Competitive intramolecular nucleophilic aromatic substitution: A new route to coumarins

Article abstract of DOI:10.1039/b009172n

4-Hydroxy-3-(2′-pyridyl)coumarins (4) (R = 6-Cl, H, 6-NO₂, 8-NO₂) were prepared in moderate to good yields by the intramolecular nucleophilic aromatic substitution reaction of β-ketoesters (I) in refluxing xylenes; evidence for the r

Full text of DOI:10.1039/b009172n

View Article Online / Journal Homepage / Table of Contents for this issue
Competitive intramolecular nucleophilic aromatic substitution: a new route to
coumarins
Concepción Alonso, Marilyn M. Olmstead, Michael H. Nantz and Mark J. Kurth*
Department of Chemistry, University of California, One Shields Avenue, Davis, CA 95616 USA.
E-mail: mjkurth@ucdavis.edu
Received (in Corvallis, OR, USA) 10th November 2000, Accepted 17th January 2001
First published as an Advance Article on the web 14th March 2001
oxide.8 Generation of the lithium salt of 1 and subsequent
treatment with various o-halobenzoyl chlorides gave the
corresponding 3-oxo-2-(2A-pyridyl)(o-halophenyl)propanoates
(2). Yields for 1 ? 2 are generally moderate (60–75%) for a
variety of halobenzoyl chlorides.
Heating 2a in xylenes at reflux for 2 h delivered a crystalline
product which we initially assumed was benzo[c]quinolizinium
salt III (82% yield, Scheme 2). However, single crystal X-ray
crystallographic analysis‡ (Fig. 2) revealed that the product was
in fact isolated as the neutral benzo[c]quinolizine 3a replete
with 1-carboalkoxy and 2-oxo substituents on the newly formed
ring. The observation that 2a ? 3a via III, which was consistent
with the results reported by Fozard and Bradsher for cycloqua-
ternization of cis-2A-chloro-2-stilbazole, suggested that ipso
substitution in our 3-oxo-2-(2A-pyridyl)(o-halophenyl)propa-
noate series would provide a general route to the benzo[c]quino-
lizine ring system.
4-Hydroxy-3-(2A-pyridyl)coumarins (4) (R = 6-Cl, H, 6-NO2,
8-NO2) were prepared in moderate to good yields by the
intramolecular nucleophilic aromatic substitution reaction of b-
ketoesters (I) in refluxing xylenes; evidence for the reversible
formation of benzo[c]quinolizinium III from I (X = 4-Cl), with
eventual formation of 4 (R = 6-Cl), is also presented.
Cystic fibrosis (CF) results from defects in the gene encoding a
cyclic adenosine monophosphate-dependent chloride ion chan-
nel known as the cystic fibrosis transmembrane conductance
regulator1 (CFTR) and is characterized by defective chloride
transport across epithelia of the airways, exocrine ducts, and
intestine as well as viscous epithelial mucous secretions.2
Becq’s recent report3 that the benzo[c]quinolizinium derivative
MPB-07 (II, Fig. 1) activates wild-type CFTR membrane
protein4 in a variety of cell systems, coupled with our interests5
in developing small molecule drugs capable of modulating
chloride-selective ion channels,6 led us to explore the use of
3-oxo-2-(2A-pyridyl)(o-halophenyl)propanoates as precursors to
CFTR-active compounds. The synthesis of benzo[c]quinolizin-
ium salts via an intramolecular cyclization reported by Fozard
and Bradsher7 led us to consider cyclization of I to MPB-07
analog III by an intramolecular ipso substitution reaction. We
report here a wider perspective on the intriguing and useful
intramolecular nucleophilic aromatic substitution chemistry of
these a-(2-pyridyl)-b-ketoesters.
Our work started with C-acylation of benzyl 2-(2-pyr-
idyl)acetate (1, Scheme 1),† in turn generated by the transester-
ification of methyl 2-(2-pyridyl)acetate with lithium benz-
Fig. 1 3-Oxo-2-(2A-pyridyl)(o-halophenyl)propanoates as precursors to
benzo[c]quinolizinium derivatives.
Scheme
1
Preparation of 3-oxo-2-(2A-pyridyl)(o-halophenyl)propa-
Scheme 2 Intramolecular ipso substitution in 3-oxo-2-(2A-pyridyl)(o-
noates.
halophenyl)propanoates.
DOI: 10.1039/b009172n
Chem. Commun., 2001, 639–640
This journal is © The Royal Society of Chemistry 2001
639

View Article Online
We thank the National Science Foundation and Cystic
Fibrosis Foundation for financial support of this research as
well as the National Science Foundation CRIF program (CHE-
9808183) for Varian Inova 400 MHz and Mercury 300 MHz
NMR instrument purchases. C. A. thanks the Departamento de
Educación, Universidades e Investigación del Gobierno Vasco
(Spain) for a Postdoctoral Fellowship.
Notes and references
† General procedure for C-acylation of benzyl 2-(2A-pyridyl)acetate (1): to
a cold (0 °C) solution of diisopropyl amine (0.33 mL, 2.4 mmol) in dry ethyl
ether (6.0 mL) under inert atmosphere was added dropwise n-BuLi (1.63 M
in hexane, 1.47 mL, 2.4 mmol). The mixture was cooled to 278 °C and a
solution of phenylmethyl 2-(2A-pyridyl)acetate (0.456 g, 2.0 mmol) in dry
ethyl ether (2.0 mL) was added dropwise. This mixture was stirred at
278 °C for 60 min at which time a solution of the appropriate benzoyl
chloride (2.0 mmol) in dry ethyl ether (1.0 mL) was added dropwise.
Stirring was continued at 0 °C for an additional 30 min at which time the
reaction was treated with 1 M aq. HCl and extracted with CH2Cl2. The
organic layer was dried (MgSO4), concentrated, and the oily residue was
recrystallized from hexane.
Fig. 2 X-Ray crystallographic structure of 3a.
However, the next substrate investigated, 2b, underwent ring-
closing ipso substitution with loss of the benzyl ester moiety.†
On closer inspection, it became apparent that ipso substitution
of the o-chloro substituent in 2b had occurred via nucleophilic
attack of the carboalkoxy giving 4-hydroxycoumarin 4b as the
sole isolated product (72%). We speculate that 2 exists in a
highly enolized, hydrogen-bonded (N…H–O) conformation9
which biases the system to intramolecular carboalkoxy—rather
than pyridyl—nucleophilic attack resulting in 2b ? 4b. In the
case of 2a, the p-chloro substituent apparently deactivates the
ring toward nucleophilic attack to the extent that only the
2-pyridyl moiety is nucleophilic enough to participate in ipso
substitution, leading to formation of 3a.
These observations led us to speculate that benzo[c]quinoli-
zine 3a might represent the kinetic product in this reaction and
raised the question whether further heating of the 2a ? 3a
reaction mixture might lead to formation of the corresponding
4-hydroxycoumarin derivative. This would presumably occur
by reversion of III to 2a by intermolecular ipso attack by
chloride followed by slow intramolecular ring-closing by
carboalkoxy ipso attack to [4a]. Once formed, irreversible loss
of BnCl from [4a] would deliver 4a. To test this idea, the 2a ?
3a reaction was performed in toluene-d8 (110 °C) and
intermittently monitored by 1H-NMR. As anticipated, we
observed the fairly rapid formation of 3a (2a consumed in 72 h)
followed by its slow disappearance and matched by the slow
appearance of both benzyl chloride and 4-hydroxycoumarin 4a
(intermittent monitoring over 12 d). Moreover, when the
laboratory scale reaction of compound 2a was performed in
refluxing xylene, formation of benzo[c]quinolizine 3a was
detected early on (monitored by TLC). Continued heating for
10 d afforded 4-hydroxycoumarin 4a in 65% yield.
General procedure for 4-hydroxycoumarin formation (2 ? 4): a solution
of phenylmethyl 3-aryl-3-oxo-2-(2A-pyridyl)propanoate 2 (0.2 mmol) in
xylenes (2.0 mL) was stirred at reflux under N2 for 2 h. After removal of the
solvent in vacuum, the 4-hydroxycoumarin product was purified by silica
gel chromatography (hexanes+EtOAc).
‡ Crystallographic data: 3a (R = 2C6H5) C21H14ClNO3, M = 363.78,
¯
triclinic, space group P1, a = 7,1272(9), b = 9.5407(12), c = 13.1738(10)
Å, a = 77.273(8), b = 79.502(9), g = 68.701(10)°, U = 808.95(16) Å3, Z
= 2, m = 2.280 mm21, T = 133(2) K, a unique data set of 2105 independent
reflections was collected, R1 = 0.0355 for all data. CCDC 152984. See
http://www.rsc.org/suppdata/cc/b0/b009172n/ for crystallographic data in
.cif or other electronic format.
1 A. L. Gibson, L. M. Wagner, F. S. Collins and D. L. Oxender, Science,
1991, 254, 109.
2 J. R. Riordan, J. M. Rommens, B. S. Kerem, N. Alon, R. Rozmahel, Z.
Grzelezak, J. Zielenski, S. Lok, N. Plavsic, J. L. Chou, M. L. Drumm,
M. C. Iannuzzi, F. S. Collins and L. C. Tsui, Science, 1989, 245, 1059.
3 F. Becq, Y. Mettey, M. A. Gray, L. J. Galietta, R. L. Dormer, M. Merten,
T. Metaye, V. Chappe, C. Marvingt-Mounir, O. Zegarra-Moran, R.
Tarran, L. Bulteau, R. Dérand, M. M. C. Pereira, M. A. McPherson, C.
Rogier, M. Joffre, B. E. Argent, D. Sarrouilhe, W. Kammouni, C.
Figarella, B. Verrier, M. Gola and J.-M. Verfond, J. Biol. Chem., 1999,
274, 27 415.
4 J. R. Riordan, J. M. Rommens, B. S. Kerem, N. Alon, R. Rozmahel, Z.
Grzelczak, J. Zeilenski, S. Lok, N. Plavsic, J. Chou, M. L. Drumm, M. C.
Iannuzzi, F. S. Collins and L. C. Tsui, Science, 1989, 245, 1066; B.
Kerem, J. M. Rommens, J. A. Buchanan, D. Markiewicz, T. K. Cox, A.
Chakravarti, M. Buchwald and L. C. Tsui, Science, 1989, 245, 1073.
5 M. Eda, M. J. Kurth and M. H. Nantz, J. Org. Chem., 2000, 65, 5131;
M. J. Haddadin, M. J. Kurth and M. Olmstead, Tetrahedron Lett., 2000,
41, 5613.
6 M. Li, J. D. MacCann, M. P. Anderson, J. P. Clancy, C. M. Liedtke, A. C.
Nairn, P. Greengard and M. J. Welssch, Science, 1989, 244, 1353.
7 A. Fozard and C. K. Bradsher, J. Org. Chem., 1966, 31, 2346.
8 B. M. Bhawal, S. P. Khanapure and E. R. Biehl, Synthesis, 1991, 113.
9 Remarkably, the enolic proton in 2a appears as a singlet at 18.4 ppm
(CDCl3) which we believe is indicative of a pyridine H-bonded
conformation.
Three additional 3-oxo-2-(2A-pyridyl)(o-halophenyl)propa-
noates were also investigated (2c–e). In each of these, the o-halo
substituent was a bromine and, in two of these, a strongly
activating nitro group was incorporated at C5 (2d) or C3 (2e). In
each of these cases, only carboalkoxy nucleophilic attack was
observed. The yields for 2 ? 4 are generally quite good
(70–96%), with the more electron deficient C-ring systems
affording higher yields.
The method reported here provides a general and useful route
for the production of 4-hydroxy-3-(2A-pyridyl)coumarin deriva-
tives. While both pyridyl and carboalkoxy moieties can
participate in this reaction, reversible formation of the benzo-
[c]quinolizinium coupled with irreversible loss of benzyl
chloride during coumarin formation leads to the exclusive
formation of the 4-hydroxy-3-(2A-pyridyl)coumarin derivative.
640
Chem. Commun., 2001, 639–640