Asymmetric total synthesis of antitumour antibiotic A83586C

Article abstract of DOI:10.1039/a706206k

The first asymmetric total synthesis of antitumour antibiotic A83586C has been accomplished.

Full text of DOI:10.1039/a706206k

View Article Online / Journal Homepage / Table of Contents for this issue
Asymmetric total synthesis of antitumour antibiotic A83586C
Karl J. Hale* and Jiaqiang Cai
Department of Chemistry, University College London, 20 Gordon Street, London, UK WC1H 0AJ
The first asymmetric total synthesis of antitumour antibiotic
A83586C has been accomplished.
group was cleaved from 10 with TFA, and the acyl hydrazine
oxidised with N-bromosuccinimide (NBS) in aqueous THF.8
Without further purification, the crude acid was esterified with
Ph2CN24 and the Fmoc group excised to deliver 11.
A83586C 1 is a highly ornate hexadepsipeptide isolated from
the fermentation broths of Streptomyces karnatakensis by
Smitka et al.1 It shows pronounced antimicrobial activity in
vitro against Gram-positive strains of bacteria, and can combat
the growth of a CCRF-CEM human T-cell leukaemia line
The synthesis of 16 was achieved from (2S,3S)-hydroxyleu-
cine.9 After Troc protection of the amino group,6 a selective
O-allylation was performed with NaHCO3 and allyl bromide in
DMF;10 13 was isolated in 92% yield. A range of coupling
conditions were examined for establishing the O(19)–C(1) ester
bond between 13 and 14.11 Success finally came using DCC and
DMAP in CH2Cl2, but 5–10% epimerisation was always
encountered at the a-carbon atom of the threonine unit.
Notwithstanding this, 15 was usually prepared in 83–88% yield.
The next two steps involved removal of the allyl ester group
from 15 using palladium(0) catalysis,12 and acid chloride
formation with excess oxalyl chloride in benzene.
To our delight, [4 + 2]-fragment condensation between 11
and 16 was complete after 2 min when conducted in benzene at
60 °C in the presence of AgCN;6 17 was formed in yields of
73–86%.
Our next goal was to convert 17 into 18 and then
macrolactamise at the (3S)-piperazic acid and d-threonine
residues. The Troc group in 17 was substituted with a Z group
by reacting 17 with excess zinc in acetic acid,6 and acylating the
crude amine with BnO2CCl. TFA was then used to simul-
taneously deprotect the Boc and Ph2CH ester groups to give 18
in 78% overall yield for the three steps.
(IC50
=
0.0135 mg ml21). From a synthetic standpoint,
A83586C 1 presents a formidable challenge owing to its
ominous expanse of diverse functionality, which looks set to
overwhelm most protecting group endgames conceived for its
total synthesis. In light of this, we thought it prudent to dispense
with protecting groups for the final stages, and some time ago,
we synthesised the C(1)–C(47) sequence 2 via a chemoselective
union between 3 and 4 (Scheme 1).2 Unfortunately, difficulties
later arose not only in the macrolactonisation of 2, but also its
seco-acid. This led us to consider new ways of completing this
synthetic venture. One plan that looked particularly worthwhile
featured a coupling between 6 and 3 to obtain 5 and a
subsequent glycal hydration at C(30) to produce 1. We now
report the first total synthesis of A83586C based on this
strategy.
Initial attention focused on constructing 11 and 16 (Scheme
2). Tetrapeptide 11 was assembled in eight steps from known
dipeptide 7.3 Our first manoeuvres replaced the Ph2CH ester4
group in 7 with a Boc-NHNH group,5 to prevent diketopiper-
azine formation occurring when the Fmoc group was removed
with Et2NH. Fortunately, this tactic worked very well, 8 being
isolated uneventfully after Fmoc deprotection. Compound 8
coupled readily to acid 93,6 in CH2Cl2 at 0 °C after sequential
treatment with Et3N and bis(2-oxo-3-oxazolidinyl)phosphinic
chloride (BOP-Cl);7 10 was obtained in 75% yield. The Boc
A significant number of peptide coupling reagents were
evaluated for instigating the desired macrolactamisation of 18 to
19, the majority of which performed poorly. Success was
eventually attained with O-(7-azabenzotriazol-1-yl)-N,N,NA,N-
tetramethyluronium hexafluorophosphate (HATU)13 and
N-ethylmorpholine (NEM) in CH2Cl2 under conditions of very
44
36
43
O
O
O
33
40
30
41
HO
HO
HO
H
O
O
O
38
HO
O
H
H
HO
O
H
H
NH N
45
46
47
N
BtO
NH
O
28
19
13
HO
N
3
HO
N
N
N
17
18
23
i
12
O
26
+
O
O
O
O
1
O
OH
10
–
Cl
+
CO2Me
2
O
N
O
N
27
8
NH3
NH
21
O
N
O
N
HO
N
HN
22
N
HN
20
A83586C
1
3
OH
OH
O
HN
O
HN
O
OH
O
4
CO2Me
O
N
2
O
O
N
HN
HO
O
H
OH
O
HN
NH
NH2
H
N
HO
N
N
19
O
NH
18
HO
N
4
N
O
O
O
O
O
O
O
O
O
O
N
HO
O
N
+
O
O
N
HN
H
O
N
HN
OH
BtO
O
HN
O
3
OH
HN
O
6
5
Scheme 1 Reagents and conditions: i, Et3N, CH2Cl2, 278 to 25 °C, SiO2, 65%
Chem. Commun., 1997
2319

View Article Online
OBn
N
NZ
H
NZ
NZ
OBn
N
OBn
N
N
N
N
Troc
Fmoc
H
Fmoc
O
O
H
Fmoc
O
N
N
NZ NH
O
O
v–viii
i–iii
9
O
OBn
N
O
N
O
O
N
O
N
O
iv
N
O
O
N
O
O
O
O
ZN
NHNHBoc
O
O
ZN
OCHPh2
N
O
N
N
ZN
OCHPh2
ZN
NHNHBoc
O
OBn
BocHN
O
7
8
11
10
N
CO2H
ZN
OCHPh2
xiv
Troc
Troc
NH
NH
BocHN
Troc
17
NH
OBn
AllO
Cl
NH2
14
xi
xii, xiii
ix, x
AllO
HO
O
O
O
O
O
O
xv–xvii
O
OH
O
OH
13
12
Z
OBn
BocHN
OBn
BocHN
NZ NH
15
16
OBn
–
N
Cl
Z
Z
+
N
NH3
NH
NH
N
O
O
O
O
HO
N
N
BnO
N
N
O
N
O
O
O
+
xx, xxi
O
O
O
O
O
O
xviii
xix
OBn
–
NH3
1
O
O
N
O
N
CF3CO2
N
O
N
O
N
ZN
OH
HN
HN
OH
OBn
18
HN
O
ZN
O
19
20
Scheme 2 Reagents and conditions: i, TFA–CH2Cl2 (10:1), PhOH (2 equiv.), 25 °C; ii, BocNHNH2 (2 equiv.), DCC (2 equiv.), 1-hydroxybenzotriazole
hydrate (HOBT) (2 equiv), THF (ca. 0.33 m), 0 to 25 °C, 93% (2 steps); iii, Et2NH–MeCN (1:1) (ca. 0.55 m); iv, CH2Cl2 (ca. 0.15 m), add Et3N (2.3 equiv.)
at 210 °C, stir 5 min, then add BOP-Cl (1.2 equiv.), warm to 0 °C, 4 h, 75% (2 steps); v, TFA–CH2Cl2 (1:1) (ca. 0.2 m), 0 °C; vi, NBS (2 equiv.), THF–H2O
(1:1) (ca. 0.04 m), 0 to 25 °C; vii, Ph2CN2 (1.9 equiv.), Me2CO (ca. 0.16 m), 25 °C, 12 h, 68% (3 steps); viii, Et2NH–MeCN (1:1) (ca. 0.24 m), 25 °C, 40
min; ix, Troc-Cl (1.05 equiv.), NaOH (2 equiv.), THF, H2O, 25 °C, 90%; x, NaHCO3 (4.3 equiv.), allyl bromide (6.2 equiv.), DMF (ca. 0.31 m), 25 °C, 92%;
xi, DCC (1.2 equiv.), DMAP (1.25 equiv.), CH2Cl2 (ca. 0.322 m), 0 to 25 °C, 83–88%; xii, morpholine (6.5 equiv.), (Ph3P)4Pd (5.9 mol%), THF (ca. 0.22
m), 0 to 25 °C, 59%; xiii, (COCl)2 (20 equiv.), C6H6 (ca. 0.4 m), 25 °C, 2.5 h; xiv, AgCN (1.3 equiv.), C6H6 (ca. 0.12 m), 60 °C for 2 min, 73–86% (for viii,
xiii, xiv); xv, Zn (70 equiv.), AcOH–H2O (10:1) (ca. 0.0685 m), 25 °C, 3 h; xvi, ZCl (3 equiv.), 10% aq. NaHCO3, CH2Cl2, 25 °C, 2 h, 78% (2 steps); xvii,
TFA–CH2Cl2 (2:1) (ca. 0.06 m), PhOH (2.2 equiv.), 0 °C, 1 h, 100%; xviii, HATU (10 equiv.), CH2Cl2 (500 ml), 0 °C, then slow addition of 18 (1.4 g) and
NEM (13.5 equiv.) in CH2Cl2 (500 ml) over 3 h, then 0 °C, 2 h, then 25 °C, 30 h, 25%; xix, 10% Pd–C ( = weight of 19 used), MeOH (ca. 0.01 m), HCl
(1 equiv.), H2 (1 atm.), 25 °C, 24 h; xx, 3 (1 equiv.), CH2Cl2 (ca. 0.03 m), 278 °C, add Et3N (9.3 equiv.), warm to 25 °C, stir 10 min, 31% (from 19); xxi,
CDCl3 (wet), 72 h, 0 °C, 100%
† All new compounds reported in this synthetic route gave satisfactory 400
high dilution. This produced 19 in a reproducible 25% yield
after flash chromatography. Compound 19 was then deprotected
by catalytic hydrogenolysis over palladium on carbon. We
MHz 1H and 100 MHz 13C NMR and IR spectra, as well as HRMS and/or
microanalyses within 0.4%.
deliberately conducted this hydrogenation in methanolic HCl,
so as to minimise the occurrence of O(19) to N(18) acyl
rearrangement during this lengthy reaction. Without further
purification, crude 20 was suspended in CH2Cl2 along with 3.2
After cooling to 278 °C, Et3N was added, and the reaction
warmed to room temperature where it was stirred for 10 min. A
rapid coupling ensued to deliver 5 in 31% yield after
chromatographic purification. Upon standing in undried CDCl3
(Aldrich) for 72 h, 5 readily hydrated to give A83586C in
1 T. A. Smitka, J. B. Deeter, A. H. Hunt, F. P. Mertz, R. M. Ellis,
L. D. Boeck and R. C. Yao, J. Antibiot., 1988, 41, 726.
2 K. J. Hale, J. Cai and V. M. Delisser, Tetrahedron Lett., 1996, 37,
9345.
3 K. J. Hale, V. M. Delisser, L.-K. Yeh, S. A. Peak, S. Manaviazar and
G. S. Bhatia, Tetrahedron Lett., 1994, 35, 7685.
4 G. C. Stelakatos, A. Paganou and L. Zervas, J. Chem. Soc. C, 1966,
1191.
5 R. A. Boissonnas, St. Guttmann and P.-A. Jaquenoud, Helv. Chim. Acta,
1960, 43, 1349.
6 P. L. Durette, F. Baker, P. L. Barker, J. Boger, S. S. Bondy,
M. L. Hammond, T. J. Lanza, A. A. Pessolano and C. G. Caldwell,
Tetrahedron Lett., 1990, 31, 1273.
7 R. D. Tung and D. H. Rich, J. Am. Chem. Soc., 1985, 107, 4342.
8 H. T. Cheung and E. R. Blout, J. Org. Chem., 1965, 30, 315.
9 K. J. Hale, S. Manaviazar and V. M. Delisser, Tetrahedron, 1994, 50,
9181.
10 V. Bocchi, G. Casnati, A. Dossena and R. Marchelli, Synthesis, 1979,
961.
11 Compound 14 was prepared by the same method described for its
l-enantiomer by: Y. Nakamura, M. Hirai, K. Tamotus, Y. Yonezawa
and C. Shin, Bull. Chem. Soc. Jpn., 1995, 68, 1369.
12 H. Kunz and H. Waldmann, Angew. Chem., Int. Ed. Engl., 1984, 23,
71.
1
quantitative yield. The H and 13C NMR spectra of synthetic
A83586C in CDCl3 were identical in every respect with the
spectra of authentic A83586C. In addition, synthetic A83586C
also gave rise to an (M + Na)+ ion at m/z 999.5360 in its high
·
resolution FAB mass spectrum;† the calculated value for
C47H76N8O14Na is 999.5379.
We thank the EPSRC (Project Grant GR/J92590), Zeneca,
Pfizer and Rhone-Poulenc Rorer for financial support. We
thank Dr Tim Smitka of Eli Lilly for kindly supplying us with
1
the original H and 13C NMR spectra of natural 1. We are
indebted to Dr Glyn Williams of Roche Products for obtaining
1
the 500 MHz H NMR spectrum of synthetic 1, and to his
colleague Dr B. K. Handa for urging us to try the HATU
reagent. We thank the ULIRS MS Service for HRMS
measurements.
13 L. A. Carpino, J. Am. Chem. Soc., 1993, 115, 4937.
Footnotes and References
* E-mail: k.j.hale@ucl.ac.uk
Received in Glasgow, UK, 26th August 1997; 7/06206K
2320
Chem. Commun., 1997