Concise total synthesis of (-)-calycanthine, (+)-chimonanthine, and (+)-folicanthine

Article abstract of DOI:10.1002/anie.200700705

Conjoined twins: An efficient and convergent strategy for the synthesis of the dimeric hexahydropyrroloindole alkaloids, (+)-chimonanthine, (+)-folicanthine, and (-)-calycanthine, is described. The simultaneous formation of the vicinal quaternary stereocenters by using a reductive dimerization reaction provides an access to the optically active key intermediate on a gram scale. (Chemical Equation Presented).

Full text of DOI:10.1002/anie.200700705

Angewandte
Chemie
DOI: 10.1002/anie.200700705
Natural Products
Concise Total Synthesis of (À)-Calycanthine, (+)-Chimonanthine, and
(+)-Folicanthine**
Mohammad Movassaghi* and Michael A. Schmidt
Hexahydropyrroloindole alkaloids constitute a large class of
natural compounds that are formally derived from trypto-
phan.[1] A fascinating array of dimeric and oligomeric
derivatives exist, a large subset of which contain vicinal
quaternary stereocenters at C3a and C3a’ of the hexahydro-
pyrroloindole substructure (Scheme 1). The alluring structure
lective syntheses that take advantage of dialkylation or
double Heck cyclizations to establish the quaternary stereo-
centers in these targets.[4] Herein we describe a concise
enantioselective synthesis of (+)-chimonanthine (1), (+)-
folicanthine (2), and (À)-calycanthine (3) by a convergent
reductive dimerization strategy to simultaneously secure the
vicinal quaternary stereocenters.
Our retrosynthetic analysis of (+)-chimonanthine (1) is
illustrated in Scheme 2. As part of a program directed toward
Scheme 2. Retrosynthetic analysis of (+)-1.
Scheme 1. Representative alkaloids derived from hexahydropyrroloin-
doles. Bn=benzyl.
the synthesis of dimeric hexahydropyrroloindole alkaloids we
sought an efficient and convergent strategy for the introduc-
tion of the two quaternary stereocenters with complete
stereoselectivity. A powerful disconnection that has here-
tofore been unexplored in the synthesis of these alkaloids
would involve the homocoupling of the enantiomerically
enriched tricyclic free radical 6, or a related derivative, to
of these alkaloids combined with their biological activity has
led to significant interest in their total syntheses. Important
contributions in this area have established a biosynthetically
inspired oxidative dimerization[2] of oxindoles and trypt-
amines for accessing derivatives in racemic and meso form.[3]
Overman and co-workers have reported elegant enantiose-
provide the crucial C3a C3a’ bond (Scheme 2).[5] We
À
expected the stereochemistry at the C8a-position of the
tricycle 6 to impose a high level of diastereoselectivity in the
bond-forming event and secure the vicinal quaternary ste-
reocenters in (+)-1. Tricyclic bromide 7 was predicted to serve
as a versatile precursor to 6, and was envisioned to be
prepared based on the elegant reports on the synthesis of
hexahydropyrroloindole alkaloids functionalized at the C3a-
position by the research groups of Hino, Crich, and Dan-
ishefsky.[1d,6]
Treatment of the commercially available Na-methoxycar-
bonyl-l-tryptophan methyl ester (8, Scheme 2) with neat
phosphoric acid followed by N sulfonylation provided the
desired tricyclic hexahydropyrroloindole (+)-9 (> 99% de,
> 99% ee) on multigram scale (Scheme 3).[1d,7] Hydrolysis of
the methyl ester followed by decarboxylation at the C2-
position[8] of tricycle (+)-9 afforded the desired hexahydro-
[*] Prof. Dr. M. Movassaghi, M. A. Schmidt
Department of Chemistry
Massachusetts Institute of Technology
Cambridge, MA 02139 (USA)
E-mail: movassag@mit.edu
Homepage: http://web.mit.edu/movassag/www/index.html
[**] M.M. is a Firmenich Assistant Professor of Chemistry and a
Beckman Young Investigator. We thank Dr. P. Mueller for assistance
with the X-ray analysis of (+)-1. We thank Professor M. J. Lear of the
National University of Singapore for a helpful discussion which
concerned a challenging decarboxylation step. We acknowledge
generous financial support by the Damon Runyon Cancer Research
Foundation, GlaxoSmithKline, Merck, and Boehringer Ingelheim
Pharmaceutical Inc.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2007, 46, 3725 –3728
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3725

Communications
evidenced by predominant isolation of the corresponding
tryptophan derivatives. We reasoned that photolytic cleavage
À
of the weak Mn Mn bond in dimanganese decacarbonyl
complex [Mn2(CO)10] could lead to bromide abstraction by
{Mn(CO)5} under mild reaction conditions that could afford
the desired dimer 14.[10] In the event, irradiation of a mixture
of bromide (+)-11 and [Mn2(CO)10] (0.5 equiv) by visible light
(CH2Cl2, 16 h) provided the desired dimer (+)-14[11] in 18%
yield and greater than 99% ee along with the corresponding
product 9 with reduction at the C3a-position (ca. 5%) and
several by-products. The yield of the isolated hexacycle (+)-
14 was increased to 32% by using an equimolar quantity of
[Mn2(CO)10]; however, despite examination of a wide range
of reaction conditions, none were found to be superior.
Given the expected second-order dependence of the
dimerization rate on the concentration of the activated
monomer, we reasoned that more rapid activation of the
bromide might provide a solution. Gratifyingly, the use of the
highly reactive tris(triphenylphosphine)cobalt(I) chloride
[CoCl(PPh3)3][12] as a stoichiometric reducing agent in ben-
zene was found to provide rapidly the desired dimer (+)-14 in
34% yield with a marked decrease in the number of by-
products. A dramatic solvent and concentration dependence
was also noted for this reductive dimerization with the best
solvents being acetone and 2-butanone. Treatment of the
tricyclic bromide (+)-11 with [CoCl(PPh3)3] under optimum
reaction conditions (acetone, 238C, 0.1m, 15 min) provided
the desired dimer (+)-14 in 60% yield and greater than
99% ee on a 3-g scale (Scheme 4).[13] This dimerization
process may involve a bromide abstraction from (+)-11,
thus allowing the union of two free radicals, which had
escaped from the solvent cage, to give (+)-14.[12] Alterna-
tively, the process may involve a net oxidative insertion of the
cobalt(I) reagent to afford a tertiary benzylic cobalt(III)
complex that could afford (+)-14 by homodimerization
Scheme 3. Synthesis of the tricyclic bromide 7: a) 1. aq KOH/MeOH
(1:1), 238C, 30 min; 2. oxalyl chloride, DMF, CH2Cl2, 238C, 1 h;
3. (Me3Si)3SiH, AIBN, toluene, 808C, 4 h, 84% from 9. b) 1,3-dibromo-
5,5-dimethylhydantoin, CCl4, 238C, 5 h, see text. AIBN=2,2’-azobisiso-
butyronitrile, DMF=dimethylformamide.
pyrroloindole (+)-10 in 84% yield and > 99% ee. Unfortu-
nately, the benzylic bromination of 10 was found to be
capricious, and afforded bromide 7 in variable yields and with
significant loss of optical activity (45% ee, Scheme 3). We
speculated that 10 had been converted into 7 in part via the
corresponding achiral N-benzenesulfonyl tryptamine tauto-
mer.[6] The sensitivity of 10 to trace amounts of acid was
confirmed independently. Examination of a wide range of
reaction conditions, which included bromination attempts in
the presence of scavengers of Brønsted acids, failed to
completely suppress loss in optical activity or increase the
overall efficiency of the process. Importantly, the marked
stability of the derivative 11 with the endo-methoxycarbonyl
group at C2[1d] (Scheme 4) prompted us to explore a
postdimerization decarboxylation at that position.[9]
Benzylic bromination of 9 was optimally achieved using
1,3-dibromo-5,5-dimethylhydantoin on a greater than 4-g
scale to afford the bromide (+)-11 as a single diastereomer in
77% yield and greater than 99% ee.[1d,7] We next focused our
attention on the development of the reaction conditions for
the vital homodimerization step. Photochemical or thermal
activation of hexaalkyldistannanes in the presence of tricyclic
bromide (+)-11 did not afford the desired dimeric hexacycle
14 and predominantly returned the bromide. A variety of
reducing conditions were either ineffective in providing the
desired dimeric product 14 and led to slow decomposition of
bromide 11, or promoted an anionic fragmentation as
À
followed by fragmentation of the C Co bonds and formation
[14]
À
of a C C bond within a solvent cage.
With an efficient strategy for the preparation of multi-
gram quantities of enantiomerically enriched hexacycle
Scheme 4. Concise total synthesis of (+)-1: a) 1,3-dibromo-5,5-dimethylhydantoin, AIBN, CCl4, 808C, 1 h, 77%. b) [CoCl(PPh3)3], acetone, 238C,
15 min, 60%. c) aq KOH/MeOH, 238C, 30 min, 90%. d) oxalyl chloride, DMF, CH2Cl2, 238C, 1.5 h; (Me3Si)3SiH, AIBN, toluene, 808C, 3.5 h, 64%.
e) Na(Hg), Na2HPO4, MeOH, 238C, 3.5 h, 99%. f) Sodium bis(2-methoxyethoxy)aluminum hydride, toluene, 1108C, 1.5 h, 82%.
3726
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 3725 –3728

Angewandte
Chemie
(+)-14 in four steps from the commercially available Na-
methoxycarbonyl-l-tryptophan methyl ester (8), we focused
on the completion of the synthesis of (+)-chimonanthine (1,
Scheme 4). Under optimal reaction conditions, treatment of
the diester (+)-14 with a mixture of methanol and aqueous
potassium hydroxide provided the corresponding dicarbox-
ylic acid in 90% yield. Decarboxylation at the C2- and C2’-
position was achieved by sequential conversion into the
dicarboxylic acid chloride followed by treatment with tris-
(trimethylsilyl)silane and AIBN in toluene at 808C,[8] to
afford the hexacycle (+)-15 in 64% yield. At this point
photochemical[15] conditions for the removal of the benzene-
sulfonyl groups of 15 led to decomposition whereas strong
reductive conditions resulted in cleavage of the benzenesul-
enriched hexacycle (+)-14 provided a ready access to the
optically active alkaloids 1–3. This chemistry simultaneously
secures the vicinal quaternary stereocenters directed by the
stereochemistry at the C8a-position, and offers the shortest
enantioselective synthesis of these alkaloids from commer-
cially available materials. The application of this chemistry to
more complex hexahydropyrroloindole alkaloids will be
reported in due course.
Received: February 15, 2007
Published online: April 5, 2007
Keywords: alkaloids · dimerization · enantioselectivity · indoles ·
.
total synthesis
À
fonyl groups and fragmentation of the C3 C3’ bond, as
supported by the isolation (61%) of N-methoxycarbonyl-
tryptamine. Significantly, treatment of a methanolic solution
of hexacycle (+)-15 and sodium phosphate dibasic with
freshly prepared sodium amalgam[16] afforded the desired
diamine (+)-16 in 99% yield and greater than 99% ee.
Reduction of the two N-methoxycarbonyl groups of 16 with
sodium bis(2-methoxyethoxy)aluminum hydride (Red-Al)
gave (+)-chimonanthine (1) in 82% yield. The structure of
synthetic (+)-1 was confirmed by single-crystal X-ray analysis
(Scheme 4).[7]
[1] a) “Bisindole Alkaloids”: G. A. Cordell, J. E. Saxton in The
Alkaloids: Chemistry and Physiology, Vol. 20, (Eds.: R. H. F.
Manske, R. G. A. Rodrigo), Academic Press, New York, 1981,
pp. 3 – 294; b) “Chemistry and Reactions of Cyclic Tautomers of
Tryptamines and Tryptophans”: T. Hino, M. Nakagawa in The
Alkaloids: Chemistry and Pharmacology, Vol. 34 (Ed.: A.
Brossi), Academic Press, New York, 1989, pp. 1 – 75; c) “Alka-
loids from the Medicinal Plants of New Caledonia”: T. SØvenet,
J. Pusset in The Alkaloids: Chemistry and Pharmacology, Vol. 48
(Ed.: G. A. Cordell), Academic Press, New York, 1996, pp. 58 –
59; d) D. Crich, A. Banerjee, Acc. Chem. Res., 2007, 40, 151 –
161.
[2] a) R. B. Woodward, N. C. Yang, T. J. Katz, V. M. Clark, J.
Harley-Mason, R. F. J. Ingleby, N. Sheppard, Proc. Chem. Soc.
1960, 76; b) R. Robinson, H. J. Teuber, Chem. Ind. 1954, 783.
[3] a) J. B. Hendrickson, R. Rees, R. Göschke, Proc. Chem. Soc.
1962, 383; b) T. Hino, S. Yamada, Tetrahedron Lett. 1963, 4, 1757;
c) A. I. Scott, F. McCapra, E. S. Hall, J. Am. Chem. Soc. 1964, 86,
302; d) M. Nakagawa, H. Sugumi, S. Kodato, T. Hino, Tetrahe-
dron Lett. 1981, 22, 5323; e) C.-L. Fang, S. Horne, N. Taylor, R.
Rodrigo, J. Am. Chem. Soc. 1994, 116, 9480; f) M. Somei, N.
Osikiri, M. Hasegawa, F. Yamada, Heterocycles 1999, 51, 1237;
g) H. Ishikawa, H. Takayama, N. Aimi, Tetrahedron Lett. 2002,
43, 5637; h) Y. Matsudz, M. Kitajima, H. Takayama, Hetero-
cycles 2005, 65, 1031; i) for their synthesis and resolution, see: L.
Verotta, F. Orsini, M. Sbacchi, M. A. Scheidler, T. A. Amador,
E. Elisabetsky, Bioorg. Med. Chem. 2002, 10, 2133.
Dissolution of (+)-1 in a mixture of [D4]acetic acid and
deuterium oxide followed by heating to 958C led to formation
of the isomeric calycanthine (3, Scheme 5).[17] We monitored
Scheme 5. Synthesis of (À)-calycanthine and (+)-folicanthine:
a) [D4]acetic acid, D2O, 958C, 24 h; 54% of (À)-3 and 5% of (+)-1
isolated; b) aq formaldehyde, NaBH(OAc)3, MeCN, 238C, 30 min,
100%.
[4] a) J. T. Link, L. E. Overman, J. Am. Chem. Soc. 1996, 118, 8166;
b) L. E. Overman, D. V. Paone, B. A. Stearns, J. Am. Chem. Soc.
1999, 121, 7702; c) L. E. Overman, J. F. Larrow, B. A. Stearns,
J. M. Vance, Angew. Chem. 2000, 112, 219; Angew. Chem. Int.
Ed. 2000, 39, 213; .
1
this transformation in situ by H NMR spectroscopy, which
revealed an equilibrium was established between 1 and 3
within 24 h in favor of calycanthine (ca. 85:15 3/1). Chroma-
tographic separation of the mixture afforded a 54% yield of
(À)-3 along with recovered chimonanthine (5%). We
obtained clear evidence for this equilibrium ratio when
(À)-3 was subject to identical conditions and the same ratio
was reached between 1 and 3 within 24 h. N-Methylation of
(+)-1 using formalin and sodium triacetoxyborohydride
provided the first synthetic sample of (+)-folicanthine (2) in
quantitative yield (Scheme 5).[7,18]
The concise and efficient total synthesis of (+)-chimo-
nanthine, (+)-folicanthine, and (À)-calycanthine is described.
The convergent assembly of these alkaloids used a reductive
CoI-promoted dimerization of the readily available endo bro-
mide (+)-11. The gram-scale synthesis of the enantiomerically
[5] For an important precedent on the isolation (10–15%) of a
dimeric by-product during prenylation studies at the C3a-
position of an exo-hexahydropyrroloindole-derived selenide
directed towards amauromine, see: K. M. Depew, S. P. Marsden,
D. Zatorska, A. Zatorski, W. G. Bornmann, S. J. Danishefsky, J.
Am. Chem. Soc. 1999, 121, 11953.
[6] a) M. Taniguchi, A. Gonsho, M. Nakagawa, and T. Hino, Chem.
Pharm. Bull. 1983, 31, 1856; b) G. T. Bourne, D. Crich, J. W.
Davies, D. C. Horwell, J. Chem. Soc. Perkin Trans. 1 1991, 1693;
c) M. Bruncko, D. Crich, R. Samy, Heterocycles 1993, 36, 1735;
d) M. Bruncko, D. Crich, R. Samy, J. Org. Chem. 1994, 59, 5543;
e) S. P. Marsden, K. M. Depew, S. J. Danishefsky, J. Am. Chem.
Soc. 1994, 116, 11143.
[7] For details, please see the Supporting Information.
[8] a) C. Chatgilialoglu, Acc. Chem. Res. 1992, 25, 188; b) M.
Ballestri, C. Chatgilialoglu, N. Cardi, A. Sommazzi, Tetrahedron
Lett. 1992, 33, 1787.
Angew. Chem. Int. Ed. 2007, 46, 3725 –3728
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3727

Communications
À
[9] We attribute the greater stability of (+)-9 toward acids in part to
labile C Co bond; mechanistic studies are planned to investigate
this hypothesis.
attenuation of the Lewis basicity of the carbamate at Na by the
ester at the C2-position.
[10] B. C. Gilbert, C. I. Lindsay, P. T. McGrail, A. F. Parsons, D. T. E.
Whittaker, Synth. Commun. 1999, 29, 2711.
[11] As a result of the slow conformational equilibration at ambient
temperature, NMR spectra in this report were collected at 50–
1008C.
[12] a) M. Aresta, M. Rossi, A. Sacco, Inorg. Chim. Acta 1969, 3, 227;
b) Y. Yamada, D.-i. Momose, Chem. Lett. 1981, 1277; c) S. K.
Bagal, R. M. Adlington, J. E. Baldwin, R. Marquez, J. Org.
Chem. 2004, 69, 9100.
[13] A minor competing pathway is the disproportionation of 13 as
suggested by the isolation of 9 (10%) from the dimerization
reaction.
[14] While we have no experimental evidence for this hypothesis at
this time, we expect such cobalt(III) complexes to contain a
[15] T. Hamada, A. Nishida, O. Yonemitsu, J. Am. Chem. Soc. 1986,
108, 140.
[16] B. M. Trost, H. C. Arndt, P. E. Strege, T. R. Verhoeven, Tetrahe-
dron Lett. 1976, 17, 3477.
[17] a) J. B. Hendrickson, R. Göschke, R. Rees, Tetrahedron 1964, 20,
565; b) E. S. Hall, F. McCarpa, A. I. Scott, Tetrahedron 1967, 23,
4131; c) F. GuØritte-Voegelein, T. SØvenet, J. Pusset, M. T.
Adeline, B. Gillet, J. C. Beloeil, D. Guenard, P. Potier, J. Nat.
Prod. 1992, 55, 923; d) A. D. Lebsack, J. T. Link, L. E. Overman,
B. A. Stearns, J. Am Chem. Soc. 2002, 124, 9008.
[18] The sign of optical rotation and the absolute stereochemistry for
(+)-2 conform with revision of the absolute stereochemistry
of (À)- and (+)-chimonanthine (1) by Overman et al.; see
Ref. [4b–c].
3728
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 3725 –3728