Metalation-cyclisation sequence on N-(o-halobenzyl)pyrroles. Synthesis of pyrrolo[1,2-b]isoquinolones

Article abstract of DOI:10.1016/S0040-4039(00)00817-0

Parham-type cyclisation of N-(o-halobenzyl)pyrroles has been investigated. Aryllithiums generated from metalation of N-(o- iodobenzyl)pyrroles undergo intramolecular cyclisation to give pyrrolo[1,2- b]isoquinolinones in moderate to good yields, if the N,N

Full text of DOI:10.1016/S0040-4039(00)00817-0

Tetrahedron Letters 41 (2000) 5211±5214
Metalation±cyclisation sequence on N-(o-halobenzyl)pyrroles.
Synthesis of pyrrolo[1,2-b]isoquinolones
Ainhoa Ardeo, Esther Lete* and Nuria Sotomayor
Â
Departamento de Quõmica Organica, Facultad de Ciencias, Universidad del Paõs Vasco, Apdo. 644,
Â
Â
48080 Bilbao, Spain
Received 26 April 2000; accepted 17 May 2000
Abstract
Parham-type cyclisation of N-(o-halobenzyl)pyrroles has been investigated. Aryllithiums generated from
metalation of N-(o-iodobenzyl)pyrroles undergo intramolecular cyclisation to give pyrrolo[1,2-b]iso-
quinolinones in moderate to good yields, if the N,N-diethylcarbamoyl group is used as internal electrophile
and the aromatic ring is activated. # 2000 Elsevier Science Ltd. All rights reserved.
Keywords: lithiation; cyclisation; pyrroles; pyrroloisoquinolines.
The aromatic metalation±cyclisation sequence is a valuable protocol for the regiospeci®c con-
struction of carbocyclic and heterocyclic systems.1 The intramolecular cyclisations that employ
aryllithiums generated by lithium±halogen exchange are known as Parham cyclisations and they
have allowed the preparation of many carbocycles and heterocycles.2 In fact, lithium±halogen
exchange,3 though mechanistically controversial, is a particularly useful tactic for the metalation
of aromatic substrates, because metal±halogen exchange can e€ectively compete with the organo-
lithium reaction with internal electrophiles.4 However, to our knowledge, this kind of strategy has
not been applied to the synthesis of pyrrolo[1,2-b]isoquinolones. The presence of this structural
framework in some natural products such as the lycorine class of Amaryllidaceae alkaloids5
and the phenanthroindolizidine alkaloids6 were among the considerations that prompted the
design of a strategy for its construction, as an extension of our earlier work7 involving the
synthesis of isoquinoline alkaloids from N-phenethylimides. In this context, we envisioned that a
N-(o-halobenzyl)pyrrole would be metalated to form an intermediate aryllithium, which could
then cyclise onto the carbonyl substituent of the pyrrole ring to give the isoquinolone nucleus of
the desired tricyclic system. A related, though intermolecular, approach has been reported8 that
relied on the addition of ortho-lithiated benzamides to the carbonyl group of the N-benzyl-
pyrrole-2-carbaldehyde. Other approaches to this skeleton imply a tandem cyclisation of amidyl
* Corresponding author. Tel: 34-946012576; fax: 34-944648500; e-mail: qopleexe@lg.ehu.es
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00817-0

5212
radicals derived from O-acyl hydroxamic acid derivatives,9 or photocyclisation reactions of 1-benzyl-
1-pyrrolinium salts.10
Herein, we describe our investigations on the Parham-type cyclisation of functionalised N-(o-
halobenzyl)pyrroles.
To initiate our investigations we ®rst evaluated the Parham cyclisation of iodinated N-benzyl-
pyrroles with acetyl or methoxycarbonyl groups at the C-2 position of the pyrrole ring as internal
electrophiles. However, the use of halogenated ketones in Parham-type cyclisations, while suc-
cessfully exploited by Aidhen and by Kihara,4b or esters did not lead to the pyrroloisoquinolones.
In our case, lithium±iodine exchange was not fast enough to compete with addition of the indu-
cing organolithium and/or enolisation, precluding cyclisation.
In view of these results, we decided to test the N,N-diethylcarbamoyl group as an internal
electrophile. The N-(o-iodobenzyl)pyrrole 3a was prepared by alkylation of N,N-diethylpyrrole-2-
carboxamide with benzyl bromide 1a under standard conditions (Scheme 1). We next carried out
the metalation±cyclisation sequence under several conditions (Scheme 1, Table 1). After some
experimentation, we found that lithium±iodine exchange took place very eciently when the
reaction mixture was treated with t-BuLi (2.2 equiv.) at ^78ꢀC for 3 h, and allowed to reach rt,
a€ording pyrroloisoquinolone 5a in good yield (79%)11 (entry 1).
Scheme 1.
To study the in¯uence of the aromatic ring substitution, we prepared a series of N-(o-iodo-
benzyl)pyrroles 3b±g from the corresponding benzyl bromides 1 (Scheme 1).12 These iodinated
pyrroles were submitted to Parham cyclisation under the previously tested conditions a€ording
the desired pyrroloisoquinolones 5 in variable yields. For some of them (Table 1, entries 2±5),
reaction conditions were optimised.13 As shown in Table 1, moderate to good yields of pyrroloiso-
quinolines 5 were obtained when the aromatic ring is activated with donor groups (entries 1, 5, 7 and
8). An exception was trimethoxy substituted benzylpyrrole 3d, which only a€orded the correspond-
ing deiodinated product in high yield (81%, entry 6). In this case, lithium±iodine exchange took
place eciently, but further cyclisation was precluded, presumably, by steric hindrance of the
ortho-methoxy group. Signi®cantly lower yields were obtained when no such activation of the ring
was present (entries 3 and 9). We also tested the feasibility of using N-(o-bromobenzyl)pyrroles 4

5213
Table 1
Parham cyclisation of benzylpyrroles 3 and 4
as substrates. Bromine±lithium exchange took place less eciently than lithium±iodine exchange,
a€ording lower yields of pyrroloisoquinoline 5a (entry 2 versus 1), though this e€ect is not
apparent with 5b (entry 4).
Although in related Parham-type cyclisations no in¯uence of aromatic substitution pattern has
been observed,4d the di€erent behaviour of iodinated N-(o-iodobenzyl)pyrroles could be under-
stood in the light of the reaction mechanism. Although few di€erent mechanisms have been sug-
gested for the lithium±halogen exchange reaction,3 the intermediacy of an ate complex or a SN2
mechanism are nowadays the most generally accepted proposal for those cases that involve aryl
halides and alkyllithiums.14 In our case, iodine±lithium exchange seems to show a leaving group
e€ect, which could be rationalised by both mechanistic possibilities.
In summary, it has been shown that N-(o-iodobenzyl)pyrrole-2-carboxamides tolerate lithium±
halogen exchange reaction conditions, allowing the ecient construction of the isoquinoline
nucleus. Though the metalation±cyclisation sequence is limited to aromatic rings activated with
donor groups (methoxy, methylenedioxy), two features make it synthetically useful. First, the fact
that it allows access to pyrrolo[1,2-b]isoquinolinones with the aromatic substitution pattern present
in natural lycorine-type alkaloids. Second, this anionic equivalent of a Friedel±Crafts reaction
could compete with previously reported strategies.8 10
Acknowledgements
Financial support from DGICYT (PB95-0499) and the Universidad del Paõs Vasco is gratefully
acknowledged. We also thank the Departamento de Educacion, Universidades e Investigacion
(Gobierno Vasco) for a grant (A.A.).

5214
References
1. For a review on aromatic metalation, see: Gray, M.; Tinkl, M.; Snieckus, V. In Comprehensive Organometallic
Chemistry II; Abel, E. W.; Stone, F. G. A.; Wilkinson, G., Eds.; Pergamon: Exeter, 1995; Vol. 11, p. 2.
2. (a) For a review, see: Parham, W. E.; Bradsher, C. K. Acc. Chem. Res. 1982, 15, 305. (b) For some more recent
examples see Ref. 1 and 4.
3. (a) Negishi, N.; Chaneiry, D. In Comprehensive Organic Functional Group Transformations; Katrizky, A. R.;
Meth-Cohn, O.; Rees, C. W., Eds.; Pergamon: Cambridge, 1995; Vol. 2, p. 951. (b) Schlosser, M. In
Organometallics in Synthesis; Schlosser, M., Ed.; John Wiley & Sons: Chichester, 1995; p. 2.
4. For some representative examples, see: (a) imide carbonyl: Hendi, M. S.; Natalie, K. J.; Hendi, S. B.; Campbell,
J. A.; Greenwood, T. D.; Wolfe, J. F. Tetrahedron Lett. 1989, 30, 275. (b) ketone carbonyl: Aidhen, I. S.;
Narashimhan, N. S. Tetrahedron Lett. 1991, 32, 2171; Kihara, M.; Kashimoto, M.; Kobayashi, Y. Tetrahedron
1992, 48, 6. (c) amide carbonyl: Aidhen, I. S.; Ahuja, J. R. Tetrahedron Lett. 1992, 33, 5431; Quallich, G. J.; Fox,
D. E.; Friedmann, R. C.; Murtishaw, C. W. J. Org. Chem. 1992, 57, 761. (d) ester carbonyl: Paleo, M. R.;
Castedo, L.; Domõnguez, D. J. Org. Chem. 1993, 58, 2763. (e) dithioester carbonyl: Beak, P.; Park, Y. S.; Reif,
L. A.; Liu, C. J. Org. Chem. 1994, 59, 7410. (f) imine: Bradsher, C. K.; Hunt, D. A. J. Org. Chem. 1981, 46, 327.
(g) alkene: Zhang, D.; Liebeskind, L. S. J. Org. Chem. 1996, 61, 2594.
5. Hoshino, O. In The Alkaloids; Cordell, G. A., Ed.; Academic Press: San Diego, 1998; Vol. 51, p. 324.
6. Herbert, R. B.; Jackson, F. B.; Nicolson, I. T. J. Chem Soc., Perkin Trans. 1 1984, 825.
7. (a) Collado, M. I.; Sotomayor, N.; Villa, M. J.; Lete, E. Tetrahedron Lett. 1996, 37, 6193. (b) Collado, M. I.;
Manteca, I.; Sotomayor, N.; Villa, M. J.; Lete, E. J. Org. Chem. 1997, 62, 2080.
8. Iwao, M.; Mahalanabis, K. K.; Watanabe, M.; de Silva, S. O.; Snieckus, V. Tetrahedron 1983, 39, 1955.
9. Clark, A. J.; Filik, R. P.; Peacock, J. L.; Thomas, G. H. Synlett 1999, 441.
10. Cho, I.-S.; Tu, C.-L.; Mariano, P. S. J. Am. Chem. Soc. 1990, 112, 3594.
11. All new compounds gave satisfactory spectroscopic and analytical data. Typical procedure for 5a: To a solution of
the iodinated benzylpyrrole 3a (169 mg, 0.4 mmol) in dry THF (20 mL), t-BuLi (0.9 mL of a 0.92 M solution in
pentane, 0.83 mmol) was added at ^78ꢀC. The resulting mixture was stirred at this temperature for 3 h, allowed to
warm to 20ꢀC, and further stirred for 4 h. The reaction was quenched by the addition of H2O (5 mL). After
standard work-up, ¯ash column chromatography (silica gel, 70% hexane:AcOEt) a€orded 7,8-dimethoxypyrrolo-
[1,2-b]tetrahydroisoquinolin-10-one (5a) as a yellowish solid (74 mg, 79%): mp (Et2O) 139±140ꢀC (dec.); IR (neat)
1
1650 cm^1; H NMR (CDCl3) 3.90 (s, 3H), 3.92 (s, 3H), 5.17 (s, 2H), 6.35 (dd, J=4.0, 2.4 Hz, 1H), 6.64 (s, 1H),
6.94±6.98 (m, 1H), 7.08 (dd, J=4.0, 1.5 Hz, 1H), 7.63 (s, 1H); 13C NMR (CDCl3) 46.4, 56.0, 56.0, 107.1, 107.9,
111.1, 112.7, 123.7, 125.2, 129.2, 129.7, 148.7, 152.8, 174.1; MS (EI) m/z (rel. intensity) 243 (M+, 100), 242 (22), 228
(21), 212 (28), 200 (13), 199 (13). Anal. calcd for C14H13NO3: C, 69.12; H, 5.39; N, 5.76. Found: C, 69.24; H, 5.41;
N, 5.84.
12. Bromides 1 and 2 were prepared from benzylic alcohols. All aromatic iodination reactions were completely
regioselective, except case f, which a€orded a 2:1 mixture of 6-iodo and 5-iodo alcohols, resulting in a low yield of
the desired 3f, after separation of isomers.
13. In all cases, variable amounts (7±13%) of the corresponding deiodinated benzylpyrroles were isolated together
with pyrrolisoquinolones 5.
14. Beak, P.; Allen, D. J. J. Am. Chem. Soc. 1992, 114, 3420. (b) Reich, J. H.; Green, D. P.; Phillips, N. H. J. Am.
Chem. Soc. 1991, 113, 1414.