Total synthesis of (-)-sessilifoliamide C and (-)-8-epi-stemoamide

Article abstract of DOI:10.1021/ol200743u

A convergent route featuring [3,3]-sigmatropic rearrangements of a linchpin azepinopyrrolidine served to install two of the four contiguous stereocenters present in the tricyclic Stemona alkaloids sessilifoliamide and stemoamide. In addition to the first total synthesis of (-)-sessilifoliamide C, a potential biosynthetic relationship between the sessilifoliamides and previously reported Stemona alkaloids is presented.

Full text of DOI:10.1021/ol200743u

ORGANIC
LETTERS
2011
Vol. 13, No. 10
2634–2637
Total Synthesis of (À)-Sessilifoliamide
C and (À)-8-epi-Stemoamide
Adam T. Hoye and Peter Wipf*
Department of Chemistry, University of Pittsburgh, Pittsburgh, Pennsylvania 15260,
United States
pwipf@pitt.edu
Received March 19, 2011
ABSTRACT
A convergent route featuring [3,3]-sigmatropic rearrangements of a linchpin azepinopyrrolidine served to install two of the four contiguous
stereocenters present in the tricyclic Stemona alkaloids sessilifoliamide and stemoamide. In addition to the first total synthesis of
(À)-sessilifoliamide C, a potential biosynthetic relationship between the sessilifoliamides and previously reported Stemona alkaloids is presented.
Natural products from Stemona and Croomia plants
have served as an inspiration for chemical, biological, and
synthetic studies since at least the 1930s, when the first
derivatives were described in Western references.1À3 Ex-
tracts from these plants have been used for centuries in
Eastern cultures for the treatment of various respiratory
problems, such as pertussis, bronchitis, and tuberculosis.
However, with the exception of their well-documented
insecticidal activities, validated evidence for other benefi-
cial human health effects of the pure natural products is
just beginning to emerge.4
In 2003, Takeya and co-workers isolated sessilifolia-
mides AÀD (1À4) from the roots of the perennial herb
Stemona sessilifolia.5 These alkaloids possess the charac-
teristic pyrrolo[1,2-a]azepine core attached to a butenolide
substituent (Scheme 1). The relative configuration of 3 was
confirmed by a chemical degradation to a derivative in
common with sessilifoliamide A, for which an X-ray
crystal structure had been obtained.5 Interestingly, a pos-
sible biosynthetic relationship between the sessilifolia-
mides and the previously identified parvistemoline (5)6 is
apparent upon C(16)-oxidation and Michael addition.
Accordingly, we decided to investigate a unified and
possibly biomimetic strategy toward the sessifoliamides
and prepare sessilifoliamide C first, because of its similarity
to the more complex ring system present in parvistemoline.
Our retrosynthetic strategy for 3 was focused on the
construction of the C(9)ÀC(10) adjacent stereocenters
prior to butenolide formation (Scheme 2). We desired an
approach that could accomplish this task in a single
operation and envisioned that a [3,3]-sigmatropic Claisen
rearrangement7 with precise stereocontrol in the transition
(1) (a) Lee, H. M.; Chen, K. K. J. Am. Pharm. Assoc. 1940, 29, 391.
(b) Edwards, O. E.; Feniak, G.; Handa, K. L. Can. J. Chem. 1962, 40,
455. (c) Uyeo, S.; Irie, H.; Harada, H. Chem. Pharm. Bull 1967, 15, 768.
(d) Alibes, R.; Figueredo, M. Eur. J. Org. Chem. 2009, 2421. (e) Pilli,
R. A.; Rosso, G. B.; Ferreira De Oliveira, M. D. C. Nat. Prod. Rep 2010,
27, 1908.
(2) For recent synthetic studies reported on Stemona alkaloids, see,
for example: (a) Chen, Z.-H.; Tu, Y.-Q.; Zhang, S.-Y.; Zhang, F.-M.
Org. Lett. 2011, 13, 724. (b) Chen, Z.-H.; Zhang, Y.-Q.; Chen, Z.-M.; Tu,
Y.-Q.; Zhang, F.-M. Chem. Commun. 2011, 47, 1836. (c) Frankowski, K. J.;
Setola, V.; Evans, J. M.; Neuenswander, B.; Roth, B. L.; Aube, J., Proc.
Nat. Acad. Sci. U.S.A. 2011, early edition.
(3) For previous studies on Stemona alkaloids from our laboratory,
see: (a) Wipf, P.; Kim, Y. Tetrahedron Lett. 1992, 33, 5477. (b) Wipf, P.;
Kim, Y.; Goldstein, D. M. J. Am. Chem. Soc. 1995, 117, 11106. (c)
Goldstein, D. M.; Wipf, P. Tetrahedron Lett. 1996, 37, 739. (d) Wipf, P.;
Li, W. J. Org. Chem. 1999, 64, 4576. (e) Wipf, P.; Mareska, D. A.
Tetrahedron Lett. 2000, 41, 4723. (f) Wipf, P.; Spencer, S. R. J. Am.
Chem. Soc. 2005, 127, 225.
(5) Kakua, D.; Hitotsuyanagi, Y.; Matsuura, N.; Fukaya, H.;
Takeya, K. Tetrahedron 2003, 59, 7779.
(6) Lin, W.; Xu, R.; Zhong, Q. Huaxue Xuebao 1991, 49, 927.
(7) Wipf, P. Claisen Rearrangements. In Comprehensive Organic
Synthesis; Trost, B. M., Fleming, I., Paquette, L. A., Eds.; Pergamon: Oxford,
1991; Vol 5, pp 827À874.
(4) Greger, H. Planta Med. 2006, 72, 99.
r
10.1021/ol200743u
Published on Web 04/21/2011
2011 American Chemical Society

Scheme 1. Biosynthetic Hypothesis Based on the Structural
Similarities between Sessilifoliamides and Parvistemoline
Scheme 3. Preparation of Vinylpyrrolidinone 14
recovered cleanly despite modifications in the reaction time,
catalyst loading, and solvent. Pyrrolidinone protection facili-
tated the desired transformation. Immediate exposure of the
aldehyde 13 to Wittig olefination delivered a 96% yield of
alkene 14. Boc-protected 14 was found to be stable to bench-
top storage, whereas amide 10 decomposed within hours.
However, because of partial racemization prior to and/or
during olefination, 14 was obtained in 72% ee, as determined
by chiral HPLC analysis on a Chiralcel OD-H column.
The second segment, iodide 9, was prepared by an enzy-
matic resolution strategy which provided the t-butyl ester 15
on a large scale (Scheme 4).10 A two-step reduction procedure
was used to generate alcohol 16 in 87% yield; in the presence
of 2 equiv of DIBAL-H, 16 was isolated in lower yield (ca.
70%). Finally, a straightforward iodination of 16 afforded 9.
state could meet these requirements. The precursor to this
key transformation, alcohol 8, was accessible by ring-
closing metathesis (RCM) of the diene formed from the
alkylation of pyrrolidinone 10 with iodide 9.
Scheme 2. Retrosynthetic Approach for Sessilifoliamide C
Scheme 4. Preparation of Iodide 9
In preparation for the segment condensation, the N-Boc
group in 14 was cleaved with TFA (Scheme 5). After some
optimization, we discovered that alkylation of amide 10 under
phase-transfer conditions was more reproducible and man-
ageable on scale than NaH/DMF conditions, providing us
with a reliable access to diene 17. Ring closure to the pyrrolo-
[1,2-a]azepine core characteristic of Stemona alkaloids by
RCM of 17 afforded 19 in 91% yield as an inseparable 6:1
mixture of C(9a) diastereomers resulting from partially race-
mized 14. Removal of the TBS group under acidic conditions
led to the allylic alcohol 8 as a hygroscopic solid.
Prior difficulties with [3,3]-rearrangements on bicycles
motivated us to probe whether this reaction could occur
on the sterically congested concave face of pyrrolo-
[1,2-a]azepine 8. We selected an EschenmoserÀClaisen
reaction7 to address this question, in analogy to our
stenine3f and tuberostemonine3b syntheses (Scheme 6).
The desired rearrangement occurred readily, and amide
21 was obtained in 92% yield under optimized conditions.
While several methods for the preparation of vinylpyrro-
lidinone 10 are known in the literature,8 preliminary RCM
studies established that we needed an N-protective group on
10, and therefore, we pursued a new, scaleable synthesis of
the Boc derivative 14 (Scheme 3). (S)-Pyroglutamic acid 11
was converted to the thioester with ethanethiol in the pre-
sence of carbodiimide, followed by N-Boc protection to
afford carbamate 12. Direct conversion of 12 to the aldehyde
13 was successful by a Fukuyama reduction.9 Fukuyama
conditions prior to N-Boc protection did not lead to the
desired reduction product, and starting material was
(8) (a) Wei, Z.-Y.; Knaus, E. Synlett 1993, 295–296. (b) Wei, Z.-Y.;
Knaus, E. Tetrahedron 1994, 50, 5569. (c) Napoletano, M.; Della Bella,
D.; Fraire, C.; Grancini, G.; Masotto, C.; Ricciardi, S.; Zambon, C.
Bioorg. Med. Chem. Lett. 1995, 5, 589. (d) Gheorghe, A.; Schulte, M.;
Reiser, O. J. Org. Chem. 2006, 71, 2173. (e) Mo, F.; Li, F.; Qiu, D.;
Wang, J. Tetrahedron 2010, 66, 1274.
(10) (a) Ghosh, A. K.; Kulkarni, S. Org. Lett. 2008, 10, 3907. (b)
Ghosh, A. K.; Yuan, H. Tetrahedron Lett. 2009, 50, 1416. (c) Vrielynck,
S.; Vandewalle, M. Tetrahedron Lett. 1995, 36, 9023.
(9) Tokuyama, H.; Yokoshima, S.; Lin, S.-C.; Li, L.; Fukuyama, T.
Synthesis 2002, 1121.
Org. Lett., Vol. 13, No. 10, 2011
2635

Having demonstrated that the [3,3]-sigmatropic rearrange-
ment and chain extension from C(7) to C(9) on 8 was indeed
feasible, we studied the IrelandÀClaisen rearrangement7,13
for the stereoselective installation of the tertiary methine
C(11) in the sessilifoliamides. Acylation of alcohol 8 (still as
a 6:1 mixture of C(9a) diastereomers derived from 19) with
butyric acid under carbodiimide coupling conditions pro-
vided ester 25 in 92% yield (Scheme 7). Enolization with
LiHMDS in a THF/HMPA mixture14 followed by trapping
with TBSCl provided the (Z)-silyl ketene acetal as the sole
Scheme 5. Segment Condensation, RCM, and Synthesis of
Linchpin Intermediate 8
1
stereoisomer according to H NMR analysis of the crude
reaction product. The subsequent thermal rearrangement in
toluene led to an approximately 2:1 ratio of silyl esters 26a
and 26b, which were treated with TBAF followed by TMS-
diazomethane to give the corresponding methyl esters. Un-
fortunately, both the yield of the four-step sequence and the
ratio of diastereomeric products 27a and 27b remained low
(21% yield, 2.2:1 dr) despite attempts to optimize the reaction
time, temperature, and solvent polarity.
Becauseof thestructural similarity of21 and stemoamide,11
we decided to complete the remaining three steps to the
tricyclic core of this natural product. Iodolactonization of
21 and reduction of the resulting iodide proceeded in 98%
and 97% yield, respectively, to give 23 as a 6:1 mixture of
C(9a) diastereomers. Chromatographic removal of the
remaining C(9a) isomer provided the major isomer of 23
in 71% yield for this step. Finally, lactone R-methylation
led to 8-epi-stemoamide (24). While many syntheses of
stemoamide and its stereoisomers have been reported,12
our effort represents the first synthesis of the C(8) epimer.
Scheme 7. TBS-Mediated IrelandÀClaisen Rearrangement of 8
Scheme 6. Conversion of 8 to (À)-8-epi-Stemoamide
A significant improvement in this key transformation
could be achieved after an analysis of the competing transi-
tion states of the IrelandÀClaisen rearrangement (Scheme 8).
Specifically, we noticed that the orientation of the trialkylsilyl
group was remarkably different in 29q and 30q and therefore
offered an opportunity to influence the course of the reaction.
We hypothesized that a significant increase in the size of the
alkyl groups on silicon would likely selectively destabilize
chair transition state 30q, where the silyl group is positioned
underneath the 7-membered ring during the CÀC bond
formation, and be more readily tolerated in boat transi-
tion state 29q, in which the silyl group is kept distant from
the 5,7-ring system. This hypothesis could be readily put
to test (Scheme 9). IrelandÀClaisen rearrangement of the
TIPS-silyl ketene acetal, formed by enolization of 25 in
the presence of TIPSCl, required heating at reflux in
degassed15 xylenes for 3 h to go to completion. The
resulting 31a and 31b were again converted to the methyl
esters to simplify analysis.
Additionally, among the plethora of synthetic approaches
to stemoamide, this is the first example of a [3,3]-rearrange-
ment to install the C(9) stereocenter in this ring system.
(11) Lin, W.; Ye, Y.; Xu, R. J. Nat. Prod. 1992, 55, 571.
(12) (a) Williams, D. R.; Reddy, J. P.; Amato, G. S. Tetrahedron Lett.
1994, 35, 6417. (b) Khim, S.-K.; Schultz, A. G. J. Org. Chem. 2004, 69,
7734. (c) Bogliotti, N.; Dalko, P. I.; Cossy, J. J. Org. Chem. 2006, 71,
9528. (d) Kohno, Y.; Narasaka, K. Bull. Chem. Soc. Jpn. 1996, 69, 2063.
(e) Bates, R. W.; Sridhar, S. Synlett 2009, 12, 1979. (f) Jacobi, P. A.; Lee,
K. J. Am. Chem. Soc. 1997, 119, 3409. (g) Jacobi, P. A.; Lee, K. J. Am.
Chem. Soc. 2000, 122, 4295. (h) Kinoshita, A.; Mori, M. J. Org. Chem.
1996, 61, 8356. (i) Kinoshita, A.; Mori, M. Heterocycles 1997, 46, 287.
(j) Sibi, M. P.; Subramanian, T. Synlett 2004, 1211. (k) Olivo, H. F.;
ꢀ
Tovar-Miranda, R.; Barragan, E. J. Org. Chem. 2006, 71, 3287.
(13) Ireland, R. E.; Mueller, R. H.; Willard, A. K. J. Am. Chem. Soc.
1976, 98, 2868.
(14) (a) Ireland, R. E.; Wipf, P.; Armstrong, J. D. J. Org. Chem. 1991,
56, 650. (b) Ireland, R. E.; Wipf, P.; Xiang, J. N. J. Org. Chem. 1991, 56,
3572.
(l) Torssell, S.; Wanngren, E.; Somfai, P. J. Org. Chem. 2007, 72,
4246. (m) Gao, P.; Tong, Z.; Hu, H.; Xu, P.-F.; Liu, W.; Sun, C.; Zhai,
H. Synlett 2009, 2188. (n) Wang, Y.; Zhu, L.; Zhang, Y.; Hong, R.
Angew. Chem., Int. Ed. 2011, 50, 2787. (o) Honda, T.; Matsukawa, T.;
Takahashi, K. Org. Biomol. Chem. 2011, 9, 673.
2636
Org. Lett., Vol. 13, No. 10, 2011

Therefore, the 6:1 mixture of 27a,b was hydrogenated and
converted to the corresponding aldehyde by sequential treat-
ment with Pd/C/H2, LiBH4, and Swern reagent (Scheme 10).
The minor C(10)-stereoisomer could be removed chromato-
graphically from either alcohol or aldehyde intermediate after
the latter two steps to afford pure 6. Exposure of 6 to
vinylmagnesium bromide afforded the allylic alkoxides 32a
and 32b, which were acylated in situ with methacryloyl
chloride to afford a 2:1 mixture of the labile 33a and 33b. A
final RCM successfully assembled the butenolide ring in 84%
yield, and MPLC separation revealed the major isomer to be
(À)-sessilifoliamide C (3) by comparison of the spectroscopic
data with the reported literature values of the natural product.17
Scheme 8. Transition-State Analysis of the IrelandÀClaisen
Rearrangement
Scheme 10. Conversion of 27 to (À)-Sessilifoliamide C
Gratifyingly, esters 27a and 27b were isolated in 51%
yield from 25 in an improved 6:1 ratio of C(10) diaster-
eomers, in support of our transition-state analysis.16
Chromatography on SiO2 was used to remove the
diastereomers at C(9a) from the 6:1 mixture of 27a
and 27b. Confirmation of the assignment of the major
isomer 27a was gained upon completion of the natural
product and comparison of the spectroscopic data.
Scheme 9. TIPS-Mediated IrelandÀClaisen Rearrangement of 8
In summary, we have completed the first synthesis
of the Stemona alkaloid (À)-sessilifoliamide C, which
also represents the first member of the sessilifoliamide family
to succumb to total synthesis. Noteworthy aspects of our
approach include a convergent strategy featuring a [3,3]-
rearrangement to install two contiguous stereocenters at
C(9)ÀC(10) and the ability to access the stemoamide ring
system in a unified strategy from linchpin intermediate 8.
The synthetic efficiency of the key IrelandÀClaisen rearran-
gement benefited from increasing the steric bulk of the ketene
acetal silyl substituent to proceed through a preferred boat-
like transition state while providing a level of diastereoselec-
tion that is notable in a bicyclic system.18
Surprisingly, methyl esters 27a and 27b proved to be resis-
tant to nucleophilic addition of Grignard reagents and meta-
lated dimethylhydroxylamine, even at elevated temperatures.
We speculated that increased conformational flexibility and a
lower oxidation state at carbonyl C(11) could promote reactivity.
Acknowledgment. This work was supported by the NIH/
NIGMS CMLD program (GM067082). We thank Dr. John
Williams (University of Pittsburgh) for high resolution mass
spectrometry and Mr. Mike Delk (University of Pittsburgh) for
NMR spectrometer maintenance. A.T.H. thanks the Goldblatt
family for a University of Pittsburgh graduate fellowship.
(15) Degassing was necessary to suppress the formation of oxidative
byproducts.
(16) (a) Transition states were identified using a PM3 transition state
search in Spartan 10 (Wavefunction, Inc., Irvine, CA). A substituent-
simplified (R = Me, Me in place of Et) boat analogue of 29q was found to
be approximately equal in energy to the corresponding analog of chair 30q
(Scheme 8). (b) For another observation of the effect of the steric bulk of the
silyloxy group on the diastereoselectivity of the IrelandÀClaisen rearrange-
ment, see: Chen, C.-L.; Namba, K.; Kishi, Y. Org. Lett. 2009, 11, 409.
(17) The synthetic product was spectroscopically identical to the natural
product,5 with the exception of the magnitude of the specific rotation (natural:
[R]26D À140 (c 0.17, CHCl3), synthetic: [R]23D À84.6 (c 2.5, CHCl3).
Supporting Information Available. Experimental proce-
dures and spectral data for all new compounds, including
copies of 1H and 13C NMR spectra. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
€
(18) (a) Castro, A. M. M. Chem. Rev. 2004, 104, 2939. (b) Gul, S.;
Schoenebeck, F.; Aviyente, V.; Houk, K. N. J. Org. Chem. 2010, 75, 2115.
Org. Lett., Vol. 13, No. 10, 2011
2637