: Asymmetry 1994, 5, 1999–2008.
´
´
13. Kokotos, G.; Padron, J. M.; Martın, T.; Gibbons, W. A.;
In conclusion, a simple and efficient, 11-step synthesis of
blastidic acid has been achieved from commercially
available b-alanine giving the target molecule in 30%
overall yield. This route also provides intermediate 13,
which is suitably protected for coupling with a cytosi-
nine derivative for the total synthesis of (+)-blasticidin
S. Further efforts towards the total synthesis of (+)-blas-
ticidin S are currently underway.
´
´
Martın, V. S. J. Org. Chem. 1998, 63, 3741–3744; Padron,
J. M.; Kokotos, G.; Martın, T.; Markidis, T.; Gibbons, W.
´
´
A.; Martın, V. S. Tetrahedron: Asymmetry 1998, 9, 3381–
3394.
14. Ireland, R. E.; Norbeck, D. W. J. Org. Chem. 1985, 50,
2198–2200.
15. For a recent example, see: Bunnage, M. E.; Davies, S. G.;
Roberts, P. M.; Smith, A. D.; Withey, J. M. Org. Biomol.
Chem. 2004, 2, 2763–2776.
16. Langenhan, J. M.; Gellman, S. H. J. Org. Chem. 2003, 68,
6440–6443.
Acknowledgements
17. Bernatowicz, M. S.; Wu, Y.; Matsueda, G. R. Tetrahedron
Lett. 1993, 34, 3389–3392; Aurelio, L.; Brownlee, R. T. C.;
Hughes, A. B. Org. Lett. 2002, 4, 3767–3769.
18. Hamilton, D. J.; Sutherland, A. Tetrahedron Lett. 2004,
45, 5739–5741.
The authors gratefully acknowledge the University of
Glasgow and AstraZeneca, for financial support.
19. Jung, M. E.; Lyster, M. A. J. Chem. Soc., Chem. Commun.
1978, 315–316.
References and notes
20
20. Optical rotation for 2: ½aꢁD +21.5 (c 1.0, H2O); Lit.9
18
½aꢁD +21.0 (c 1.0, H2O). Spectroscopic data was entirely
1. Takeuchi, S.; Hirayama, K.; Ueda, K.; Sakai, H.; Yone-
consistent with that published for blastidic acid 2.9–11
hara, H. J. Antibiot. Ser. A. 1958, 11A, 1–5.