Synthetic studies on the bryostatins: Preparation of a truncated BC-ring intermediate by pyran annulation

Article abstract of DOI:10.1021/ol050511w

(Chemical Equation Presented) A synthesis of a potential BC-ring subunit (C₉-C₂₇) for bryostatin 1, a remarkably potent anticancer agent, has been developed in 16 steps and 18% overall yield. The key features of this route include a BITIP-catalyzed asymmetric allylation reaction, chelation-controlled allylations, a hydroformylation reaction, and a pyran annulation reaction.

Full text of DOI:10.1021/ol050511w

ORGANIC
LETTERS
2005
Vol. 7, No. 11
2149-2152
Synthetic Studies on the Bryostatins:
Preparation of a Truncated BC-Ring
Intermediate by Pyran Annulation
Gary E. Keck* and Anh P. Truong
Department of Chemistry, UniVersity of Utah, 315 South 1400 East RM 2020,
Salt Lake City, Utah 84112-0850
keck@chemistry.utah.edu
Received March 9, 2005
ABSTRACT
A synthesis of a potential BC-ring subunit (C9−C27) for bryostatin 1, a remarkably potent anticancer agent, has been developed in 16 steps and
18% overall yield. The key features of this route include a BITIP-catalyzed asymmetric allylation reaction, chelation-controlled allylations, a
hydroformylation reaction, and a pyran annulation reaction.
The bryostatins were originally isolated by Pettit and co-
workers in 1968 and fully characterized in 1982.1 Their
unique biological effects have led to the entry of bryostatin
1 into human clinical trials as a single-agent cancer chemo-
therapeutic, as well as in studies of combination therapies.2
Due to the impressive biological profile of this agent coupled
with its remarkable molecular structure, bryostatin 1 has been
an attractive target to the synthetic, biological, and medical
communities.3
Since the initial elucidation of the bryostatin structures,
three successful syntheses have been accomplished by
Masamune, Evans, and Yamamura, which provide important
precedent for further synthetic efforts.4,5 For example, both
Evans and Yamamura have demonstrated that the exocyclic
enoate on the B-ring can be installed with a high degree of
stereoselectivity by means of a Horner-Emmons reaction
using a BINOL-based reagent developed by Fuji.6 Moreover,
in an important series of papers, Wender and co-workers
have reported the first simplified bryostatin analogues that
retain activity against human cancer cell lines.7 The scarcity
of bryostatin 1 from natural sources and difficulties in the
isolation of these materials impact negatively upon further
clinical trials and detailed studies of its mode of action. This
situation has prompted us to set up a program aimed at the
development of a convergent and efficient synthesis of
bryostatin 1 and structurally simplified bryostatin analogues,
which hopefully will exhibit biological activity similar to
(5) For additional recent synthetic work in this area, see: (a) Voight, E.
A.; Seradj, H.; Roethle, P. A.; Burke, S. D. Org. Lett. 2004, 6, 4045. (b)
Voight, E. A.; Roethle, P. A.; Burke, S. D. J. Org. Chem. 2004, 69, 4534.
(c) Hale, K. J.; Frigerio, M.; Hummersone, M. G.; Manaviazar, S. Org.
Lett. 2003, 5, 503. (d) Hale, K. J.; Frigerio, M.; Manaviazar, S.;
Hummersone, M. G.; Fillingham, I. J.; Barsukov, I. G.; Damblon, C. F.;
Gescher, A.; Roberts, G. C. K. Org. Lett. 2003, 5, 499. (e) Ball, M.; Baron,
A.; Bradshaw, B.; Omori, H.; MacCormick, S.; Thomas, E. J. Tetrahedron
Lett. 2004, 45, 8747.
(1) Pettit, G. R.; Herald, C. L.; Doubek, D. L.; Arnold, E.; Clardy, J. J.
Am. Chem. Soc. 1982, 104, 6846.
(2) Mutter, R.; Wills, M. Bioorg. Med. Chem. 2000, 8, 1841.
(3) Hale, K. J.; Hummersone, M. G.; Manaviazar, S.; Frigerio, M. Nat.
Prod. Rep 2002, 19, 413 and references therein.
(4) (a) Evans, D. A.; Carter, P. H.; Charette, A. B.; Prunet, J. A.; Lautens,
M. J. Am. Chem. Soc. 1999, 121, 7540. (b) Ohmori, K.; Ogawa, Y.; Obitsu,
T.; Ishikawa, Y.; Nishiyama, S.; Yamamura, S. Angew. Chem., Int. Ed.
2000, 39, 2290. (c) Kageyama, M.; Tamura, T.; Nantz, M. H.; Roberts, J.
C.; Somfai, P.; Whritenour, D. C.; Masamune, S. J. Am. Chem. Soc. 1990,
112, 7407.
(6) (a) Tanaka, K.; Ohta, Y.; Fuji, K.; Taga, T. Tetrahedron Lett. 1993,
34, 4071. (b) Tanaka, K.; Otsubo, K.; Fuji, K. Tetrahedron Lett. 1996, 37,
3735.
(7) Wender, P. A.; Baryza, J. L.; Bennett, C. E.; Bi, F. C.; Brenner, S.
E.; Clarke, M. O.; Horan, J. C.; Kan, C.; Lacote, E.; Lippa, B.; Nell, P. G.;
Turner, T. M. J. Am. Chem. Soc. 2002, 124, 13648.
10.1021/ol050511w CCC: $30.25
© 2005 American Chemical Society
Published on Web 04/28/2005

that of bryostatin 1. Toward the goal of developing a practical
synthetic route to these materials, we describe herein a
concise and efficient synthesis of the BC-ring portion
(C9-C27) of bryostatin 1.
Scheme 2. Synthesis of the C17-C27 Fragment
Our initial retrosynthetic analysis of bryostatin 1 is shown
in Scheme 1. Disconnections of the macrolactone ester
Scheme 1. Initial Retrosynthetic Analysis of Bryostatin 1
BOM-protected ester afforded aldehyde 5. Chelation-
controlled allylation by reaction with allyltri-n-butylstannane
and MgBr2‚Et2O provided the C25 alcohol as a single
diastereomer. After protection of the resulting alcohol as a
PMB ether using PMBBr and KH in THF, oxidative cleavage
of the olefin using ozonolysis in CH2Cl2/MeOH (4:1) at -78
°C in the presence of NaHCO3 (to prevent the formation of
the dimethylacetal) furnished aldehyde 6 in good yield. A
second allylstannane addition to aldehyde 6, again with
MgBr2‚Et2O as the Lewis acid, gave rise to a new stereo-
center at C21 with essentially complete selectivity (>95:5).8
The hydroxyl group was then protected as a TBS ether using
TBSCl and imidazole in DMF to give alkene 7 in 97% yield.
Initial attempts at hydroformylation of the terminal alkene
using dicarbonylacetylacetonato rhodium(I) and P(OPh)3 in
toluene resulted in decomposition. However, it was found
that when the reaction conditions and ligand developed by
Buchwald were employed, hydroformylation of alkene 7
proceeded smoothly to give an 11:1 (linear:branched)
mixture, which could be separated to give the desired
aldehyde 8 in 92% yield.9 Treatment of aldehyde 8 with a
prenyl indium reagent generated in situ in DMF at ambient
temperature afforded the desired alcohol as a mixture of
linkage and the C16-C17 trans olefinic bond provide an
AB-ring fragment and a C-ring fragment. It was initially
envisioned that the olefinic unit might be constructed via
olefin metathesis, with the Julia coupling utilized in all other
approaches to date available as a backup alternative from
the same advanced intermediates. Given the apparent im-
portance of the C-ring to the biological activity of bryostatin,
we sought a strategy that would allow for systematic
modifications in this region to provide for efficient access
to new C-ring analogues. Moreover, for our purposes,
differential protection of the C26 and C27 hydroxyl groups
was essential. Since this requirement adds additional steps
over approaches where both hydroxyls of this diol can be
masked simultaneously, for example, as an acetonide, a high
degree of efficiency in the construction of this segment was
necessary.
Synthesis of the C17-C27 Fragment. We anticipated that
the C-ring exocyclic enoate could be incorporated from a
ketone via an aldol condensation with methyl glyoxylate, as
previously documented in the Wender and Evans studies.
Due to the potential instability of the enoate at C21, we
decided to install this R,â-unsaturated subunit at a late stage
in the synthesis.
(8) (a) Keck, G. E.; Castellino, S.; Wiley, M. R. J. Org. Chem. 1986, 5,
5478. (b) Evans, D. A.; Dart, M. J.; Duffy, J. L.; Yang, M. G.; Livingston,
A. B. J. Am. Chem. Soc. 1995, 117, 6619. (c) Evans, D. A.; Dart, M. J.;
Duffy, J. L.; Yang, M. G. J. Am. Chem. Soc. 1996, 118, 4322.
Our synthesis (Scheme 2) commenced with BOM protec-
tion of the hydroxyl group of (R)-isobutyl lactate (4) using
BOMCl and Hu¨nig’s base. DIBAL-H half-reduction of the
(9) Cuny, G. D.; Buchwald, S. L. J. Am. Soc. Chem. 1993, 115, 2066.
2150
Org. Lett., Vol. 7, No. 11, 2005

diastereomers in excellent yield (92%). This hindered alcohol
was then oxidized with a combination of PCC, 4 Å MS, and
NaOAc in CH2Cl2 to afford ketone 9 (94%). Multigram
quantities of analytically pure ketone 9 (>20 g) have been
prepared using this reaction sequence in 57% overall yield
from the starting aldehyde 5. Thus, this overall sequence is
remarkably efficient.
Scheme 4. Attempted Homologation
With ketone 9 in hand, we set out to close the C-ring.
Removal of the TBS ether using HF‚py or TBAF in THF at
room temperature provided the corresponding hydroxy
ketone in good yield (87%). This hydroxy ketone was then
cyclized with concomitant dehydration under acidic condi-
tions (CSA, PhH, azeotropic removal of water at 85 °C) to
give glycal 10 in 90% yield. Epoxidation of this glycal with
MMPP proceeded with methanolysis to give an intermediate
methoxyketal-alcohol that was immediately oxidized with
TPAP to furnish ketone 11 (67% over two steps). It is at
this stage that the exocylic ester at C21 can be installed using
a known three-step protocol that was reported by Evans and
co-workers or a one-step procedure that was published by
both Wender and Hale.3,4,10,11
We have recently reported the development of methodol-
ogy that allows for a very concise asymmetric construction
of 4-methylene dihydropyrans, which we termed the pyran
annulation reaction.12 This process was developed specifically
to provide access to the B-ring of the bryostatins. To
investigate the viability of such a strategy, in the present
case, we would require access to a C-ring enal such as 14.
This could then potentially be employed in a pyran annulation
reaction with a â-hydroxy allylsilane to incorporate a
functionalized B-ring subunit in just one additional linear
step.
hindered aldehyde. Thus, we were led to explore a different
approach that elaborated a less sterically encumbered alde-
hyde.
We envisioned that the R,â-unsaturated aldehyde could
come from an R,â-unsaturated thiol ester (Scheme 5). Thus,
Scheme 5. Retrosynthetic Analysis of the C-Ring Enal
We initially attempted to access this enal intermediate via
homologation of the aldehyde that arose from an oxidative
cleavage of the alkene moiety in 11 (Scheme 4).
Scheme 3. Potential New Approach
we decided to reorder two independent operations: instead
of forming the C-ring first and then extending the chain at
C17, we decided to elaborate at C17 prior to closing the C-ring.
This timing of the sequence would permit the homologation
to be performed on a less sterically hindered substrate. In
addition, this approach would provide for a more convergent
route to the R,â-unsaturated aldehyde utilizing intermediates
that were already in hand.
The investigation of this alternative approach began with
the ozonolysis of alkene 9 (Scheme 6). The resulting alde-
hyde 23 was immediately subjected to a Horner-Emmons
reaction to give the desired R,â-unsaturated thiol ester 20
with both excellent stereoselectivity and yield (92%). Depro-
tection of the TBS ether using HF/py buffer was followed
by cyclization to give the desired glycal 19 in 89% yield
over two steps. Half-reduction of the thiol ester with
DIBAL-H at -78 °C then provided the desired R,â-un-
saturated aldehyde 14 in 84% yield. Utilizing the sequence
Conditions similar to those that were employed by Wender
and co-workers did not afford the 2-carbon homologated
product.5 As shown in Scheme 4, even the vinyl zincate used
successfully by the Wender group failed to add into this
(10) Wender, P. A.; De Brabander, J.; Harran, P. G.; Jimenez, J.-M.;
Koehler, M. F. T.; Lippa, B.; Park, C.-M.; Shiozaki, M. J. Am. Chem. Soc.
1998, 120, 4534.
(11) For a different route to a similar intermediate (with a different
protecting group at C26), see: Keck, G. E.; Yu, T.; McLaws, M. J. Org.
Chem. 2005, 70, 2543.
(12) Keck, G. E.; Covel, J. A.; Schiff, T.; Yu, T. Org. Lett. 2002, 4,
1189.
Org. Lett., Vol. 7, No. 11, 2005
2151

Several aspects of the annulation reaction were potentially
of concern. First of all, we had not examined any cases of
this reaction using R,â-unsaturated aldehydes. Although we
had no reason to suspect a problem with this sort of substrate,
the compatibility of such unsaturation with the pyran
annulation process was not known. Of more concern was
the presence of the highly reactive glycal moiety in the
substrate, as well as the PMB and BOM protecting groups,
which are also reasonably labile under acidic conditions. In
the event, reaction of enal 14 with the hydroxy allylsilane
13 in ether at -78 °C, under the influence of TMS triflate
as previously described, was found to afford the desired
bicyclic product 24 in 66% isolated yield. Although this is
a lower yield than typically obtained for this process with
less complex substrates, it is reasonable given the nature of
the C-ring glycal substrate.13
Scheme 6. Completion of the C-Ring Enal (C15-C27)
It should be noted that this annulation reaction is very
reminiscent of the Diels-Alder reaction in terms of the
structural changes accomplished. Thus, in structural terms,
one can consider that the pyran is constructed by a four-
carbon annulation across the π-bond of an aldehyde, leading
to incorporation of the B-ring pyran in a single step.14
In summary, a short (15 steps), efficient (27% overall
yield), and practical synthesis of the C-ring enal intermediate
14 has been established. One additional step incorporates a
functionalized B-ring fragment. The resulting BC-ring
subunit is suitably functionalized to serve as an advanced
intermediate in the synthesis of bryostatin 1. In addition, as
the following letter describes,15 a number of highly con-
vergent approaches to tricylic macrocyclic analogues of
bryostatin are made possible utilizing this C-ring intermediate
with the pyran annulation reaction as a key strategic element.
outlined above and shown in Scheme 6 allowed the
C15-C27 fragment 14 to be assembled in 15 steps (27%
overall yield) from (R)-isobutyl lactate.
Scheme 7. Pyran Annulation with Enal 14
Supporting Information Available: Spectral (1H and
13C NMR) data for compounds described herein. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Acknowledgment. Financial assistance provided by the
National Institutes of Health (through Grant GM-28961) is
gratefully acknowledged.
OL050511W
(13) The somewhat lower yield seems to be associated with the use of
this unsaturated aldehyde as a substrate. In a similar case containing the
glycal and the same protecting groups, but a saturated aldehyde, a yield of
80% was obtained. (See following paper in this issue.)
(14) Mechanistically, this is incorrect, as the pyran oxygen originates
from the hydroxyl group in the hydroxy allylsilane component; however,
for purposes of analysis of target structures, it is convenient to think of the
reaction in these terms.
Having accomplished the synthesis of the desired C-ring
enal, we next attempted to couple this fragment with a simple
hydroxy allylsilane to incorporate the B-ring using the pyran
annulation reaction. The requisite allylsilane 13 was prepared
enantioselectively via the BITIP-catalyzed CAA protocol as
previously described.12
(15) Keck, G. E.; Truong, A. P. Org. Lett. 2005, 7, 2153.
2152
Org. Lett., Vol. 7, No. 11, 2005