Synthesis of the novel antibacterial 6,8-dihydroxy-7-propyl-9H-pyrrolo [1,2-b][1,3]-benzoxazin-9-one

Article abstract of DOI:10.1016/S0040-4039(00)00590-6

The total synthesis of 6,8-dihydroxy-7-propyl-9H-pyrrolo[1,2- b][1,3]benzoxazin-9-one, a novel antibacterial agent, is described. The key step in the synthesis is achieved via an oxidative cyclization of a phenol to the 2-position of a pyrrole using stoichiometric Pd(OAc)₂. This provides a straightforward approach to the tricyclic structure of this previously unreported template and future derivatives. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00590-6

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 4061–4064
Synthesis of the novel antibacterial 6,8-dihydroxy-7-propyl-9H-
pyrrolo[1,2-b][1,3]-benzoxazin-9-one
Liren Huang,a Edmund L. Ellsworth,b,∗ Sab Randhawa,a Michael A. Stier,b Yanting Huang,a
Paul Bird,a Mikhail Lebedev,a Kun Wu,a Ron G. Micetich,a John M. Domagala,b
Rajeshwar Singh a and Norm L. Colbry c
aSynPhar Laboratories Inc., #2, 4290–91 A Street, Edmonton, Alberta T6E 5V2, Canada
bThe Department of Chemistry, Parke-Davis Pharmaceutical Research Division of the Warner-Lambert Company,
2800 Plymouth Road, Ann Arbor, Michigan 48105, USA
cThe Department of Chemical Development, Parke-Davis Pharmaceutical Research Division of the Warner-Lambert
Company, 2800 Plymouth Road, Ann Arbor, Michigan 48105, USA
Received 24 January 2000; revised 22 March 2000; accepted 23 March 2000
Abstract
The total synthesis of 6,8-dihydroxy-7-propyl-9H-pyrrolo[1,2-b][1,3]benzoxazin-9-one, a novel antibacterial
agent, is described. The key step in the synthesis is achieved via an oxidative cyclization of a phenol to the 2-
position of a pyrrole using stoichiometric Pd(OAc)2. This provides a straightforward approach to the tricyclic
structure of this prevously unreported template and future derivatives. © 2000 Elsevier Science Ltd. All rights
reserved.
Increasing resistance to existing antibacterial agents necessitates the identification of new templates
with activity against resistant strains of bacteria.1 The novel natural product 2-chloro-6,8-dihydroxy-
7-propyl-9H-pyrrolo[1,2-b][1,3]benzoxazin-9-one (1, XR 587) was isolated from Streptomyces rimosus
during a microbial extract screening program to identify inhibitors of bacterial histidine kinase.2 This
compound demonstrates promising Gram-positive antibacterial activity against both wild-type and
resistant strains of bacteria.3–5 The dechlorination of 1 with Raney nickel in DMF provides 6,8-
dihydroxy-7-propyl-9H-pyrrolo[1,2-b][1,3]benzoxazin-9-one (2), which exhibits similar antibacterial
activity.5
∗
Corresponding author.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00590-6
tetl 16869

4062
In the course of studying the structure–activity relationships of 2, we needed to develop a rapid
synthetic pathway into this previously unreported ring system. We envisioned the key step into this
heterocyclic system to be ring closure of 3 to provide 2 (Scheme 1). The reaction of a phenol coupling to
the 2-position of an N-acylpyrrole is, to the best of our knowledge, unknown in the literature.
Scheme 1.
Treatment of 1-benzoylpyrroles and 1-acetylindoles with arenes and stoichiometric Pd(OAc)2 is
known to give rise to 2-aryl-substituted pyrroles and indoles.6a These reactions are proposed to occur via
palladation of the 2-position of the electron rich pyrrole or indole rings followed by subsequent attack
of the aroyl moiety generating Pd(0) as a by-product as exemplified by the cyclization of 4 to provide 5
(Scheme 2).
Scheme 2.
We anticipated that a phenol, under similar conditions, might also serve as an effective coupling partner
based on the work of Buchwald and co-workers (Scheme 3).6b It was anticipated that 7 would cyclize,
via 8 and 9, to afford 10 upon treatment with a Pd(II) catalyst, the rate determining step presumably
being the palladation of the pyrrole ring. We found that treatment of a THF solution of 7, prepared
by the coupling of 6 with pyrrole and subsequent deprotection, with one equivalent of Pd(OAc)2 and
50 equivalents of lithium chloride7a,b at 125°C provides the desired cyclized 10 in 54% yield. Other
palladium reagents such as Pd(TFA)2 and PdCl2 were less efficient, each producing 10 in less than 10%
yield. It was observed that using fewer numbers of equivalents of LiCl (0–10 equiv.) that significantly
lower yields were obtained (0–10% yield) suggesting that the ionic strength of the solvent7b,c rather than
the Lewis acidity of LiCl influences the efficiency of the reaction. The use of bases such as K2CO3,
KO-t-Bu or amine bases provided little reaction and/or decomposition of the substrate. Attempts to make
this process catalytic by using 10 mol% of Pd(OAc)2 with Cu(OAc)2 and/or air as oxidants have been
unsuccessful.8
Using this methodology, our synthesis of 2 began with MOM-protected phloroglucinol (11).9 ortho-
Directed metallation10 with n-BuLi, followed by treatment with n-propyl iodide provided 12 that was then
metallated and quenched with 1-pyrrolecarboxylic anhydride11 to provide the benzoylpyrrole derivative
13 in high yield. Acidic deprotection afforded 3. Intramolecular cyclization of 3 was carried out as
described above to provide a mixture of 2 and 14 in a 3:1 ratio (54% yield). Recrystallization from
CHCl3 gave the desired isomer (2) as a white crystalline solid (Scheme 4).12
In order to avoid isomer formation of the palladium mediated cyclization step, an alternative approach
introducing the n-propyl group late in the synthesis was undertaken. This synthesis of 2 also began
with the ortho-directed metallation10 of MOM-protected phloroglucinol (11),9 followed by treatment

4063
Scheme 3.
Scheme 4.
with 1-pyrrolecarboxylic anhydride11 to provide the benzoylpyrrole derivative 15 in 93% yield. Acidic
deprotection of 15 afforded 16. The key intramolecular cyclization13 was carried out to provide the
desired ring system 17 in 63% yield. The n-propyl side-chain was then introduced by allylation of both
phenols to provide 18 that upon treatment with ZnCl2 underwent a Claisen rearrangement generating 19
in good yield. The allyl ether at the 8-position was subsequently cleaved by treatment with Bu3SnH/5%
Pd(Ph3P)4 to give 20, which was hydrogenated to generate 2. Compound 2 was spectroscopically
identical to that prepared by semisynthesis (vide supra).12

4064
In summary, we have accomplished the first total synthesis of the novel antibacterial agent 2. The
key step involved a palladium mediated cyclization of 1-(2-hydroxybenzoyl)pyrrole to provide the
previously unknown pyrrolobenzoxazine ring system. This synthetic methodology has allowed us to
explore structure–activity relationships associated with the antibacterial activities of this ring system.
These will be reported in due course.
References
1. Domagala, J. M.; Sanchez, J. P. Ann. Rep. Med. Chem. Bristol, J. A., Ed.; Academic Press: New York 1997, 32, 111–120.
2. (a) Trew, S. J.; Wrigley, S. K.; Pairet, L.; Sohal, J.; Shanu-Wilson, P.; Hayes, M. A.; Martin, S. M.; Manohar, R. N.;
Chicarelli-Robinson, M. I.; Kau, D. A.; Byrne, C. V.; Wellington, E. M. H.; Howard, J.; Hupe, D.; Olson, E. Journal of
Antibiotics 2000, 53(1), 1–11. (b) Raggatt, M. E.; Simpson, T. J.; Wrigley, S. K. Chem. Commun. 1999, 11, 1039–1040.
3. Demuth, H.; Breinholt, J.; Nielsen, S.; Gurtler, H. Patent PCT 09723486.
4. Breinholt, J.; Gurtler, H.; Kjar, A.; Nielsen, S. E.; Olsen, C. E. Acta Chem. Scand. 1998, 52, 1040–1044.
5. Domagala, J. M.; Ellsworth, E. L.; Huang, L.; Micetich, R. G.; Stier, M. A.; Singh, R.; Wrigley, S. K.; Wang, S. PCT Int.
Appl. WO 98 25934; Chem. Abstr. 129:67790h.
6. (a) Itahara, T. J. Org. Chem. 1985, 50, 5272–5275 and the references cited therein. (b) Widenhoefer, R. A.; Zhong, H. A.;
Buchwald, S. L. J. Am. Chem. Soc. 1997, 119, 6787–6795 and the references cited therein.
7. (a) Stille, J. K. Angew. Chem., Int. Ed. Engl. 1986, 25, 508–524. (b) Thaler, A.; Seebach, D.; Cardinaux, F. Helv. Chim.
Acta 1991, 74, 617–627. (c) Solvents such as diethyl ether, dimethylformamide and acetic acid were ineffective.
8. Auerbach, J.; Weinreb, S. M. J. Chem. Soc. Chem. Commun. 1974, 298–299.
9. Townsend, G. A.; Davis, S. G.; Christensen, S. B.; Link, J. C.; Lewis C. P. J. Am. Chem. Soc. 1981, 103, 6885–6888.
10. Winkle, M. R.; Ronald, R. C. J. Org. Chem. 1982, 47, 2101–2108.
11. Boger, D. L.; Patel, M J. Org. Chem. 1987, 52, 2319–2323.
1
12. H NMR (CDCl3) of 2: δ 11.48 (s, 1H, 8-OH), 7.19 (m, 1H), 6.46 (m, 1H), 6.29 (s, 1H), 5.82 (m, 1H), 5.48 (s, 1H, 6-OH),
2.64 (t, J=7.6 Hz, 2H), 1.61 (sex., J=7.6 Hz, 2H), 0.99 (t, J=7.6 Hz, 3H).
13. Intramolecular cyclization of 16 was carried out in the following fashion: A pressure bottle charged with 1-(2,4,6-
trihydroxybenzoyl)pyrrole (200 mg, 0.912 mmol), Pd(OAc)2 (205 mg, 0.912 mmol), anhydrous LiCl (1.933 g, 45.6 mmol)
and THF (5 ml) was heated and stirred at 125°C for 4 h. The mixture was filtered and the solid residue washed with EtOAc.
The combined organic phase was washed with brine and dried on Na2SO4. Flash chromatography on a silica column with
35% EtOAc/hexane as the eluent afforded the product 17 (124 mg, 63%) as a white crystalline solid. Mp: 201–202°C. 1H
NMR (CDCl3): δ 7.20 (dd, J=3.8 Hz, J=1.8 Hz, 1H), 6.47 (t, J=3.7 Hz, 1H), 6.29 (d, J=2.4 Hz, 1H), 6.26 (d, J=2.4 Hz,
1H), 5.84 (dd, J=3.8 Hz, J=1.8 Hz, 1H), 5.48 (s, 1H, -OH).