Straightforward synthesis of 11H-indolo[3,2-c]isoquinoline and benzofuro[3,2-c]isoquinoline by ring transformation

Article abstract of DOI:10.1016/S0040-4039(02)01244-3

An efficient method was established for the synthesis of 11H-indolo[3,2-c]isoquinoline and benzofuro[3,2-c]isoquinoline using thermal ring transformation of a benzisoxazolo[2,3-a]isoquinoline salt.

Full text of DOI:10.1016/S0040-4039(02)01244-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 6035–6038
Straightforward synthesis of 11H-indolo[3,2-c]isoquinoline and
benzofuro[3,2-c]isoquinoline by ring transformation†
Mariann Be´res, Ge´za Tima´ri* and Gyo¨rgy Hajo´s
Institute of Chemistry, Chemical Research Center, Hungarian Academy of Sciences, PO Box 17, H-1525 Budapest, Hungary
Received 25 March 2002; revised 29 May 2002; accepted 24 June 2002
Abstract—An efficient method was established for the synthesis of 11H-indolo[3,2-c]isoquinoline and benzofuro[3,2-c]isoquinoline
using thermal ring transformation of a benzisoxazolo[2,3-a]isoquinoline salt. © 2002 Elsevier Science Ltd. All rights reserved.
A series of tetracyclic heteroaromatic compounds based
on the indoloquinoline framework have been isolated
from Cryptolepis sanguinolenta1 which has been used in
traditional medicine in Central and West Africa. More
recently one of these alkaloids, cryptosanguinolentine
2b, was synthetized by us2 and Molina.3 Cryptolepis
alkaloids and their derivatives possess a number of
reported bioactivities, including antibacterial,4 anti-
inflammatory5 and antiplasmodial6 activity. These
findings as well as some recent reports on the effective-
ness of various tetracyclic compounds as anticancer
drugs,7 prompted us to focus our attention in this area.
Thus, we report in this paper the synthesis of indolo-
and two benzofuro-fused isoquinolines. Thermal
cyclization of azidophenyl-quinoline2 1 (Fig. 1) was
performed in boiling dichlorobenzene and resulted
exclusively in the formation of 4-substituted quinoline
derivative 2a, however, when isoquinoline derivative 4
was subjected to the same reaction a new ring system 5
was formed8 through NꢀN bond formation, indicating
the higher reactivity of the forming nitrene towards the
nitrogen atom of the isoquinoline rather than the car-
bon at position 4. We decided to prepare the isoquino-
line analogue 3 of cryptosanguinolentine, which differs
Figure 1.
Keywords: indolo–isoquinoline; benzofuro–isoquinoline; ring transformation; oxenium ion.
* Corresponding author. Present address: CHINOIN Pharmaceutical and Chemical Works Co. Ltd, H-1325 Budapest, PO Box 110, Hungary.
Fax: +36-1/3705412; e-mail: geza.timari@sanofi-synthelabo.com
† Dedicated to Professor Andra´s Messmer on the occasion of his 80th birthday.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01244-3

6036
M. Be´res et al. / Tetrahedron Letters 43 (2002) 6035–6038
from the natural product only in the position of one
nitrogen atom.
yield. This reaction can be interpreted by cyclisation of
the azido-compound through formation of an interme-
diate nitrene followed by thermal deoxygenation of the
ring-closed product 11. This assumption seemed to be
supported by our earlier finding8 that thermal cyclisa-
tion of the deoxy derivative of 10 (i.e. 4) gave exclu-
sively 5 and formation of 3 was not detected.
In order to block the reactivity of the isoquinoline
nitrogen atom in compound 6, prepared8 by a Pd-
catalysed cross-coupling reaction of 2-(N-pivaloyl-
amino)phenylboronic acid and 3-hydroxyisoquinoline-
O-triflate, in the ring-closure reaction, the N-oxide
function was introduced 7 using mCPBA. It was
expected that this would be an easily removable pro-
tecting group, which would not influence the reactivity
of the isoquinoline ring at position 4 (Scheme 1). After
hydrolysis, the amino-compound 8 was transformed to
the corresponding azide 10 via diazotisation followed
by reaction with sodium azide. However, besides the
azido compound 10 a new product, benzisoxazolo[2,3-
b]isoquinoline salt 12 was also isolated, the details of
which are discussed later. It was anticipated that the
methodology successfully used previously9 for the syn-
thesis of polyfused heterocycles could be employed for
the thermal cyclisation of azide 10, however, the
expected N-oxide derivative 11 was not formed.
Instead, the desired final product, tetracyclic 11H-
indolo[3,2-c]isoquinoline10 3 was isolated in moderate
Formation of the new ring-system, benzisoxazolo[2,3-
b]isoquinoline 12 can be interpreted by a nucleophilic
attack of the N-oxide at the electrophilic carbon atom
bearing diazonium group, followed by nitrogen elimina-
tion. A similar reaction was earlier observed by
Abramovitch11 and most recently it has been proved
that tertiary propargylamine N-oxides easily undergo a
new type of rearrangement12 through formation of an
isoxazoline intermediate. Compound 12 was obtained
in a higher yield when the diazonium salt was used in
the stable tetrafluoroborate form prepared by adding a
cold aqueous ammonium tetrafluoroborate solution to
the diazotized mixture. It is known from the litera-
ture,13 that the generation of aryloxenium ions has been
proposed in the thermolysis of the benzisoxazolo[2,3-
Scheme 1. Reagents and conditions: (a) mCPBA, CHCl3, r.t., 1 d, 76%; (b) 20% H2SO4, reflux, 7 h, 92%; (c) conc. HCl, NaNO2,
0°C, 1 h, then NH+4BF4−, 0oC, 1 h, 65%; (d) NaN3, 0°C, 1 h, 63%; (e) o-dichlorobenzene, 150°C, 6 h, 43%; (f) CH3CN, r.t., 8 h,
82%; (g) o-dichlorobenzene, MW, 170°C, 30 min, 65% (14), 13% (15).

M. Be´res et al. / Tetrahedron Letters 43 (2002) 6035–6038
6037
Claeys, M.; Vlietinck, A. Tetrahedron Lett. 1996, 37,
1703–1707.
2. Tima´ri, G.; Soo´s, T.; Hajo´s, G. Synlett. 1997, 1067–1768.
3. (a) Fresneda, P. M.; Molina, P.; Delgado, S. Tetrahedron
2001, 57, 6197–6202; (b) Molina, P.; Fresneda, P. M.;
Delgado, S. Synthesis 1999, 326–329.
4. Cimanga, K.; De Bruyne, T.; Lasure, A.; Van Poel, P.;
Pieters, L.; Claeys, M.; Vanden Berghe, D.; Kambu, K.;
Tona, L.; Vlietinck, A. J. Planta Med. 1996, 62, 22–28.
5. Noamesi, B. K.; Bamgbose, S. O. A. Planta Med. 1983,
47, 100–105.
Figure 2.
6. Cimanga, K.; De Bruyne, T.; Pieters, L.; Claeys, M.;
Vlietinck, A. J. Nat. Prod. 1997, 60, 688–692.
a]pyridinium 16 ring system (debenzologue of 12). It
was hoped that the generation of the aryloxenium ion
from this salt would lead to intramolecular substitution
of the pyridine ring.
7. (a) Julio, M.; Stevens, M. F. G. J. Chem. Soc., Perkin
Trans. 1 1998, 1677 and references cited therein; (b)
Deady, L. W.; Kaye, A. J.; Finlay, G. J.; Baguley, B. C.;
Denny, W. A. J. Med. Chem. 1997, 40, 2040–2051 and
references cited therein; (c) Takeuchi, Y.; Oda, T.; Chang,
M.; Okamota, Y.; Ono, J.; Oda, Y.; Harada, K.; Hashi-
gaki, K.; Yamato, M. Chem. Pharm. Bull. 1997, 45,
406–410.
8. Tima´ri, G.; Soo´s, T.; Hajo´s, G.; Messmer, A.; Nacsa, J.;
Molnar, J. Bioorg. Med. Chem. Lett. 1996, 6, 2831–2835.
9. (a) Csa´nyi, D.; Tima´ri, G.; Hajo´s, G. Synth. Commun.
1999, 29, 3959–3969; (b) Csa´nyi, D.; Hajo´s, G.; Riedl, Z.;
Tima´ri, G.; Bajor, Z.; Cochard, F.; Sapi, J.; Laronze,
J-Y. Bioorg. Med. Chem. Lett. 2000, 10, 1767–1769; (c)
Trecourt, F.; Mongin, F.; Mallet, M.; Queguiner, G.
Synth. Commun. 1995, 25, 4011–4017.
Thermolysis of 16a failed to yield benzofuro[3,2-
b]pyridine 17a. It was reasoned that this could be due
to the proposed oxenium ion intermediate not being
electrophilic enough to attack a pyridine b-position.
Introduction of a nitro group para to the oxygen atom
should remedy this problem. Indeed, thermolysis of 16b
gave 17b in 20% yield (Fig. 2). The fact that oxenium
ions are isoelectronic with nitrenes adds another dimen-
sion to the interest in these species. Because of the high
reactivity at C-4 of the isoquinoline ring towards elec-
trophiles,14 an increase in the amount of ring transfor-
mation product 14 compared to 17 would be expected
without activation of aryloxenium ion by the nitro
group. Indeed, when linearly fused isoxazolium salt 12
was heated in o-dichlorobenzene facile ring transforma-
tion took place to give the expected benzofuro[3,2-
10. (a) Govindachari, T. R.; Sudarsanam, V. Indian J. Chem.
1967, 5, 16; (b) Martin, M. J.; Trudell, M. L.; Diaz
Arauzo, H.; Allen, M. S.; LaLoggia, A. J.; Deng, L. J.
Med. Chem. 1992, 35, 4105–4117.
c]isoquinoline15 14 together with
a little of the
11. Abramovitch, R. A.; Inbasekaran, M. N. J. Chem. Soc.,
para-fluorophenol derivative 15, the formation of which
can be interpreted by a nucleophilic attack of fluoride
at the para position of the phenyloxenium cation 13
similar to the Schiemann reaction.16 This consideration
was supported by replacing the BF−4 anion of 12 with
HSO−4; in this case the formation of phenol derivative
15 was not detected.
Chem. Commun. 1978, 149–150.
12. Szabo´, A.; Hermecz, I. J. Org. Chem. 2001, 66, 7219–
7223.
13. (a) Abramovitch, R. A.; Alvernhe, G.; Bartnik, R.; Das-
sanayake, N. L.; Inbasekaran, M. N.; Kato, S. J. Am.
Chem. Soc. 1981, 103, 4558–4565; (b) Abramovitch, R.
A.; Inbasekaran, M. N.; Kato, S.; Radzikowska, T. A.;
Tomasik, P. J. Org. Chem. 1983, 48, 690–694.
14. Be´res, M.; Hajo´s, G.; Riedl, Z.; Soo´s, T.; Tima´ri, G.;
Messmer, A. J. Org. Chem. 1999, 64, 5499–5503.
15. Selected data for compounds in Scheme 1: (a) Data for 9:
In summary, we have elaborated a concise synthesis of
two closely related isomers (3 and 14) of natural cryp-
tosanguinolentine 2b. This route seems to provide a
general method for the preparation of various substi-
tuted derivatives of c-fused indolo- and benzofuro-iso-
quinolines using appropriately substituted isoquinolines
and substituted N-pivaloylamino arylboronic acids and
opens the way for their biological evaluation.
1
BF−4 salt, H NMR (CD3CN) l 9.05 (1H, s), 8.61 (1H, d,
J=8 Hz), 8.4 (1H, s), 8.38 (1H, m), 8.12–7.94 (4H, m),
7.84–7.26 (2H, m). IR (KBr) w=3102, 2284, 1561, 1310,
1060, 770 cm−1; (b) Data for 10: 1H NMR (CDCl3) l 8.86
(1H, s), 7.79 (1H, d, J=8 Hz), 7.74 (1H, d, J=8 Hz),
7.70 (1H, s), 7.62–7.58 (2H, m), 7.53 (1H, m), 7.45 (1H,
d, J=7.5 Hz), 7.30 (1H, d, J=7.5 Hz), 7.27 (1H, m). IR
(KBr) w=2086, 1435, 1315, 1285, 1127, 760 cm−1; (c)
Data for 3: 1H NMR (DMSO-d6) l 12.3 (1H, s), 9.12
(1H, s), 8.51 (1H, d, J=7.7 Hz), 8.29 (1H, d, J=8 Hz),
8.24 (1H, d, J=8Hz), 7.91 (1H, m), 7.70 (1H, m), 7.69
(1H, d, J=7.6 Hz), 7.49 (1H, m), 7.31 (1H, m), 13C
(DMSO-d6) l 144.5, 138.4, 133.4, 129.8, 128.4, 127.2,
126.4, 126.1, 125.4, 123.4, 122.6, 121.1, 119.7, 119.1,
Acknowledgements
Fund OTKA 26476 is gratefully appreciated.
References
1
111.7 ppm; (d) Data for 12: H NMR (CD3CN) l 10.31
1. (a) Sharaf, M. H. M.; Schiff, P. L.; Tackie, A. N.;
Phoebe, C. H.; Martin, G. E. J. Heterocycl. Chem. 1996,
33, 239–243; (b) Cimanga, K.; De Bruyne, T.; Pieters, L.;
(1H, s), 9.22 (1H, s), 8.52 (1H, d, J=8 Hz), 8.45–8.44
(2H, m), 8.23 (1H, m), 8.10 (1H, m), 8.01 (1H, m), 7.90
(1H, d, J=8.0 Hz), 7.79 (1H, m). 13C (CD3CN) l 156.2,

6038
M. Be´res et al. / Tetrahedron Letters 43 (2002) 6035–6038
1
7.64 (1H, m), 7.54 (1H, m); (f) 21Data for 15: H NMR
(DMSO-d6) l 9.82 (1H, s), 9.18 (1H, s), 8.54 (1H, d,
J=8.0 Hz), 8.18 (1H, d, J=8.0 Hz), 8.10 (1H, m), 8.09
(1H, d, J=10 Hz), 7.90 (1H, m), 7.30 (1H, dd, J=12Hz,
9 Hz), 7.07 (1H, dd, J=5 Hz, 9 Hz).
137.5, 136.9, 135.3, 134.6, 132.6, 130.1, 128.9, 128.1,
126.3, 123.9, 120.1, 118.3, 118.1, 110.3 ppm. IR (KBr)
w=1643, 1460, 1371, 1225, 1084, 1058, 769 cm−1; (e) Data
for 14: H NMR (DMSO-d6) l 9.38 (1H, s), 8.41 (1H, d,
J=7.7 Hz), 8.39 (1H, d, J=7.7 Hz), 8.22 (1H, d, J=8.0
Hz), 8.0 (1H, m), 7.91 (1H, d, J=8.0 Hz), 7.82 (1H, m),
1
16. Suschitzky, H. Adv. Fluorine Chem. 1965, 4, 1.