Enantioselective total synthesis of (+)- and (-)-nigellamine A2

Article abstract of DOI:10.1021/ja061559n

The nigellamine alkaloids are dolabellane diterpenes displaying potent lipid metabolism-promoting activity. Total synthesis of (+)- and (-)-nigellamine A₂ has been accomplished. Absolute stereochemistry of synthetic nigellamine A₂ was established through an intramolecular asymmetric allylic alkylation using a Pd(phosphinooxazoline) catalyst. Other notable transformations include a radical alkynylation, a diastereoselective Nozaki-Hiyama-Kishi cyclization, and a regio- and stereoselective catalytic epoxidation. On the basis of X-ray crystallographic analysis of an optically active intermediate, we have confirmed the assigned absolute stereochemistry of the natural product. Minor modifications of the synthetic sequence outlined here should provide access to the other nigellamine alkaloids. Copyright

Full text of DOI:10.1021/ja061559n

Published on Web 05/23/2006
Enantioselective Total Synthesis of (+)- and (-)-Nigellamine A2
Jianwei Bian, Matthew Van Wingerden,1 and Joseph M. Ready*
Department of Biochemistry, The UniVersity of Texas Southwestern Medical Center at Dallas,
5323 Harry Hines BouleVard, Dallas, Texas 75390-9038
Received March 6, 2006; E-mail: joseph.ready@utsouthwestern.edu
Dolabellane diterpenes have been isolated from marine and
to replace the electron-withdrawing alkoxy group with an electron-
donating surrogate. Accordingly, allylic ester 10 was prepared as
outlined in Scheme 2. Performing the sequence without chromato-
graphic purification of any intermediates facilitated the preparation
of multigram quantities. Exposure of the lithium enolate of 10 to
a Pd(phosphinooxazoline) complex7 resulted in the production of
diene 12 in 95% ee as the only reaction product.
terrestrial sources and are produced by animals, plants, and fungi.
That their broad distribution reflects the privileged status of the
dolabellane ring system is supported by the correspondingly
impressive array of biological activities associated with these natural
products and by the fact that both enantiomeric forms occur in
nature.2 The ubiquity of these natural products combined with their
structural complexity has stimulated substantial activity in the
synthetic community.2,3
Scheme 2
Scheme 1
Yoshikawa and co-workers recently disclosed the two- and three-
dimensional structures of a series of alkaloids possessing the
dolabellane 5,11-trans-fused bicyclic ring system (1a-e). Isolated
from Nigella satiVa (black cumin), these nigellamine alkaloids are
associated with potent lipid metabolism-promoting activity.4 In
considering a total synthesis of the nigellamines, we recognized
that an intramolecular addition of a substituted malonate onto a
metal-coordinated allyl moiety (2) might generate the isopropylidene
cyclopentane core appended with functionality to install the 11-
membered ring (Scheme 1). The successful implementation of this
strategy would require the formation of a tetrasubstituted M(π-
allyl) from the corresponding allyl acetate followed by a regio-
and enantioselective anti-Baldwin cyclizationsa series of events
with little literature precedent.5
Iodolactonization of diene 12 differentiated the two carbonyls
and installed the C10 hydroxyl with good diastereoselectivity
(Scheme 3). X-ray crystallographic analysis of the minor diaste-
reomer (C10-epi-13) established the stereochemical outcome of the
Pd-catalyzed cyclization and, by extension, the absolute stereo-
chemistry of synthetic 1b.8 Nucleophilic substitution of the primary
iodide in 13 proved untenable, and failure met all attempts to
convert the iodolactone to a terminal epoxide. Eventually, a high
yielding functionalization of C9 was found in Fuchs’ radical
alkynylation reaction using triflone 14.9
Desilylation and reduction yielded terminal alkyne 16, a substrate
for Negishi methyliodination.10 Under anhydrous conditions, Cp2-
ZrCl2 did not mediate addition of Me3Al to the terminal alkyne.
Instead, addition to the lactol occurred, resulting in a mixture of
methyl-tetrahydrofuran diastereomers. Remarkably, the reagent
derived from a 1:1:3 mixture of water, Cp2ZrCl2, and Me3Al
effected methylalumination without competing addition to the lactol.
In situ iodination and subsequent silylation afforded vinyl iodide
17 in good yield. While the rate acceleration offered by water in
the methylalumination was described by Wipf,11 the modulation
of chemoselectivity has not been documented previously.
In initial attempts to execute the plan outlined above, we focused
on substrates incorporating C10 oxygenation (nigellamine number-
ing). To this end, a series of allylic acetates 3 was cyclized in the
presence of a palladium catalyst (eq 1). While we had anticipated
malonate addition to the less-hindered C11 terminus of the presump-
tive Pd(π-allyl) intermediate (2, M ) Pd), in fact attack occurred
at the more substituted carbon to return the undesired isomer 4 as
the sole cyclization product. Reasoning that this regiochemical
outcome was dictated by electronics rather than sterics,6 we sought
The remaining carbon atoms of the nigellamine skeleton were
installed through cross-coupling with alkyl zinc reagent 18 and a
second methylalumination/iodination sequence. The 11-membered
ring was subsequently formed by means of a Cr-mediated cycliza-
tion involving the C3 vinyl iodide and a C2 aldehyde.12 The high
yield and complete diastereoselectivity of the Nozaki-Hiyama-
Kishi (NHK) cyclization are notable, even more so given the
hindered nature of the aldehyde component and the formation of a
9
7428
J. AM. CHEM. SOC. 2006, 128, 7428-7429
10.1021/ja061559n CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Scheme 3
trans-fused ring system. The conformational bias imposed by
embedding the C10 stereocenter in a rigid lactone likely minimizes
the entropic penalty associated with macrocyclization. Unfortu-
nately, circumstantial evidence convinced us that the newly formed
C2 hydroxyl (20) possessed the incorrect relative stereochemistry,
a conclusion born out by subsequent transformations. In particular,
Dess-Martin oxidation provided an intermediate enone. Reduction
with the reagent derived from 1:1 mixture of (iBu)2AlH and tBuLi
produced an allylic alcohol displaying spectroscopic data consistent
with the desired C2 stereochemistry (21).13 In contrast, reduction
with NaBH4/CeCl3, CBS catalyst, and Li(OtBu)3AlH returned the
original diastereomer (20) while the Selectrides favored 1,4 addition.
Reductive opening of the lactone and selective acylation of the
primary alcohol led to a substrate for epoxidation (22). Concerned
about competition among the three reactive olefins, we reasoned
that the C2 alcohol might deactivate the C3-C4 olefin while the
ketone catalyst 23 might be unreactive toward tetrasubstituted ole-
fins.14 In fact, preliminary studies showed rapid oxidation of all
three olefins with reagents such as mCPBA. Using the Shi catalyst
and oxone, however, oxidation proceeded regio- and stereoselec-
tively to produce the desired epoxide as the major product. Only a
single report has detailed the epoxidation of a macrocyclic olefin
using ketone catalyst 23,15 and we were therefore interested in deci-
phering substrate versus catalyst control. In this regard, substrate
conformation appears to dictate facial preference as both 23 and
ent-23 generated the same product as a single diastereomer. Inter-
estingly, the enantiomeric catalysts displayed slight differences in
olefin preference. Epoxidations employing ketone 23 yielded a 7:1
ratio of monoepoxides, favoring epoxidation of the C7-C8 olefin
over the C3-C4 olefin. In contrast, otherwise identical epoxidations,
save for the use of ent-23, resulted in a 2:1 ratio of the same
epoxides.
Acknowledgment. We acknowledge a generous gift of ent-23
from Yian Shi (Colorado State). X-ray analysis was performed by
Radha Akella (UTSW) and Richard Staples (Harvard). Financial
support was from the Welch Foundation and UT Southwestern
(J.M.R. is a Southwestern Medical Foundation Scholar in Biomedi-
cal Research). This investigation was conducted in a facility
constructed with support from the Research Facilities Improvement
Program (C06 RR-15437). M.V.W. was supported by an NIH grant
for undergraduate research fellowships (R25 GM072832-02).
Supporting Information Available: Complete experimental details
and characterization data. This material is available free of charge via
the Internet at http://pubs.acs.org.
References
(1) Summer Undergraduate Research Fellow.
(2) Rodriguez, A. D.; Gonzalez, E.; Ramirez, C. Tetrahedron 1998, 54,
11683-11729.
(3) (a) Miyaoka, H.; Baba, T.; Mitome, H.; Yamada, Y. Tetrahedron Lett.
2001, 42, 9233-9236. (b) Miyaoka, H.; Isaji, Y.; Mitome, H.; Yamada,
Y. Tetrahedron 2003, 59, 61-75. (c) Williams, D. R.; Heidebrecht, R.
W., Jr. J. Am. Chem. Soc. 2003, 125, 1843-1850. (d) Kingsbury, J. S.;
Corey, E. J. J. Am. Chem. Soc. 2005, 127, 13813-13815. (e) Snyder, S.
A.; Corey, E. J. J. Am. Chem. Soc. 2006, 128, 740-742.
(4) (a) Morikawa, T.; Xu, F.; Kashima, Y.; Matsuda, H.; Ninomiya, K.;
Yoshikawa, M. Org. Lett. 2004, 6, 869-872. (b) Morikawa, T.; Xu, F.;
Ninomiya, K.; Matsuda, H.; Yoshikawa, M. Chem. Pharm. Bull. 2004,
52, 494-497.
(5) Reviews on allylic alkylation in natural product synthesis: (a) Trost, B.
M.; Crawley, M. L. Chem. ReV. 2003, 103, 2921-2943. (b) Helmchen,
G.; Ernst, M.; Paradies, G. Pure Appl. Chem. 2004, 76, 495-506.
(6) Pretot, R.; Pfaltz, A. Angew. Chem., Int. Ed. 1998, 37, 323-325.
(7) (a) Dawson, G. J.; Frost, C. G.; Williams, J. M. J.; Coote, S. J. Tetrahedron
Lett. 1993, 34, 3149-3150. (b) Helmchen, G.; Pfaltz, A. Acc. Chem. Res.
2000, 33, 336-345.
(8) See Supporting Information.
(9) Xiang, J.; Fuchs, P. L. Tetrahedron Lett. 1998, 39, 8597-8600.
(10) Negishi, E. Acc. Chem. Res. 1987, 20, 65-72.
(11) Wipf, P.; Lim, S. Angew. Chem., Int. Ed. Engl. 1993, 32, 1068-1071.
(12) (a) Conditions modified from Nicolaou, K. C.; Theodorakis, E. A.; Rutjes,
F. P. J. T.; Tiebes, J.; Sato, M.; Untersteller, E.; Xiao, X.-Y. J. Am. Chem.
Soc. 1995, 117, 1171-1172. (b) NHK reaction in synthesis of 5,11-trans-
fused bicycles: Luker, R.; Whitby, R. J. Tetrahedron Lett. 1996, 37,
7661-7664.
(13) Trost, B. M.; Nishimura, Y.; Yamamoto, K.; McElvain, S. S. J. Am. Chem.
Soc. 1979, 101, 1328-1330.
(14) Shi, Y. Acc. Chem. Res. 2004, 37, 488-496.
Finally, acylation of the two secondary alcohols with nicotinic
acid provided ent-nigellamine A2. The synthetic material was fully
1
consistent with the natural product by H NMR, 13C NMR, and
MS analysis. The optical rotation ([R]20D ) +19.6, c ) 0.5 CHCl3)
was opposite in sign and similar in magnitude to that reported for
the natural product ([R]27D ) -24.2, c ) 1.00 CHCl3),16 confirming
the stereochemical assignment made by the Yoshikawa group.
Natural nigellamine A2 was produced from ent-12 following a
sequence analogous to that outlined in Scheme 3.8,17 Minor
modifications of this synthesis should provide other members of
the nigellamine family. Having access to numerous congeners of
both enantiomeric series should facilitate detailed biological studies.
(15) Altmann, K.-H.; Bold, G.; Caravatti, G.; Denni, D.; Florsheimer, A.;
Schmidt, A.; Rihs, G.; Wartmann, M. HelV. Chim. Acta 2002, 85, 4086-
4110.
(16) CD (MeOH) showed a positive Cotton effect at 261 nm. Natural 1b shows
a negative Cotton effect at 256 nm.
(17) Diene 10 was cyclized in the presence of a catalyst incorporating ligand
(R)-11.
JA061559N
9
J. AM. CHEM. SOC. VOL. 128, NO. 23, 2006 7429