Synthesis of the C(1)-C(16) fragment of bryostatins

Article abstract of DOI:10.1016/S0040-4039(02)01054-7

A synthesis of the C(1)-C(16) fragment 43 of the bryostatins is reported which features a stereoselective equivalent of an 'ene' reaction between the allylsilane 35 and the alkynone 33 and the stereoselective conjugate addition-cyclisation of the dienyl k

Full text of DOI:10.1016/S0040-4039(02)01054-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 5491–5494
Synthesis of the C(1)ꢀC(16) fragment of bryostatins
Matthew O’Brien, Nicholas H. Taylor and Eric J. Thomas*
The Department of Chemistry, The University of Manchester, Manchester M13 9PL, UK
Received 19 April 2002; revised 29 May 2002; accepted 30 May 2002
Abstract—A synthesis of the C(1)ꢀC(16) fragment 43 of the bryostatins is reported which features a stereoselective equivalent of
an ‘ene’ reaction between the allylsilane 35 and the alkynone 33 and the stereoselective conjugate addition–cyclisation of the dienyl
ketone 36 to give the acetal 39 after acetalisation. © 2002 Elsevier Science Ltd. All rights reserved.
Me Me
The bryostatins are a group of complex natural prod-
ucts first isolated from the marine filter feeding bry-
ozoan Bugula neritina and shown to have potent
antineoplastic activity against a number of cell lines.1
Presently the bryostatins are involved in over 40 phase
I and phase II clinical trials either alone or in combina-
tion with other drugs. Many studies have been pub-
lished concerned with approaches to the synthesis of
bryostatins and several total syntheses have been com-
pleted.2–4 Nevertheless, the scarcity of natural material
and the promising biological activity of analogues
ensures that synthetic studies in this area are still of
importance.
H
O
HO
OAc
MeO2C
H
H
H
O
OH
O
16
17
OH
O
O
Me
Me
20
HO
CO2Me
Me
1
SiMe3
Me
O
Me
Me
O
Me
H
H
O
A well established strategy for the assembly of bryo-
statins involves a Julia reaction between a C(1)ꢀC(16)
aldehyde and a C(17)ꢀC(27) sulphone followed by func-
tional group modification and macrolactonisation as
indicated for bryostatin 11 1, a 20-deoxybryostatin
which has not been synthesised to date. We here report
the evolution of a synthesis of the bicyclic acetal 43
which corresponds to the C(1)ꢀC(16) fragment of the
bryostatins. In our approach the 4-methylenetetra-
hydropyran unit 3 is assembled by a stereoselective
conjugate addition–cyclisation of a hydroxyenone 2,
followed by development of the exocyclic methoxycar-
bonylmethylene group later in the synthesis.5
OH
OP
OP
2
3
these alcohols, with propargylic ketones as a route to
5-methylenealk-2-enones ready for cyclisation. How-
ever, in the presence of zinc iodide, the allylsilanes 10
and 11 were found to react with tert-butyl ethynyl
ketone to give the 5-[(Z)-trimethylsilylmethylene]-alk-2-
enones 12 and 13 with excellent stereocontrol of both
double-bonds.7 Although these reactions are formally
equivalent to ‘ene reactions, it is likely that two step
processes are involved.8,9
Preliminary studies to evaluate the proposed cyclisation
are outlined in Scheme 1. Copper-catalysed reactions of
the Grignard reagent derived from the allylsilane 5,6
which had been prepared from the dibromide 4, with
the racemic protected epoxyalcohols 6 and 7, gave the
alcohols 8 and 9 in good yield. It was intended to
examine conjugate addition reactions of the tert-
butyldimethylsilyl ethers 10 and 11, prepared from
Cyclisations of the dienes 12 and 13 were examined
under acidic conditions. Thus, treatment with aqueous
hydrogen fluoride in acetonitrile at room temperature
removed the protecting and trimethylsilyl groups and
induced cyclisation, to give mixtures of the correspond-
ing 2,6-cis- and -trans-disubstituted-4-methylenetetra-
hydropyrans 14 and 16, which were characterised and
separated as their acetates 15 and 17.10 When the
* Corresponding author.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01054-7

5492
M. O’Brien et al. / Tetrahedron Letters 43 (2002) 5491–5494
deprotection and cyclisation reactions were carried out
at room temperature overnight, the cyclisations were
reasonably stereoselective and gave the 2,6-cis- and
-trans-isomers in ratios of ca. 85:15 in favour of the
cis-isomer 14. However, when milder conditions were
used, e.g. to remove the tert-butyldimethylsilyl group
selectively from the SEM-ether 13, the stereoselectivity
of cyclisation dropped suggesting that the 85:15 mix-
tures were the result of thermodynamic control.
O
O
OH
O
RO
BnO
Me
Me
HO
Me
ii
iii
Me
Me
OBn
Me
i
18 R = H
22
20
19 R = Bn
iv
O
Me3Si
Me
Me
as
These studies indicated that the combination of the
‘ene’ reactions between the allylsilanes and the propar-
gylic ketones followed by conjugate addition–cyclisa-
tion provides a short and convergent entry to the
C(9)ꢀC(15) component of the bryostatins. The use of
this approach to prepare an intermediate for a bryo-
statin synthesis is outlined in Scheme 2.
+
OTBS
OSEM
H
Scheme 1
OSEM
O
OBn
(S)-7
(R)-11
23
v
SiMe3
Me Me
Me Me
H
(R)-Pantolactone 18 was protected as its benzyl ether
19 which was reduced to the lactol 20.11 A Wittig
reaction using phosphonium salt 21 then gave the
alkenol 22 which was converted into the alkynone 23 by
oxidation to the corresponding aldehyde, addition of
ethynylmagnesium bromide and further oxidation using
Dess–Martin periodinane. (S)-Glycidol was protected
vi
2
O
H
O
OBn
OTBS O
OSEM
OBn
OR
25 R = H
26 R = Ac
24
i
(2 mol equiv. Pr2NEt, SEMCl, dichloromethane, 20°C,
Scheme 2. Reagents and conditions: (i) BnOC(NH)CCl3,
DCM, TFA, rt, 15 h (98%); (ii) DIBAL-H, THF, −35°C, 5 h,
MeOH quench at −78°C (88%); (iii) Ph3PMeBr 21, n-BuLi,
THF, from −78°C to rt, 18 h (82%); (iv) a. (COCl)2, DMSO,
DCM, 1 h −78°C then Et3N with warming to rt; b.
HCCMgBr, THF, −78°C to rt, 18 h; c. Dess–Martin period-
O
Me3Si
ii
X
+
OH
OP
Br
OP
,
inane, DCM, 0°C to rt, 3 h (87% from 22); (v) ZnI2, 4 A MS,
4 X = Br
8 P = TIPS
6 P = TIPS
7 P = SEM
i
DCM, rt, 24 h (75%); (vi) a. 60% aq. HF, aq. MeCN, rt, 18
h; b. Ac2O, py, rt, 18 h [64% from 24; 85:15 mixture of
epimers at C(2)].
5 X = SiMe3
9 P = SEM
iii
SiMe3
Me
Me
16 h) as its trimethylsilylethoxymethyl (SEM) ether
(S)-7 which was taken through to the allylsilane (R)-11.
This allylsilane and the alkynone 23 were then coupled
using zinc(II) iodide to give the trienylketone 24 with
excellent stereoselectivity at the di- and trisubstituted
double-bonds. Deprotection–cyclisation was again
accomplished using aqueous HF in acetonitrile to give
the 2,6-cis-substituted tetrahydropyran 25 together with
15% of its epimer at C(2) which were conveniently
separated as their acetates.
Me3Si
iv
Me
OTBS
OTBS O
OP
OP
10 P = TIPS
11 P = SEM
12 P = TIPS
13 P = SEM
v
Me
Me
H
Me
Me
H
Me
Me
2
To complete a synthesis of the C(1)ꢀC(16) fragment of
a bryostatin from acetate 26 it remained to introduce
the C(1)ꢀC(5) fragment and the exocyclic methoxycar-
bonyl–methylene group, stereoselectively. However,
preliminary investigations indicated that epoxidation,
hydroboration and hydroxylation of the acetate 26 all
took place regioselectively on the exocyclic methylene
group. Moreover, the ketone carbonyl group compli-
cated hydroboration of the terminal double-bond.12
For these reasons it was decided to assemble the
C(1)ꢀC(9) fragment before the ‘ene’ and cyclisation
reactions.
2
H
H
O
O
+
O
O
85 : 15
OR
OR
14 R = H
16 R = H
15 R = Ac
17 R = Ac
Scheme 1. Reagents and conditions: (i) Cl3SiH, Et3N, cat.
CuCl, Et2O, 0°C to rt, 15 h, then 3 M MeMgBr in Et2O, 0°C,
15 h (70%); (ii) Mg, THF, reflux, 0.5 h then add to CuI and
6 or 7 in THF, −10°C to rt, 1 h (8, 95%; 9, 88%); (iii)
TBSOTf, 2,6-lutidine, dichloromethane (DCM), rt, 18 h (77–
90%); (iv) HCC·CO·CMe3, ZnI2, DCM, protected from light,
rt, 24 h (12, 52%; 13, 62%); (v) a. 60% aq. HF, MeCN, 18 h,
rt; b. Ac2O, py, DCM (72% from 12/13).
To this end, the alcohol 22 was protected as its tert-
butyldimethylsilyl ether which was hydroborated to

M. O’Brien et al. / Tetrahedron Letters 43 (2002) 5491–5494
5493
give the alcohol 27, see Scheme 3. Oxidation gave the
aldehyde 28 which, following a literature precedent,13
was subjected to an aldol condensation with the lithium
enolate of 4-benzyloxybutan-2-one 34 to give the anti-
adduct 29,14 the product of chelation control, with
excellent stereoselectivity. The configuration of adduct
29 at C(5) was assigned by analogy with the literature13
and was confirmed by comparison of the 1H NMR
spectra of its (R)- and (S)-acetylmandelates. Reduction
of hydroxyketone 29 with tetramethylammonium tri-
acetoxyborohydride gave a good yield of the anti-diol
30 which was protected as its acetonide 31, the relative
chemical shifts of the acetonide methyl groups confirm-
ing the assigned 3,5-anti-stereochemistry. Selective
deprotection followed by oxidation then gave the alde-
hyde 32 which, by addition of ethynylmagnesium bro-
mide and oxidation, gave the alkynone 33.
see Scheme 4. This was deprotected and cyclised using
aqueous HF in acetonitrile at room temperature to give
the triacetate 38, after treatment with an excess of
acetic anhydride, so confirming participation of the
intermediate triol 37.
Attempts were now made to trap the initially formed
trihydroxyketone 37 as an acetal. Indeed treatment of
the crude product from the cyclisation–deprotection
with trimethyl orthoformate in the presence of an acid
gave the acetal 39. This was further protected as its
bis-silyl ether 40.
It now remained to introduce the exocyclic methoxy-
carbonylmethylene group and this was achieved as out-
lined in Scheme 5. Hydroxylation of 40 and cleavage of
the diol 41 using sodium periodate gave the ketone 42.
This was condensed with the chiral phosphine oxide 44,
a reagent used previously in bryostatin synthesis,3 to
give the required ab-unsaturated ester 43 together with
its geometrical isomer, ratio 72:28, in favour of 43.
The zinc iodide-promoted reaction of the alkynone 33
with the (R)-allyl silane 35 prepared from (S)-glycidol
gave the dienylketone 36 with excellent stereocontrol,
Me
Me
Me
Me
Me
Me
Me
Me
ii
iv
i
5
OH
iii
CHO
22
TBSO
OBn
TBSO
OBn
TBSO
OBn OH
O
OBn
TBSO
OBn OH
OH
OBn
27
28
30
29
v
Me
Me
Me
OHC
Me
Me
Me
5
Me
vi
3
vii
O
OBn
O
OBn
O
O
OBn
OBn
O
O
OBn
TBSO
OBn
O
O
OBn
Me Me
Me Me
Me Me
31
33
34
32
Scheme 3. Reagents and conditions: (i) a. TBSCl, imid., DMAP (cat.), TBAI (cat.), rt, 30 min (97%), b. 1 M BH3 in THF, −18°C
to rt, 18 h, then NaOH, 30% aq. H2O2, 50°C, 3 h (67%); (ii) (COCl)2, DMSO, DCM, −78°C, 20 min then Et3N, −78°C to rt; (iii)
34, LDA, −78°C, 30 min, then add 28, −78°C, 2 min, MeOH quench (49%); (iv) Me4NBH(OAc)3, AcOH, MeCN (1:1), −20°C,
18 h (95%); (v) Me2C(OMe)2, PPTS (cat.), rt, 18 h (74%); (vi) a. TBAF, THF, rt, 18 h (95%), b. (COCl)2, DMSO, −78°C, 20 min,
then Et3N, −78°C to rt (ca. 100%); (vii) a. HCCMgBr, THF, −78°C to rt (ca. 100%), b. Dess–Martin periodinane, DCM, rt 4 h
(99%).
Me Me
SiMe3
Me Me
H
i
ii
Me3Si
H
O
O
OBn OH
OH
OBn
OTBS
OTBS
OTBS O
OBn
O
O
OBn
Me Me
36
OH
OTBS
35
37
iii
Me Me
Me Me
MeO
H
MeO
H
O
Me Me
OBn
H
OBn
iv
H
H
H
H
O
O
H
O
O
O
OBn OAc OAc OBn
OTBS
OH
OAc
OTBS OBn
OH
OBn
40
39
38
,
Scheme 4. Reagents and conditions: (i) 33, DCM, ZnI2, 4 A MS, protect from light, rt, 18 h (55%); (ii) a. 56% aq. HF, MeCN,
rt, 18 h, b. Ac2O, py, DMAP (cat.), rt, 18 h (65% from 36); (iii) a. 56% aq. HF, MeCN, rt 24 h, b. HC(OMe)3, MeOH, PPTS
(cat.), rt, 3 h (34% from 36); (iv) TBSCl, imid., TBAI, DMAP, DCM, rt (ca. 100%).

5494
M. O’Brien et al. / Tetrahedron Letters 43 (2002) 5491–5494
OH
Me Me
MeO
Me Me
Me Me
MeO
OH
H
O
H
O
H
O
MeO
OBn
O
OBn
OBn
MeO2C
H
H
H
H
H
H
ii
i
iii
O
O
O
40
OTBS
OTBS
OTBS
OTBS OBn
OTBS OBn
OTBS OBn
43
42
41
Scheme 5. Reagents and conditions: (i) Na2CO3, OsO4 (cat.), NMO, acetone–tBuOH–H2O, rt, 18 h (55%); (ii) NaIO4, Na2CO3,
MeOH–THF–H2O, rt 30 min (69%); (iii) 44, NaHMDS, −78°C, 30 min, then add 42, −50°C, 18 h, then −15°C, 18 h (49%; 43 plus
geometrical isomer 72:28).
This synthesis of the ester 43 completes a synthesis of
the C(1)ꢀC(16) fragment of the bryostatins and features
the formal ‘ene’ reaction between the allylsilane 35 and
the alkynone 33 together with the stereoselective conju-
gate addition–cyclisation of 36. Further work is in
progress to incorporate the exocyclic trisubstituted dou-
ble-bond earlier in the synthesis, to replace the benzyl
with more labile protecting groups, and to attempt to
complete a synthesis of a natural bryostatin.
7. The geometries of all the allylsilanes were confirmed by
NOE studies.
8. (a) Audran, G.; Monti, H.; Leandri, G.; Monti, J.-P.
Tetrahedron Lett. 1993, 34, 3417–3418; (b) Monti, H.;
Audran, G.; Leandri, G.; Monti, J.-P. Tetrahedron Lett.
1994, 35, 3073–3076.
9. Monti, H.; Audran, G.; Feraud, M.; Monti, J.-P.; Leandri,
G. Tetrahedron 1996, 52, 6685–6698.
10. The cis- and trans-configurations of 15 and 17 were
established by NOE studies.
11. The benzyl ether was chosen because of its stability to the
acidic conditions used in the thermodynamic cyclisation
reactions. Partial deprotection was observed during the
cyclisation if, for example, p-methoxybenzyl protection
was used.
CO2Me
P
O
12. The acetate 26 has been converted into the fully protected
polyol i which also corresponds to the C(1)ꢀC(17) fragment
of a bryostatin, albeit lacking the exocyclic methoxycar-
bonylmethylene group, by cleavage of the exocyclic meth-
ylene group, protection of the ketone so formed and
reduction/protection of the 2%-ketone, followed by addition
of the C(1)ꢀC(4) fragment via a hydroboration/oxidation,
aldol condensation, reduction and acetalisation sequence.
44
Acknowledgements
We thank the EPSRC for studentships (to M.O’B. and
N.D.T.) and Duncan Gill for early experiments on the
cyclisation.
References
1. Mutter, R.; Wills, M. Bioorg. Med. Chem. 2000, 8, 1841–
1860.
2. (a) Kageyama, M.; Tamura, T.; Nantz, M. H.; Roberts,
J. C.; Somfrai, P.; Whritenour, D. C.; Masamune, S. J. Am.
Chem. Soc. 1990, 112, 7407–7408; (b) Ohmori, K.; Ogawa,
Y.; Obitsu, T.; Ishikawa, Y.; Nishiyama, S.; Yamamura,
S. Angew. Chem., Int. Ed. 2000, 39, 2290–2294.
3. Evans, D. A.; Carter, P. H.; Carreira, E. M.; Charette, A.
B.; Prunet, J. A.; Lautens, M. J. Am. Chem. Soc. 1999, 121,
7540–7552.
4. (a) Hale, K. J.; Frigerio, M.; Manaviazar, S. Org. Lett.
2001, 3, 3791–3794; (b) Vakalopoulos, A.; Lampe, T. F.
J.; Hoffmann, H. M. R. Org. Lett. 2001, 3, 929–932; (c)
Almendros, P.; Rae, A.; Thomas, E. J. Tetrahedron Lett.
2000, 41, 9565–9568.
13. De Brabander, J.; Vanderwalle, M. Synthesis 1994, 855–
865.
14. This aldol reaction of ketone 34 was complicated by
competing deprotonation of the ketone at the a-methylene
group leading to the formation of regioisomers as minor
products (ca. 20%). This could be avoided by using the
analogous 4-tert-butyldiphenylsilyloxybutan-2-one which
gave a better, 76%, yield of aldol product, but this led to
the formation of the tetra-acetate ii on cyclisation followed
by acetylation since the tert-butyldiphenylsilyloxy group
was not stable under the acidic cyclisation conditions. The
chemistry of ii has not been studied further at present.
Me Me
H
5. A related conjugate addition has recently been used in a
related approach to this fragment of bryostatins: Yadav,
J. S.; Bandyopadhyay, A.; Kunwar, A. C. Tetrahedron
Lett. 2001, 42, 4907–4911.
H
O
O
OBn OAc OAc OAc
6. Trost, B. A.; Grese, T. A.; Chan, D. M. T. J. Am. Chem.
ii
OAc
Soc. 1991, 113, 7350–7362.