Total synthesis of bryostatins: The development of methodology for the atom-economic and stereoselective synthesis of the ring C subunit

Article abstract of DOI:10.1002/chem.201002898

Bryostatins, a family of structurally complicated macrolides, exhibit an exceptional range of biological activities. The limited availability and structural complexity of these molecules makes development of an efficient total synthesis particularly impor

Full text of DOI:10.1002/chem.201002898

DOI: 10.1002/chem.201002898
Total Synthesis of Bryostatins: The Development of Methodology for the
Atom-Economic and Stereoselective Synthesis of the Ring C Subunit
Barry M. Trost,* Alison J. Frontier, Oliver R. Thiel, Hanbiao Yang, and
Guangbin Dong[a]
Abstract: Bryostatins, a family of struc-
turally complicated macrolides, exhibit
an exceptional range of biological ac-
tivities. The limited availability and
structural complexity of these mole-
cules makes development of an effi-
cient total synthesis particularly impor-
tant. This article describes our initial
efforts towards the total synthesis of
bryostatins, in which chemoselective
and atom-economical methods for the
stereoselective assembly of the ring C
subunit were developed. A Pd-cata-
lyzed tandem alkyne–alkyne coupling/
6-endo-dig cyclization sequence was ex-
plored and successfully pursued in the
synthesis of
a
dihydropyran ring
system. Elaboration of this methodolo-
gy ultimately led to a concise synthesis
of the ring C subunit of bryostatins.
Keywords: alkynes · atom econo-
my · bryostatins · cross-coupling ·
dihydropyran synthesis · palladium
Introduction
these outstanding biological activities, bryostatin 1 has been
examined in numerous phase I and II clinical trials, includ-
ing for the treatment of melanoma, non-Hodgkinꢀs lympho-
ma, chronical lyphocytic leukemia, sarcomas, and more re-
cently for the treatment of Alzheimerꢀs disease.[14–19] Al-
though their exact mode of action is still under debate,
bryostatinsꢀ efficacy can be attributed to their strong affinity
(picomolar) for protein kinase C (PKC) isozymes,[20] en-
zymes that phosphorylate serine and threonine residues in
many target proteins that are often involved in cell-signal
transduction pathways. Bryostatins activate PKC through
binding to the same active sites as phorbol esters,[21,22] how-
ever, unlike many phorbol esters, which are tumor promot-
ers, bryostatins act as antitumor agents.
Isolation and biology: Since their first isolation in 1968 by
Pettit and co-workers from the marine bryozoan Bugula ner-
itina,[1,2] bryostatins 1–20 (1a–t), a class of structurally com-
plex macrolides, have continuously attracted the attention of
both chemists and biologists. These marine natural products
exhibit an exceptional range of biological activities, the most
notable being that bryostatins can act as antineoplasic
agents and exhibit excellent anticancer activity.[3] Biological
experiments indicate that bryostatin 1 (1a) significantly in-
hibits the growth of a large number of tumor cell lines in
vivo with low toxicity, including murine leukemia, ovarian
carcinoma, reticulum cell sarcoma, and B16 melanoma.[4,5,6,7]
Recently, the clinical application of bryostatin 1 in combina-
tion with other chemotherapeutic agents has shown signifi-
cant potential to treat some cancers with high potency.[8,9]
Unlike other antineoplasic agents, however, bryostatin 1
stimulates the immune and hematopoietic system,[10] which
has been suggested as a key factor for its antitumor
action.[11] Furthermore, recent efforts have revealed that
bryostatin 1 significantly enhances both cognition and
memory in animals, which suggests a potential use of bryos-
tatin 1 for the treatment of Alzheimerꢀs disease, depression,
and other cognitive impairments.[12] Remarkably, Sun et al.
recently suggested that bryostatin 1 could be a potential new
treatment to reverse brain damage after a stroke.[13] Due to
Synthetic challenge and previous efforts: The structures of
the bryostatins constitute significant synthetic challenges,
which include a 26-membered lactone fused by three heavily
substituted polyhydropyran (PHP) rings, two acid/base-sen-
sitive exo-cyclic unsaturated esters, and one congested C16–
C17 trans-olefin, as well as numerous oxygen-containing
functionalities and stereogenic centers. Previously, three of
the twenty bryostatins have been accessed by total synthe-
ses,[23,24,25] and formal syntheses of bryostatins have also
been reported.[26,27] These elegant syntheses have illustrated
the power of organic synthesis for the creation of molecules
of extreme complexity, however, their lengths (>40 steps in
the longest linear sequence and >70 steps in total) have so
far restrained them from serving as a practical source for
these natural products.
[a] Prof. Dr. B. M. Trost, Prof. Dr. A. J. Frontier, Dr. O. R. Thiel,
Dr. H. Yang, Dr. G. Dong
Department of Chemistry, Stanford University
Stanford, California 94305-5080 (USA)
Fax : (+1)650-725-0002
One of the main reasons for the length of the reported
synthetic sequences is the large number of steps devoted to
protecting group manipulations and functional group trans-
formations, necessitated by the complexity of the bryostatin
structure. We envisioned that the development of conver-
E-mail: bmtrost@stanford.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201002898.
9762
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9762 – 9776

FULL PAPER
gent and chemoselective methods would enable us to syn-
thesize the PHP rings in a more efficient fashion, and ulti-
mately to shorten the syntheses. Herein, we report detailed
studies on the development of methods and strategies for
the chemo- and diastereoselective formation of bryostatin
ring C.[28]
of them has a unique substitution pattern (Scheme 1). Our
initial strategy for the assembly of ring C (see Scheme 1)
was based on the unique reactivity of alkynes. We envi-
sioned that the ring-C fragment could ultimately come from
functionalization of dihydropyran intermediate I. The key
dihydropyran (I) could be prepared through a highly atom-
economic transformation from alkynes III and IV. By taking
advantage of the different reactivities of a terminal alkyne
and an ynoate, we could develop a Pd-catalyzed tandem se-
quence: alkyne–alkyne coupling to give enyne intermediate
II followed by a 6-endo-dig O-cyclization reaction to gener-
ate the dihydropyran in a one-pot process.
Results and Discussion
Initial design: As a key structural feature, bryostatins con-
tain three multi-substituted polyhydropyran rings, and each
Chem. Eur. J. 2011, 17, 9762 – 9776
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9763

B. M. Trost et al.
chosen as the model system and
examined under the conditions
reported for furan formation.
Initial experiments by using Pd-
AHCTUNRTGEG(NNNU OAc)2 (5 mol%) and tri(2,6-
dimethoxyphenyl)phosphine
(TDMPP; 2 mol%) gave dihy-
dropyran 5a, although the reac-
tion time was rather long if the
reaction was run at room tem-
perature
(7 days,
Table 1,
entry 1). It was possible to iso-
late the simple cross-coupling
product 4a after ꢀ24 h. Elevat-
ed temperatures shortened the
reaction time, but at a cost of
selectivity and yield (Table 1,
Scheme 1. The initial design of the synthesis of ring C in bryostatins. Pg=protecting group.
Our initial efforts were inspired by a previous method de-
veloped in this group, in which furans were synthesized
through the palladium-catalyzed cross-coupling of terminal
alkynes and ynoates with a propargyl alcohol, followed by
intramolecular 5-endo-dig oxypalladation and then olefin
migration (Scheme 2).[29] If an ynoate bearing a hydroxyl
group at the homopropargylic position was used in the
tandem alkyne–alkyne coupling/6-endo-dig oxycyclization
reaction, a dihydropyran ring system would be obtained
(Scheme 3). Two possible competing pathways would be ex-
pected: 1) the alcohol could attack the alkyne in a 5-exo
fashion to give polyhydrofurans; or 2) the alcohol could
Scheme 3. The formation of dihydropyrans through Pd-catalyzed tandem
alkyne–alkyne coupling/6-endo-dig oxycyclization. i) PdACHTNUTRGNEUNG(OAc)2 (5 mol%),
TDMPP (2 mol%).
attack the ester group to provide a 6-membered lactone
through transesterification.
To test the feasibility of the desired cyclization, the reac-
tion of 1-heptyne (2a) and homopropargylic ynoate 3a was
entries 3–6). As the reaction temperature was raised, pro-
gressively higher proportions of lactone 6a were observed.
Fortunately, a higher catalyst loading both accelerated the
reaction to a reasonable time interval (60 h) and eliminated
Table 1. Optimization of the dihydropyran formation reaction.[a]
Pd
A
(OAc)2/TDMPP
T
[8C]
t
Ratio of
5a:6a
Yield
[%]
[h]
1
2
3
4
5
6
5:2
RT
RT
135
60
60
54
36
24
>20:1
>20:1
5.5:1
5.5:1
1.8:1
61
61
61
57
30
23
10:4
5:2
5:2
5:2
5:2
RT–50[b]
50
RT–80[b]
80
1.6:1
Scheme 2. The formation of furans through Pd-catalyzed tandem alkyne–
[a] All reactions were performed in benzene at a concentration of 0.7m
of substrate 3a. [b] Reaction run at ambient temperature for the first
24 h, followed by heating to the stated temperature.
alkyne coupling/5-exo-dig oxycyclization. i) Pd
ACHTUNGRTEN(NUNG OAc)2 (5 mol%),
TDMPP (2 mol%); ii) 1,8-diazabicycloundec-7-ene (DBU).
9764
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9762 – 9776

Synthesis of Bryostatin Ring C
FULL PAPER
Table 2. The Pd-catalyzed tandem alkyne–alkyne coupling/6-endo-dig oxycycliza
(PMB=p-methoxybenzyl)
ACHTUNGTRENNUNG
tion.[a]
the formation of the lactone. By using 10 mol% Pd-
ACHTUNGTRENNUNG(OAc)2 and 4 mol% TDMPP in benzene (0.7m in
ynoate) at room temperature, dihydropyran 5a was
isolated in 61% yield (Table 1, entry 2).
For its application in the total synthesis of bryo-
statins, the chemoselectivity and substrate scope of
this method were next explored. Alkynes and
ynoates[30] with various functional groups and sub-
stitution patterns were subjected to the tandem
coupling/cyclization reaction under the optimal re-
action conditions (Tables 2 and 3).
Alkyne Ynoate
T
Product
Yield
[%][b]
1
2
3
4
5
2a
3a
3b
3c
3d
60 h
5a 61
As indicated in Table 2, ynoates bearing primary,
secondary, and tertiary alcohols all underwent cycli-
zation without incident. It is not unexpected that
the reaction with a tertiary alcohol substrate (3g)
was markedly slower, but gratifyingly, still able to
provide the desired dihydropyran in 44% yield, de-
spite the high levels of steric hindrance (Table 2,
entry 8). The excellent chemoselectivity was re-
vealed by the observation that the relatively reac-
tive primary chloride substituent was tolerated
under these conditions (Table 2, entry 4). Ynoates
with the homopropargylic alcohol chain restricted
within a ring (3e and f)[31] failed to undergo the
subsequent O-cyclization reaction; only the cross-
coupled products (4e and f) were isolated (Table 2,
entries 5 and 7). In the case of 3e, a longer reaction
time (7 d) gave the corresponding lactone (6e)
without any detectable amount of the dihydropyran
in the reaction mixture (Table 2, entry 6). This diffi-
culty is likely caused by the trans-orientation of the
substituents on the cyclohexane ring, which presents
an unfavorable geometry for the endo cyclization
reaction. A donor alkyne with a branch at the prop-
argylic position (2b) was also examined (Table 2,
entry 9), and the cross-coupling/cyclization se-
quence was comparable to the unsubstituted cases.
Next, we investigated the functional group com-
patibility and electronic effects of substitution on
this transformation. Studies shown in Table 3 reveal
that free alcohols, nitriles, acetals, and vinyl silanes
are compatible with the reaction conditions. The
electronic nature of the substituents on the donor
alkynes proved to affect the regioselectivity of the
oxypalladation step. A minor amount of 5-exo
product (7) was observed if a donor alkyne bearing
2a
2a
2a
2a
60 h
7 d
5b 57
5c 57
5d 42
4e 50
7 d
3e
(rac)
48 h
3e
(rac)
6
7
8
2a
2a
2a
7 d
13 d
7 d
6e 58
4 f 69
5g 44
3 f
(rac)
3g
3b
9
2b
60 h
5 f 57
[a] Pd
ACHTUNGRTEN(NUNG OAc)2 (5 mol%), TDMPP (2 mol%), benzene, 0.7m in ynoate 3, RT. [b] Isolat-
ed yield.
(Scheme 4). Variation of the reaction temperature, concen-
tration, and order of addition had little to no effect on the
endo/exo ratio of the reaction
depicted in Table 3, entry 1.
However, use of 2-methyl-3-
butyne-2-ol 2k as the alkyne
partner led to a reversal of the
product ratios, presumably due
to a combination of both elec-
an electron-withdrawing group was employed. The highest
proportions of the 5-exo products were observed if an elec-
tronegative group was installed at the propargylic position
(e.g., Table 3, entries 1, 2, and 9), with lesser amounts ob-
served if the substitution was at the homoallylic position
(Table 3, entries 4 and 5), and none if the group was four
carbons removed from the alkyne (Table 3, entry 8). This
unusual regioselectivity was presumably caused by the in-
ductive effect of the electronegative substituents adjacent to
the donor alkyne, as well as dipole repulsion between the al-
cohol nucleophile and the electronegative substituent
tronic
and
steric
effects
Scheme 4. The 5-exo-dig versus
6-endo-dig oxycyclization path-
(Table 3, entry 9). Vinylsilane
substitution[32] also caused for- ways.
Chem. Eur. J. 2011, 17, 9762 – 9776
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9765

B. M. Trost et al.
Table 3. Pd-catalyzed tandem alkyne–alkyne coupling/6-endo-dig oxycyc-
lization.[a]
ACHTUNGTRENNUNG
Donor alkyne
Product ratio
Yield[b]
(5+7) [%]
AHCTUNGTRENNUNG
To address this problem of slow cyclization, methods for
accelerating the oxypalladation step of the reaction se-
quence were next examined. Envisioning that the cyclization
was also a palladium-catalyzed reaction, its rate should
depend on the electrophilicity of the palladium(II) species,
and thus increasing the electrophilicity should increase the
rate of cyclization. We have previously noted a dramatic en-
hancement in the rate of formation of p-allylpalladium com-
plexes from less nucleophilic alkenes when palladium tri-
fluoroacetate was employed.[33] In the initial experiment, the
trifluoroacetate salt was generated in situ by simply adding
trifluoroacetic acid to palladium acetate. Indeed, the cycliza-
tion rate increased significantly (complete cyclization within
24 h), but a 1:1 mixture of the exocyclic alkene isomers of
5q was isolated in 59% yield (Table 4, entry 2).[34]
1
2
2c
2d
2.3:1 (5h/7h)
1.9:1 (5i/7i)
50
18
3
4
5
2e
2 f
2g
–
–
7.5:1 (5j/7j)
9.3:1 (5k/7k)
69
35
6
2h
6.2:1 (5l/7l)
59
7
8
2i
2j
6.9:1 (5m/7m)
>20:1 (5n)
52
51
9
2k
2l
1:4.2 (5o/7o)
2:1 (5p/7p)
41
55
Ancillary experiments revealed that E/Z isomerization of
the double bond was catalyzed by Brønsted acids. To solve
the olefin isomerization problem, the more electrophilic pal-
ladium trifluoroacetate was employed directly as the oxypal-
ladation catalyst. Indeed, by addition of palladium trifluoro-
10[c]
[a] PdACHTUNGTRENNUNG
3b RT. [b] Isolated yield. [c] Pd
then Pd(OCOCF3)2 (6 mol%).
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
able to obtain the desired dihydropyran in 62% yield after
36 h without isomerization of the exocyclic alkene (Table 4,
entry 3), which is a significant improvement compared to a
25% yield over 14 days (Table 4, entry 1).
mation of the exo-product (Table 3, entry 10), likely due to
the ability of silicon to stabilize a positive charge at the b-
position.
As the synthesis of bryostatins requires the use of a donor
alkyne with a propargylic all-carbon quaternary center (see
Scheme 1), 3,3-dimethylbutyne (2m) was then employed as
the model substrate. Although enyne (4q) was produced in
70% yield within 48 h [Eq. (1)], the oxypalladation step was
quite sluggish. The 6-endo-dig product dihydropyran 5q
formed very slowly (25% yield) after exposure to the typical
reaction conditions for 14 days [Eq. (1) and Table 4,
entry 1].
In general, this new procedure provides rapid direct con-
version of the two alkyne starting materials into the dihy-
dropyrans in a one-pot fashion. A direct comparison of
these two methods (Table 5) demonstrates that a considera-
bly reduced reaction time and increased yield was observed
with this new two-stage, one-pot method. Furthermore, this
procedure permits the cyclization of substrates that had un-
dergone only the cross-coupling reaction under the Meth-
od A reaction conditions (Table 5, entries 6 and 7).
By using this two-stage pro-
Table 4. Stepwise dihydropyran formation.[a]
tocol, synthetically useful bicy-
clic systems were generated
from cyclic ynoates 3e and f
[Eq. (2)], a cyclization that did
not proceed in the absence of
palladium trifluoroacetate (see
Table 2, entries 5 and 7). The
Reagents and conditions
T
Yield
[%]
formation of seven-membered
rings [Eq. (3)] also became pos-
sible by using this combination
of catalysts.
1
2
3
Pd
Pd(OAc)2 (10 mol%), TDMPP (4 mol%), benzene, RT then TFA (20 mol%)
Pd(OAc)2 (4 mol%), TDMPP (4 mol%), Pd(OCOCF3)2 (6 mol%), benzene, RT
A
14 d
24 h
36 h
25
N
59[b]
62
R
ACHTUNGTRENNUNG
A few substrates produced
unusual results when subjected
[a] Intermediate enyne 4q was prepared by using Pd
bient temperature. [b] E/Z ratio was 1:1
ACHTUNGRTENUN(NG OAc)2 (10 mol%), TDMPP (4 mol%) in benzene at am-
9766
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9762 – 9776

Synthesis of Bryostatin Ring C
FULL PAPER
Table 5. Tandem versus stepwise reaction conditions for cyclization.
were conducted. Attempts to trigger the reaction sequence
without Pd
occur. If a cross-coupled enyne product of type 4 was treat-
ed with Pd(O2CCF3)2 without a ligand, no cyclization prod-
ACHTUNGRTENUN(NG OAc)2 failed, as the cross-coupling step did not
AHCTUNGTRENNUNG
uct was detected, which indicates the importance of ligands
in the oxypalladation step. Finally, if an enyne 4 was subject-
ed to PdACHTUNRTGNEUNG(O2CCF3)2 (5 mol%) and a ligand (2 mol%), the
Alkyne R’
Product
3a[d] 5a
Yield [%][a] (t)
Method A[b] Method B[c]
oxypalladation occurs, but the reaction rate is slow. An in-
vestigation into palladium to phosphine ligand ratios, as well
as palladium acetate to palladium trifluoroacetate ratios, re-
vealed that the best results are obtained by using the origi-
1
2
3
4
5
6
7
2a
H
61 (5a; 60 h) 63 (12 h)
57 (5b; 52 h) 66 (30 h)
2a
Me
3b
5b
5c
5r
5q
5s
5t
2a
CH2OPMB 3c
52 (5c; 7 d)
52 (5r; 7 d)
58 (24 h)
61 (36 h)
nal conditions (i.e., Pd
ACHTUNGTRENNUNG
2a
CH2OTBS
Me
CH2OPMB 3c
CH2OTBS 3h
3h
3b
2m
2m
2m
25 (5q; 14 d) 62 (78 h)
– (5s; 7 d)[e] 69 (9 d)
– (5t; 7 d)[e] 50 (7 d)[f]
for the cross-coupling and then Pd
AHCTUNGTRENNUNG
the oxypalladation). It was also found that the use of an
excess of the terminal alkyne (>1.1 equiv) accelerated the
cross-coupling step, but seriously compromised the efficency
of the oxypalladation step, making the competing lactoniza-
tion a problem.
[a] Isolated yields. [b] Method A: Pd
(4 mol%), benzene. [c] Method B: PdAHCUTNGTRENNUNG
(4 mol%), then PdACHTUNGTRENNUNG(OCOCF3)2 (6 mol%), benzene. [d] The methyl ester
was used. [e] No oxycyclization was observed after 7 days (enyne product
of the cross-coupling only). [f] 95% conversion. The yield of this reaction
was 58% after 5 days (82% conversion).
ACHTUNGTRENNUNG
Synthesis of the ring C subunit of bryostatins: With this new
Pd-catalyzed tandem coupling/cyclization method in hand,
the stage was set to apply this method to the synthesis of
the ring C subunit of bryostatins. We envisioned that bryo-
AHCTUNGTREGsNNNU tatins could be divided into two fragments: the northern
fragment containing rings A and B and the southern frag-
ment (12) containing ring C (Scheme 5). These two frag-
ments could then be sewn together through esterification to
form the C1 ester and an olefination reaction to form the
C16–C17 alkene. The ring C fragment 12 could be ultimately
prepared from donor alkyne 13 and ynoate 14.
To this end, a variety of donor alkynes with different func-
tional handles were synthesized to examine in the tandem
dihydropyran formation protocol targeting ring C of bryo-
AHCTUNGTREGsNNUN tatins (Scheme 6). 2,2-Dimethyl-3-butyn-1-ol (13a) was syn-
thesized in two steps from commercially available 2-methyl-
but-3-yn-2-ol 2k according to a literature protocol,[36] and
then elaborated in various ways to form other substituted
terminal alkynes. Simple silylation of the primary alcohol
gave tert-butyldimethylsilyl ether 13b. SN2 displacement of
the corresponding tosylate with potassium thioacetate gave
thioacetate 13c and with thiophenol provided the sulfide,
which was subsequently oxidized to produce sulfone 13d.
Sulfone 13e was produced by the Mitsunobu reaction with
1-phenyl-1H-tetrazol-5-thiol followed by oxidation.[37] All of
these alkynes possess substituents appropriate for future
coupling reactions with the northern fragments, as outlined
in Scheme 5.
to the tandem reaction protocol. The ortho-alkynyl phenol
10[35] failed to undergo the initial cross-coupling reaction
[Eq. (4)]. It is also surprising that terminal enyne 11 did not
undergo the oxycyclization, and only the lactone product
was observed [Eq. (5)].
To gain insight into the role of each reagent during the
two-stage, one-pot sequence, a series of control experiments
The synthesis of the target ynoate partners of type 14 for
the projected tandem cross-coupling/oxycyclization se-
quence is shown in Scheme 7. Silyl ether 15a was prepared
in high enantiomeric purity over two steps from commer-
cially available 1,4-trans-hexadiene according to a literature
procedure.[38]
After some experimentation, conditions were found to
convert the terminal olefin into the corresponding diol in a
diastereoselective fashion (Table 6). Suprisingly, when TBS
substrate 15a was subjected to the Sharpless asymmetric di-
Chem. Eur. J. 2011, 17, 9762 – 9776
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9767

B. M. Trost et al.
Scheme 5. Strategy for the synthesis of ring C in bryostatins.
Table 6. Dihydroxylation of compound 15.[a]
R
Chiral ligand
Ratio
(anti:syn)
ACHTUNGTRENNUNG
Scheme 6. The synthesis of donor alkynes for the ring C study. i) PBr3,
66%; ii) Al, HgCl2 (cat.), (CH2O)n, 37%; iii) tert-butyldimethylsilyil chlo-
ride (TBSCl), imidazole, 99% yield of 13b; iv) p-toluenesulfonyl chloride
(TsCl), pyridine, then AcSK, 61% yield of 13c; v) TsCl, pyridine, then
NaSPh followed by m-chloroperoxybenzoic acid (mCPBA), 60% yield of
13d over the two steps; vi) 1-phenyl-1H-tetrazol-5-thiol, diisopropyl azo-
dicarboxylate (DIAD), PPh3 followed by [(H4N)6Mo7O24]·H2O2, 71%
yield of 13e over the two steps.
1
2
3
4
5
6
7
TBS
TBS
acetonide
acetonide
TBS
15a
15a
15b
15b
15a
15a
15a
N
6:1
1:1.2
1:2.5
2:1
1.8:1
2.7:1
1.5:1
TBS
TBS
[a] The Sharpless AD-mix protocols were used when possible for the di-
hydroxylation reactions. [b] OsO4 (10 mol%), N-methyl morpholine-N-
oxide (NMO), acetone/H2O.
a 6:1 d.r. (Table 6, entry 1). It is likely that 15a is too steri-
cally crowded to fit into the chiral pocket properly. On the
other hand, the stereochemistry of the diol after SAD of
acetonide substrate 15b was consistent with the prediction,
albeit with a relatively poor diastereoselectivity (Table 6, en-
tries 3 and 4). Examination of other asymmetric dihydroxyl-
AHCTUNGTREGaNNNU tion ligands was not fruitful (Table 6, entries 5 and 6), de-
spite evidence that the anthraquinone ligand is often superi-
or to the DHQD series for terminal alkenes.[40] In addition,
substrate-controlled dihydroxylation of substrate 15a (no
chiral ligand present) resulted in a 1.5:1 mixture of isomers,
slightly favoring the desired diastereomer (Table 6, entry 7).
With pure diol 16a in hand, conversion to epoxide 17 was
uneventful (Scheme 7). Opening the epoxide with the lithi-
um anion of methyl propiolate provided target ynoate 18a
in 80% yield. Ynoates 18b and c were also synthesized to
examine their performance in the subsequent tandem
alkyne coupling/cyclization sequence. The silyl groups were
removed under acidic conditions to give triol 18b. Subse-
quently, triol 18b was selectively reprotected at the 1,2-diol
by treatment with 2,2-dimethoxypropane to give acetonide
18c (72% yield from bis(silyl ether) 18a).
Scheme 7. The synthesis of ynoates for the synthesis of ring C. i) AD-mix
a, tBuOH/H2O; ii) TsCl, pyridine; iii) NaOMe, MeOH; iv) LiCCCO2Me,
BF3·OEt2, THF, À788C; v) aqueous HF, CH3CN; vi) p-toluenesulfonic
acid (PTSA).
hydroxylation (SAD) conditions, the observed stereochemi-
cal outcomes were exactly the opposite of the prediction
based on the Sharpless mnemonic.[39] By using the ꢂmatchedꢀ
DHQD ligand, a poor diastereoselectivity was obtained that
slightly favored the undesired diastereomer (Table 6,
entry 2); however, use of the ꢂmismatchedꢀ DHQ ligand pro-
vided the desired diol isomer as the major product in about
Attempts were made to streamline the synthesis of ynoate
18a from diol 16a by adopting a protocol developed by For-
syth and Cink [Eq. (6)].[41] This transformation proceeded
9768
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9762 – 9776

Synthesis of Bryostatin Ring C
FULL PAPER
very well on a small scale (0.5 mmol; 90%); however, the
yield was diminished on a larger scale (12.2 mmol, 45%)
when problems arose with the control of the temperature of
the reaction. Furthermore, efficient stirring during the epox-
ide-opening stage was difficult to achieve, due to the pres-
ence of the salts formed in the first stage of the reaction.
Ynoates 18a–c were next examined in the context of the
tandem cross-coupling oxypalladation process with donor al-
kynes, sulfone 13d and silyl ether 13b (Table 8). Ynoate
18a, the most directly accessible ynoate (see Scheme 7), was
unsuitable for two reasons: 1) even by using the Pd-
AHCTNUGTER(GNNUN O2CCF3)2 protocol, the reaction times were unreasonably
long, and 2) a high propor-
tion of lactone 22 was ob-
served in the product mixture
(Table 8, entries 1 and 2).
The formation of this unde-
sired byproduct, as well as
the sluggishness of the reac-
Table 8. Palladium-catalyzed tandem coupling/cyclization of alkynes 13
and 18.[a]
With routes to access both the donor alkyne and the
ynoate partners, the stage was set to test them in the Pd-cat-
alyzed tandem coupling/cyclization reaction. Donor alkynes
were examined first by using 3b as the model ynoate
(Table 7). Subjecting sulfone 13d to the two-stage, one-pot
Table 7. Palladium-catalyzed tandem reaction between alkynes 22 and
13b.[a]
R
X
t [d]
Products (ratio)
1
2
3
4
5
6
TBS
TBS
H
18a SO2Ph 13d >14
18a OTBS 13b >24
12a:21a:22a (4:1:4)
12b:21b:22b (6:1:2)[b]
18b SO2Ph 13d ꢁ2.5 isomerization of the exo-
X
t [d]
Yield
[%][b]
Products (ratio)
H
18b OTBS 13b ꢂ1.5 enoate
acetonide 18c SO2Ph 13d
acetonide 18c OTBS 13b
3
4
12c:21c:22c (3:1:1)
1
2
SO2Ph
SAc
13d
13c
4d
–
54
–
19d:20d (3:1)
12d:21d:22d (12:1:1)[c]
[c]
–
[a] Pd
ACHTUNGRTEN(NUNG OAc)2 (4 mol%), TDMPP (4 mol%), benzene, RT, then Pd-
AHCTUNGTRENNUNG
3
13e
3
38
19e:20e (1:3.5)
yield of 12d.
[d]
4
5
OH
OTBS
13a
13b
2
6
–
19a:20a (1:1)
19b only
58
tion process may be attributed to the extremely hindered
nature of the homopropargylic alcohol in 18a, due to the
two neighboring TBS ethers.
[a] PdACHTUNGTRENNUNG(OAc)2 (4 mol%), TDMPP (4 mol%), benzene, RT, then Pd-
ACHTUNGTRENNUNG(CO2CF3)2, (6 mol%). [b] Isolated yield. [c] Cross-coupling product
(enyne) only [d] Not determined.
Substrate 18b underwent rapid cyclization with very high
selectivity (no byproducts 21 or 22 were detected by
1H NMR spectroscopy of the crude products). However, di-
hydropyran 12 could not be isolated without isomerization
of the exocyclic enoate (Table 8, entries 3 and 4). In addi-
tion, isolation of the enoate mixture revealed that the yield
of the reaction was lower than for other cases (ꢀ40%). It
was decided that ynoate 18b was too reactive to perform re-
liably in the cyclization. Acetonide ynoate (18c) proved to
have the proper reactivity and provided the best results in
the tandem cyclization process (Table 8, entries 5 and 6). In
general, the reaction with silyl ether 13b as the donor
alkyne proceeded without significant formation of the 5-exo
cyclization product. It seems that sulfone 13d was inferior
with respect to the formation of the undesired products 21
and 22 (Table 8, entries 1 and 5). Therefore, donor alkyne
13b and ynoate 18c clearly acted as the best combination as
reaction conditions produced the desired dihydropyran 19d,
along with 5-exo byproduct 20d in a 3:1 ratio (Table 7,
entry 1). In contrast, sulfone 13e produced mainly 20e
(3.5:1 ratio; Table 7, entry 3). Thioacetate 13c underwent
cross-coupling with ynoate 3b (Table 7, entry 2), but no di-
hydropyran 19c was formed, even after an extended reac-
tion period; the only product isolated was the lactone by-
product. When alkyne 13a (free alcohol) was subjected to
the reaction conditions, a mixture of 19a and 20a was ob-
served (Table 7, entry 4). However, when silyl ether 13b was
the donor alkyne, the sequence was selective for 19b
(Table 7, entry 5). From these studies, it seemed appropriate
to select 13d and b (Table 7, entries 1 and 5) as the terminal
alkyne partners for use during the synthesis of the fully
functionalized dihydropyran 12 (see Scheme 5) that is neces-
sary for the synthesis of the bryostatins.
Chem. Eur. J. 2011, 17, 9762 – 9776
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9769

B. M. Trost et al.
indicated in Table 8, entry 6 for the synthetic route targeting
bryostatin ring C.
this strategy (see the Supporting Information for details). To
this end, commercially available d-galactono-1,4-lactone was
brominated at the primary hydroxyl functional group and
then peracetylated.[45] Subsequent global hydrogenation cat-
alyzed by palladium on charcoal reduced the bromine and
led to intermediate 26, which contains all three stereogenic
centers of the C22–C27 moiety of the bryostatins. Exhaus-
tive reduction with lithium borohydride was followed by
acetonide formation to give bisacetonide 27. The sterically
less hindered terminal acetonide was chemoselectively
cleaved, and the ensuing diol was converted to the known
epoxide 28 by activation of the primary alcohol with N-
(2,4,6-triisopropylbenzenesulfonyl)imidazole followed by
base-mediated ring closure. Finally, synthesis of ynoate 18c
was furnished by BF3-mediated epoxide ring opening with
the lithium acetylide derived from methyl propiolate. Multi-
gram quantities of ynoate 18c can be obtained from this
chiral-pool approach, the product of which is spectroscopi-
cally identical to the one that was derived from tetraol 16a;
furthermore, this approach also proves our previous stereo-
chemical assignment of the SAD intermediates from the
first generation route.
Second generation approaches for the synthesis of donor
alkyne 13b and ynoate 18c: Having identified the optimal
coupling partners for the tandem cyclization, we next opti-
mized the syntheses of alkyne 13b and ynoate 18c, improv-
ing both efficiency and scalability. A more effective synthe-
sis of alkyne 13b on a large scale is described in Scheme 8.
Scheme 8. An improved synthesis of alkyne 13b. i) CH2O, CF3COOH;
ii) TBSCl, imidazole, DMF; iii) lithium diisopropylamide (LDA), THF,
À
À788C; iv) ClPO(OEt)2, À788C to RT; v) LDA, 788C to RT.
G
With high yielding and scalable routes to access both cou-
pling fragments, our efforts were directed towards improving
the efficiency of the formation of dihydropyran 12d. As in-
Commercially available 3-methyl-2-butanone was subjected
to an acid-mediated aldol reaction with paraformaldehyde
to provide alcohol 23,[42] which was then protected as a TBS
ether. Subsequently, methyl ketone 24 was converted to ter-
minal alkyne 13b by using a one-pot protocol developed by
Negishi.[43]
dicated in Table 8, the use of the Pd
AHCTUNGTRENNUNG
ACHTUNGRTENUN(NG OAc)2 and Pd-
in an acceptable yield. This protocol was reproducible at
scales of 0.3 mmol; unfortunately, upon further scaling up,
sluggish reactions and the formation of different isomers
was observed.
To avoid the cumbersome purification encountered when
running the reaction on a large scale by using the SAD
route, we next developed an efficient ꢂchiral-poolꢀ-based
route to access ynoate 18c (Scheme 9). Although it was par-
tially adapted from a known route,[44] the reaction parame-
ters were carefully optimized for successful completion of
A closer investigation of the reaction products of the
tandem reaction revealed the presence of four different
products (Table 9). Although the desired dihydropyran 12d
from a 6-endo cyclization was the major product, we also
observed varying amounts of its double-bond isomer 29, the
furan derivative 21d, resulting from the 5-exo cyclization,
and the lactone 22d. Consequently, we attempted to deter-
Table 9. NMR study on the ratio of the products from the one-pot cou-
pling/cyclization of alkyne 13b and ynoate 18c.
Scheme 9. A chiral-pool-based synthesis of alkynoate 18c. i) HBr, AcOH
then Ac2O; ii) H2, Pd/C, Et3N; iii) LiBH4, THF then Amberlyst 15;
iv) 2,2-dimethoxypropane, PTSA, THF; v) I2, CH3OH, 66% (87% based
on recovered starting material (BRSM)); vi) NaH, THF, then N-(2,4,6-
triisopropylbenzenesulfonyl)imidazole; vii) LiCCCO2CH3, BF3·OEt2,
t [h]
30
[%]
12d
[%]
29
[%]
21d
[%]
22d
[%]
1
2
3
24
48
96
56
19
0
15
35
49
1
2
5
2
4
5
26
40
41
À
THF, 788C.
9770
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9762 – 9776

Synthesis of Bryostatin Ring C
FULL PAPER
mine the relative kinetics of the formation of each product
through NMR studies. Alkyne 13b and ynoate 18c were
treated with PdACHTUNGTRENNUNG(OAc)2 (4 mol%) and TDMPP (4 mol%) in
molecular addition of alcohols to alkynes,[46] but silver catal-
ysis was not effective for this cyclization. High catalyst load-
ings of silver trifluoroacetate (60 mol%) were necessary to
achieve cyclization (Table 10, entries 4 and 5), and cycliza-
tion with AgOTf was inefficient and unselective. Interesting
results were obtained by using PtCl2 as the catalyst.[47] Com-
plete conversion was observed in toluene at 608C without
formation of lactone 22d. Interestingly, this reaction showed
a strong preference for the 5-exo versus the 6-endo cycliza-
tion pathway (Table 10, entry 6). No cyclization was ob-
benzene (0.5m) at RT. Complete conversion to enyne 30
was observed after 20 h. Palladium trifluoroacetate
(8 mol%) was added to the reaction mixture and aliquots of
the reaction were analyzed by NMR spectroscopy after 24,
48, and 96 h (Table 9). The results suggested that lactone
formation was a very competitive pathway, mostly occurring
directly after addition of palladium trifluoroacetate. The 6-
endo-cyclization was favored over the other two competing
pathways, but the reaction was relatively slow. Catalytic
turnover was still observed after 2 days, indicating good cat-
alyst stability.
served in experiments that used [CpRuACTHUNGTRENNNUG
cyclopentadienyl), [{RuCl
as the catalyst.
N
ACHTUNGTRENNUNG
Utimoto and co-workers reported that [PdCl2ACTHNUTRGNEN(UG PhCN)2]
Having identified all of the side products, we performed a
more detailed catalyst survey for the oxycyclization step. To
this end, enyne 30 was isolated in 89% yield after quenching
the reaction at the first stage [Eq.(7)], and subsequently
subjected to different catalytic conditions (Table 10).
The use of palladium trifluoroacetate as the catalyst led
to only low selectivity towards the desired product 12d, and
significant amounts of olefin isomer 29 and 5-exo product
21d were observed (Table 10, entries 1 and 2). The addition
of TDMPP as a ligand afforded similar selectivities as in the
two-stage, one-pot sequence (Table 10, entry 3). Silver salts
have been described in the literature to catalyze the inter-
was able to catalyze intramolecular alcohol addition to al-
kynes (Table 10).[48] Reactions with this catalyst system were
very fast (complete conversions within one hour) with no
detectable formation of lactone 22d by H NMR spectrosco-
py (Table 10, entries 7–11). Ethereal solvents provided the
best results (Table 10, entries 8–11), and THF was chosen
1
for further optimization. Cyclization with [PdCl2ACTHNUTRGNEN(UG MeCN)2]
led to similar results (Table 10, entry 12). One major side re-
action for this catalyst system was isomerization of the exo-
cyclic olefin. An initial hypothesis was that this isomeriza-
tion was mediated by protons or chloride ions generated
from the catalyst in the course of the catalytic cycle. The use
of amine bases as a buffer re-
sulted in complete inhibition of
the oxycyclization, although
ethylvinylether was somewhat
effective as an acid trap
(Table 10, entry 13). The addi-
tion of solid inorganic bases
seemed more promising, but
only barium oxide and silver
Table 10. Metal-catalyzed 6-endo-dig cyclization of compound 30.[a,b]
oxide gave complete conversion
(Table 10, entries 14 and 15).
Finally, inspired by the results
in the previous one-pot proce-
dure, phosphines were investi-
Catalyst
G
Additive
([mol%])
t
30
12d 29
21d 22d
gated as additives to this
[PdCl2A(RCN)2] catalyst system.
N
[h]
[%] [%] [%] [%] [%]
CTHUNGTRENNUNG
1
2
3
Pd
Pd
Pd
U
–
–
12
14
12
12
16
14
–
–
–
41
–
–
–
–
–
–
–
–
–
–
–
–
18
33
49
32
47
29
30
45
47
45
50
44
53
50
52
62
32
22
15
5
20
16
8
4
3
57
49
31
27
27
20
26
28
35
34
28
30
29
28
18
24
–
The addition of TDMPP (1:1 to
Pd) dramatically slowed the re-
action, but the use of a smaller
amount of phosphine (3 mol%)
provided complete conversion
within 7 h and suppressed the
formation of both 21d and 29
(Table 10, entry 16). The use of
other phosphines, such as 1,1’-
bis(diphenylphosphino)ferro-
cene (dppf) and PPh3, resulted
in extremely low reactivity.
A
–
–
–
–
–
–
–
–
5[c] Ag
E
26
14
21
24
26
28
30
35
19
15
14
10
6[d] PtCl2
7
8
9
10
11
12
13
14
15
16
PdCl
PdCl
PdCl
PdCl
PdCl
PdCl
PdCl
PdCl
PdCl
PdCl
G
0.75
0.75
1
0.5
0.5
1
1
1
1
7
–
–
–
–
–
–
–
–
–
–
–
ethylvinylether (25)
BaO (50)
AgO (20)
TDMPP (3)
During the original oxycycli-
zation studies on 30, the most
practical set of conditions
[a] Reactions were run at 258C unless otherwise noted. [b] The ratio of products was determined by 1H NMR
spectroscopy. [c] This reaction was run at 408C. [d] This reaction was run at 608C.
Chem. Eur. J. 2011, 17, 9762 – 9776
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9771

B. M. Trost et al.
found for dihydropyran formation involved the use of
[PdCl2A(MeCN)2] (5 mol%)/TDMPP (3 mol%), on a 7 mmol
scale [Eq. (8)]. This protocol provided a good isolated yield
of the product mixture (91%) with a selectivity that reflect-
ed our small-scale experiments (Table 10, entry 16, 12d/29/
21d=6:1:3).
With scalable access to dihydropyran 12d, the remaining
challenge was to install the C19 and C20 oxygen functionali-
ties in a chemoselective and diastereoselective fashion
[Eq. (9)].
the conditions. The NMR spectra indicated that peracid oxi-
dations at ambient temperature provided exclusive forma-
tion of the C20 diastereomer depicted as structure 32, with
the ratio of diastereomers at the anomeric position between
3:1 and 2:1 depending on the reaction conditions employed.
The reactions at lower temperatures and the epoxidations
with peroxyimidates gave diastereomeric mixtures that also
contained the alcohols inverted at C20. The absolute stereo-
chemistry at C20 was assigned by converting 32 into tetrahy-
dropyran 33 through reduction of the anomeric methoxy
CHTUNGTRENNUNG
Pure alcohol 31 was isolated after removal of the TBS
group with TBAF from the mixture of TBS ethers obtained
in the [PdCl2ACHTUNGTRENNUNG(MeCN)2]/TDMPP cyclization reaction. To
access a-hydroxyl ketal 32, chemoselective epoxidation of
the C19–C20 alkene, followed by an in situ epoxide ring-
opening procedure with a methanol nucleophile was investi-
gated. Use of mCPBA as the oxidant was inefficient
(Table 11, entry 1). Switching to the more reactive trifluoro-
peracetic acid (TFPAA),[49a] the dihydropyran was oxidized
group with triACTHUNGETRNNUeG thylsilane [Eq. (10) and Scheme 10]. The ste-
reochemistry at the ketal center (C19) was tentatively as-
signed based on a similar epoxidation outcome in the Evans
bryostatin synthesis.[24b]
As the stereochemistry of the secondary alcohol at C20
had to be inverted to synthesize bryostatins, an oxidation–
reduction approach was explored next (Scheme 11). Howev-
er, diol 32 is a very sensitive compound and it decomposed
quickly in the presence of acid (even in CDCl3), and even
on standing as a neat sub-
stance (08C, under argon).
Nonetheless, the oxidation of
32 to 34 was attempted under
a variety of conditions. The
use of 2-iodoxybenzoic acid
(IBX)[50a] at RT in THF/
DMSO led to no conversion,
and oxidation with IBX in
acetonitrile at 808C caused
complete decomposition of
instantly at 08C. Using acetonitrile/DCM/MeOH as the
mixed-solvent system, alcohol 32 was isolated as the major
product in 34–48% yield (Table 11, entries 2–4). Increasing
the amount of TFPAA increased the formation of byprod-
diol 32. Moffatt–Swern oxidation[50b] (DMSO, trifluoroacetic
anhydride (TFAA), triethylamine (TEA)) gave an apparent-
ly clean reaction by TLC, however, the crude NMR spec-
trum showed the presence of five different aldehyde prod-
ucts. Switching to oxalyl
chloride as the activating
agent in the Moffatt–Swern
oxidation afforded no im-
provement.
Somewhat
promising results were ob-
tained by using the Dess–
ucts (Table 11, entry 3). The selectivity of the reaction
seemed to be slightly improved when it was performed at
À208C (Table 11, entry 4), although incomplete conversion
was observed. When MTO was used as a catalyst along with
urea hydrogen peroxide (UHP) as the oxidant, the desired
product was obtained but the yield was poor (Table 11, en-
try 7).[49b–d] Switching to trichloroacetonitrile gave similar re-
sults (Table 11, entry 8). Oxidations using Mo(CO)6 and
[49e]
VO
G
as catalysts gave mainly recovered 31, as did
Martin periodinane (DMP).[51] Aldehyde 34 was obtained in
44% yield in an impure form; however, subjection of 34 to
the standard Luche reduction conditions (CeCl3, NaBH4,
MeOH)[52] only led to decomposition. Therefore, this ap-
proach was not pursued further.
experiments using dioxiranes (Table 11, entries 5 and 6).
The findings are shown in Table 11.
Interpretation of the NMR spectrum of 32 was complicat-
ed by the presence of diastereomers. The stereoisomers at
C19 and C20 were formed in varying amounts depending on
9772
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9762 – 9776

Synthesis of Bryostatin Ring C
FULL PAPER
Table 11. Oxidation of compound 31. TBAF=tetra-n-butylammonium fluoride, TFPAA=trifluoroperacetic
acid, DMDO=dimethyldioxirane, UHP=urea hydrogen peroxide, MTO=methyltrioxorhenium.
Due to the labile nature of
diol 32, attempts were next
made to oxidize the TBS-pro-
tected dihydropyran 12d direct-
ly (Scheme 12). Subjecting 12d
(60% purity, 22d and 29 as
minor components) to the opti-
mal epoxidation conditions (tri-
fluoroperacetic acid;[53] see
Table 11) provided alcohol 36
in 30% yield, which corre-
sponds to a 50% yield for this
step adjusting for the purity of
the starting material. The ste-
reochemistry of the C19 and
C20 centers was assigned in
analogy with compound 32. Ox-
idation of the secondary alcohol
with DMP cleanly afforded
ketone 37. Finally, a highly dia-
stereoselective Luche reduc-
Conditions
Product
(yield [%])[a]
AHCTUNGTRENNUNG
[b,c]
1
2
3
4
5
6
7
8
mCPBA (1.5 equiv), CH2Cl2/MeOH, RT, 2 h
–
TFPAA (1.5 equiv), Na2HPO4, CH2Cl2/MeOH/CH3CN, RT, 1 h
TFPAA (3.5 equiv), Na2HPO4, CH2Cl2/MeOH/CH3CN, 08C, 15 min
TFPAA (1.5 equiv), Na2HPO4, CH2Cl2/MeOH/CH3CN, À208C, 30 min
DMDO, acetone, MeOH, RT, 3 h
CF3COCH3, UHP, Na2HPO4, CH2Cl2/MeOH, RT, 20 h
MTO (10 mol%), UHP (3 equiv), CH2Cl2/MeOH, RT, 3 h
CCl3CN (4 equiv), UHP (2 equiv), Na2HPO4, CH2Cl2/MeOH, RT, 4 h
32 (48)
32 (27)
32 (41)[d]
no reaction
no reaction
32 (28)
32 (52)
[a] The yields in parenthesis are isolated yields. [b] A complex mixture of products was observed. [c] Poor con-
version was observed. [d] Yield of 32 was 53% based on recovered 31.
tion[54] of 37, followed by acylation of the resulting secon-
dary alcohol (38), furnished ring C fragment 39 with the de-
sired C20 stereochemistry in 45% yield over four steps.[55]
Scheme 10. The nOe data for tetrahydropyran 33.
Conclusion
During the initial stage of our efforts towards the total syn-
thesis of bryostatins, we have developed a distinct method
for the stereoselective assembly of the ring C subunit, which
features a Pd-catalyzed tandem alkyne–alkyne coupling/6-
endo-dig cyclization sequence. This method is both chemo-
selective and atom economic, and utilization of this method
successfully resulted in a concise enantioselective synthesis
of the ring C fragment.
Experimental Section
Scheme 11. The oxidation–reduction approach to C20 inversion.
General procedure A for dihydropyran synthesis: A solution of Pd-
AHCTUNGTRENNUNG
A
(1.0 equiv) and alkyne (1.1 equiv) in
benzene was added (concentration of
the reaction mixture: 0.7m). The reac-
tion was stirred until the addition
product was consumed and the solvent
was removed in vacuo. The residue
was purified by flash column chroma-
tography on Florisil eluting with a di-
ethyl ether/petroleum ether mixture.
General procedure B for dihydropyran
synthesis:
A solution of PdACHTUNGTRENNUNG(OAc)2
(4 mol%) and TDMPP (4 mol%) in
benzene was stirred at RT for 20 min.
A solution of ynoate (1.0 equiv) and
alkyne (1.1 equiv) in benzene was
added (concentration of reaction the
Scheme 12. The synthesis of ring C fragment 39. i) TFPAA, Na2HPO4, CH3CN/DCM, MeOH, 08C; ii) DMP;
iii) NaBH4, CeCl3, CH3OH, À308C; iv) Ac2O, 4-dimethylaminopyridine (DMAP), pyridine, CH2Cl2.
Chem. Eur. J. 2011, 17, 9762 – 9776
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9773

B. M. Trost et al.
mixture: 0.7m). The reaction was stirred until the ynoate was consumed,
palladium trifluoroacetate (6 mol%) was then added and the reaction
stirred until the cross-coupling product was consumed. The solvent was
removed in vacuo and the residue was purified by flash column chroma-
tography on Florisil eluting with a diethyl ether/petroleum ether mixture.
1002, 925, 854, 838, 776 cmÀ1; elemental analysis calcd for C25H44O6Si: C
64.06, H 9.46; found: C 63.92, H 9.65.
Synthesis of 36: Trifluoroacetic anhydride (TFAA, 0.99 mL, 7.05 mmol)
was added to
7.49 mmol) in CH3CN (23.4 mL) at 08C and stirred for 50 min to obtain
homogeneous trifluoroperacetic acid solution. Na2HPO4 (1.33 g,
9.4 mmol) was added to
solution of 12d (1.1 g, ꢀ50% purity,
a suspension of urea hydrogen peroxide (0.705 g,
a
Synthesis of 13b: Ketone 24 (4.60 g, 20 mmol) in THF (20 mL) was
added to a solution of LDA [prepared by addition of nBuLi (13.1 mL,
21 mmol, 1.6m in hexanes) to diisopropylamine (2.94 mL, 21 mmol) at
À788C and warming to RT for 1 h] in THF (40 mL) at À788C. After the
solution was stirred for 1 h at this temperature, diethyl chlorophosphate
(3.18 mL, 22 mmol) was added. After the reaction mixture was warmed
to RT over 90 min, it was added to another solution of LDA (45 mmol,
prepared as above) in THF (40 mL) at À788C. Stirring was continued at
this temperature for 1 h. After additional stirring for 1 h at RT the reac-
tion mixture was quenched by addition of saturated aqueous NH4Cl
(50 mL). Extraction with Et2O (3ꢃ100 mL), washing of the combined or-
ganic phases with saturated aqueous NaHCO3, drying over MgSO4, filtra-
tion, and evaporation of the solvents afforded a crude oil. Flash column
chromatography on silica gel (petroleum ether/Et2O, 50:1) afforded 13b
(2.64 g, 62%) as a colorless oil. 1H NMR (300 MHz, CDCl3): d=3.46 (s,
2H), 2.06 (s, 1H), 1.19 (s, 6H), 0.90 (s, 9H), 0.06 ppm (s, 6H); 13C NMR
(75 MHz, CDCl3): d=90.6, 71.1, 68.0, 33.3, 25.9, 25.3, 18.3, À5.4 ppm; IR
(film): n˜ =3313, 2957, 2931, 2899, 2859, 1472, 1410, 1391, 1362, 1254,
1106, 838, 776, 632 cmÀ1; elemental analysis calcd for C12H24OSi: C 67.86,
H 11.39; found: C 67.97, H 11.60.
a
1.18 mmol) in a mixture of CH2Cl2 (23.4 mL) and CH3OH (23.4 mL) at
08C. The suspension was stirred for 5 min before being treated with the
above prepared trifluoroperacetic acid solution. This mixture was stirred
at 08C for 40 min and then quenched with saturated Na2S2O3. The mix-
ture was diluted with saturated aqueous NaHCO3 and extracted with
CH2Cl2. The organic extracts were dried over MgSO4, filtered, and con-
centrated. The residue was subjected to flash column chromatography on
silica gel to give impure 36 (0.94 g). An analytically pure sample was ob-
tained by further column chromatography on silica gel. Rf =0.26 (petrole-
um ether/Et2O, 2:1); [a]2D9 =À5.7 (c=1.20 in CH2Cl2); 1H NMR
(400 MHz, C6D6): d=6.64 (t, J=1.8 Hz, 1H), 4.40 (dd, J=13.7, 2.3 Hz,
1H), 4.35 (dd, J=7.2, 1.8 Hz, 1H), 3.88–3.82 (m, 1H), 3.81 (d, J=9.9 Hz,
1H), 3.64 (d, J=7.2 Hz, 1H), 3.52–3.46 (m, 1H), 3.43 (d, J=8.2 Hz, 1H),
3.39 (s, 3H), 3.28–3.25 (m, 1H), 3.26 (s, 3H), 1.83–1.76 (m, 1H), 1.46–
1.36 (m, 2H), 1.36 (s, 6H), 1.16 (s, 3H), 1.15 (s, 3H), 1.09 (d, J=6.0 Hz,
3H), 0.95 (s, 9H), 0.03 (s, 3H), 0.02 ppm (s, 3H); 13C NMR (100 MHz,
C6D6): d=167.3, 158.5, 127.9, 113.2, 107.8, 103.7, 78.7, 77.1, 73.6, 69.4,
68.7, 51.2, 50.5, 45.8, 38.4, 35.4, 27.5, 27.4, 26.0, 25.9, 22.9, 22.3, 18.5, 17.1,
À5.5, À5.6 ppm; IR (film): n˜ =3339, 2949, 2886, 2859, 1714, 1660, 1474,
1438, 1371, 1364, 1252, 1225, 1166, 1089, 999, 836, 772 cmÀ1; HRMS: m/z
calcd for C26H48O8Si: 539.3016 [M+Na]+; found: 539.3018.
Synthesis of 18c: HF (aq. 40%, 4 mL) was added to a solution of silyleth-
er 18a (2.52 g, 5.85 mmol) in acetonitrile (60 mL). After stirring for 4 h
at RT, the reaction was quenched by careful addition of solid NaHCO3.
Dilution with CH2Cl2 (250 mL), drying over MgSO4, filtration through a
pad of Celite and evaporation of the solvents afforded a crude oil that
was used directly in the next reaction. This oil was dissolved in CH2Cl2
(50 mL) and then 2,2-dimethoxypropane (1.44 mL, 11.7 mmol) and p-tol-
uenesulfonic acid (111 mg, 0.585 mmol) were added. The reaction mix-
ture was stirred for 12 h at RT and quenched by addition of triethylamine
(1 mL). Evaporation of the solvents and flash column chromatography
on silica gel (petroleum ether/EtOAc, 2:1) afforded 18c (1.07 g, 72%,
over the 2 steps) as a colorless oil. [a]2D3 =18.3 (c=1.20 in CH2Cl2);
1H NMR (400 MHz, CDCl3): d=4.17–4.09 (m, 1H), 3.82–3.74 (m, 2H),
3.77 (s, 3H), 3.05 (d, J=5.0 Hz, 1H), 2.59 (d, J=6.2 Hz, 2H), 1.89–1.82
(m, 1H), 1.76–1.69 (m, 1H), 1.40 (s, 3H), 1.39 (s, 3H), 1.27 ppm (d, J=
5.7 Hz, 3H); 13C NMR (100 MHz, CDCl3): d=153.9, 108.2, 85.8, 79.0,
76.4, 74.6, 67.0, 52.6, 37.1, 27.4, 27.2, 27.1, 16.9 ppm; IR (film): n˜ =3454,
2986, 2937, 2240, 1716, 1436, 1381, 1257, 1172, 1076, 999, 931, 869, 836,
754 cmÀ1; elemental analysis calcd for C13H20O5: C 60.92, H 7.87; found:
C 60.87; H 7.61.
Synthesis of 37: The Dess–Martin periodinane (DMP) was added in two
portions (1.16 g and 0.39 g) to a solution of impure 36 (0.94 g) and pyri-
dine (1.47 mL) in CH2Cl2 (30 mL). The reaction mixture was stirred at
RT for 1 h and then quenched with a mixture of saturated aqueous
NaHCO3 and saturated aqueous Na2S2O3 (1:1). The mixture was extract-
ed with CH2Cl2. The combined organic extracts were dried over Na2SO4,
filtered, and concentrated. The residue was subjected to flash column
chromatography on silica gel to give impure ketone 37 (0.39 g). An ana-
lytically pure sample was obtained by further column chromatography on
silica gel. Rf =0.32 (petroleum ether/Et2O, 4:1); [a]2D4 =À54.1 (c=0.84 in
CH2Cl2); 1H NMR (400 MHz, C6D6): d=6.89–6.87 (m, 1H), 4.17–4.09
(m, 1H), 3.92–3.85 (m, 2H), 3.55 (d, J=10.0 Hz, 1H), 3.54–3.48 (m, 1H),
3.45–3.39 (d, J=17.8 Hz, 1H), 3.34 (s, 3H), 3.30 (s, 3H), 2.74 (ddd, J=
17.8, 12.7, 3.4 Hz, 1H), 1.42–1.22 (m, 2H), 1.35 (s, 3H), 1.34 (s, 3H), 1.26
(s, 3H), 1.14 (s, 3H), 1.12 (s, 3H), 0.91 (s, 9H), 0.01 (s, 3H), 0.00 ppm (s,
3H); 13C NMR (100 MHz, C6D6): d=195.9, 166.2, 148.3, 122.4, 108.0,
104.7, 78.4, 77.2, 70.3, 68.5, 52.0, 51.2, 46.6, 38.7, 36.1, 27.5, 27.4, 26.2,
26.0, 20.3, 19.9, 18.7, 17.1, À5.4, À5.6 ppm; IR (film): 2932, 2858, 1727,
1710, 1634, 1471, 1435, 1379, 1252, 1215, 1171, 1130, 1093, 1043, 1004,
936, 838, 779 cmÀ1; elemental analysis calcd for C26H46O8Si: C 60.67, H
9.01; found: C 60.8, H 8.97.
Synthesis of 12d: Pd
(6.2 mg, 0.014 mmol, 4 mol%) were stirred for 30 min in benzene
(0.5 mL) to obtain homogenous solution. Alkyne 13b (82 mg,
ACHTUNGTRENNUNG(OAc)2 (3.1 mg, 0.014 mmol, 4 mol%) and TDMPP
a
0.38 mmol) and ynoate 18c (87 mg, 0.34 mmol) are then added. The reac-
tion mixture was stirred for 1 day at RT, that is, until complete conver-
sion was observed (as judged by thin layer chromatography). Pd-
ACHTUNGTRENNUNG(OCOCF3)2 (8.7 mg, 0.027 mmol, 8 mol%) was then added and stirring
was continued for 2 days. Evaporation of the solvent and flash column
chromatography on silica gel (petroleum ether/Et2O, 10:1, 2% triethyl-
Synthesis of 38: CeCl3·7H2O (141 mg, 0.38 mmol) was added to a solution
of the impure ketone 37 (0.39 g) in CH3OH (15 mL) at À308C. The mix-
ture was stirred for 20 min before being treated with NaBH4 (57 mg,
1.5 mmol). The yellow solution turned clear within 10 min. After stirring
for another 10 min at À308C, brine was added. The mixture was warmed
to room temperature and extracted with EtOAc. The combined organic
extracts were dried over MgSO4, filtered, and concentrated. The residue
was subjected to flash column chromatography on silica gel to give
impure alcohol 38 (0.28 g). An analytically pure sample was obtained by
further column chromatography on silica gel. Rf =0.20 (petroleum ether/
ether, 2:1); [a]2D6 =À9.8 (c=0.98 in CH2Cl2); 1H NMR (400 MHz, C6D6):
d=6.57 (m, 1H), 5.89 (d, J=7.4 Hz, 1H), 4.62 (d, J=7.4 Hz, 1H), 4.21–
4.14 (m, 1H), 3.98 (d, J=10.0 Hz, 1H), 3.92–3.86 (m, 1H), 3.55–3.48 (m,
1H), 3.43 (s, 3H), 3.40–3.34 (m, 1H), 3.30 (s, 3H), 2.91–2.82 (m, 1H),
1.43–1.26 (m, 3H), 1.37 (s, 6H), 1.13 (d, J=5.9 Hz, 3H), 1.06 (s, 3H),
1.05 (s, 3H), 0.94 (s, 9H), 0.00 (s, 3H), À0.01 ppm (s, 3H); 13C NMR
(100 MHz, C6D6): d=167.0, 161.0, 127.9, 113.6, 107.8, 104.2, 78.7, 77.2,
70.0, 69.8, 68.2, 50.5, 49.7, 47.2, 39.0, 36.2, 27.5, 27.5, 25.9, 22.0, 21.5, 18.4,
17.1, À5.6, À5.8 ppm; IR (film): 3339, 2931, 2859, 1719, 1469, 1455, 1433,
ACHTUNGTRENNUNGamine) afforded 12d (90 mg, 55%) as a pale yellow oil. Minor impurities,
the double-bond isomer and the 5-exo isomer, were detected in the NMR
spectrum. Rf =0.32 (petroleum ether/Et2O, 4:1); [a]2D6 =14.3 (c=1.82 in
CH2Cl2); 1H NMR (400 MHz, CDCl3): d=5.44 (s, 1H), 5.40 (s, 1H),
4.16–4.08 (m, 1H), 3.82–3.76 (m, 1H), 3.74–3.68 (m, 1H), 3.66 (s, 3H),
3.56 (dd, J=17.2, 3.2 Hz, 1H), 3.46 (d, J=11.2 Hz, 1H), 3.43 (d, J=
11.2 Hz, 1H), 2.45 (ddd, J=17.2, 12.0, 2.0 Hz, 1H), 1.86 (ddd, J=14.0,
10.0, 2.0 Hz, 1H), 1.70 (ddd, J=14.0, 9.6, 2.8 Hz, 1H), 1.38 (s, 3H), 1.37
(s, 3H), 1.26 (d, J=6.4 Hz, 3H), 1.07 (s, 3H), 1.06 (s, 3H), 0.86 (s, 9H),
0.00 ppm (s, 6H); 13C NMR (100 MHz, CDCl3): d=168.5, 167.9, 150.3,
108.0, 107.8, 102.0, 78.4, 76.8, 73.3, 69.4, 50.7, 41.7, 37.5, 31.8, 27.3, 27.2,
25.8, 22.5, 22.4, 18.2, 17.1, À5.4, À5.6 ppm; IR (film): n˜ =2949, 2922,
2858, 1710, 1644, 1472, 1434, 1379, 1327, 1248, 1226, 1153, 1096, 1045,
9774
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9762 – 9776

Synthesis of Bryostatin Ring C
FULL PAPER
1374, 1223, 1170, 1091, 997, 948, 838, 780 cmÀ1; elemental analysis calcd
for C26H48O8Si: C 60.43, H 9.36; found: C 60.51, H 9.18.
[3] For some recent reviews, see: a) K. J. Hale, S. Manaviazar, Chem.
Asian J. 2010, 5, 704–754, and reviews cited therein; b) P. A.
Wender, J. L. Baryza, M. K. Hilinski, J. C. Horan, C. Kan, V. A.
Verma in Drug Discovery Research: New Frontiers in the Post-Ge-
nomic Era (Ed.: Z. Huang), Wiley, New York, 2007, pp. 127–162;
c) R. Mutter, M. Wills, Bioorg. Med. Chem. 2000, 8, 1841–1860.
[4] A. S. Kraft, J. Natl. Canc. Inst. 1993, 85, 1790.
Synthesis of 39: DMAP (6 mg, 0.049 mmol), pyridine (1.1 mL), and
acetic anhydride (0.55 mL) were sequentially added to a solution of alco-
hol 38 (280 mg) in CH2Cl2 (6 mL). The mixture was stirred at room tem-
perature for 3.5 h, and then directly purified by flash column chromatog-
raphy (Et2O/petroleum ether, 1:3) to give acetate 39 (260 mg, 45% yield
over four steps) as a clear oil. Rf =0.39 (Et2O/petroleum ether, 1:3);
[a]2D3 =À2.6 (c=1.1 in CH2Cl2); 1H NMR (500 MHz, CDCl3): d=5.88 (s,
1H), 5.58 (s, 1H), 4.13 (t, J=10.7 Hz, 1H), 3.85 (td, J=2.2, 9.2 Hz, 1H),
3.69 (s, 3H), 3.69 (m, 1H), 3.56 (d, J=9.4 Hz, 1H), 3.54 (d, J=9.4 Hz,
1H), 3.44 (dd, J=1.5, 14.1 Hz, 1H), 3.34 (s, 3H), 2.39 (t, J=13.3 Hz,
1H), 2.11 (s, 3H), 1.77–1.72 (m, 1H), 1.70–1.64 (m, 1H), 1.36 (s, 3H),
1.35 (s, 3H), 1.27 (d, J=6.0 Hz, 1H), 1.01 (s, 3H), 0.97 (s, 3H), 0.88 (s,
9H), 0.01 ppm (s, 6H); 13C NMR (125 MHz, CDCl3): d=168.9, 166.5,
152.7, 116.4, 107.8, 102.7, 78.2, 76.9, 72.2, 68.1, 67.3, 51.1, 50.8, 47.1, 38.4,
33.4, 27.3, 27.1, 25.9, 21.1, 20.6, 18.4, 17.0, À5.5 ppm (2); IR (film): 2988,
2955, 2929, 1755, 1721, 1658, 1468, 1434, 1227, 1155, 1092, 855, 834,
775 cmÀ1; HRMS calcd for C25H45O7Si: 485.2934 [MÀC2H3O2]+; found:
485.2952.
[5] G. R. Pettit, J. Nat. Prod. 1996, 59, 812.
[6] R. M. Stone, Leuk. Res. 1997, 21, 399.
[7] L. M. Schuchter, A. H. Esa, W. S. May, M. K. Laulis, G. R. Pettit,
A. D. Hess, Cancer Res. 1991, 51, 682.
[8] a) For a review, see: D. J. Newman, G. M. Cragg, J. Nat. Prod. 2004,
67, 1216; b) for ongoing studies, see http://clinicaltrials.gov/.
[9] A. Dowlati, H. M. Lazarus, P. Hartman, J. W. Jacobberger, C. Whit-
AHCTUNGERTGaNNUN cre, S. L. Gerson, P. Ksenich, B. W. Cooper, P. S. Frisa, M. Gottlieb,
A. J. Murgo, S. C. Remick, Clin. Cancer Res. 2003, 9, 5929; A. C.
Pavlick, J. Wu, J Roberts, M. A. Rosenthal, A. Hamilton, S. Wadler,
K. Farrell, M. Carr, D. Fry, A. J. Murgo, R. Oratz, H. Hochster, L.
Liebes, F. Muggia, Cancer Chemother. Pharmacol. 2009, 64, 803.
[10] a) R. J. Gulakowski, J. B. McMahon, R. W. Buckheit, Jr., K. R. Gus-
tafson, M. R. Boyd, Antiviral Res. 1997, 33, 87; b) K. Eisemann, A.
Totola, K. Jurcic, G. R. Pettit, H. Wagner, Pharm. Pharmacol. Lett.
1995, 1, 45; c) A. S. Kraft, V. Adler, P. Hall, G. R. Pettit, W. H. Ben-
jamin, Jr., D. E. Briles, Cancer Res. 1992, 52, 2143; d) T. M. Tuttle,
T. H. Inge, D. S. Lind, H. D. Bear, Surgical Oncology 1992, 1, 299.
[11] a) C. Scheid, J. Prendiville, G. Jayson, D. Crowther, B. Fox, G. R.
Pettit, P. L. Stern, Cancer Immunol. Immunother. 1994, 39, 223;
b) G. Trenn, G. R. Pettit, H. Takayama, J. Hu-Li, M. V. Sitkovsky, J.
Immunol. 1988, 140, 433; c) A. D. Hess, M. K. Silanskis, A. H. Esa,
G. R. Pettit, W. Stratford May, J. Immunol. 1988, 141, 3263; d) A. H.
Esa, J. T. Warren, A. D. Hess, W. Stratford May, Res. Immunol.
1995, 146, 351.
[12] a) R. Etcheberrigaray, M. Tan, I. Dewachter, C. Kuiperi, I. Van der
Auwera, S. Wera, L. Qiao, B. Bank, T. J. Nelson, A. P. Kozikowski,
F. Van Leuven, D. L. Alkon, Proc. Natl. Acad. Sci. USA 2004, 101,
11141; b) D. L. Alkon, M.-K. Sun, T. J. Nelson, Trends Pharmacol.
Sci. 2007, 28, 51; c) D. L. Alkon, H. Kuzirian, A. Epstein, M. C. Ben-
nett, T. J. Nelson, Proc. Natl. Acad. Sci. USA 2005, 102, 16432;
d) M.-K. Sun, D. L. Alkon, Eur. J. Pharmacol. 2005, 512, 43; e) J.
Hongpaisan, D. L. Alkon, Proc. Natl. Acad. Sci. USA 2007, 104,
19571.
[13] M.-K. Sun, J. Hongpaisan, T. J. Nelson, D. L. Alkon, Proc. Natl.
Acad. Sci. USA 2008, 105, 13620.
[14] C. M. Wendtner, B. F. Eichhorst, M. J. Hallek, Semin. Hematol.
2004, 41, 224.
[15] J. Kortmansky, G. K. Schwartz, Cancer Invest. 2003, 21, 924.
[16] G. Schwartsmann, A. B. da Rocha, J. Mattei, R. M. Lopes, Expert
Opin. Invest. Drugs 2003, 12, 1367.
[17] A. Clamp, G. C. Jayson, Anti-Cancer Drugs 2002, 13, 673.
[18] A. B. Da Rocha, R. M. Lopes, G. Schwartsmann, Curr. Opin. Phar-
macol. 2001, 1, 364.
[19] M. Jaspars, Advances in Drug Discovery Techniques (Ed.: A. L.
Harvey), Wiley, New York, 1998, p. 65.
[20] M. L. DellꢀAquilla, C. L. Harold, Y. Kamano, G. R. Pettit, P. M.
Blumberg, Cancer Res. 1988, 48, 3702.
[21] a) J. B. Smith, L. Smith, G. R. Pettit, Biochem. Biophys. Res.
Commun. 1985, 132, 939; b) D. J. DeVries, C. L. Herald, G. R.
Pettit, P. M. Blumberg, Biochem. Pharmacol. 1988, 37, 4069.
[22] R. L. Berkow, A. S. Kraft, Biochem. Biophys. Res. Commun. 1985,
131, 1109.
Acknowledgements
We thank the National Institutes of Health (GM33049) and NSF for
their generous support of our programs. We also thank the Alexander-
von-Humboldt foundation (O.R.T.), the NIH (A.J.F.), Amgen (H.Y.), and
Bristol-Myers Squibb (H.Y.) for their generous financial support. G.D. is
a Stanford Graduate Fellow. Mass spectra were provided by the Mass
Spectrometry Regional Center of the University of California, San Fran-
cisco, supported by the NIH Division of Research Resources. Palladium
and ruthenium salts were generously supplied by Johnson–Matthey.
[1] G. R. Pettit, J. F. Day, J. L. Hartwell, H. B. Wood, Nature 1970, 227,
962.
[2] a) G. R. Pettit, C. L. Herald, D. L. Doubek, D. L. Herald, E. Arnold,
J. Clardy, J. Am. Chem. Soc. 1982, 104, 6846; b) bryostatin 2: G. R.
Pettit, C. L. Herald, Y. Kamano, D. Gust, R. Aoyagi, J. Nat. Prod.
1983, 46, 528–531; c) bryostatins 1–3: G. R. Pettit, C. L. Herald, Y.
Kamano, J. Org. Chem. 1983, 48, 5354–5356; bryostatin 3: D. E.
Schaufelberger, G. N. Chmurny, J. A. Beutler, M. P. Koleck, A. B.
Alvarado, B. W. Schaufelberger, G. M. Muschik, J. Org. Chem. 1991,
56, 2895–2900; G. N. Chmurny, M. P. Koleck, B. D. Hilton, J. Org.
Chem. 1992, 57, 5260–5264; d) bryostatin 4: G. R. Pettit, Y.
Kamano, C. L. Herald, M. Tozawa, J. Am. Chem. Soc. 1984, 106,
6768–6771; e) bryostatins 5–7: G. R. Pettit, Y. Kamano, C. L.
Herald, M. Tozawa, Can. J. Chem. 1985, 63, 1204–1208; f) bryostatin
8: G. R. Pettit, Y. Kamano, R. Aoyagi, C. L. Herald, D. L. Doubek,
J. M. Schmidt, J. J. Rudloe, Tetrahedron 1985, 41, 985–994; g) bryo-
ACHTUNGTRENNUNGstatin 9: G. R. Pettit. Y. Kamano, C. L. Herald, J. Nat. Prod. 1986,
49, 661–664; h) bryostatins 10 and 11; G. R. Pettit, Y. Kamano,
C. L. Herald, J. Org. Chem. 1987, 52, 2848–2854; i) bryostatins 12
and 13: G. R. Pettit, J. E. Leet, C. L. Herald, Y. Kamano, F. E.
Boettner, L. Baczynskyj, R. A. Nieman, J. Org. Chem. 1987, 52,
2854–2860; j) bryostatins 14 and 15: G. R. Pettit, F. Gao, D. Sengup-
ta, J. C. Coll, C. L. Herald, D. L. Doubek, J. M. Schmidt, J. R. Van
Camp, J. J. Rudloe, R. A. Nieman, Tetrahedron 1991, 47, 3601–3614;
k) bryoACHTUNGTRENNUNGstatins 16–18: G. R. Pettit, F. Gao, P. M. Blumberg, C. L.
[23] M. Kageyama, T. Tamura, M. H. Nantz, J. C. Roberts, P. Somfai,
D. C. Whritenour, S. Masamune, J. Am. Chem. Soc. 1990, 112, 7407,
and references cited therein.
Herald, J. C. Coll, Y. Kamano, N.E Lewin, J. M. Schmidt, J.-C. Cha-
puis, J. Nat. Prod. 1996, 59, 286–289; l) bryostatin 19: H. Lin, X.
Yao, Y. Yi, X. Li, H. Wu, Zhongguo Haiyang Yaowu 1998, 17, 1–6;
[24] a) D. A. Evans, P. H. Carter, E. M. Carreira, J. A. Prunet, A. B.
Charette, M. Lautens, Angew. Chem. 1998, 110, 2526; Angew. Chem.
Int. Ed. 1998, 37, 2354; b) D. A. Evans, P. H. Carter, E. M. Carreira,
A. B. Charette, J. A. Prunet, M. Lautens, J. Am. Chem. Soc. 1999,
121, 7540, and references cited therein.
m) bryoACHTUNGTRENNUNGstatin 20: N. Lopanik, K. R. Gustafson, N. Lindquist, J. Nat.
Prod. 2004, 67, 1412–1414; n) for some of our recent efforts, see:
B. M. Trost, H. Yang, C. S. Brindle, G. Dong, Chem. Eur. J. 2011, 17,
9777–9788; B. M. Trost, H. Yang, G. Dong, Chem. Eur. J. 2011, 17,
9789–9805.
Chem. Eur. J. 2011, 17, 9762 – 9776
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9775

B. M. Trost et al.
[25] a) K. Ohmori, Y. Ogawa, T. Obitsu, Y. Ishikawa, S. Nishiyama, S.
Yamamura, Angew. Chem. 2000, 112, 2376; Angew. Chem. Int. Ed.
2000, 39, 2290; b) for a review of this synthesis, see: K. Ohmori,
Bull. Chem. Soc. Jpn. 2004, 77, 875–885.
[26] G. R. Pettit, D. Sengupta, C. L. Herald, N. A. Sharkey, P. M. Blum-
berg, Can. J. Chem. 1991, 69, 856.
[27] For a formal synthesis of bryostatin 7, see: S. Manaviazar, M. Friger-
io, G. S. Bhatia, M. G. Hummersone, A. E. Aliev, K. J. Hale, Org.
Lett. 2006, 8, 4477, and references therein.
[40] H. Becker, K. B. Sharpless, Angew. Chem. Int. Ed. Eng. 1996, 35,
448.
[41] R. D. Cink, C. J. Forsyth, J. Org. Chem. 1995, 60, 8122.
[42] M. Boeykens, N. De Kimpe, T. K. Abbaspour, J. Org. Chem. 1994,
59, 6973.
[43] E. Negishi, A. O. King, W. L. Klima, J. Org. Chem. 1980, 45, 2526.
[44] The synthesis of epoxide 28 followed a route first reported by Roy
and co-workers in connection with their synthetic studies on bryo-
AHCTUNGTREGsNNNU tatin ring C, see: R. Roy, A. W. Rey, M. Charron, R. Molino, J.
[28] For previous synthetic strategies directed towards ring C of bryosta-
tins, see: a) A.P Green, S. Hardy, A. T. L. Lee, E. J. Thomas, Phyto-
chem. Rev. 2010, 9, 501–513; b) J. V. Allen, A. P. Green, S. Hardy,
N. M. Heron, A. T. L. Lee, E. J. Thomas, Tetrahedron Lett. 2008, 49,
6352–6355, and references cited therein; c) P. A. Wender, B. Lippa,
Tetrahedron Lett. 2000, 41, 1007, and references cited therein;
d) G. E. Keck, A. P. Truong, Org. Lett. 2005, 7, 2149–2152; e) G. E.
Keck, T. Yu, M. D. McLaws, J. Org. Chem. 2005, 70, 2543–2550, and
references cited therein; f) E. A. Voight, P. A. Roethle, S. D. Burke,
J. Org. Chem. 2004, 69, 4534–4537; g) J. De Brabander, A. Kulkarni,
R. Garcia-Lopez, M. Vandewalle, Tetrahedron: Asymmetry 1997, 8,
1721.
Chem. Soc. Chem. Commun. 1989, 1308.
[45] K. Bock, I. Lundt, C. Pedersen, Carbohydr. Res. 1979, 68, 313.
[46] Y. Kataoka, O. Matsumoto, K. Tani, Chem. Lett. 1996, 727.
[47] a) A. Fꢄrstner, F. Stelzer, H. Szillat, J. Am. Chem. Soc. 2001, 123,
11863; b) M. Mꢅndez, M. P. Munoz, A. M. Echavarren, J. Am.
Chem. Soc. 2000, 122, 11549.
[48] K. Utimoto, Pure Appl. Chem. 1983, 55, 1845.
[49] a) P. von Zezschwitz, K. Voigt, A. Lansky, M. Noltemeyer, A. de
Meijere, J. Org. Chem. 1999, 64, 3806.
[50] a) For a recent review, see: H. Tohma, Y. Kita, Adv. Synth. Catal.
2004, 346, 111; b) For a review, see: T. T. Tidwell, Org. React. 1990,
39, 297.
[29] B. M. Trost, M. C. McIntosh, J. Am. Chem. Soc. 1995, 117, 7255.
[30] Ynoates were synthesized by opening the appropriate epoxide with
ethyl propiolate under BF3·OEt2 catalysis (see the Experimental
Section).
[31] C. Gravier-Pelletier, Y. Le Merrer, J.-C. Depezay, Tetrahedron 1995,
51, 1663.
[32] For the preparation of alkyne 2l, see: J. Ohshita, K. Furumori, A.
Matsuguchi, M. Ishikawa, J. Org. Chem. 1990, 55, 3277.
[33] B. M. Trost, P. J. Metzner, J. Am. Chem. Soc. 1980, 102, 3572.
[34] The structure of the isomerized exocyclic enoate (E geometry) was
assigned on the basis of 1H NMR data: the chemical shift of the
vinyl ring proton was shifted downfield relative to the Z-enoate,
while the ring protons allylic to the exocyclic enoate were shifted
upfield relative to the Z-enoate.
[51] D. B. Dess, J. C. Martin, J. Am. Chem. Soc. 1991, 113, 7277.
[52] J. L. Luche, J. Am. Chem. Soc. 1978, 100, 2226.
[53] P. von Zezschwitz, K. Voigt, A. Lansky, M. Noltemeyer, A. de Mei-
jere, J. Org. Chem. 1999, 64, 3806; b) W. A. Herrmann, R. W. Fisch-
er, D. W Marz, Angew. Chem. 1991, 103, 1706; Angew. Chem. Int.
Ed. Engl. 1991, 30, 1638; c) T. R. Boehlow, C. D. Spilling, Tetrahe-
dron Lett. 1996, 37, 2717d) G. Soldaini, F. Cardona, A. Goti, Tetra-
hedron Lett. 2003, 44, 5589; e) S. Tanaka, H. Yamamoto, H. Nozaki,
K. B. Sharpless, R. C. Michaelson, J. D. Cutting, J. Am. Chem. Soc.
1974, 96, 5254; f) W. Adam, C. R. Saha-Mçller, P. A. Ganeshpure,
Chem. Rev. 2001, 101, 3499; g) D. Yang, M.-K. Wong, Y.-C. Yip, J.
Org. Chem. 1995, 60, 3887.
[54] Reduction of a similar substrate has been reported by Evans et al.,
see Ref. [24].
[35] O. Haglund, M. Nilsson, Synlett 1991, 723.
[36] F. Barbot, Ph. Miginiac, J. Organomet. Chem. 1992, 440, 249; see the
Supporting Information for modifications.
[55] The analytically pure alcohol 38 could be obtained after epoxida-
tion; however, careful and even repetitive chromatography was nec-
essary, during which the compound decomposed, resulting in a low
yield. Thus, a practical solution was developed as follows: carrying
out this four-step transformation in one day by using rapid silica gel
flash chromatographic purifications to remove as many impurities as
possible after each step, and finally purifying acetate 39 by careful
silica gel column chromatography.
[37] For a recent review on the Mitsunobu reaction, see: K. C. Ku-
ACHTUNGTRENNUNGmara Swamy, N. N. Bhuvan Kumar, E. Balaraman, K. V. P. Pavan
Kumar, Chem. Rev. 2009, 109, 2551.
[38] a) K. J. Hale, J. A. Lennon, S. Manaviazar, M. H. Javaid, C. J.
Hobbs, Tetrahedron Lett. 1995, 36, 1359; b) D. Xu, G. A. Crispino,
K. B. Sharpless, J. Am. Chem. Soc. 1992, 114, 7570.
[39] K. B. Sharpless, W. Amberg, Y. L. Bennani, G. A. Crispino, J. Har-
tung, K.-S. Jeong, H.-L. Kwong, K. Morikawa, Z.-M. Wang, D. Xu,
X.-L. Zhang, J. Org. Chem. 1992, 57, 2768.
Received: October 8, 2010
Revised: May 12, 2011
Published online: July 26, 2011
9776
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 9762 – 9776