Total synthesis of (-)-fusarisetin A and reassignment of the absolute configuration of its natural counterpart

Article abstract of DOI:10.1021/ja211444m

The first total synthesis of (-)-fusarisetin A, the enantiomer of naturally occurring acinar morphogenesis inhibitor (+)-fusarisetin A, was accomplished in 13 steps, leading to the reassignment of the absolute configuration of the natural product. The synthesis featured a Lewis acid-promoted intramolecular Diels-Alder reaction, a Pd-catalyzed O→C allylic rearrangement, a chemoselective Wacker oxidation, and a Dieckmann condensation/hemiketalization cascade.

Full text of DOI:10.1021/ja211444m

Communication
pubs.acs.org/JACS
Total Synthesis of (−)-Fusarisetin A and Reassignment of the
Absolute Configuration of Its Natural Counterpart
†
†
Jun Deng, Bo Zhu, Zhaoyong Lu, Haixin Yu, and Ang Li*
State Key Laboratory of Bioorganic and Natural Products Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of
Sciences, 345 Lingling Road, Shanghai 200032, China
*
S Supporting Information
elucidated. The molecular structure of 1 was determined by
employing X-ray crystallographic analysis (relative stereo-
ABSTRACT: The first total synthesis of (−)-fusarisetin
A, the enantiomer of naturally occurring acinar morpho-
genesis inhibitor (+)-fusarisetin A, was accomplished in 13
steps, leading to the reassignment of the absolute
configuration of the natural product. The synthesis
featured a Lewis acid-promoted intramolecular Diels−
Alder reaction, a Pd-catalyzed O→C allylic rearrangement,
a chemoselective Wacker oxidation, and a Dieckmann
condensation/hemiketalization cascade.
chemistry) and the exciton chirality circular dichroism method
3
(
absolute configuration). From a structural perspective, 1
exhibits a 6,6,5,5,5-fused pentacyclic ring system (Figure 1)
bearing 10 stereogenic centers. Of particular interest is the
intricate 5,5,5-angular tricycle motif, reminiscent of the
molecular scaffolds of certain other natural products, such as
4c
4a
4b
chaetochalasin A, phomopsichalasin, and diaporthichalasin
(
2−4, Figure 1). The interesting chemical structure as well as
biological activity of fusarisetin A made it an attractive target for
total synthesis. Herein, we report the first total synthesis of the
proposed structure of this molecule.
atural products serve as an abundant source in the search
N
for anticancer agents due to unparalleled structural
diversity and accompanying molecular modes of action.
From a strategic point of view, a successful synthesis of the
6,6,5,n,5-fused (n = 5 or 6) pentacyclic system shown in Figure
1 must solve the following problems: (1) construction of a
decalin building block for the left-hand side of the molecule;
(2) assembly of the 5,5-spirobicyclic structure, especially
introduction of the quaternary stereogenic center on the
angular position of the 5,n,5-fused tricyclic moiety; and (3)
formation of the right-hand side n-membered ring (n = 5 or 6).
Problem 2, the most severe synthetic challenge posed by the
whole class of structurally related natural products, might
ideally be solved together with problem 3 in one step or pot.
Based on this general guideline, a retrosynthetic analysis of 1
was undertaken, as shown in Figure 2. We considered the initial
disassembly of the 5-membered lactol and its neighboring
spirobicycle through a retro hemiketalization/Dieckmann
condensation sequence, to render tricyclic intermediate 5.
The next obvious disconnection at the amide bond of 5,
followed by dehydration of its secondary hydroxyl functionality,
resulted in simplified tricyclic compound 6, which could be
further disconnected to a trans-decalin derivative 7 by cleavage
1
Recently, small molecules that inhibit cancer cell metastasis
have received increasing interest in related drug discovery
2
process, because such action may complement that of existing
anticancer drugs such as the tubulin stabilizers/destabilizers and
topoisomerase inhibitors. In May 2011, Ahn et al. reported the
isolation of a biologically intriguing natural product, fusarisetin
A (1, Figure 1), from a soil fungus, Fusarium sp. FN080326,
of its allylic C−C bond. Thus, an intramolecular S 2′ or a
N
transition-metal-catalyzed allylation reaction was expected to
give 6 in the forward direction. Compound 7 was envisioned to
derive from a linear precursor such as 8 through a
diastereoselective intramolecular Diels−Alder (D-A) reaction.
The desired D-A substrate could be assembled from known
aldehyde 9 and phosphonates 10 and 11 through double
Horner−Wadsworth−Emmons (H-W-E) olefinations.
Based on the above analysis, we first investigated the
preparation of trans-decalin 7 employing an intramolecular D-A
Figure 1. Fusarisetin A (1) and structurally related natural products
2−4.
5
−7
reaction as the key transformation,
as shown in Scheme 1.
which displays significant inhibition of acinar morphogenesis as
well as cell migration and invasion without apparent
cytotoxicity. The mechanism of action remains to be
Received: December 7, 2011
Published: December 26, 2011
3
©
2011 American Chemical Society
920
dx.doi.org/10.1021/ja211444m | J. Am. Chem.Soc. 2012, 134, 920−923

Journal of the American Chemical Society
Communication
Starting with aldehyde 9, which is readily prepared from
8
6
(
−)-citronellal or (−)-citronellol, we carried out H-W-E
9
reaction with lithiated phosphonate 10. Reduction with
DIBAL-H and acetylation of the resulting primary alcohol led
to triene 12 with good overall efficiency (67%, 3 steps).
Desilylation of 12 with HF·py and subsequent oxidation with
Dess−Martin periodinane (79%, 2 steps) set the stage for a
second H-W-E olefination. Aldehyde 13 was readily converted
to ketothioester 8 as a single E isomer in 66% yield by
6
,10
treatment with phosphonate 11
and KHMDS. With 8 in
hand, we examined a variety of conditions for the intra-
molecular D-A reaction. To our delight, BF ·OEt was found to
3
2
be an efficient promoter for this transformation, and trans-
decalin 7 was obtained as a single isolable diastereomer in 63%
6b
yield. The structure of 7 was confirmed by X-ray crystallo-
graphic analysis of a later intermediate (vide infra). As we
expected, a similar D-A reaction employing the methyl ester
counterpart of 8 as substrate also proceeded smoothly to give
the corresponding desired product, although the preparation of
the methyl ester substrate from the corresponding phosphonate
(
counterpart of 11) and aldehyde 13 was much less effective.
Interestingly, analogous D-A precursors with monocarbonyl
dienophile (aldehyde and ester) led to a mixture of
diastereomers, the major components of which were two
endo cycloadducts, under thermal or Lewis acidic (only effective
for aldehyde substrate) conditions. The diastereochemical
output of the above D-A processes indicated an unusual
mode of interaction between the dicarbonyl substrate and
Figure 2. Retrosynthetic analysis of fusarisetin A (1).
commonly monocoordinate Lewis acid BF , which remains
interesting for further investigations.
3
a
11
Scheme 1. Preparation of Decalin Derivative 7
With 7 in hand, we turned our attention to the synthesis of a
,6,5-fused tricyclic intermediate such as 6 (Figure 2),
6
according to our retrosynthetic analysis. The allylic hydroxyl
functionality of 7 was released by treatment with K CO in
2
3
MeOH to afford alcohol 14, which was subsequently activated
under mesylation conditions. However, without observation of
the corresponding mesylate, O-allylation compound 15 was
immediately generated in 93% yield, as shown in Scheme 2. In
an alternative attempt, the re-acetylated product from 14 was
subjected to standard Pd-catalyzed allylic substitution con-
ditions [Pd(PPh ) , LiOAc] but also rapidly converted into
3
4
hydrofuran 15. Interestingly, under very mild transesterification
12
conditions [Ag(TFA), Et N, MeOH, or trifluoroethanol],
3
cyclization product 15 or 16 was readily formed from 7 in one
pot in high yield. The structure of 16 was unambiguously
13
proven by X-ray crystallographic analysis (Figure 3), which
also confirmed the stereochemistry of D-A product 7. At this
point, it became obvious that the O-allylation product was
kinetically more favored than the desired C-allylation product
in this particular case. Thus, proper conditions to activate the
allylic C−O bond are desired, since the spontaneous and
irreversible reconstruction of an allylic C−C bond would drive
14
the “equilibrium” to the C-allylation side. Considering that
the basicity and leaving ability of β-ketoester enolate are similar
to those of phenoxide, we rationalized that a transition metal
a
Reagents and conditions: (a) 10 (1.1 equiv), LiHMDS (1.0 equiv), 9
1.0 equiv), THF, −78 °C → 0 °C, 30 min; then 0 °C, 30 min, 71%;
15
(
(
complex that promotes the substitution of a phenyl allyl ether
b) DIBAL-H (2.2 equiv), THF, −78 °C, 30 min, 98%; (c) Ac O (1.4
2
would fit the requirement. Much to our delight, treatment of 15
or 16 with Pd(OAc) and electron-rich ligand PBu cleanly
equiv), Et N (1.6 equiv), CH Cl , 22 °C, 2 h, 96%; (d) HF·py/THF
3
2
2
2
3
(
8
1:4), 0 °C, 2 h, 93%; (e) DMP (1.2 equiv), CH Cl , 22 °C, 30 min,
2 2
rendered the expected C-allylation product 17 or 18,
respectively, in satisfactory yield. The vinyl group of these
compounds was postulated to repose in the desired orientation
based on the diastereoselectivity of the O-allylation reaction,
5%; (f) 11 (1.1 equiv), KHMDS (2.1 equiv), 13 (1.0 equiv), THF,
−
78 °C, 1 h, 66%; (g) BF ·OEt (3.0 equiv), CH Cl , −78 °C, 10 min;
3
2
2
2
then −40 °C, 40 min, 63%.
9
21
dx.doi.org/10.1021/ja211444m | J. Am. Chem.Soc. 2012, 134, 920−923

Journal of the American Chemical Society
Communication
Scheme 2. Pd-Catalyzed O→C Allylic Rearrangement
Approach to 6,6,5-Tricycles 17 and 18
Scheme 3. Completion of the Total Synthesis of
a
(−)-Fusarisetin A and Reassignment of the Absolute
a
Configuration of Natural Fusarisetin A
a
Reagents and conditions: (a) K CO (2.0 equiv), MeOH/CH Cl
2
2
3
2
(1:1), 22 °C, 3 h, 89%; (b) MsCl (3.0 equiv), Et N (5.0 equiv),
3
CH Cl , 0 °C, 5 min, 93%; (c) (i) Ac O (1.5 equiv), Et N (2.0 equiv),
2
2
2
3
CH Cl , 0 °C, 1 h, 96%; (ii) Pd(PPh ) (0.25 equiv), LiOAc (2.0
2
2
3 4
equiv), THF, 22 °C, 15 min, 80%; (d) Ag(TFA) (2.0 equiv), Et N
3
(
4.0 equiv), MeOH or CF CH OH, 0 °C, 15 min; then 22 °C, 3 h,
3 2
9
0% for 15, 91% for 16; (e) Pd(OAc) (0.50 equiv), PBu (1.25
2
3
equiv), 22 °C, 1 h, 70% for 17, 75% for 18.
Figure 3. X-ray-derived ORTEPs of 16 and 23. Non-hydrogen atoms
are shown as 30% ellipsoids.
confirmed at a later stage by X-ray crystallographic analysis
(
vide infra).
a
Having successfully assembled the left half of 1, we entered
Reagents and conditions: (a) 19 (5.4 equiv), 4-DMAP (2.5 equiv),
the last stage of our total synthesis journey: construction of the
spirobicyclic moiety and the 5-membered lactol, as shown in
Scheme 3. The amino acid side chain was introduced by
toluene, 90 °C, 2 h, 72%; (b) PdCl
2
(0.25 equiv), CuCl (1.5 equiv),
O (balloon), DMF/water (10:1), 22 °C, 12 h, 79%; (c) CeCl •7H O
2
3
2
(
1.3 equiv), NaBH (1.1 equiv), MeOH, −20 °C, 20 min; (d) NaOMe
4
16
(5.0 equiv), MeOH, 0 °C, 10 min; then 22 °C, 1 h, 41% (2 steps) for
1
aminolysis of trifluoroethyl ester 18 with methyl amine 19
derived from D-serine at elevated temperature, furnishing 20 in
2% yield. The next challenge would be the chemo- and
, 8% (2 steps) for 23.
7
good yield while sparing the endocyclic CC bond.
Unfortunately, the next move toward alcohol 5 through
chemo- and diastereoselective reduction also proved to be
problematic. Treatment of 21 with sterically hindered hydride
sources such as L-Selectride, Red-Al, and LiAlH(Ot-Bu)3
resulted in diol 22, the undesired diastereomer of 5, as the
major product. 22 was further subjected to Dieckmann
condensation conditions (NaOMe/MeOH) to render the 5-
epimer of fusarisetin A (23) with good overall efficiency from
regioselective Markovnikov hydration of the terminal CC
bond of 20. However, oxymercuration of 20 with a variety of
Hg(II) species proved to be fruitless, while electrophilic
activators such as cationic halogen and selenium reagents
preferentially reacted with the more electron-rich and distorted
endocyclic CC bond. Although a similar example of selective
dihydroxylation of the terminal CC bond in the presence of
endocyclic competitor using Sharpless AD-mix-β was re-
1
7
ported, 20 decomposed into an intractable mixture under
such conditions. Despite the failure of this attempt, we were
inspired that transition-metal-mediated reactions, which are
often more sensitive to steric rather than electronic effects, may
offer the solution to this selectivity problem. To our delight,
Wacker oxidation of 20 furnished the desired ketone 21 in
21. The structure of 23 was unambiguously determined by X-
13
ray crystallographic analysis (Figure 3), which also confirmed
the vinyl group direction of 18. Commonly used asymmetric
ketone reduction methods also proved to be unsuccessful:
18
Corey−Bakshi−Shibata reduction conditions led to complete
9
22
dx.doi.org/10.1021/ja211444m | J. Am. Chem.Soc. 2012, 134, 920−923

Journal of the American Chemical Society
Communication
19
decomposition of the substrate, while Noyori reduction gave
a ca. 1:1 mixture of 5 and 22 in poor yield. Finally, we were
pleased to find Luche reduction as a suitable mean for the
selective reduction, affording a ca. 5:1 mixture of 5 and 22. This
mixture underwent Dieckmann condensation in the presence of
NaOMe in MeOH, followed by a spontaneous hemiketaliza-
tion, furnishing a synthetic sample of the proposed structure of
fusarisetin A (1) in 41% yield over 2 steps, together with a small
portion of its 5-epimer (23). The physical properties o3f
synthetic 1 matched those reported for the natural material,
Ngamrojnavanich, N.; Sriubolmas, N.; Chaichit, N.; Ohta, T.
Tetrahedron Lett. 2007, 48, 651.
(5) Systematic studies on intramolecular D-A reaction for synthesiz-
ing bicyclic systems: (a) Roush, W. R.; Hall, S. E. J. Am. Chem. Soc.
1
1
981, 103, 5200. (b) Roush, W. R.; Peseckis, S. M. J. Am. Chem. Soc.
981, 103, 6696. (c) Roush, W. R.; Gillis, H. R.; Ko, A. I. J. Am. Chem.
Soc. 1982, 104, 2269. (d) Roush, W. R.; Gillis, H. R. J. Org. Chem.
982, 47, 4825.
6) Elegant syntheses of decalin natural product equisetin using D-A
1
(
reaction: (a) Burke, L. T.; Dixon, D. J.; Ley, S. V.; Rodríguez, F. Org.
Lett. 2000, 2, 3611. (b) Burke, L. T.; Dixon, D. J.; Ley, S. V.;
Rodríguez, F. Org. Biomol. Chem. 2005, 3, 274.
27
except for the sign of its optical rotation {synthetic: [α]
=
D
2
5
(7) Excellent review ofD-A reaction in total synthesis: Nicolaou, K.
−
88.0 (c = 0.15 in MeOH); natural: [α] = +84.6 (c = 0.2 in
D
C.; Snyder, S. A.; Montagnon, T.; Vassilikogiannakis, G. Angew. Chem.,
Int. Ed. 2002, 41, 1668.
MeOH)}. Thus, the absolute configuration of naturally
occurring 1 was reassigned as that of 24 based on our total
synthesis.
(
8) Inoue, A.; Kanematsu, M.; Yoshida, M.; Shishido, K. Tetrahedron
Lett. 2010, 51, 3966.
9) (a) Nicolaou, K. C.; Daines, R. A.; Chakraborty, T. K.; Ogawa, Y.
In conclusion, we developed an efficient synthetic strategy
for the total synthesis of the enantiomer of fusarisetin A, a
newly discovered acinar morphogenesis inhibitor possessing an
intricate structure, and reassigned the absolute configuration of
the natural product through our synthesis. The synthesis
featured an intramolecular Diels−Alder reaction, a Pd-mediated
O→C allylic rearrangement, a chemoselective Wacker
oxidation, and a Dieckmann condensation/hemiketalization
cascade. The reported synthetic strategy and methods are
expected to be applicable to the construction of other
structurally or biosynthetically related natural products, as
well as designed analogues of fusarisetin A, and thus to facilitate
the exploration of its mechanism of action on a molecular level.
(
J. Am. Chem. Soc. 1988, 110, 4685. (b) Amans, D.; Bellosta, V.; Cossy,
J. Angew. Chem., Int. Ed. 2006, 45, 5870.
(10) Hayashi, Y.; Narasaka, K. Chem. Lett. 1998, 313.
(11) Dicarbonyl/BF2 complexcould possibly be generated in situ
from BF ·OEt and dicarbonyl substrateand serve as an activated
3 2
intermediate for D-A reaction. Preparationand characterization of such
class of complex: Maeda, H.; Yohei, H.; Nakanishi, T. J. Am. Chem. Soc.
2
007, 129, 13661.
(12) (a) Masamune, S.; Hayase, Y.; Schilling, W.; Chan, W. K.; Bates,
G. S. J. Am. Chem. Soc. 1977, 99, 6756. (b) Booth, P. M.; Fox, C. M. J.;
Ley, S. V. Tetrahedron Lett. 1983, 24, 5143.
(13) CCDC-854787 and CCDC-854788 contain the supplementary
crystallographic data for 16 and 23, respectively, and are available free
of charge from The Cambridge Crystallographic Data Centre via www.
ccdc.cam.ac.uk/data_request/cif.
ASSOCIATED CONTENT
■
(14) Mechanistically similarPd-catalyzed O→C allylic migration
*
S
Supporting Information
reactions: Mohr, J. T.; Stoltz, B. M. Chem. Asian J. 2007, 2, 1476
and referencestherein.
(15) Hosokawa, T.; Kono, T.; Uno, T.; Murahashi, S.-I. Bull. Chem.
Soc. Jpn. 1986, 59, 2191.
Experimental procedures and compound characterization. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(16) White, K. N.; Konopelski, J. P. Org. Lett. 2005, 7, 4111.
(
17) Wipf, P.; Kim, Y.; Goldstein, D. M. J. Am. Chem. Soc. 1995, 117,
AUTHOR INFORMATION
■
1
1106.
18) Comprehensivereview: Corey, E. J.; Helal, C. J. Angew. Chem.,
Int. Ed. 1998, 37, 1986.
19) Fujii, A.; Hashiguchi, S.; Uematsu, N.; Ikariya, T.; Noyori, R. J.
Am. Chem. Soc. 1996, 118, 2521.
Corresponding Author
ali@sioc.ac.cn
(
(
Author Contributions
These authors contributed equally.
†
ACKNOWLEDGMENTS
■
This paper is dedicated to Prof. K. C. Nicolaou. We thank
Zhunzhun Yu and Shuhang Wu for technical contributions,
Prof. Chunyang Cao for NMR assistance, and Dr. D. J.
Edmonds and Prof. Dawei Ma for helpful discussions. Financial
support was provided by State Key Laboratory of Bioorganic
and Natural Products and Shanghai Institute of Organic
Chemistry, Chinese Academy of Sciences.
REFERENCES
■
(
1) (a) Gordaliza, M. Clin. Transl. Oncol. 2007, 9, 767. (b) Newman,
D. J.; Cragg, G. M. J. Nat. Prod. 2007, 70, 461.
(
(
2) Hulkower, K. I.; Herber, R. L. Pharmaceutics 2011, 3, 107.
3) Jang, J.-H.; Asami, Y.; Jang, J.-P.; Kim, S.-O.; Moon, D. O.; Shin,
K.-S.; Hashizume, D.; Muroi, M.; Saito, T.; Oh, H.; Kim, B. Y.; Osada,
H.; Ahn, J. S. J. Am. Chem. Soc. 2011, 133, 6865.
(
4) Selected natural products that are structurally related to
fusarisetin A: (a) Oh, H.; Swenson, D. C.; Gloer, J. B. Tetrahedron
Lett. 1998, 39, 7633. (b) Horn, W. S.; Simmonds, M. S. J.; Schwartz, R.
E.; Blaney, W. M. Tetrahedron 1995, 51, 3969. (c) Pornpakakul, S.;
Roengsumran, S.; Deechangvipart, S.; Petsom, A.; Muangsin, N.;
9
23
dx.doi.org/10.1021/ja211444m | J. Am. Chem.Soc. 2012, 134, 920−923