Total synthesis of (+)-lyconadin A and related compounds via oxidative C-N bond formation

Article abstract of DOI:10.1021/ja903868n

The formation of carbon-nitrogen (C-N) bonds is a fundamental bond construction in organic synthesis and is indispensable for the synthesis of alkaloid natural products. In the context of the synthesis of the architecturally complex Lycopodium alkaloid lyconadin A, we have discovered a highly efficient oxidative C-N bond forming reaction that relies on the union of a nitrogen anion and a carbon anion. Empirical evidence amassed during our synthetic studies suggests that the mechanism of the C-N bond forming process encompasses polar as well as radical processes. Herein, we present our study of this novel C-N bond forming reaction and its application to the enantioselective total synthesis of lyconadin A and related derivatives.

Full text of DOI:10.1021/ja903868n

Published on Web 07/10/2009
Total Synthesis of (+)-Lyconadin A and Related Compounds
via Oxidative C-N Bond Formation
Scott P. West, Alakesh Bisai, Andrew D. Lim, Raja R. Narayan, and
Richmond Sarpong*
Department of Chemistry, UniVersity of California, Berkeley, California 94720
Received May 12, 2009; E-mail: rsarpong@berkeley.edu
Abstract: The formation of carbon-nitrogen (C-N) bonds is a fundamental bond construction in organic
synthesis and is indispensable for the synthesis of alkaloid natural products. In the context of the synthesis
of the architecturally complex Lycopodium alkaloid lyconadin A, we have discovered a highly efficient
oxidative C-N bond forming reaction that relies on the union of a nitrogen anion and a carbon anion.
Empirical evidence amassed during our synthetic studies suggests that the mechanism of the C-N bond
forming process encompasses polar as well as radical processes. Herein, we present our study of this
novel C-N bond forming reaction and its application to the enantioselective total synthesis of lyconadin A
and related derivatives.
Introduction
endeavors can play a critical role in the bioactivity elucidation
studies of the Lycopodium alkaloids by providing access to
Since the isolation of the first Lycopodium alkaloid natural
product (lycopodine (1, Figure 1)) by Bo¨deker in 18811 and
the structural elucidation of annotinine (2) by Wiesner in 1957,2
synthetic organic chemists have remained fascinated with
the architecture of this family of compounds. The majority of
these natural products are believed to arise in Nature from the
phlegmarine skeleton (see 3). The reigning postulate for the
biosynthesis of these compounds was advanced by Spenser3 and
is well described in the literature.4,5 Over the past three decades,
there has been an increase in the number and diversity of
alkaloids identified as constituents of the Lycopodium family.6
Several members, exemplified by huperzine A (4),7 have been
shown to be potent inhibitors of acetylcholinesterase and have
begun to find use in the treatment of Alzheimer’s disease in
China.8 In addition, biological activity ranging from neurotrophic
activity to anticancer properties has been reported for others in
this family.9
significant quantities of the natural products and related deriva-
tives. We have been interested in tracing the connections among
a subset of Lycopodium alkaloids referred to as the “miscel-
laneous” group10 because they possess unique frameworks
distinct from the traditional structural classes (i.e., lycopodine,
lycodine, and fawcettimine). The development of a unified
strategy for the synthesis of the “miscellaneous” group would
afford opportunities to study their biosynthetic relationships as
well as provide unambiguous structural characterization and
satisfactory quantities for biological studies.
On the basis of the structural resemblance between dihydro-
lycolucine (5), lucidine A (6), and oxolucidine A (7), we
reasoned that these compounds could arise synthetically from
a common precursor (8) related to the tetracyclic core of 5 via
a series of hydrogenation and/or oxygenation reactions. Impor-
tantly, installation of a C11-N bond (lucidine numbering; see
5)11 could offer an opportunity to access lyconadin A (9)12 and
related compounds such as lyconadin B (10).13 Moreover, in
line with the proposed biogenesis of spirolucidine (11) from an
oxolucidine relative,14 ring-contractive rearrangement of the
R-hydroxyimine moiety of 7 could form the spirotetracycle of
11. Additionally, the spirocyclic nankakurine skeleton (see 12),
Despite the reported bioactivity of select congeners, a
comprehensive biological screen of the majority of the Lyco-
podium alkaloids has yet to be undertaken. Total synthesis
(1) Bo¨deker, K. Ann. Chem. 1881, 208, 363.
(2) Wiesner, K.; Ayer, W. A.; Fowler, L. R.; Valenta, Z. Chem. Ind.
(London) 1957, 564–565.
(3) For leading references, see: (a) Castillo, M.; Gupta, R. N.; MacLean,
D. B.; Spenser, I. D. Can. J. Chem. 1970, 48, 1893–1903. (b) Castillo,
M.; Gupta, R. N.; Ho, Y. K.; MacLean, D. B.; Spenser, I. D. Can.
J. Chem. 1970, 48, 2911–2918. (c) Heimscheidt, T.; Spenser, I. D.
J. Am. Chem. Soc. 1993, 115, 3020–3021. (d) Heimscheidt, T.;
Spenser, I. D. J. Am. Chem. Soc. 1996, 118, 1799–1800.
(4) Ayer, W. A. Nat. Prod. Rep. 1991, 8, 455–463.
(8) Jiang, H.; Luo, X.; Bai, D. Curr. Med. Chem. 2003, 10, 2231–2252.
(9) For a recent review on Lycopodium alkaloids, which discusses their
biological activity in detail, see: Ma, X.; Gang, D. R. Nat. Prod. Rep.
2004, 21, 752–772.
(10) This word was introduced to describe a subset of Lycopodium alkaloids
of unclassified biogenesis. See ref 9.
(5) Ayer, W. A.; Trifonov, L. S. Lycopodium Alkaloids. In The Alkaloids;
Cordell, G. A., Brossi, A., Eds.; Academic Press: New York, 1994;
Vol. 45, pp 233-266.
(11) For lucidine numbering, see: 6 (Figure 1) and ref 9.
(12) Kobayashi, J.; Hirasawa, Y.; Yoshida, N.; Morita, H. J. Org. Chem.
2001, 66, 5901–5904.
(6) For a recent review, see: Hirasawa, Y.; Kobayashi, J.; Morita, H.
Heterocycles 2009, 77, 679–729.
(13) Ishiuchi, K.; Kubota, T.; Hoshino, T.; Obara, Y.; Nakahata, N.;
Kobayashi, J. Biorg. Med. Chem. 2006, 14, 5995–6000.
(14) (a) Tori, M.; Shimoji, T.; Takaoka, S.; Nakashima, K.; Sono, M.; Ayer,
W. A. Tetrahedron Lett. 1999, 40, 323–324. (b) Tori, M.; Shimoji,
T.; Shimura, E.; Takaoka, S.; Nakashima, K.; Sono, M.; Ayer, W. A.
Phytochemistry 2000, 53, 503–509.
(7) (a) Liu, J. S.; Zhu, Y. L.; Yu, C. M.; Zhou, Y. Z.; Han, Y. Y.; Wu,
F. W.; Qi, B. F. Can. J. Chem. 1986, 64, 837–839. (b) For a recent
review on huperzine A, see Ma, X.; Tan, C.; Zhu, D.; Gang, D. R.;
Xiao, P. J. Ethnopharmacol. 2007, 113, 15–34.
9
10.1021/ja903868n CCC: $40.75  2009 American Chemical Society
J. AM. CHEM. SOC. 2009, 131, 11187–11194 11187

A R T I C L E S
West et al.
Figure 1. Selected congeners of the Lycopodium alkaloid family.
Scheme 1
which was recently revised by Kobayashi to be epimeric at C13
with respect to 1115 and confirmed by Overman and co-workers
via total synthesis,16 could be constructed by a related R-hy-
droxyimine rearrangement.
Prior to our initial communication on lyconadin A, only one
total synthesis of this Lycopodium alkaloid, by Beshore and
Smith, had been reported.17 Smith’s elegant approach to the
lyconadins also provided a synthesis of (-)-lyconadin B and
unambiguously established the absolute configurations of these
natural products. In addition, the groups of Castle18 and Hsung19
have disclosed creative approaches to the lyconadins.
Our synthetic studies of lyconadin A (9), reported herein,
have culminated in a concise, enantioselective synthesis of
tetracycle 8, which has been applied to a total synthesis of (+)-
lyconadin A using an unprecedented oxidative C-N bond
forming reaction.20 Additionally, we report studies that have
enabled us to gain insight into the C-N bond forming process
for the synthesis of the lyconadin pentacycle. The general
synthetic strategy that was developed for the synthesis of
lyconadin A sets the stage for exploring the total syntheses of
many of the other Lycopodium alkaloids depicted in Figure 1.
Initial Synthetic Plan. Given our focus on the development
of a unified synthetic strategy, the initial goal was to achieve a
concise synthesis of tetracyclic amine 8, which could be
transformed into a number of different Lycopodium alkaloids
including lyconadin A. Our retrosynthetic approach to lyconadin
A is shown in Scheme 1. We anticipated that late stage C-N
bond formation to construct the lyconadin pentacycle could be
accomplished via lateral functionalization of the methoxypyri-
dine moiety of 8. Tetracycle 8 could be accessed from tricycle
13 using standard amination reactions. We imagined tricycle
13 could derive from pyridine-annulated cycloheptadiene 14.21
An important, albeit naive, assumption we made at this stage
was that the resident lone stereocenter in 14 would correctly
direct the installation of the four additional stereocenters present
in pyridine-annulated cycloheptane 13. Thus, the racemic
synthesis of 13 was an important milestone that would enable
a test of this hypothesis. We envisioned cycloheptadiene 14
(15) (a) Hirasawa, Y.; Morita, H.; Kobayashi, J. Org. Lett. 2004, 6, 3389–
3391. (b) Hirasawa, Y.; Kobayashi, J.; Obara, Y.; Nakahata, N.;
Kawahara, N.; Goda, Y.; Morita, H. Heterocycles 2006, 68, 2357–
2364.
(16) Nilsson, B. L.; Overman, L. E.; de Alaniz, J. R.; Rohde, J. M. J. Am.
Chem. Soc. 2008, 130, 11297–11299.
(17) (a) Beshore, D. C.; Smith, A. B., III. J. Am. Chem. Soc. 2007, 129,
4148–4149. (b) Beshore, D. C.; Smith, A. B., III. J. Am. Chem. Soc.
2008, 130, 13778–13789.
(18) Grant, S. W.; Zhu, K.; Zhang, Y.; Castle, S. L. Org. Lett. 2006, 8,
1867–1870.
(19) Tracey, M. R.; Hsung, R. Abstract of Papers, Presented at the 226th
National Meeting of the American Chemical Society, New York,
September 2003; paper ORGN-721.
(21) For other examples of syntheses of natural products that contain a
seven-membered ring using cycloheptadiene intermediates from our
group, see: (a) Simmons, E. M.; Sarpong, R. Org. Lett. 2006, 8, 2883–
2886. (b) Simmons, E. M.; Yen, J. R.; Sarpong, R. Org. Lett. 2007,
9, 2705–2708.
(20) For an initial communication, see: Bisai, A.; West, S. P.; Sarpong,
R. J. Am. Chem. Soc. 2008, 130, 7222–7223.
9
11188 J. AM. CHEM. SOC. VOL. 131, NO. 31, 2009

Synthesis of (+)-Lyconadin A and Related Compounds
A R T I C L E S
hydrogenation conditions employing Brown’s,26 Wilkinson’s,
or Crabtree’s catalysts were unsuccessful in promoting reduction
of the diene from the desired face to provide the correct relative
stereochemistry for the natural product.27 We believe that the
pseudoequatorial orientation of the C13 hydroxyl group in the
diene substrate prevents productive metal-hydride orientation.
Additionally, coordination of the methoxypyridine moiety of
the substrate to the homogeneous hydrogenation catalysts could
lead to reduced activity. These factors most likely account for
the lack of directing group effect in the hydrogenation and
instead result in sterically controlled reduction of the diene from
the face opposite the C15 methyl group to yield 20 as the major
product.
Scheme 2
Nonetheless, 20 provided a serviceable model system to
investigate the synthesis of the advanced tetracycle of the
lyconadins and other related lucidine alkaloids. In this regard,
a sequence involving MOM protection of the C13 hydroxyl
group of 20, saponification of the ethyl ester, and Curtius
rearrangement using diphenylphosphoryl azide (DPPA)28 in the
presence of benzyl alcohol afforded Cbz-protected amine 21 in
63% overall yield. Cleavage of the MOM ether was followed
by Swern oxidation to provide an intermediate ketone, which
was subjected to reductive amination conditions to afford
tetracycle 22. Protection of the secondary amine with a Boc
group gave 23, which furnished X-ray quality crystals that
enabled an unequivocal assignment of this late stage tetracycle.
Secondary amine 22 differed from desired tetracycle 8 (Scheme
1) only in the C15 stereochemistry.
To obtain tetracycle 8 with the desired C15 stereochemistry
for the synthesis of lyconadin A, we began with vinylogous
ester 2529 and bromopicoline 16 (Scheme 4). Following a
sequence similar to that previously described in Scheme 3
provided access to alcohol 26, which was converted to enone
27 by Swern oxidation and subsequent Saegusa-Ito oxidation.30
Conjugate addition of the Gilman reagent to 27 proceeded from
the convex face (R-face) to install the methyl group with the
desired stereochemistry at C15 (see 28). Tricycle 28 was carried
on to tetracycle 8 following synthetic steps analogous to those
already established for the closely related conversion of 20 to
22 (Scheme 3).
Forging the Key C-N Bond en Route to Lyconadin A. With
tetracycle 8 in hand, we initially investigated its conversion to
the lyconadin pentacycle by functionalization at C6 using the
Boekelheide variant31 of the Polonovski rearrangement32 em-
ploying the pyridine N-oxide of 29 (Scheme 5). Treatment of
29 with m-chloroperbenzoic acid generated the N-oxide inter-
mediate, but subsequent reaction with trifluoroacetic anhydride
(TFAA) or acetic anhydride did not lead to the desired product
(30) bearing a nucleofuge at C6. Inspired by Shibanuma’s
examples of building highly caged alkaloid structures,33 an
alternative approach to form pentacycle 32 via a Hofmann-
arising from a union of vinylogous ester 15 and bromomethox-
ypicoline 16.
Results and Discussion
Construction of the Pyridine-Annulated Cycloheptane 13.
Our synthesis of 13 began with the preparation of bromopicoline
derivative 16 (Scheme 2) starting from commercially available
methoxypicoline 17. Following literature precedent, bromination
of 17 using dibromodimethylhydantoin (DBDMH) gave 16 in
97% yield.22 The corresponding coupling partner (15) was
prepared by a sequence that began with allylation of com-
mercially available 18,23 followed by formation of the vinylo-
gous ester using trimethyl orthoformate in 64% yield over the
two steps.
The union of 15 and 16 (Scheme 3) was accomplished by
initial lateral deprotonation of 16 followed by reaction with 15
in a Stork-Danheiser sequence to give 19 in 63% yield.24
Initially, we investigated the oxidative cleavage of the terminal
double bond of 19 to afford an aldehyde followed by a Wittig
reaction to install an R,ꢀ-unsaturated ester. However, be-
cause of the pronounced instability of the requisite intermediate
aldehyde, this proved to be a capricious route at best. These
stability concerns were obviated by using an alternative route
employing cross-metathesis of the allyl group of 19 with ethyl
acrylate, which was accomplished using the Grubbs-Hoveyda
second generation catalyst.25 Intramolecular Heck reaction of
the intermediate ester and concomitant double bond migration
furnished cycloheptadiene 14 in 77% yield over the two steps.
Luche reduction of the carbonyl group to the corresponding
alcohol (>14:1 dr) and ensuing hydrogenation afforded tricycle
20 as a single diastereomer, setting three stereocenters in a single
operation. Unfortunately, X-ray analysis of 24 (Figure 2)
revealed that the diastereoselectivity of the reduction (with
respect to the C15 stereocenter and the newly introduced
hydrogens; see 20) was anti, which is opposite to the relative
stereochemistry needed to access the natural product. In an
attempt to reverse the diastereoselectivity of the hydrogenation,
we sought to utilize the hydroxyl group resulting from carbonyl
reduction of 14 as a directing group. A variety of directed
(26) Evans, D. A.; Morrissey, M. M. J. Am. Chem. Soc. 1984, 106, 3866–
3868.
(27) For a review on directed hydrogenations, see: Hoveyda, A. H.; Evans,
D. A.; Fu, G. C. Chem. ReV. 1993, 93, 1307–1370.
(28) For a Curtius rearrangement using DPPA, see: Shiori, T.; Ninomiya,
K.; Yamada, S. J. Am. Chem. Soc. 1972, 94, 6203–6205.
(29) For the synthesis of 25, see: Patterson, J. W. Tetrahedron 1993, 49,
4789–4798.
(22) For a synthesis of bromopicoline 16, see: Haudrechy, A.; Chassaing,
C.; Riche, C.; Langlois, Y. Tetrahedron 2000, 56, 3181–3187.
(23) Rajamannar, T.; Palani, N.; Balasubramanian, K. K. Synth. Commun.
1993, 23, 3095–3108.
(30) Ito, Y.; Hirao, T.; Saegusa, T. J. Org. Chem. 1978, 43, 1011–1013.
(31) For pertinent accounts of the Boekelheide reaction, see: (a) Traynelis,
V. J. In Mechanisms of Molecular Migrations; Thyagarjan, B. S., Ed.;
Interscience: New York, 1969; Vol. 2, pp 1-42. (b) Cohen, T.; Deets,
G. L. J. Am. Chem. Soc. 1972, 94, 932–938.
(24) Stork, G.; Danheiser, R. L. J. Org. Chem. 1973, 38, 1775–1776.
(25) For a recent review, see: Schrodi, Y.; Pederson, R. L. Aldrichimica
Acta 2007, 40, 45–52.
(32) For a review, see: Grierson, D. Org. React. 1990, 39, 85–295.
9
J. AM. CHEM. SOC. VOL. 131, NO. 31, 2009 11189

A R T I C L E S
West et al.
Scheme 3
Scheme 4
Lo¨ffler-Freytag (HLF) reaction34 was explored. HLF substrate
31a was easily prepared by treatment of secondary amine 8 with
N-chlorosuccinimide (NCS). Unfortunately, under a variety of
conditions known to effect the HLF reaction,35 a complex
mixture of products was always obtained. Imine byproducts
resulting from elimination of HCl were observed in many cases.
Additionally, N-nitrosoamine 31b was synthesized from amine
8 by nitrosation with NOCl and pyridine to pursue a variant of
the Barton nitrite ester oxidation.36 In an attempt to functionalize
C6, photolysis of 31b was conducted and resulted in nonspecific
decomposition of the substrate. Alternatively, following the Rabe
example in the synthesis of quinine recently revisited by
Williams,37 we envisioned that lateral deprotonation of 31a
followed by intramolecular nucleophilic attack on the chloro-
amine nitrogen with loss of chloride could provide the desired
pentacycle. However, attempted deprotonation of 31a with a
variety of bases (LDA, NaHMDS, NaH, NaNH2, KOt-Bu) led
to complex mixtures of products. However, we were encouraged
to find that treatment of chloroamine 31a with KOH in refluxing
methanol effected the desired transformation to afford pentacycle
32, albeit in low yield.
deprotonation of 8 using excess n-BuLi38 at low temperature
(-78 °C) followed by oxidation of the resulting dianion to
deliver pentacycle 32 (Scheme 6). Initially, reductive elimination
strategies for C-N bond formation utilizing reagents such as
Pd(OAc)2, FeCl3, Cu(OAc)2, or Cu(OTf)2 with excess iodine
as an oxidant were attempted from the presumed dianion
intermediate 34.39 The eventual use of iodine alone as an oxidant
was fortuitous and emerged only after control experiments for
these reactions revealed that iodine in the absence of metal salts
accomplished the desired transformation. Demethylation of 32
using NaSEt proceeded without event to provide (()-lyconadin
A (9) in 76% yield.
Investigation of the Key C-N Bond Forming Reaction. Bond
forming reactions that rely on the linking of two organolithium
anions via oxidative processes are experiencing a renaissance
in modern organic synthesis. This powerful “umpolung” tactic
serves as an impetus for the design of new strategies in complex
molecule synthesis. While many notable examples that feature
C-C bond forming oxidative processes have appeared,40 studies
detailing C-N bond formation via oxidation of dianionic
intermediates have, to our knowledge, not been pursued.
Ultimately, we found that the desired C-N bond can be
formed cleanly in high yield by a one-pot process that entails
(33) Shibanuma, Y.; Okamoto, T. Chem. Pharm. Bull. 1985, 33, 3187–
3194.
(34) For pertinent reviews, see: (a) Majetich, G.; Wheless, K. Tetrahedron
1995, 51, 7095–7129. (b) Wolff, M. E. Chem. ReV. 1963, 63, 55–64.
(35) Corey, E. J.; Hertler, W. R. J. Am. Chem. Soc. 1960, 82, 1657–1668.
(36) For a recent review, see: Suginome, H. In CRC Handbook of Organic
Photochemistry and Photobiology, 2nd ed.; Horspool, W. M., Lenci,
F., Eds.; CRC Press: Boca Raton, FL, 2004; pp 101-116.
(37) (a) Rabe, P.; Kindler, K. Ber. Dtsch. Chem. Ges. 1918, 51, 466–467.
(b) Smith, A. C.; Williams, R. M. Angew. Chem., Int. Ed. 2008, 47,
1736–1740.
(38) Subsequent optimization studies have shown that these reactions require
only 2 equiv of n-BuLi.
(39) Dieter, R. K.; Li, S.; Chen, N. J. Org. Chem. 2004, 69, 2867–2870.
Figure 2. ORTEP representation of 24.
9
11190 J. AM. CHEM. SOC. VOL. 131, NO. 31, 2009

Synthesis of (+)-Lyconadin A and Related Compounds
A R T I C L E S
Scheme 5
Scheme 6
group in 8 (with Boc, allyl, Cbz, etc.) followed by attempted
lateral deprotonation of the “picolinic” methylene position and
trapping with a variety of electrophiles (e.g., D2O, PhSeCl, I2)
returned only the starting material (Scheme 7). Additionally,
we believe that deprotonation to form lithium dianion 34 may
be favorable due to the added stabilization provided by formation
of a six-membered chelate of the nitrogen lone pair with the
C6-bound lithium (see 34, Scheme 6), in line with earlier
examples.43,44
Primarily due to its ease of access, model tetracyclic amine
3645 (Scheme 8), which lacks the methyl substituent at C15,
was used for further investigation of the key C-N bond
formation. When 36 is treated with n-BuLi (2 equiv), it
presumably generates the dianion 37 which is converted to
pentacyclic amine 39 upon treatment with iodine (1 equiv).
Mechanistically, the C-N bond formation may proceed via
C-iodination at the picolinic position of dianion 37 to form
iodide 38, which is rapidly converted to pentacycle 39. The
stereochemistry at C6 upon reaction of 37 with electrophiles
deserves further comment. Because of the anticipated configu-
rational stability of dianion 37 at -78 °C, reaction with sterically
small, “hard” electrophiles (e.g., D2O) should yield the products
of “endo” addition (see 40). Indeed, quenching 37 with D2O
did result in deuterium incorporation at the C6 endo position.46
Subjection of deuterated amine 40a (72% D) to n-BuLi followed
by addition of H2O or D2O revealed that quenching occurs
exclusively at the endo position of the picolinic methylene
carbon (see 40a (72% D) f 40b (85% D) and 40a f 36,
Scheme 8). Conversely, the addition of a halogen electrophile
(e.g., iodine) likely proceeds with inversion about C6 by virtue
of the steric size of the electrophile and its “soft” nature. This
potential stereodivergence, based on the electrophile, in the
In the context of our synthesis of lyconadin A (9), we viewed
a direct C-N bond forming approach as a powerful simplifying
transform. Presumably, upon treatment of secondary amine 8
with n-BuLi, the resulting lithium amide anion serves as an
intramolecular base to effect lateral deprotonation of the
methoxypyridine moiety to ultimately yield dianion 34.41
Because of the relative acidity of the picoline pseudobenzylic
position (pKa ≈ 34 in THF),42 deprotonation of the secondary
amine of 8 might not be viewed as an important requirement
for lateral deprotonation of the pseudobenzylic (“picolinic”)
position. However, our investigations of the direct C-N bond
formation suggest that deprotonation of the secondary amine
group is indeed a vital component of the overall process. The
requisite initial deprotonation of the secondary amine is sup-
ported by the observation that protection of the secondary amine
(40) For pertinent examples, see: (a) Martin, C. L.; Overman, L. E.; Rohde,
J. M. J. Am. Chem. Soc. 2008, 130, 7568–7569. (b) Baran, P. S.;
Hafensteiner, B. D.; Ambhaikar, N. B.; Guerrero, C. A.; Gallagher,
J. D. J. Am. Chem. Soc. 2006, 128, 8678–8693. (c) Paquette, L. A.;
Bzowej, E. I.; Branan, B. M.; Stanton, K. J. J. Org. Chem. 1995, 60,
7277–7283. (d) Cohen, T.; McNamara, K.; Kuzenko, M. A.; Ramig,
K.; Landi, J. J., Jr.; Dong, Y. Tetrahedron 1993, 49, 7931–7942.
(41) For a discussion of lateral deprotonation and dianion chemistry, see:
(a) Thompson, C. M. Dianion Chemistry in Organic Synthesis; CRC
Press: Boca Raton, FL, 1994. (b) Brandsma, L. PreparatiVe Polar
Organometallic Chemistry 2; Springer-Verlag: Berlin-New York, 1987;
p 127. (c) Wheatley, A. E. H. Eur. J. Inorg. Chem. 2003, 3291–3303.
(42) Fraser, R. R.; Mansour, T. S.; Savard, S. J. Org. Chem. 1985, 50,
3232–3234.
(43) Li-amide aggregates have been extensively characterized. See: Chad-
wick, S. T.; Ramirez, A.; Gupta, L.; Collum, D. B. J. Am. Chem. Soc.
2007, 129, 2259–2268.
(44) Basu, A.; Thayumanavan, S. Angew. Chem., Int. Ed. 2002, 41, 716–
738.
(45) Tetracyclic amine 36 was obtained from alcohol 26 in 46% yield over
six steps analogous to the conversion of 20 to 22 (Scheme 3).
(46) Determined on the basis of 2D NMR studies of 36 and 40. See
Supporting Information for details.
9
J. AM. CHEM. SOC. VOL. 131, NO. 31, 2009 11191

A R T I C L E S
West et al.
Scheme 7
Scheme 8
Scheme 9
Scheme 10
reactions of sterically encumbered lithium anions is supported
by observations made in the 1960s by Applequist47 and Glaze.48
Additionally, the studies on the “conducted tour mechanism”
for epimerization of heteroatom stabilized carbanions first
suggested by Cram and Gosser49 also support reaction with
inversion using halogen electrophiles.50
In addition to iodine, other oxidants such as NCS, NBS, NIS,
and (diacetoxyiodo)benzene (PIDA; Scheme 9) can effect the
9
11192 J. AM. CHEM. SOC. VOL. 131, NO. 31, 2009

Synthesis of (+)-Lyconadin A and Related Compounds
A R T I C L E S
Scheme 11
desired C-N bond formation to provide 39 in comparable yield
(65% yield for PIDA, 80-90% yield for NBS, NCS, and NIS).51
Recrystallization of the HCl salt of tertiary amine pentacycle
39 provided crystals suitable for X-ray analysis (see 42),
providing unambiguous support for the connectivity of 39.
To gain further insight into the electrophilic trapping of
carbanion 37, studies with diphenyl disulfide were conducted
(eqs 1-4). Treatment of 36 with 2 equiv of n-BuLi followed
by 12 equiv of PhSSPh yielded disulfide 4352 in 40% yield (eq
1), whereby reaction with inversion of the carbanion was
observed.53 Along with 43, 22% of pentacycle 39 and trace
amounts of 45 (eq 4) were recovered. Conversely, generation
of dianion 37 using 3 equiv of n-BuLi and trapping with 2 equiv
of PhSSPh yielded desired pentacycle 39 as the major product
in 64% yield (eq 2) along with trace amounts of 45. To better
understand the C-N bond formation using PhSSPh, we
independently prepared N-sulfide 44 (eq 3),54 which upon
treatment with n-BuLi yielded pentacycle 39 in 71% yield. This
preliminary observation lends support to 44 as a potential
precursor to the C-N bond-forming event.55 In an interesting
development, treating 36 with 3 equiv of n-BuLi followed by
addition of 8 equiv of PhSSPh yielded ketosulfide 45 (eq 4) as
the major product in 55% yield following aqueous workup. The
structure of 45 was unambiguously established by X-ray
crystallography as illustrated by the ORTEP picture in eq 4.56
Although the observations associated with the reaction of
dianion 37 using PhSSPh as an electrophile are suggestive of a
polar process, the possibility of single electron transfer (SET)
processes cannot be ruled out. The SET reactions of metalated
organic compounds have been well documented and appear to
have some dependency on the nature of the electrophile.57 On
the basis of this existing precedent, we decided to test the
likelihood of SET processes in the C-N bond formation from
dianion 37. In the key test, subjecting 36 to 2 equiv of n-BuLi
at -78 °C followed by treatment with TEMPO (2.5 equiv) led
to clean formation of pentacycle 39 in 70% yield (eq 5). These
observations (eqs 1-5) taken together support a complementary
polar and radical nature to the C-N bond forming process and
accentuate the continuum of reactivity that is possible with these
dianion intermediates depending on the choice of electrophile/
oxidant.
Enantioselective Total Synthesis of (+)-Lyconadin A. We
designed the synthesis of lyconadin A to highlight a unified
approach to the “miscellaneous” group of Lycopodium alkaloids.
Furthermore, to initiate highly convergent syntheses of dihy-
drolycolucine (5) and spirolucidine (11) via fragment coupling
of two enantioenriched pieces, we targeted the enantioselective
synthesis of the common tetracyclic intermediate 8 (Scheme
10). This exercise commenced with the Corey-Bakshi-
Shibata (CBS) reduction58 of the previously prepared cyclo-
heptadienone 46 (from 16 and 25) to afford allylic alcohol 48
in 85% yield and 98% ee. Cycloheptadienol 48 served as a
substrate for hydrogenation using Pd/C and H2 (1 atm) to afford
pyridine-annulated cycloheptane 26 as the major diastereomer
(8:1 dr). A screen of various hydrogenation catalysts and
solvents did not lead to any improvement in the diastereocon-
trol.59 However, recrystallization of the product mixture pro-
vided 26 as a single diastereomer in 60% yield and 99% ee.
(51) A mechanism involving the formation of a carbene intermediate (i)
followed by N-H insertion to form 39 cannot be discounted.
(47) Applequist, D. E.; Chmurny, G. N. J. Am. Chem. Soc. 1967, 89, 875–
880.
(48) Glaze, W. H.; Selman, C. M.; Ball, A. L.; Bray, L. E. J. Org. Chem.
1969, 34, 641–644.
(49) Cram, D. J.; Gosser, L. J. Am. Chem. Soc. 1964, 86, 2950–2952.
(50) For a pertinent example, see: Gawley, R. E.; Zhang, Q. J. Org. Chem.
1995, 60, 5763–5769.
(52) Stereochemistry was determined by NOESY measurements.
9
J. AM. CHEM. SOC. VOL. 131, NO. 31, 2009 11193

A R T I C L E S
West et al.
The absolute and relative stereochemistry of alcohol 26 was
confirmed by X-ray crystallographic analysis.60,61 Swern oxida-
tion of 26 proceeded without event to afford ketone 49.62 With
49 in hand, the enantioselective synthesis of (+)-lyconadin A
was achieved by a sequence analogous to that described in
Schemes 4 and 6. Spectral data for synthetic (+)-lyconadin A
(9) were in agreement with spectroscopic (1H, 13C, IR, MS) and
chiroptical data obtained for natural (+)-lyconadin A (9) as well
as the synthetic material prepared by Smith and Beshore.
Synthesis of C15-epi-lyconadin A and C15-nor-Me-lyconadin
A. As a further testament to the generality of our approach to
lyconadin A, we have applied an analogous synthetic protocol
to the preparation of analogues of lyconadin A that differ at
the C15 position (Scheme 11).63 Tetracycle 22, prepared as
detailed in Scheme 3, smoothly underwent C-N bond formation
and was advanced to C15-epi-lyconadin A (50). Cleavage of
the methyl ether of pentacycle 39 (Scheme 11) provided C15-
nor-Me-lyconadin A (51). These compounds will be evaluated
alongside lyconadin A in a comprehensive screen for neu-
rotrophic biological activity.
Conclusion
In summary, an enantioselective total synthesis of the
structurally complex Lycopodium alkaloid (+)-lyconadin A has
been achieved in 17 steps. A key aspect of this synthesis was
the development of an operationally simple C-N bond-forming
reaction to efficiently forge the pentacyclic core of the lycona-
dins. Empirical evidence amassed during our synthetic studies
suggests the C-N bond formation may proceed via polar or
SET mechanisms. The route to (+)-lyconadin A features a
highly enantioselective Corey-Bakshi-Shibata reduction and
a diastereoselective hydrogenation which sets three stereocenters
in a single operation. Our approach lays the foundation for the
enantioselective total syntheses of many other “miscellaneous”
Lycopodium alkaloids including dihydrolycolucine, spiroluci-
dine, and the nankakurines. These efforts are currently underway
in our laboratories.
(53) The formation of ii along with 43 upon treatment of 36 with n-BuLi
(2 equiv) followed by the addition of PhSSPh (3 equiv) and subsequent
reaction of the crude product mixture with Boc2O supports initial
reaction of electrophiles with the carbanionic position of dianion 37.
Acknowledgment. The authors are grateful to UC Berkeley,
Abbott Laboratories, Eli Lilly, GlaxoSmithKline and Johnson and
Johnson for generous, unrestricted financial support. The authors
thank Materia (Pasadena, CA) for a gift of the Grubbs-Hoveyda
Generation II catalyst and BristolMyersSquibb (Dr. Rodney Par-
sons) for a gift of cyclohexanedione starting material. The authors
are indebted to Dr. Fred Hollander and Dr. Antonio DiPasquale
for crystallographic data. R.S. is a 2009 Alfred P. Sloan Fellow
and a 2009 Camille Dreyfus Teacher-Scholar.
Supporting Information Available: Experimental details and
characterization data for select compounds, in addition to
spectral comparison of synthetic and natural (+)-lyconadin A.
This material is available free of charge via the Internet at http://
pubs.acs.org.
(54) For the preparation of 44, see the Supporting Information.
(55) This is consistent with our initial observations regarding the reaction
of 31a to give 32 (Scheme 5).
(56) A proposed mechanism for the formation of 45 may proceed via iii-
vii as intermediates.
JA903868N
(58) (a) Corey, E. J.; Bakshi, R. K. Tetrahedron Lett. 1990, 31, 611–614.
(b) Corey, E. J.; Bakshi, R. K.; Shibata, S. J. Am. Chem. Soc. 1987,
109, 5551–5553. (c) Corey, E. J.; Bakshi, R. K.; Shibata, S.; Chen,
C. P.; Singh, V. K. J. Am. Chem. Soc. 1987, 109, 7925–7926.
(59) The diastereomers resulting from hydrogenation would lead to
enantiomers in subsequent oxidation steps, which would diminish the
overall ee of the final compound.
(60) Cu radiation was used to determine the absolute stereochemistry of
alcohol 26.
(61) (a) Flack, H. D. Acta Crystallogr., Sect. A 1983, 39, 876–881. (b)
Bernardinelli, G.; Flack, H. D. Acta Crystallogr., Sect. A 1985, 41,
500–511.
(57) For pertinent examples, see: (a) Meyers, A. I.; Edwards, P. D.; Reiker,
W. F.; Bailey, T. R. J. Am. Chem. Soc. 1984, 106, 3270. (b) Maji,
M. S.; Pfeifer, T.; Studer, A. Angew. Chem., Int. Ed. 2008, 47, 9547–
9550.
(62) If alcohol 26 is not recrystallized, Swern oxidation of the diastereomeric
mixture of alcohols from the hydrogenation led to ketone 51 in a
diminished 77% ee.
(63) For details, see the Supporting Information.
9
11194 J. AM. CHEM. SOC. VOL. 131, NO. 31, 2009