Indole alkaloids by a chemoenzymatic approach: Two convergent routes for the first enantioselective synthesis of (+)-20R-15,20-dihydrocleavamine

Article abstract of DOI:10.1016/S0040-4039(00)00404-4

A stereocontrolled total synthesis of the title compound is described, starting from enantiopure intermediates. Two alternative strategies have been developed to ensure the critical formation of the nine-membered ring of 3. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00404-4

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 3489–3492
Indole alkaloids by a chemoenzymatic approach: two convergent
routes for the first enantioselective synthesis of (+)-20R-15,20-
dihydrocleavamine
Bruno Danieli, Giordano Lesma,∗ Daniele Passarella and Alessandra Silvani
Dipartimento di Chimica Organica e Industriale, Università degli Studi di Milano, Centro CNR di Studio per le Sostanze
Organiche Naturali, via Venezian 21-20133 Milano, Italy
Received 1 February 2000; accepted 7 March 2000
Abstract
A stereocontrolled total synthesis of the title compound is described, starting from enantiopure intermediates.
Two alternative strategies have been developed to ensure the critical formation of the nine-membered ring of 3.
© 2000 Elsevier Science Ltd. All rights reserved.
Our continuing interest in the enantioselective synthesis of indole alkaloids by a chemoenzymatic
approach1 led us to explore the enzymatic desymmetrization of some meso-piperidine-3,5-dimethanols
and their diacetates. By this method, the enantiopure, advanced derivatives 1 and 2 (Fig. 1) were
provided in both enantiomeric forms and they are presently under investigation as potentially useful
intermediates for the asymmetric synthesis of Iboga alkaloids possessing the tacaman or the pseudo-
aspidosperma skeleton. In this communication we report the preliminary results of our study on the
first enantioselective synthesis of (+)-20R-15,20-dihydrocleavamine 3,2 a tetracyclic alkaloid structurally
related to 16-β-carbomethoxyvelbanamine, the indole ‘upper half’ of the antitumoral bisindole alkaloids
occurring in Catharantus roseus, such as vinblastine and vincristine. Our plan for the construction of
the dihydrocleavamine framework starting from 1 and 2, was based upon two convergent strategies, as
outlined in Fig. 1, wherein the construction of the central nine-membered C-ring plays a key role.
In the approach a (Fig. 1), we first planned the formation of the C(2)–C(16) bond of 3 (biogenetic
numbering), by coupling tryptophol, in a Pictet–Spengler-like reaction, with a piperidine A, derived from
1, bearing a masked aldehyde function. Then a N(4)→C(5) intramolecular alkylation would complete the
construction of the C-ring. In this strategy, we decided to postpone the elaboration of the C(20) side
chain of 3 until the later stages of the synthesis. In this way, the required chirality at C(20) and C(14) of
3 depended upon the stereogenic centres at C(3)-R and C(5)-R (piperidine numbering) of 1, respectively.
∗
Corresponding author.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00404-4
tetl 16684

3490
Fig. 1.
In a different approach (b, Fig. 1), the first stage of the synthesis would be the formation of the
C(5)–N(4) bond of 3, by coupling indoleacetic acid with a suitable piperidine B, derived from 2.
After elongation of the C(3) side-chain to a carboxymethylene group, an intramolecular Friedel–Crafts
acylation with the formation of the C(2)–C(16) bond of 3 would afford the nine-membered ring. As a
consequence of this sequence, the chirality at C(14) and C(20) of 3 results from that at positions C(3)-S
and C(5)-S of 2, respectively.
It should be noted that the overall strategy would allow the conversion of both the intermediates 1 and
2, of opposite configuration at the stereogenic centres, into the same chiral product, by complementary
modifications of the two different functional groups in the C(3) and C(5) side chains.
In the first synthetic pathway (Scheme 1) the fully protected piperidine 4 was initially prepared
smoothly by benzoylation of 1. Building on the precedent reported by Wilson et al.,3 we then developed
an efficient procedure for the installation of the indole unit which took advantage of the enol methyl ether
function present in 4. In the event, treatment of 4 with tryptophol in the presence of a catalytic amount of
TFA and of 4 Å molecular sieves, cleanly generated the tricyclic indole derivative 5 in 74% yield, which
was isolated as an inseparable 1:1 mixture of diastereoisomers.
Scheme 1. Reagents and conditions: (a) BzCN, DIPEA, CH2Cl2, rt, 98%; (b) tryptophol, TFA (0.13 mol equiv.), 4 Å mol.
sieves, CH2Cl2, rt, 45 h, 74%; (c) Et3SiH, CF3SO3H, CH2Cl2, −50°C, 6 h, 44%; (d) MsCl, DIPEA, CH2Cl2, rt, 3 h, 98%; (e)
H2, 5% Pd(C), EtOAc, rt, 17 h, then chlorobenzene, DIPEA, reflux, 4 h, 28%; (f) NaOH, MeOH/H2O, rt; (g) TsCl, DIPEA,
CH2Cl2, rt, 6 h; (h) CuI, MeLi, Et2O, −40°C, 62% from 8

3491
The ring opening to give the C(2)-substituted tryptophol derivative 6, proved challenging. Most of
the available methods for this reduction, based on the use of boron hydrides in acidic conditions [e.g.
6
NaBH4/TFA;4 NaBH3CN/ZnI2;5 NaBH4/PdCl2 ], were also known to reduce indole systems to the
corresponding 2,3-dihydroderivatives, and indeed when we examined a few of these procedures, they did
not provide regioselective control in the reduction of 5. Turning our attention to other reducing systems,
the required 6 was finally accessed by the action of Et3SiH/CF3SO3H, following the protocol described
by Olah et al.7 The reaction required 6 h at −50°C and provided 6 in 44% yield.
The final cyclization step required the conversion of the hydroxy group of 6 into a leaving group,
followed by N(4)-deprotection and subsequent intramolecular nucleophilic displacement. Treatment of 6
with mesyl chloride gave the sensitive mesylate 7 in almost quantitative yield, which upon hydrogenation
over 5% Pd(C) followed by heating in refluxing chlorobenzene, afforded the tetracyclic derivative 8,
albeit in only 28% yield. Finally, 8 was converted into (+)-20R-15,20-dihydrocleavamine 38 as a single
enantiomer ([α]D=130, c 0.1, CHCl3; lit.2 [α]D=133, c 0.078, CHCl3), in 62% overall yield by a three-
step sequence involving hydrolysis of benzoate, tosylation and displacement of the tosylate with lithium
dimethylcuprate.
Evidently, the most notable difficulty of achieving our target resided in the formation of the nine-
membered ring. Therefore, in our second synthetic approach (b, Fig. 1), we looked to an intramolecular
acylation strategy, which was well precedented in the Aspidosperma alkaloid synthetic area,9 although
variable yields have been reported. It must be noted9a (Scheme 2) that the presence of the amidic
piperidine nitrogen should play a key role for the success of this transformation. The tight coordination
of N(4) with the transient acylium ion at C(16) would provide the ‘activated’ intermediate 12A, and the
consequent cyclization of a six-membered ring, so as to reduce the entropy of the system at the transition
state.
Scheme 2. Reagents and conditions: (a) CBr4, TPP, CH2Cl2, rt, 3 h, 88%; (b) Bu4NCN, CH2Cl2, rt, 3 h, 85%; (c) H2, 5% Pd(C),
EtOH, 24 h, 80%; (d) indoleacetic acid, TEA, DPPA, acetonitrile, rt, 6 h, 81%; (e) HCl(g), MeOH, rt, 4 h, then H2O, 93%; (f)
NaOH, MeOH/H2O, 50°C, 3 h, 81%; (g) PPE, CHCl3, 60°C, 4 h, 32%; (h) LAH, dioxane reflux, 8 h, 42%
The cyanopiperidine 9 was readily prepared from the (3S,5S)-benzyl-3-(hydroxymethyl)-5-(ethenyl)-
1-piperidinecarboxylate 2, via the bromo derivative, according to the procedure described by Starks,10
followed by hydrogenation. Upon treatment with indoleacetic acid in the presence of diphenylphosphoryl
azide and Et3N, 9 underwent smooth amidation to give the indolepiperidine 10. In order to effect
the subsequent cyclization, the nitrile 10 was first converted into the amido ester 11, and then, upon
hydrolysis with NaOH, into the amido acid 12, in almost quantitative yield. In spite of the large variety
of methods (e.g., PPA;9a,b Ph3PO-Tf2O;11 PPSE;12 SnCl4/CH2Cl2) attempted to induce the crucial
intramolecular Friedel–Craft acylation, only the protocol reported by Hanaoka et al.9c (PPE, CHCl3,
60°C) succeeded in giving 13 in acceptable yields (32%). When treated with LAH9 in refluxing dioxane,

3492
the ketolactam 13 underwent clean reduction of both amide and carbonyl functions, furnishing the
optically active (+)-20R-15,20-dihydrocleavamine 3 ([α]D=128, c 0.1, CHCl3),8 in 14% overall yield
from 12, identical in all respects to the natural compound.
The convergent strategies described herein provide possible solutions to the enantioselective construc-
tion of the unusual dihydrocleavamine tetracyclic framework, which is possible to achieve in both of the
enantiomeric forms. Further efforts, aimed at improving the yields in the formation of the nine-membered
ring, are currently underway.
Acknowledgements
This work was supported by Ministero dell’Università e della Ricerca Scientifica e Tecnologica
(MURST).
References
1. (a) Danieli, B.; Lesma, G.; Passarella, D.; Silvani, A. J. Org. Chem. 1998, 63, 3492–3496; (b) Danieli, B.; Lesma, G.;
Passarella, D.; Silvani, A. Current Organic Chemistry 2000, 4, 111–141, and references cited therein.
2. Van Beek, T. A.; Verpoorte, R.; Baerheim Svendsen, A. Tetrahedron 1984, 40, 737–748.
3. Wilson, R. M.; Farr, R. A.; Burlett, D. J. J. Org. Chem. 1981, 46, 3293–3302.
4. Ketcha, D. M.; Lieurance, B. A.; Homan, D. F. J. Org. Chem. 1989, 54, 4350–4356.
5. Lau, C. K.; Dufresne, C.; Langer, P. C.; Pietre, S.; Scheigetz, J. J. Org. Chem. 1986, 51, 3038–3043.
6. Satoh, T.; Mitsuo, N.; Nishiki, M.; Nanba, K.; Suzuki, S. Chemistry Lett. 1981, 1029–1030.
7. Olah, G. A.; Arvanaghi, M.; Ohannessian, L. Synthesis 1986, 770–772.
8. The spectroscopic data of synthetic (+)-20R-15,20-dihydrocleavamine were in complete agreement with those reported in
the literature.2
9. (a) Ziegler, F. E.; Kloek, J. A.; Zoretic, P. A. J. Am. Chem. Soc. 1969, 91, 2342–2346; (b) Ziegler, F. E.; Bennett, G. B.
J. Am. Chem. Soc. 1973, 95, 7458–7464; (c) Hanaoka, M.; Imanishi, T.; Nakai, A.; Yagi, N. Chem. Pharm. Bull. 1981,
901–903.
10. Starks, C. M. J. Am. Chem. Soc. 1971, 93, 195–199.
11. Hendrickson, J. B.; Hussoin, Md. S. J. Org. Chem. 1989, 54, 1144–1149.
12. Angelastro, R.; Marquat, A. L.; Podlogar, B. L.; Huber, E. W.; Demeter, D. A.; Poet, N. P.; Weintraub, H. J. R. J. Org.
Chem. 1994, 59, 2092–2100.