Pictet-Spengler reaction: Is carbonyl the best choice? A highly diastereoselective alternative approach to trans-1,3-disubstituted tetrahydro-β-carbolines

Article abstract of DOI:10.1016/S0040-4039(00)00751-6

Unprecedented 1,2,3-trisubstituted tetrahydro-β-carbolines (THBCs) have been synthesized via a short and highly diastereoselective synthetic route. The key step of the sequence is a flexible variant of the Pictet-Spengler reaction employing synthetic equivalents of several non-available carbonyl compounds. Using one such synthon, the resultant THBC could be further ring- closed to a tetracyclic indole alkaloidal skeleton. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00751-6

Tetrahedron Letters 41 (2000) 4977±4980
Pictet±Spengler reaction: is carbonyl the best choice?
A highly diastereoselective alternative approach to
trans-1,3-disubstituted tetrahydro-b-carbolines
Kamaljit Singh* and Prasant K. Deb
Department of Applied Chemical Sciences & Technology, Guru Nanak Dev University, Amritsar 143 005, India
Received 7 March 2000; accepted 8 May 2000
Abstract
Unprecedented 1,2,3-trisubstituted tetrahydro-b-carbolines (THBCs) have been synthesized via a short
and highly diastereoselective synthetic route. The key step of the sequence is a ¯exible variant of the Pictet±
Spengler reaction employing synthetic equivalents of several non-available carbonyl compounds. Using
one such synthon, the resultant THBC could be further ring-closed to a tetracyclic indole alkaloidal
skeleton. # 2000 Elsevier Science Ltd. All rights reserved.
The Pictet±Spengler reaction1 has proved to be a tool of importance in the synthesis of tetra-
hydro-b-carbolines (THBCs) and related heterocyclic systems. With the advent of asymmetric
stereocontrol,2 the importance of this method for the stereospeci®c synthesis of the THBC core
has rapidly increased culminating in enantioselective and diastereoselective preparations of this type
of compound using carbohydrates,3 (R)- and (S)-glyceraldehyde,4 (^)-8-phenylmethylcarbamates,5
chiral acetylenic sulphoxides,6 chemoselective and asymmetric reductions,7 etc. A salient feature
of these reactions has been the diversi®cation at C-1, an attribute of potential utility in the total
syntheses of indole alkaloids and lead compounds showing a variety of pharmacologically
interesting properties. With the major emphasis laid on the stereocontrol of the reaction using
relatively stable aromatic aldehydes, the prime limitations of the conventional Pictet±Spengler
condensation include the use of rather inaccessible functionalized aliphatic aldehydes and,
although widely realized,8 have not been adequately addressed. Even under non-acidic aprotic
conditions, with an excess of aldehyde (10±15 equivalents) to compensate for competitive side
reactions,8b (including aldol condensation of the aldehydes to give a,b-unsaturated carbonyl
compounds and their instability), lower yields and the presence of unreacted starting materials
plague the reaction. In keeping with our interest in a more general route to the C-1 diversi®ed title
* Corresponding author.
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00751-6

4978
compounds and in the light of the above, we wished to add a new dimension to their thermo-
dynamically controlled, high yielding and diastereoselective synthesis by employing readily
available perhydro-1,3-heterocycles: oxazines 3 and oxazolidines 4 (carbonyl equivalents),9
capable of transferring multi-functionalized carbon (C-2) fragments to biogenic amines, resulting
in variously C-1-substituted THBCs.
The oxazines 3 (R2=CH3, C2H5, CH2COOEt, CH2CH2SEt/Ph) used as reagents are readily
available from a variety of cheaply available starting materials employing literature10a or
modi®ed10b methods. Another approach that has been developed10c involves reacting 2-chloro-
methyl dihydro-oxazine (obtained from chloroacetonitrile) with the carbanion of acetonitrile
generated at low temperature (LDA, THF, ^78ꢀC) followed by reduction (NaBH4, THF:EtOH (1:1),
^40ꢀC) to give 3 (R2=CH2CH2CN). Likewise, oxazolidines 4 (R2=CH2CN, CH2COCH2COOEt,
CH2COC6H5) have also been synthesized by reacting carbanions of acetonitrile, ethyl aceto-
acetate and acetophenone (LDA/THF/^78ꢀC) with 2,4,4-trimethyl-Á2-oxazolinium iodide11
(Scheme 1).
Scheme 1. (a) Ritter synthesis of dihydro-oxazine: 2-methyl-2,4-pentanediol, 0ꢀC, quantitative; (b) reduction: NaBH4,
THF:EtOH (1:1), ^40ꢀC, quantitative; (c) oxazoline synthesis: re¯ux, 80%; (d) N-quaternisation: MeI, quantitative; (e)
carbanion addition, THF, ±78ꢀC; (f) MeCN: acid 10:1, 80ꢀC; (g) EtOH/HCl, quantitative; (h) aq. NH3, 80%
If a stoichiometric mixture of Nb-benzyl (l)-tryptophan methyl ester (TrypOMe) 1 in anhydrous
acetonitrile:acid (10:1) solution (Table 1) is reacted under re¯ux with oxazines 3 (existing mainly
in their enamine-chain tautomeric form)11 or oxazolidines 4 (existing as ring-chain tautomers)12
the iminium intermediate 5 (Scheme 1) formed in situ cyclizes spontaneously by intramolecular
electrophilic attack of the iminium carbon at C-2 or at C-3 of the indole moiety to form a spiro-
indolenine1,2b 6 which rearranges and deprotonates to yield THBCs 7 and 8 in high yield and very
high diastereomeric ratios (Table 1). The trans- Nb-substituted diastereomers are thermo-
dynamically more stable than their cis-congeners. Evidence for this observation has already been
furnished and proved by equilibration in acidic media.1

4979
Table 1
Diastereoselectivity in the Pictet±Spengler reaction between oxazines 3 or oxazolidines 4 and Nb benzyl (l)-TrypOMe 1
In all the examples recorded herein, the use of sensitive aldehydes8 is avoided. This synthesis is
particularly attractive in light of the large number of commercially available precursors of oxazines
3 and oxazolidines 4 and can be used to introduce much needed diversity at the C-1 position of
the THBCs.
The structures of all compounds were supported by 300 MHz 1H and 50.3 MHz 13C NMR and
mass spectra. All compounds give a single spot on TLC analysis and correct microanalytical data.
1
Spin decouplings (entry 9) and 2D NMR spectra (entry 8) are used to assign signals in the H
NMR spectra. The diastereomers are identi®ed by 1H and 13C NMR and the stereochemistry was
unambiguously assigned by comparison with literature data.14 Due to space restrictions only
salient chemical shifts in the NMR are presented.15
2,4-Dioxo-1,2,3,4,6,7,12,12b-octahydroindolo [2,3-a] quinolizine 11 (a ketolactam) is a
compound of interest in connection with the total synthesis of indole alkaloids and their analogs.
The known synthesis16 of 11 proceeds in low yield and requires the use of sensitive 4-ethoxy-
carbonyl±3,3-ethylenedioxybutanal. The reaction of tryptamine 2 with 4 (R2=CH2COCH2-
COOEt) (anhydrous CH3CN:CH3COOH, 10:1, re¯ux) furnished enaminoketone ester 917 and
this was converted into 11 (Scheme 1) in 80% yield. Likewise, the dihydroindole 10 (a potential
vindoline)13 intermediate, has been obtained by acetylation of the amine 9 and treatment of the
resultant acetyl derivative with BF3/Et2O to obtain 10.13
Entry 3 constitutes straightforward incorporation of the ±CH2CN functionality at C-1, to furnish
an analog of a pivotal intermediate used in the synthesis of corynantheidol and dihydrocorynantheol
and related seco-alkaloids, otherwise prepared through a cumbersome, multistep approach.18
In summary, several oxazines 3 and oxazolidines 4, synthetic equivalents of aldehydes, have
been synthesized and used in the Pictet±Spengler reaction with TrypOMe and tryptamine to give
C-1-functionalized THBCs in excellent yields and very high diastereoselectivity. This protocol is
¯exible for many variations at C-1. The application of the methodology in the synthesis of some
isoquinoline alkaloidal skeletons is in progress.
Acknowledgements
We wish to thank CSIR, New Delhi for the grant [01 (1441)/97-EMR-II], senior research
fellowship to P.K.D. and Professor Harjit Singh, Department of Chemistry for helpful suggestions.

4980
References
1. Cox, E. D.; Cook, J. M. Chem. Rev. 1995, 95, 1797, and references cited therein.
2. (a) Czerwinski, K. M.; Cook, J. M. Stereochemical Control of the Pictet±Spengler Reaction in the Synthesis of
Natural Product Synthesis; Pearson, W., Ed.; JAI Press: Greenwich, CT, 1996; Vol 3, p. 217. (b) Bailey, P. D.;
Hollinshead, S. P.; McLay, N. R.; Morgan, K.; Palmer, S. J.; Prince, S. N.; Reynolds, C. D.; Wood, S. D. J. Chem.
Soc., Perkin Trans. 1 1993, 431.
3. Piper, I. M.; MacLean, D. B.; Kvarnstrom, I.; Szarek, W. A. Can. J. Chem. 1983, 61, 2721.
4. Czarnocki, Z.; MacLean, D. B.; Szarek, W. A. Can. J. Chem. 1986, 64, 2205.
5. Bauer, T.; Chapuis, C.; Kozak, J.; Jurezak, J. Helv. Chim. Acta. 1989, 72, 482.
6. Lee, A. W. M.; Chan, W. H. Chiral Acetylenic Sulfoxides and Related Compounds in Organic Synthesis in Topics in
Current Chemistry; Springer-Verlag, 1997; Vol. 190, p. 103.
7. McNulty, J.; Still, I. W. J. Tetrahedron Lett. 1995, 36, 7965.
8. (a) Gremmen, C.; Burm, B. E. A.; Wenner, M. J.; Koomen, G. J. Tetrahedron Lett. 1998, 39, 1441. (b) Ungemach,
F.; DiPiero, M.; Weber, R.; Cook, J. M. J. Org. Chem. 1981, 46, 164. (c) Bailey, P. D.; Hollinshead, S. P.; McLay,
N. R.; Everett, J. H.; Reynolds, C. D.; Stephen, D. W.; Giordano, F. J. Chem. Soc., Perkin Trans. 1 1993, 451. (d)
Cox, E. D.; Hamaker, L. K.; Li, J.; Yu, P.; Czerwinski, K. M.; Deng, L.; Bennett, D. W.; Cook, J. M.; Watson,
W. H.; Krawiec, M. J. Org. Chem. 1997, 62, 44. [Entry 1: obtained as mixture (trans:cis, 77:23) after a 72 h
reaction in a sealed tube].
9. Singh, K.; Deb, P. K. Heterocycles 1999, 51, 1509.
10. (a) Meyers, A. I.; Nabeya, A.; Adickes, H. W.; Politzer, I. R.; Malone, G. R.; Kovelesky, A. C.; Nolen, R. L.;
Portnoy, R. C. J. Org. Chem. 1973, 38, 36. (b) Singh, K.; Singh, J.; Deb, P. K.; Singh, H. Tetrahedron 1999, 55,
12873. (c) Singh, K.; Deb, P. K. Unpublished.
11. Singh, K.; Singh, J.; Singh, H. Tetrahedron 1998, 54, 3567.
12. Lazar, L.; Lakatos, A. G.; Fulop, F.; Bernath, G.; Riddell, F. G. Tetrahedron 1997, 53, 1081.
13. Hiemstra, H. C.; Bieraugel, H.; Pandit, U. K. Tetrahedron Lett. 1982, 23, 3301.
14. (a) Ungemach, F.; Soerens, D.; Weber, R.; DiPierro, M.; Campos, O.; Mokry, P.; Cook, J. M.; Silverton, J. V.
J. Am. Chem. Soc. 1980, 102, 6976. (b) Bailey, P. D.; Hollinshead, S. P. Heterocycles 1987, 26, 389.
1
15. Entry 3 (cis-isomer): H NMR (CDCl3): ꢀ 2.82±3.50 (m, 4H, 2ÂCH2), 3.65 (s, 3H, OCH3), 3.95 (dd, J1=6.87 Hz
and J2=2.07 Hz, 1H, CH), 4.17 (s, 2H, CH2), 4.50 (dd, J1=8.50 Hz and J2=2.92 Hz, 1H, CH), 7.11±7.23 Hz (m,
2H, arom H), 7.31±7.43 (m, 6H, arom H), 7.54 (d, J=7.31 Hz, 1H arom H), 8.28 (br, 1H, NH). Entry 3 (trans-
1
isomer): H NMR (CDCl3): ꢀ 2.67±2.91 (ABX system, JAB=16.68 Hz, JAX=4.92 Hz, JBX=7.89 Hz, 2H, CH2),
3.04±3.25 (m, 2H, CH2), 3.69 (s, 3H, OCH3), 3.85 (AB quartet, J=14.42 Hz, 2H, CH2), 3.91±3.97 (m, 1H, CH),
4.44 (dd, J1=7.59 Hz and J2=5.12 Hz, 1H, CH), 7.09±7.31 (m, 2H, arom H), 7.33±7.42 (m, 6H, arom H), 7.53 (d,
J=7.72 Hz, 1H, arom H), 8.05 (br, 1H, NH).
16. Brutcher Jr., F. V.; Vanderwer€, W. D. J. Org. Chem. 1972, 37, 297.
17. The Z-con®guration of 9 was attested by the coupling constants of the vinylic protons (ꢀ 4.93, J=7.2 Hz).
18. Beard, R. L.; Meyers, A. I. J. Org. Chem. 1991, 56, 2091.