Total synthesis of the neotropical poison-frog alkaloid (-)-205B

Article abstract of DOI:10.1021/ol0510264

(Chemical Equation Presented) A stereocontrolled total synthesis of the neotropical poison-frog alkaloid (-)-205B (1) has been achieved, employing a dithiane three-component linchpin coupling, a one-pot sequential construction of the embedded indolizidine

Full text of DOI:10.1021/ol0510264

ORGANIC
LETTERS
2005
Vol. 7, No. 15
3247-3250
Total Synthesis of the Neotropical
Poison-Frog Alkaloid (−)-205B
Amos B. Smith, III* and Dae-Shik Kim
Department of Chemistry, Monell Chemical Senses Center and Laboratory for
Research on the Structure of Matter, UniVersity of PennsylVania, Philadelphia,
PennsylVania 19104
smithab@sas.upenn.edu
Received May 4, 2005
ABSTRACT
A stereocontrolled total synthesis of the neotropical poison-frog alkaloid (−)-205B (1) has been achieved, employing a dithiane three-component
linchpin coupling, a one-pot sequential construction of the embedded indolizidine ring, and ring-closing metathesis (RCM) to arrive at the
novel 8b-azaacenaphthylene ring system comprising the alkaloid. The synthesis proceeded with a longest linear sequence of 19 steps, affording
(−)-1 in 5.6% overall yield.
For nearly forty years the Daly Laboratory at the National
Institutes of Health has provided the biomedical community
with an almost bewildering array of architecturally diverse
alkaloids, isolated from the skin of neotropical poison-frogs;
examples include the steroidal batrachotoxins, the histrioni-
cotoxins, the gephyrotoxins, and the pumiliotoxins.1 One such
extract from the Panamanian frog Dendrobated pumilo
yielded alkaloid (-)-205B (1), possessing the unusual 8b-
azaacenaphthylene ring system.2 The structure, provisionally
reported in 1987,2a was assigned by Daly et al. in 1998, based
on an extensive series of FTIR, NMR, and HRMS studies,
in conjunction with molecular modeling, to comprise (-)-
1.2b The absolute stereochemistry was subsequently reported
in 2003 by Toyooka et al. based on their total synthesis of
the unnatural (+)-antipode.3a The central feature of the
Toyooka synthesis entailed a series of Michael-type additions
to enaminoesters; the longest linear sequence required 30
steps from known (S)-6-(tert-butyldiphenylsilyloxymethyl)-
piperidin-2-one.3b Although biological studies of the naturally
occurring alkaloid have not as yet been reported, due
presumably to lack of material, the unnatural (+)-antipode
was recently reported to block selectively the R7 nicotinic
receptor.4
Intrigued both by the architecture of alkaloid (-)-205B
(1) and the opportunity to provide material for biological
evaluation, we recently initiated a program to construct (-)-
1, as well as a small family of congeners. In this paper, we
report completion of the initial phase of this program: the
first total synthesis of the natural (-)-antipode of alkaloid
205B (1).
Our synthetic strategy calls upon our recently successful
three-component linchpin construction of the indolizidine
ring system (Scheme 1),5 employing an N-Ts aziridine as
(3) (a) Toyooka, N.; Fukutome, A.; Shinoda, H.; Nemoto, H. Angew.
Chem., Int. Ed. 2003, 42, 3808. (b) Toyooka, N.; Fukutome, A.; Shinoda,
H.; Nemoto, H. Tetrahedron 2004, 60, 6197. The known piperidone was
prepared in five steps and 35% overall yield from 4-methoxyphenylmeth-
ylhex-5-enoate; see: Hodgkinson, T. J.; Shipman, M. Synthesis 1998, 1141.
(4) Tsuneki, H.; You, Y.; Toyooka, N.; Kagawa, S.; Kobayashi, S.;
Sasaoka, T.; Nemoto, H.; Kimura, I.; Dani, J. A. Mol. Pharmacol. 2004,
66, 1061.
(1) Daly, J. W.; Garraffo, H. M.; Spande, T. F. In Alkaloids: Chemical
and Biological PerspectiVe; Pelletier, S. W., Ed.; Pergamon: New York,
1999; Vol. 13, p 1.
(2) (a) Tokuyama, T.; Nishimori, N.; Shimada, A.; Edwards, M. W.;
Daly, J. W. Tetrahedron 1987, 43, 643. (b) Tokuyama, T.; Garraffo, H.
M.; Spande, T. F.; Daly, J. W. An. Asoc. Quim. Argent. 1998, 86, 291.
10.1021/ol0510264 CCC: $30.25
© 2005 American Chemical Society
Published on Web 06/11/2005

Scheme 1
Scheme 2
(1.05 equiv) at -78 °C led to ketone (+)-9 in excellent yield
(92%). Protection of the carbonyl as the ethylene ketal
employing the Noyori conditions7 then furnished (+)-4.
For epoxide 6 (Scheme 3), Brown asymmetric crotylation8
of known aldehyde (-)-109 led to alcohol (-)-11 both in
good yield and with excellent diastereoselectivity (ca.
11:1). Protection of the resulting secondary alcohol, followed
by removal of the TBS groups next afforded diol (-)-12.
Completion of (-)-6 was achieved via a one-flask Fraser-
Reid epoxide protocol.10 The overall yield for the four-step
sequence was 65%.
Scheme 3
the second electrophile, followed by a one-pot sequential
cyclization to complete elaboration of the indolizidine ring
encased within the 8b-azaacenaphthylene ring. Ring-closing
metathesis (RCM) of the derived kinetic silyl enol ether
would then afford the 205B azaacenaphthylene skeleton.
Assuming success, this synthetic scenario would also hold
the promise of numerous readily available synthetic conge-
ners, requiring only alternation of the structure and/or
configuration of the epoxide and aziridine coupling partners.
With ample quantities of aziridine (+)-4 and epoxide (-)-6
in hand, we executed the multicomponent linchpin coupling
(Scheme 4). Pleasingly, lithiation of 5 in Et2O, followed in
turn by addition of epoxide (-)-6, warming the reaction
mixture to -20 °C over 0.5 h, stirring for an additional 1.5
h at -20 °C, and then addition of aziridine (+)-4 in THF
An additional site for structural diversity of the azaacenaph-
thylene skeleton in the C(4) carbonyl would be available after
RCM and removal of the dithiane.
We initiated the synthesis of alkaloid (-)-205B (1) with
construction of aziridine 4 (Scheme 2). Tosylation of the
nitrogen of commercially available serine methyl ester
hydrochloride (-)-7, followed by Mitsunobu ring closure,
furnished aziridine (+)-8,6 which upon treatment with MeLi
(6) (a) Baldwin, J. E.; Spivey, A. C.; Schofield, C. J.; Sweeney, J. B.
Tetrahedron 1993, 49, 6309. (b) Bergmeier, S. C.; Seth, P. P. J. Org. Chem.
1997, 62, 2671. (c) Fujii, N.; Nakai, K.; Tamamura, H.; Otaka, A.; Mimura,
N.; Miwa, Y.; Taga, T.; Yamamoto, Y.; Ibuka, T. J. Chem. Soc., Perkin
Trans. 1 1995, 1359.
(7) Tsunoda, T.; Suzuki, M.; Noyori, R. Tetrahederon Lett. 1980, 21,
1357.
(5) (a) Smith, A. B., III; Kim, D.-S. Org. Lett. 2004, 6, 1493. (b) Smith,
A. B., III; Pitram, S. M.; Boldi, A. M.; Gaunt, M. J.; Sfouggatakis, C.;
Moser, W. H. J. Am. Chem. Soc. 2003, 125, 14435. (c) Smith, A. B., III;
Boldi, A. M. J. Am. Chem. Soc. 1997, 119, 6925.
(8) Brown, H. C.; Bhat, K. S. J. Am. Chem. Soc. 1986, 108, 5919.
3248
Org. Lett., Vol. 7, No. 15, 2005

containing 1,2-dimethoxyethane (DME) to trigger the Brook
13 for the three-step sequence was 69%. Hydrolysis of the
acetal with 2 M HCl in acetone at reflux then provided ketone
(+)-2, which upon treatment with lithium hexamethyldisi-
lazide (LHMDS) in the presence of TMSCl led to the kinetic
silyl enol ether. Ring-closing metathesis employing the
second-generation Grubbs catalyst (14) furnished the requi-
site advanced tricyclic dithiane (+)-15 as a crystalline solid
(mp 101 °C).12 Single-crystal X-ray analysis established both
the structure and relative stereochemistry.
rearrangement, led to (+)-3 in 53% yield.11
Scheme 4
Having achieved construction of the azaacenaphthylene
ring system, we now faced what suprisingly proved to be a
difficult task, namely, introduction of an axial methyl group
at C(6) in (+)-15. From the stereoelectronic persepective, R
addition of a methyl nucleophile to the C(6) carbonyl would
be expected to furnish the equatorial alcohol. However,
deoxygenation of the latter via the Barton-McCombie13 and/
or related protocols14 involving radical mechanisms would,
in all likelihood, lead to the thermodynamically more stable
equatorial C(6) methyl substituent as a major product. We
therefore turned to the equatorial alcohol, readily available
upon reduction of (+)-15 with NaBH4. Unfortunately, all
attempts to displace the derived tosylate with a variety of
nucleophiles either proceeded in low yield or furnished
elimination products. We also examined the possibility of
hydrogenation of both the C(6) exomethylene and C(6)-
C(7) trisubstituted olefin congeners. Not suprisingly, in both
cases hydrogen was delivered from the less hindered R face
to furnish predominately the C(6) equatorial methyl congener
(ca. 5:1).
Undaunted, we next explored protonation of the enol or
enolate derived from the C(6) aldehyde. To this end, Wittig
olefination of (+)-15 (Scheme 5) with (methoxymethyl)-
triphenylphosphonium chloride employing t-BuOK to gener-
ate the ylide furnished an E/Z mixture (ca. 4:3) of methyl
enol ethers. To our delight, hydrolysis with 6 M HCl at 0
°C for 24 h furnished the desired axial aldehyde as the major
diastereomer (4:1; NMR).15 This result is explained via
electrostatic interactions; that is, axial delivery of a proton,
presumably the first step in the enol ether hydrolysis, would
be more sterically encumbered by electrostatic repulsion
between an incoming hydronium ion and the positive charge
of the protonated nitrogen than the equatorial delivery.
Without separation, reduction of the mixture of aldehydes
with NaBH4 provided alcohol (+)-16 in 74% isolated yield,
after separation from accompanying alcohol (+)-17 (18%).16
Mesylation of (+)-16, followed in turn by reduction of the
Sequential closure of the two rings that comprise the
indolizidine ring system, embedded in the 8b-azaacenaph-
thylene ring system of alkaloid (-)-1, was next achieved by
removal of the silyl groups (TBAF), bismesylation of the
resulting diol, and treatment of the bismesylate with potas-
sium carbonate in MeOH, followed without purification by
addition of sodium amalgam (5%); the overall yield of (+)-
(9) (a) Shimizu, A.; Nishiyama, S. Tetrahedron Lett. 1997, 38, 6011.
(b) Chattopadhyay, S.; Mamdapur, V. R.; Chadha, M. S. Tetrahedron 1990,
46, 3667. Aldehyde (-)-10 was prepared in four steps and 95% overall
yield from commercially available ethyl (S)-3-(2,2-dimethyl-1,3-dioxolan-
4-yl)-2-propenoate (i): see the Supporting Information for details.
(12) (a) Okada, A.; Ohshima, T.; Shibasaki, M. Tetrahedron Lett. 2001,
42, 8023. (b) Arisawa, M.; Theeraladanon, C.; Nishida, A.; Nakagawa, M.
Tetrahedron Lett. 2001, 42, 8029. (c) Scholl, M.; Ding, S.; Lee, C. W.;
Grubbs, R. H. Org. Lett. 1999, 1, 953.
(13) (a) Barton, D. H. R.; McCombie, S. W. J. Chem. Soc., Perkin Trans.
1 1975, 1574. (b) Barton, D. H. R.; Parekh, S. I.; Tse, C.-L. Tetahedron
Lett. 1993, 34, 2733.
(14) Zhang, L.; Koreeda, M. J. Am. Chem. Soc. 2004, 126, 13190 and
references therein.
(15) Hydrolysis at room temperature for 12 h led to a mixture (ca. 1:1)
of aldehydes.
(16) The stereochemistry at the newly formed position in (+)-16 was
later confirmed by comparison of the NMR data of (-)-19 with those of
known (+)-19; see ref 3.
(10) (a) Hicks, D. R.; Fraser-Reid, B. Synthesis 1974, 203. (b) Cink, R.
D.; Forsyth, C. J. J. Org. Chem. 1995, 60, 8122.
(11) In this case, use of HMPA or DMPU as cosolvents to trigger the
Brook rearrangement resulted in poor yield (<10%).
Org. Lett., Vol. 7, No. 15, 2005
3249

protocol17 [e.g., bis(trifluoroacetoxy)iodobenzene] then fur-
nished ketone (-)-19, which was subjected to the Toyooka
endgame sequence.3 Specifically, Wittig methylenation fol-
lowed by acid-catalyzed isomerization of the resultant
exomethylene alkene to the internal olefin afforded the
natural alkaloid (-)-205B as the major isomer (6.2:1; NMR)
in an isolated yield of 70%. Synthetic (-)-1 possessed
spectral data {e.g., 400 and 500 MHz 1H and 125 MHz 13C;
[R]D ) -8.3 (c 0.12, CHCl3) [lit.2 [R]D ) -8.5 (c 0.59,
CHCl3)]} identical to the data reported for both the natural
product2 and the synthetic antipode {[R]D ) +8.1 (c 1.05,
CHCl3)},3 except, of course, in the latter case for the
chiroptical properties.
Scheme 5
In summary, we have achieved an effective, stereocon-
trolled total synthesis of alkaloid (-)-205B (1). Highlights
of the synthetic strategy include an indolizidine ring con-
struction tactic, comprising a three-component linchpin
coupling of (+)-4, 5, and (-)-6, followed by a one-pot
sequential cyclization. Ring-closing metathesis completed
construction of the tricyclic 8b-azaacenaphthylene ring
system of alkaloid (-)-205B. Final stereoselective installation
of the axial methyl group at C(6) and application of the
Toyooka endgame led to the natural antipode of alkaloid (-)-
205B (1). The longest linear sequence to (-)-1 required 19
steps and proceeded in 5.6% overall yield, beginning with
known aldehyde (-)-10. Biological evaluation of (-)-1 and
the construction of related congeners will be reported in due
course.
Acknowledgment. Financial support was provided by the
National Institutes of Health (Institute of General Medical
Sciences) through Grant No. GM-29028 and an Eli Lilly
Foundation Fellowship to D.-S.K. We also thank Dr. John
W. Daly, National Institutes of Health, for kindly providing
the 400 MHz 1H NMR and the proton/carbon HMBC spectral
data for the natural antipode of alkaloid 205B.
Supporting Information Available: Experimental pro-
cedures and spectroscopic and analytical data for all new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL0510264
mesylate with Super-Hydride (LiHBEt3), afforded dithiane
(+)-18. Removal of the dithiane exploiting the Stork
(17) (a) Fleming, F. F.; Funk, L.; Altundas, R.; Tu, Y. J. Org. Chem.
2001, 66, 6502. (b) Stork, G.; Zhao, K. Tetrahedron Lett. 1989, 30, 287.
3250
Org. Lett., Vol. 7, No. 15, 2005