A tandem radical approach to the ABCE-rings of the Aspidosperma and Strychnos alkaloids

Article abstract of DOI:10.1039/b008465o

Reaction of indole 19 via the derived aryl radical gives the tetracycle 20 in 43% yield. This compound contains the ABCE-rings of both Aspidosperma and Strychnos alkaloids.

Full text of DOI:10.1039/b008465o

A tandem radical approach to the ABCE-rings of the Aspidosperma and
Strychnos alkaloids
Stephen T. Hilton,a Tim C. T. Ho,b Goran Pljevaljcic,b Marcus Schulteb and Keith Jones*ab
a School of Chemical and Pharmaceutical Sciences, Kingston University, Penrhyn Road, Kingston-upon-Thames,
Surrey, KT1 2EE UK
b Department of Chemistry, King’s College London, Strand, London, WC2R 2LS UK
Received (in Liverpool, UK) 16th October 2000, Accepted 18th December 2000
First published as an Advance Article on the web
Reaction of indole 19 via the derived aryl radical gives the
tetracycle 20 in 43% yield. This compound contains the
ABCE-rings of both Aspidosperma and Strychnos alka-
loids.
radical 6 to the indole 3-position to create the E-ring. The a-
amino radical generated in this step can then be captured in a
6-exo manner to create the C-ring. In order to generate radical 6,
it was decided to take advantage of the well-known ability of
aryl radicals to abstract hydrogen atoms via a 6-membered ring
transition state.7 This analysis led back to the tertiary amide 7 as
the substrate for a series of radical reaction steps under the usual
conditions of tributyltin hydride (TBTH) and a suitable
initiator.
The pentacyclic Aspidosperma alkaloids such as aspidospermi-
dine 1 and the heptacyclic Strychnos alkaloids such as
The synthesis of radical precursor 7 commenced with indole-
3-acetic acid 8 (Scheme 2). Treatment with sodium hydride and
methyl iodide gave the N-methylindole-3-acetic acid 9 in 93%
yield. Reaction of 9 with oxalyl chloride in THF at 0 °C gave the
acid chloride 10 which was reacted immediately with 2-bromo-
benzylamine hydrochloride in the presence of excess diisopro-
pylethylamine to give amide 11 in overall 86% yield. Alkylation
of 11 with 5-bromopentene using potassium hydride as base in
THF and tetrabutylammonium iodide as catalyst proceeded
strychnine 2 remain interesting targets for synthetic chemists
owing to their chemical architecture and their biological
activities.1 The tetracyclic sub-structure 3, consisting of the
ABCE-rings of both these series of natural products represents
the core of these molecules and functionalised examples of this
tetracycle have featured in the total synthesis of both Strychnos
and Aspidosperma alkaloids.2 Efficient approaches to 3 carry-
ing appropriate functionality would allow not only the synthesis
of these natural products but also related non-natural products.
Some years ago, we commenced a programme of research based
on a strategy of radical cyclisation to form oxindoles (indolin-
2-ones) followed by carbanion cyclisation to create the B- and
C-rings.3 Murphy has disclosed a tandem radical cyclisation
route to this tetracyclic structure in which the B- and E-rings are
created by an aryl radical addition to a double bond followed by
internal trapping of the resultant radical by an azide.4 This
provides an alternative to the ‘radical-polar crossover’ chem-
istry developed by the same group and utilised in a synthesis of
(±)-aspidospermidine 1.5 Parsons has also reported a con-
ceptually different approach to the Aspidosperma skeleton
using aryl radicals.6 We now wish to disclose a new route to the
ABCE–tetracycle using a novel tandem radical route to create
the C- and E-rings.
1
smoothly resulting in tertiary amide 7 in 73% yield. The H
NMR spectrum of 7 showed clearly the presence of two
rotamers caused by restricted rotation about the amide bond.
The minor rotamer possessed a singlet resonance at d 4.72
assigned to the benzyl group in 7a (Fig. 1) whilst the major
rotamer displayed a singlet at d 4.56 assigned to the benzyl
group in 7b. Integration of these signals revealed a ratio of 6+5
favouring 7b. It is well known that the timescale for amide bond
rotation to occur is considerably longer than the lifetime of
radical intermediates which means that cyclisation of the
translocated radical derived from major rotamer 7b is unlikely.8
The new disconnection sequence is shown in detail in
Scheme 1. The key features of this plan involve the addition of
Scheme 2 Reagents and conditions: i, NaH, THF, 0 °C, then Mel warm to
20 °C, 12 h, 93%; ii, oxalyl chloride, THF, 0 °C to 20 °C, 20 h; iii,
2-bromobenzylamine hydrochloride, N,N-diisopropylethylamine, dichloro-
methane, 0 °C, 2 h, 86%; iv, KH, THF, 0 °C, then 5-bromopentene,
tetrabutylammonium iodide, warm to 20 °C for 20 h, 73%, v, TBTH (or
TBTD), AIBN, xylene, reflux, 10 h.
Scheme 1
DOI: 10.1039/b008465o
Chem. Commun., 2001, 209–210
This journal is © The Royal Society of Chemistry 2001
209

View Article Online
With this problem in mind, reaction of 7 with TBTH at reflux in
xylene under dilute conditions (0.02 M) was carried out. A
complex mixture of products was obtained and to help
determine the nature of the products, the reaction was repeated
using tributyltin deuteride (TBTD). Extensive spectroscopy on
the isolated products allowed assignment of the structure 12
(both rotamers) to the major product (50%) and structure 13 to
one of the minor products (18%). Both products contained the
intact alkene unit indicating that tetracycle formation had not
occurred whilst the product 12 from reaction with TBTD
contained deuterium adjacent to the nitrogen indicating success-
ful radical translocation. Addition of the intermediate a-amido
radical to the indole C-2 followed by rearomatisation to give 13
is not unreasonable given the results of Moody9 and related
work by Bowman10 in other heteroaromatic systems. Product 12
could simply be arising from the major rotamer 7b which cannot
cyclise. However, the isolation of 13 indicates that the rather
nucleophilic a-amido radical prefers to add to the indole ring at
the C-2 position.
Following the precedent in the work of Fang,11 we decided to
introduce an electron-withdrawing group at the 2-position of
indole 7 to encourage addition of the nucleophilic a-amido
radical to C-3 of the indole. Reaction of 3-methylindole with
triphenylphosphine–thiocyanogen has been reported to give
2-cyano-3-methylindole in 85% yield.12 Treatment of 7 under
these conditions led only to starting material. A variety of other
studies to introduce a cyano-group at the 2-position were
undertaken but eventually it was decided to develop a new route
to the desired radical precursor (Scheme 2). The required
secondary amine was prepared following the chemistry of
Fukuyama and Bowman.13 Reaction of 2-bromobenzylamine
hydrochloride with sulfonyl chloride 14 gave sulfonamide 15 in
86% yield (Scheme 3). Alkylation using 5-bromopentene and
caesium carbonate as base gave 16 in 93% yield and removal of
the 2-nitrosulfonamide group was achieved in high yield using
potassium carbonate and benzenethiol yielding secondary
amine 17. Reaction of 17 with the acid chloride 18 (derived
from the acid14 by reaction with oxalyl chloride in THF at 0 °C)
gave the radical precursor 19 in 83% yield. Again the NMR
revealed the presence of rotamers in a ratio of 1.1+1. Reaction
of 19 with TBTH in refluxing 5-tert-butyl-m-xylene (ca.
200 °C) under syringe pump conditions gave some 50% of
reduced product (19 with Br = H) and 43% of 20 which had
undergone the desired translocation–cyclisation–cyclisation
sequence. Tetracycle 20 was isolated as a mixture of 4
diastereoisomers in a ratio of 8+3+2+1. The relative ster-
eochemistry of the four diastereoisomers obtained cannot be
assigned from their NMR spectra owing to the contiguous
quaternary centres at the BC- and CE-ring junctions. Calcula-
tions show the trans,trans stereochemistry at these ring
junctions to be of very high energy leaving 6 possible
diastereoisomers, four of which possess the natural cis CE-ring
junction stereochemistry and two of which possess the trans
CE-ring junction stereochemistry. Although the transition state
for the addition of the a-amido radical to the 3-position of the
indole would seem to favour creation of a cis CE-ring junction,
we cannot rule out the formation of isomers containing the
trans-CE, cis-BC stereochemistry.
Scheme 3 Reagents and conditions: i, 2-bromobenzylamine hydrochloride,
triethylamine, dichloromethane, 0 °C, 86%; ii, caesium carbonate, DMF,
80 °C, 30 min, then 5-bromopentene, 75 °C, 3 h, 93%; iii, potassium
carbonate, PhSH, MeCN, 80 °C, 86%; iv, dichloromethane, diisopropyl-
ethylamine, 0 °C, 12 h, 83%; v, TBTH, 5-tert-butyl-m-xylene, syringe
pump, reflux, 6 h, 43%.
products and related compounds and is under active in-
vestigation.
We should like to acknowledge financial support from
Kingston University (S. T. H.), King’s College London
(T. C. T. H.) and the EU (G. P. and M. S.).
Notes and references
1 J. E. Saxton, Nat. Prod. Rep., 1996, 14, 559.
2 J. Bonjoch and D. Sole, Chem. Rev., 2000, 100, 3455; V. H. Rawal and
S. Iwasa, J. Org. Chem., 1994, 59, 2685; M. E. Kuehne and F. Xu,
J. Org. Chem., 1993, 58, 7490; Y. Ban, T. Ohnuma, M. Nagia, Y. Sendo
and T. Oishi, Tetrahedron Lett., 1972, 13, 5023; P. J. Laronze, J.
Laronze-Fontaine, J. Lévy and J. LeMen, Tetrahedron Lett., 1974, 15,
491; S. J. Veenstra and W. N. Speckamp, J. Am. Chem. Soc., 1981, 103,
4645.
3 C. McCarthy and K. Jones, J. Chem. Soc., Chem. Commun., 1989,
1717.
4 M. Kizil, B. Patro, O. Callaghan, J. A. Murphy, M. B. Hursthouse and
D. Hibbs, J. Org. Chem., 1999, 64, 7856.
5 O. Callaghan, C. Lampard, A. R. Kennedy and J. A. Murphy, J. Chem.
Soc., Perkin Trans. 1, 1999, 995.
6 P. J. Parsons, C. S. Penkett, M. C. Cramp, R. I. West, J. Warrington and
M. C. Saraiva, Synlett, 1995, 507.
7 A. Fiumana and K. Jones, Tetrahedron Lett., 2000, 41, 4209; D. P.
Curran and J. Y. Xu, J. Am. Chem. Soc., 1996, 118, 3142; A. L. J.
Beckwith and J. M. D. Storey, J. Chem. Soc., Chem. Commun., 1995,
977.
8 O. M. Musa, J. H. Horner and M. Newcomb, J. Org. Chem., 1999, 64,
1022; K. Jones and C. McCarthy, Tetrahedron Lett., 1989, 30, 2657.
9 C. J. Moody and C. L. Norton, J. Chem. Soc., Perkin Trans. 1, 1997,
2639.
In summary, we have shown that the sequence of radical
reaction steps shown in Scheme 1 is successful if a cyano group
is present at the indole C-2. This short sequence to tetracycles
such as 20 may provide a rapid entry into indole alkaloid natural
10 F. Aldabbagh, W. R. Bowman, E. Mann and A. M. Z. Slawin,
Tetrahedron, 1999, 55, 8111.
11 C. C. Yang, H. T. Chang and J. Fang, J. Org. Chem., 1993, 58, 3100.
12 Y. Tamura, M. Adachi, T. Kawasaki, H. Yasuda and Y. Kita, J. Chem.
Soc., Perkin Trans. 1, 1980, 1132.
13 T. Fukuyama, C.-K. Jow and M. Cheung, Tetrahedron Lett., 1995, 36,
6373; W. R. Bowman and D. R. Coghlan, Tetrahedron, 1997, 53,
15 787.
14 S. T. Hilton, T. C. T. Ho, G. Pljevaljcic and K. Jones, Org. Lett., 2000,
2, 2639.
Fig. 1
210
Chem. Commun., 2001, 209–210