A novel approach to conformationally restricted analogues of nicotine and anabasine by an intramolecular Hamaguchi-Ibata reaction

Article abstract of DOI:10.1016/S0040-4039(99)02148-6

A short synthesis of methyl (SR)-8,10-dichloro- 1,2,3,4,5,6-hexahydro-6- hydroxy-1-methyl-1,9-phenanthroline-6-carboxylate (14) has been achieved en route to conformationally restricted analogues of nicotine and anabasine. The key feature of this process involves an efficient intramolecular Hamaguchi- Ibata reaction of 11. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(99)02148-6

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 759–762
A novel approach to conformationally restricted analogues of
nicotine and anabasine by an intramolecular
Hamaguchi–Ibata reaction
Tarun K. Sarkar,∗ Sankar Basak and Sunil K. Ghosh
Department of Chemistry, Indian Institute of Technology, Kharagpur 721302, India
Received 8 September 1999; revised 12 November 1999; accepted 17 November 1999
Abstract
A short synthesis of methyl (SR)-8,10-dichloro-1,2,3,4,5,6-hexahydro-6-hydroxy-1-methyl-1,9-phenanthroline-
6-carboxylate (14) has been achieved en route to conformationally restricted analogues of nicotine and anabasine.
The key feature of this process involves an efficient intramolecular Hamaguchi–Ibata reaction of 11. © 2000
Elsevier Science Ltd. All rights reserved.
Keywords: azaisobenzofuran; intramolecular cycloaddition; domino reaction; 1,9-phenanthroline.
The examination of the role of particular conformations on the pharmacological properties of important
biologically active molecules has remained an active area of investigation. One approach to such a
study is to modify the parent molecule in such a fashion that its original conformational mobility is
severely limited to one particular conformation.1 In light of this, recently a series of conformationally
restricted analogues, e.g. 1 and 2 of nicotine (3) and anabasine (4) have been synthesized and evaluated
as agonists of neuronal acetylcholine receptors (nAChRs).2 In particular, compound 1 which selectively
activates human recombinant α2β4 and α4β4 nAChRs has been shown to be active in animal models of
Parkinson’s disease and pain. The long and circuitous synthetic route2 described for the ring systems 1
and 2 does not appear to be viable for the synthesis of a large variety of bridged nicotines and anabasines
required for biological testing. Hence, the development of an alternative and more flexible route which
overcomes these limitations was felt desirable.
∗
Corresponding author.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02148-6
tetl 16076

760
We have recently reported on a Hamaguchi–Ibata reaction involving a stable azaisobenzofuran to give
functionalized isoquinolines.3 In this letter, we describe for the first time an intramolecular4 version 6→5
of this methodology which has allowed synthesis of 14, a molecule which incorporates the basic ring
system present in bridged anabasines.
The substituted diazoacetic ester 11 (IR: 2110,1715,1637 cm−1), the substrate for the intramolecular
Hamaguchi–Ibata reaction, was made from the readily available carboxylic acid 73 by a standard
synthetic protocol as shown in Scheme 1.5 When 11 was exposed to 1 mol% Rh2(OAc)4 in refluxing
benzene for 1 h, the bridged anabasine 14 (mp 165–167°C) was obtained in 52% yield as a yellow
crystalline solid via intramolecular cycloaddition 12→13 followed by ring opening of 13 and subsequent
1
proton transfer. The structure of 14 is supported by its elemental analysis, H and 13C NMR spectra6
(Scheme 1).
Scheme 1. (a) (COCl)2, PhH, reflux and then N-methylpent-4-enylamine, py, rt, 72%; (b) LDA, THF, −78°C and then CO2,
83%; (c) CH2N2, 94%; (d) 4-acetamidobenzenesulfonyl azide, Et3N, 0°C to rt, 12 h (Ref. 7), 93%; (e) Rh2(OAc)4, PhH, reflux,
1 h, 52%
Unfortunately, when the dienophile tether was shortened by one methylene unit, no intramolecular
cycloaddition leading to a bridged nicotine analogue 19 was obtained (Scheme 2).5 This difficulty may
be overcome as recently shown by Padwa et al8 by replacing the amine tether (cf. 6) by an amide tether.9
A noteworthy feature of our approach is that one of the chloro substituents flanking the nitrogen atom
in 10 can be selectively replaced by a nucleophile, if desired. It is evident that there is considerable steric
hindrance to nucleophilic attack at C-2 in 10. Indeed, exposure of 105,10 to aqueous alkali in methanol
and a follow-up treatment with diazomethane gave practically a single product to which structure 205,11
is assigned based on 1H NMR analysis12 (Scheme 3).

761
Scheme 2. (a) (COCl)2, PhH, reflux and then N-methylbut-3-enylamine, py, rt, 66%; (b) LDA, THF, −78°C and then CO2, 80%;
(c) CH2N2, 92%; (d) 4-acetamidobenzenesulfonyl azide, Et3N, 0°C to rt, 12 h (Ref. 7), 89%; (e) Rh2(OAc)4, PhH, reflux, 1 h
Scheme 3. (a) 10% aq. KOH, MeOH, 12 h, rt; H3O+; (b) CH2N2 (73% from 10)
In conclusion, the domino cascade sequence13 presented herein has allowed ready access to a bridged
anabasine analogue. Further work on bridged nicotines and their biological activities are in progress and
will be reported in due course.
Acknowledgements
Financial support from CSIR & DST, Government of India is gratefully acknowledged. We are also
grateful to Professor T. Gallagher (Bristol, UK), Dr. S. Djuric (Abbott, USA), Dr. M. K. Ghosh (Labline,
Germany) and Dr. A. Sarkar (NCL, Pune) for help.
References
1. Chavdarian, C. G.; Seeman, J. I.; Wooten, J. B. J. Org. Chem. 1983, 48, 492 and references cited therein.
2. Vernier, J.-M.; Holsenback, H.; Cosford, N. D. P.; Whitten, J. P.; Menzaghi, F.; Reid, R.; Rao, T. S.; Sacaan, A. I.; Lloyd, G.
K.; Suto, C. M.; Chavez-Noriega, L. E.; Washburn, M. S.; Urrutia, A.; McDonald, I. A. Bioorg. Med. Chem. Lett. 1998, 8,
2173.
3. Sarkar, T. K.; Ghosh, S. K.; Nandy, S. K.; Chow, T. J. Tetrahedron Lett. 1999, 40, 397.
4. Peters, O.; Friedrichsen, W. Tetrahedron Lett. 1995, 36, 8581.
5. Compounds 8–11, 15–18 and 20 exist as mixtures of rotamers in solution.
6. IR (KBr) νmax: 3378, 1745, 1616 cm−1; 1H NMR (200 MHz, CDCl3): δ 7.41 (s, 1H), 3.69 (s, 3H), 3.20 (bs, 1H), 3.19–3.09
(m, 1H), 2.90–2.75 (m, 1H), 2.65 (1H, dt, J=15 and 1.8 Hz), 2.44 (1H, d, J=15.1 Hz), 2.40 (s, 3H), 2.20–2.10 (m, 2H),
1.80–1.65 (m, 2H); 13C NMR (50 MHz, CDCl3); δ 173.4, 152.6, 147.7, 145.1, 138.2, 125.9, 118.5, 113.7, 74.9, 52.9, 51.3,
41.2, 40.8, 28.8, 18.5. Anal. calcd for C15H16N2O3Cl2: C, 52.49; H, 4.69; N, 8.16; found: C, 52.73; H, 4.62; N, 8.10.
7. Baum, J. S.; Shook, D. A.; Davies, H. M. L.; Smith, H. D. Synth. Commun. 1987, 17, 1709.
8. Padwa, A.; Kappe, C. O.; Cochran, J. E.; Snyder, J. P. J. Org. Chem. 1997, 62, 2786.
9. This possibility is under investigation.
1
10. H NMR (300 MHz, CDCl3): δ 7.34 and 7.28 (s, 1H), 5.91–5.78 and 5.71–5.60 (m, 1H), 5.12–4.90 (m, 2H), 3.80 (d,
1H, J=16.2 Hz), 3.72 and 3.71 (s, 3H), 3.56 (d, 1H, J=16.2 Hz), 3.70–3.43 and 3.17–2.96 (m, 2H), 3.11 and 2.87 (s, 3H),
2.20–1.66 (m, 4H).

762
1
11. H NMR (300 MHz, CDCl3): δ 6.69 and 6.64 (s, 1H), 5.93–5.77 and 5.76–5.60 (m, 1H), 5.11–4.90 (m, 2H), 3.95 and 3.94
(s, 3H), 3.77 (d, 1H, J=16.2 Hz), 3.70 and 3.68 (s, 3H), 3.51 (d, 1H, J=16.2 Hz), 3.72–3.42 and 3.17–3.01 (m, 2H), 3.09
and 2.88 (s, 3H), 2.35–1.65 (m, 4H).
12. Compound 20 is contaminated with a trace of its regioisomer 21. In 20, the C-5 proton resonates at δ 6.69 and 6.64 (two
rotamers), whereas the C-5 proton in 21 resonates at δ 6.96 and 6.92 (two rotamers); all these signals have moved upfield in
comparison to the C-5 signals (δ 7.34 and 7.28) in 10. The chemical shifts of C-5 proton in the two rotamers of 20 match
with the chemical shift of the C-5 proton in 22 (δ 6.6)14 and not 23 (δ 6.9).14
13. Ho, T.-L. Tandem Organic Reactions; Wiley-Interscience: New York, 1992; Frontiers in Organic Synthesis, Chem. Rev.
1996, 96, 1; Tietze, L. F.; Beifuss, U. Angew. Chem., Int. Ed. Engl. 1993, 32, 131; Padwa, A. Chem. Commun. 1998, 1417.
14. Pelisson, M. M. M.; da Silva, G. V. J.; Clive, D. L. J.; Coltart, D. M.; Hof, F. A. J. Heterocyclic Chem. 1999, 36, 653.