Concise total synthesis of (+)-brefeldin a: A combined β-lactone/ cross-metathesis-based strategy

Article abstract of DOI:10.1021/ol026438g

equation presented A highly convergent total synthesis of (+)-brefeldin A that relies on a diastereoselective, β-lactone-based cyclopentane synthesis combined with complex cross-metathesis reactions is described. The utility of β-lactones for natural product synthesis and the versatility of cross-metathesis in this context were demonstrated, including the tolerance of an epimerizable aldehyde and a β-lactone.

Full text of DOI:10.1021/ol026438g

ORGANIC
LETTERS
2002
Vol. 4, No. 19
3231-3234
Concise Total Synthesis of (+)-Brefeldin
A: A Combined â-Lactone/
Cross-Metathesis-Based Strategy
Yingcai Wang and Daniel Romo*
Department of Chemistry, Texas A&M UniVersity, P.O. Box 30012,
College Station, Texas 77842-3012
Romo@mail.chem.tamu.edu
Received June 28, 2002
ABSTRACT
A highly convergent total synthesis of (+)-brefeldin A that relies on a diastereoselective, â-lactone-based cyclopentane synthesis combined
with complex cross-metathesis reactions is described. The utility of â-lactones for natural product synthesis and the versatility of cross-
metathesis in this context were demonstrated, including the tolerance of an epimerizable aldehyde and a â-lactone.
The fungal metabolite, (+)-brefeldin A (BFA, 1), displays
potent antitumor, antifungal, antiviral, antimitotic, and im-
munosuppressive activities.1 BFA’s ability to inhibit secretory
pathways in both animal and plant cells through a process
involving disruption of the Golgi complex2 and more recently
to induce apoptosis3 continues to make it an important probe
for cell biology. Recently, one of BFA’s effects at the
molecular level was found to involve the inhibition of
activation of the GTPase, adenosine diphosphate-ribosylation
factor 1 (Arf1) by inducing a nonproductive complex with
its membrane-associated GTP exchange factor (GEF).4 These
unique and potent bioactivities have stimulated extensive
synthetic efforts from several laboratories over the past 25
years.5 BFA continues to be an excellent proving ground
for new synthetic methodologies, in particular, stereoselective
cyclopentane construction.
Our interest in a total synthesis of BFA originated from a
unique â-lactone-based approach to cyclopentanes involving
intramolecular allylsilane additions (Figure 1).6 Our approach
Figure 1. Strategy for (+)-brefeldin A (1) synthesis showing
(1) (a) Nuchtern, J. G.; Bonifacino, J. S.; Biddison, W. E.; Klausner, R.
D. Nature 1989, 339, 223-226. (b) Klausner, R. D.; Donaldson, J. G.;
Lippincott-Schwartz, J. J. Cell Biol. 1992, 116, 1071-1080.
(2) Jackson, C. L. Subcell. Biochem. 2000, 34, 233-272.
(3) (a) Fox, B. M.; Vroman, J. A.; Fanwick, P. E.; Cushman, M. J. Med.
Chem. 2001, 44, 3915-3924. (b) Niwa, S.; Ishibashi, O.; Inui, T. Life Sci.
2001, 70, 315-324. (c) Zhu, J.-W.; Nagasawa, H.; Nagura, F.; Mhoamad,
S. B.; Uto, Y.; Ohkura, K.; Hori, H. Bioorg. Med. Chem. 2000, 8, 455-
463. (d) Shao, R.-G.; Shimizu, T.; Pommier, Y. Exp. Cell. Res. 1996, 227,
190-196.
required building blocks.
to BFA also lent itself to further exploration of cross-
metathesis (CM) processes with highly functionalized sub-
strates. Furthermore, we were interested in utilizing the con-
vergency of our synthetic strategy for the generation of a
focused collection of BFA-like molecules that may display
biological activities related to known and potentially un-
(4) Robineau, S.; Chabre, M.; Antonny, B. Proc. Natl. Acad. Sci. 2000,
97, 9913-9918.
10.1021/ol026438g CCC: $22.00 © 2002 American Chemical Society
Published on Web 08/21/2002

known prostaglandins.7 Stereoisomer 2 was employed as an
intermediate to BFA due to the syn relative stereoselectivity
obtained during the tandem Mukaiyama aldol-lactonization
(TMAL) reaction leading to â-lactone 4 in addition to the
stereochemical requirements for high selectivity in the
cyclopentane-forming step (vide infra).
was improved by increasing the bulkiness of the silyl
protecting groups on both the aldehyde and ketene acetal.
TMAL reaction of aldehyde 11 and ketene acetal 12
delivered â-lactone 13 and the anti diastereomer (not shown)
with high internal (>19:1 trans:cis) and good relative (5:1
syn:anti) stereoselectivities (Scheme 2). Isolation of the major
As a first step toward applying the â-lactone-based
synthesis of cyclopentanes to BFA, we studied the factors
that govern the selectivity of this process employing R-meth-
yl-â-lactones 9 (Scheme 1). Specifically, the impact of
Scheme 2. Synthesis of the Cyclopentane of BFAa
Scheme 1. Effect of â-Lactone and Allylsilane
Stereochemistry on the Stereoselectivity of Cyclopentane
Formation
â-lactone stereochemistry and allylsilane geometry was
investigated. The requisite substrates were prepared using
diastereoselective approaches to both cis- and trans-â-
lactones developed in our laboratory.8 In general, trans-â-
lactones gave a higher (10:1, anti:syn) and opposite diaste-
reoselectivity compared to cis-â-lactones (2:3, anti:syn) upon
cyclizations to cyclopentanes. On the other hand, the
allylsilane geometry had no effect on the stereochemical
outcome. After screening several Lewis acids, TiCl4 was
found to provide the highest yields for this cyclization.
Having determined that trans-â-lactones gave the highest
selectivity for cyclopentane construction, we studied the
TMAL reaction for the synthesis of the requisite â-lactone
13 from aldehyde 11.8a-c This aldehyde was prepared in four
steps from commercially available chloroester 5.9 Initially,
the relative stereoselectivity of the TMAL process was poor
(1.3:1 syn:anti) employing the triethylsilyl thiopyridyl ketene
acetal, which had previously proven to be successful for
R-alkyl ketene acetals.8a Ultimately, the diastereoselectivity
a Reaction conditions: (a) (i) NaI, acetone, reflux; (ii) TIPSOTf,
2,6-lutidine, CH2Cl2, 0 °C; (iii) CuI, CH2dCH2MgBr, HMPA, THF,
-20 °C; (iv) DIBAl-H, CH2Cl2, -78 °C (87% overall, four steps).
(b) ZnCl2, CH2Cl2, 25 °C (68%, dr 5:1). (c) 2.5 mol % 14, 4.0
equiv CH2dCH2CH2SiMe3, CH2Cl2, reflux, 80%. (d) TiCl4, CH2Cl2,
-78 °C, 90% (dr > 19:1). (e) (i) TMSCHN2, benzene/MeOH, 0
°C; (ii) DIBAl-H, CH2Cl2, -78f0 °C; (iii) Swern oxidation (82%
overall, three steps).
â-lactone diastereomer 13 was accomplished by preparative
MPLC. The internal stereochemistry of the major diastero-
meric â-lactone 13 was determined to be trans by 1H NMR
coupling constant analysis (JHa,Hb ) 3.6 Hz),10 while the
relative stereochemistry was ultimately determined to be syn
by conversion to BFA (vide infra).
The relative selectivity obtained in this TMAL process is
opposite to that predicted on the basis of Evans’ model and
is likely due to the bulkiness of the silyl protecting groups.11
Introduction of the allylsilane moiety was accomplished by
cross-metathesis (CM) using Grubbs’ imidazolylidene cata-
lyst 14 with â-lactone 13 and allyl trimethylsilane to
smoothly deliver allylsilanes 15 as an inconsequential ∼3:1
mixture of E/Z isomers (Scheme 2).12
(5) (a) For a review describing total synthesis of the brefeldins, see:
Kobayashi, Y.; Watatani, K. J. Synth. Org. Chem. Jpn. 1997, 55, 110-
120. (b) For more recent synthetic efforts, see: Kim, D.; Lee, J.; Shim, P.
J.; Lim, J. I.; Jo, H.; Kim, S. J. Org. Chem. 2002, 67, 764-771 and refs
cited therein. (c) Suh, Y.-G.; Jung, J.-K.; Seo, S.-Y.; Min, K.-H.; Shin,
D.-Y.; Lee, Y.-S.; Kim, S.-H.; Park, H.-J. J. Org. Chem. 2002, 67, 4127-
4137. (d) Kim, D.; Lee, J.; Shim, P. J.; Lim, J. I.; Doi, T.; Kim, S. J. Org.
Chem. 2002, 67, 772-781. (e) Trost, B. M.; Crawley, M. L. J. Am. Chem.
Soc. 2002, 124, 9328-9329.
(6) Zhao, C.; Romo, D. Tetrahedron Lett. 1997, 38, 6537-6540.
(7) The structural resemblance between BFA and the prostaglandins has
been noted previously, see: (a) Mabuni, C. T.; Garlaschelli, L.; Ellison, R.
A.; Hutchinson, C. R. J. Am. Chem. Soc. 1977, 99, 7718-7720. (b) Gorst-
Allman, C. P.; Steyn, P. S. J. Chem. Soc., Perkin Trans. 1 1982, 2837-
2839.
(8) (a) Yang, H. W.; Romo, D. J. Org. Chem. 1997, 62, 4-5. (b) Yang,
H. W.; Zhao, C.; Romo, D. Tetrahedron 1997, 53, 16471-16488. (c) Yang,
H. W.; Romo, D. J. Org. Chem. 1998, 63, 1344-1347. (d) Wang, Y.; Zhao,
C.; Romo, D. Org. Lett. 1999, 1, 1197-1199.
(9) Hareau, G. P.-J.; Koiwa, M.; Hikichi, S.; Sato, F. J. Am. Chem. Soc.
1999, 121, 3640-3650.
Cyclization of â-lactone 15 with TiCl4 smoothly delivered
the desired trans-fused cyclopentane 16 as a single diaste-
reomer (>19:1, 500 MHz 1H NMR) with inversion of
(10) Mulzer, J.; Zippel, M.; Bruentrup, G.; Segner, J.; Finke, J. Liebigs
Ann. Chem. 1980, 1108-1134.
(11) Evans, D. A.; Dart, M. J.; Duffy, J. L.; Yang, M. G. J. Am. Chem.
Soc. 1996, 118, 4322-4343.
(12) Engelhardt, F. C.; Schmitt, M. J.; Taylor, R. E. Org. Lett. 2001, 3,
2209-2212 and refs cited therein. We thank Prof. Taylor for providing
procedures for this improved CM with allylsilanes prior to publication.
3232
Org. Lett., Vol. 4, No. 19, 2002

configuration at the â-carbon as previously observed (Scheme
2).6 Esterification followed by a reduction/oxidation sequence
gave aldehyde 17. Overall, this sequence from aldehyde 11
introduces three new stereogenic centers through effective
use of substrate control.
Table 1. Optimization of Cross-Metathesis of Alkenes 3
and 21
The application of olefin metathesis in natural product
synthesis has primarily focused on RCM, and this has been
particularly effective for medium-sized ring and macrocycle
synthesis,13 including BFA.5b In contrast, far fewer studies
of CM reactions applied to natural product synthesis have
appeared.14 As a means to maximize convergence in the
proposed CM, we were attracted to the possibility of
employing the fully functionalized cyclopentane 17 and
phosphonate 3 equipped for direct intramolecular HWE
olefination following coupling. A model study of the
proposed CM was undertaken using vinylcyclopentane (18)
and phosphonate 3 (Scheme 3). The latter compound was
21
3
addition
recovered 21 yield of 22
entry (equiv) (equiv) methoda catalyst
(%)b
(%)b
1
2
3
4
5
6
1.0 1.0
4.0 1.0
1.0 3.0
1.0 4.0
1.0 2.0
1.0 2.0e
A
A
B
C
A
A
2 0
2 0
2 0
2 0
1 4
1 4
53
90c
80c
45
0
17
30d
20c
50
84
100c
0
a Addition method: (A) addition of catalyst to a mixture of 21 and 3;
(B) slow addition of 3 to a mixture of catalyst and 21; (C) portionwise
addition of 3 (1.0 equiv, then 3.0 equiv 22 h later) to a mixture of catalyst
and 21. b Refers to isolated yields. c Refers to percent conversion of 21 based
on 1H NMR (500 MHz) analysis of crude reaction mixture. d Based on
percent conversion of 3. e Homodimer of 3 was utilized.
Scheme 3. Model Studies of the Cross-Metathesisa
described above, appreciable amounts of dimerization of
vinylcyclopentane (18) were observed employing catalyst 14.
This was not observed with the more sterically demanding
vinylcyclopentane 21. On the other hand, dimerization of
phosphonate olefin 3 was found to be competitive with CM
of this olefin and vinylcyclopentane 21. However, the ability
to utilize the dimer of olefin 3 for the CM renders this process
inconsequential with catalyst 14 (Table 1, entry 6).
a Reaction conditions: (a) (i) 3-butenylmagnesium bromide, CuI
(0.1 equiv), THF, -20 °C; (ii) diethylphosphonoacetic acid, EDCI,
DMF (96% overall, two steps). (b) 2.5 mol % of 14, CH2Cl2, reflux,
44% (∼4:1 E:Z); or 2.5 mol % of RuCl2(CHC6H5)[P(C6H11)3]2 (20),
CH2Cl2, reflux, 74% (4:1 E:Z).
readily prepared in two steps from commercially available
epoxide 7. Grubbs’ first generation catalyst 20 gave superior
yields (74 vs 44%) of the desired CM product 19 (4:1 E:Z)
relative to the second-generation catalyst 14 (Scheme 3).
With these results in hand, we first attempted coupling of
the fully functionalized, enantiomeric vinyl cyclopentane 21
and phosphonate 3 using catalyst 20. However, this gave
only 17% of alkene 22 (Table 1, entry 1). Analysis of the
crude reaction mixture revealed that the major product was
the dimer formed from self-metathesis of phosphonate 3.
Neither increasing the number of equivalents of either
reaction partner nor changing the order of addition of
substrates or reagents had an appreciable effect on conversion
(Table 1, entries 2-4). Pleasingly, when the CM reaction
was performed with Grubb’s second-generation catalyst 14
and 2.0 equiv of olefin 3 (Table 1, entry 5), complete
conversion to adduct 22 was possible and proceeded in high
yield and moderate selectivity (4:1 E:Z). In model studies
In a similar manner, enantiomeric aldehyde 17 was
converted to aldehyde-phosphonate 23 via CM in 86% yield
(4:1 E/Z; Scheme 4). Intramolecular HWE olefination of this
E/Z mixture of phosphonates 23 furnished the macrocycle
Scheme 4. Completion of the Synthesisa
(13) For recent reviews on olefin metathesis, see: (a) Trnka, T. M.;
Grubbs, R. H. Acc. Chem Res. 2001, 34, 18-29. (b) Fu¨rstner, A. Angew.
Chem., Int. Ed. Engl. 2000, 39, 3012-3043. (c) Maier, M. E. Angew. Chem.,
Int. Ed. Engl. 2000, 39, 2073-2077.
(14) (a) Blackwell, H. E.; O’Leary, D. J.; Chatterjee, A. K.; Washenfelder,
R. A.; Bussmann, D. A.; Grubbs, R. H. J. Am. Chem. Soc. 2000, 122, 58-
71. (b) In (-)-prosophylline synthesis, see: Cossy, J.; Willis, C.; Bellosta,
V. Synlett 2001, 1578-1580. (c) In (-)-cylindrocyclophane synthesis,
see: Smith, A. B.; Adams, C. M.; Kozmin, S. A.; Paone, D. V. J. Am.
Chem. Soc. 2001, 123, 5925-5937.
a Reaction conditions: (a) 2.5 mol % of 14, CH2Cl2, reflux, 86%.
(b) (i) LiBr, DBU, Et3N, THF; (ii) TBAF, THF, 39%, (two steps).
(c) (i) p-NO2C6H4CO2H, DEAD, PPh3, THF, 25 °C, 70% (bis-ester);
or 0 °C, 82% (mono ester); (ii) K2CO3, MeOH/H2O, 0 °C, 90%.
Org. Lett., Vol. 4, No. 19, 2002
3233

24 after desilylation with high selectivity for the C2,C3-(E)-
olefin isomer (Scheme 4).15 The minor (Z)-C10,C11-olefin
isomer obtained during the CM reaction was readily sepa-
rated at this stage. In contrast to aldehyde 22, intramolecular
HWE olefination of the diastereomeric aldehyde 23 delivered
the macrocycle in greatly reduced yield (41 vs 86%,
respectively), indicating the importance of stereochemistry
on the efficiency of macrocyclization as has been noted
previously.15c,16
Inversion of both alcohol centers of C4,C7-bis-epi-BFA
(24) was required to access the natural product. This was a
consequence of the unusual syn relative stereoselectivity
obtained during the TMAL reaction leading to â-lactone 13
in addition to the stereochemical requirements for high
selectivity in the cyclopentane-forming step (vide supra).
Pleasingly, Mitsunobu reaction of diol 24 leading to inversion
at both C4 and C7 proceeded efficiently to deliver natural
BFA (1) after hydrolysis of the resulting p-nitrobenzoate
esters.17 All spectral data of synthetic BFA correlated well
with that of the natural product. 4-epi-BFA (25) was also
accessible by selective Mitsunobu inversion of diol 24 (0
°C) at C7 followed by hydrolysis.18
The described total synthesis of BFA constitutes one of
the most concise synthetic routes reported to date (15 steps
from chloroester 5; longest linear sequence) and employs a
unique â-lactone-based strategy to construct the cyclopentane
nucleus. An allylsilane-generating cross-metathesis and a
highly complex, fragment-coupling cross-metathesis process
contributed to the brevity of the approach. Findings made
during these CM studies include the tolerance of sensitive
functionality, namely, a â-lactone and an epimerizable
aldehyde, and the ability to use a homodimer of 3 generated
during the metathesis process. This highly convergent, syn-
thetic strategy is currently being employed for the synthesis
of a focused set of BFA derivatives aimed at the discovery
of potentially novel bioactivities related to prostaglandins.7
Acknowledgment. We thank the NIH (GM-45617), the
NSF (CAREER Award CHE-9624532 and 0104457), the
Robert A. Welch Foundation (A-1280), Zeneca, and Pfizer
for financial support. D.R. is an Alfred P. Sloan Fellow and
a Camille-Henry Dreyfus Teacher-Scholar. We thank Dr.
Joe Reibenspies for X-ray structure determination using
instruments obtained with funds from the NSF (CHE-
9807975). The NSF (CHE-0077917) also provided funds for
purchase of NMR instrumentation.
(15) (a) Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.;
Masamune, S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25, 2183-
2186. (b) Rathke, M. W.; Nowak, M. J. Org. Chem. 1985, 50, 2624-2626.
(c) For studies of an intramolecular HWE reaction in an approach to BFA,
see: Raddatz, P.; Winterfeldt, E. Angew. Chem. Int. Ed. Engl. 1981, 20,
286-287.
(16) For a similar observation in macrolactonization towards BFA, see:
Corey, E. J.; Wollenberg, R. H. Tetrahedron Lett. 1976, 4701-4704.
(17) Martin, S. F.; Dodge, J. A. Tetrahedron Lett. 1991, 32, 3017-3020.
(18) 4-epi-BFA was previously prepared in two steps (16% overall) from
BFA, see: Zhu, J. W.; Nagasawa, H.; Nagura, F.; Mohamad, S. B.; Uto,
Y.; Ohkura, K.; Hori, H. Bioorg. Med. Chem. 2000, 8, 455-463.
Supporting Information Available: Selected experi-
1
mental procedures and characterization data (including H
and 13C NMR spectra) for compounds 13, 16, 23, and 25
and comparative 1H NMR of natural and synthetic BFA. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL026438G
3234
Org. Lett., Vol. 4, No. 19, 2002