A novel route to pyrrolo[2,1-c][1,4]benzodiazepin-5-ones. Formal total synthesis of (±)-DC-81

Article abstract of DOI:10.1021/ol991100g

(formula presented) Compounds 3a and 3b were synthesized from N-allylisatoic anhydrides 5a and 5b in six and seven steps, respectively. Synthesis of 3b constitutes a formal total synthesis of (±)-DC-81.

Full text of DOI:10.1021/ol991100g

ORGANIC
LETTERS
1999
Vol. 1, No. 11
1835-1837
A Novel Route to
Pyrrolo[2,1-c][1,4]benzodiazepin-5-ones.
Formal Total Synthesis of (±)-DC-81
Tiansheng Wang,* Alfred S. Lui,† and Ian S. Cloudsdale†
NoVartis Crop Protection, Palo Alto Research Center, 975 California AVenue,
Palo Alto, California 94304
twang@nitromed.com
Received September 29, 1999
ABSTRACT
Compounds 3a and 3b were synthesized from N-allylisatoic anhydrides 5a and 5b in six and seven steps, respectively. Synthesis of 3b
constitutes a formal total synthesis of (±)-DC-81.
Currently there is considerable interest in the discovery and
development of small molecules such as pyrrolo[2,1-c][1,4]-
benzodiazepines (PBD’s) as potential antitumor and gene
targeted drugs.1 The PBD class of antitumor antibiotics exert
their biological activity by covalently binding to the N-2 of
guanine in the minor groove of DNA through the imine or
imine equivalent functionality at N10-C11 of the PBD’s.
Anthramycin (1) and DC-81 (2) are two well-known and
promising members of the PBD’s (Figure 1). Synthetic
synthesis of PBD. The key feature of our synthesis is based
on the pyrrolidine C-ring formation from radical-initiated
carbon-carbon bond formation of the C1-C11a bond from
4 (Scheme 1).
Scheme 1
The N-allylisatoic anhydrides 5a and 5b were chosen as
our starting materials. While 5a is commercially available,4
Figure 1.
* Author to whom correspondence should be addressed at NitroMed Inc,
12 Oak Park Drive, Bedford, MA 01730.
† Current addresses: A. S. Lui, Roche Bioscience, 3401 Hillview Avenue,
Palo Alto, CA 94304. I. S. Cloudsdale, Novartis Agribusiness Biotechnology
and Research Institute, 3054 Cornwallis Road, P.O. Box 12257, Research
Triangle Park, NC 27709.
approaches to PBD’s have been documented.2,3 Herein we
would like to introduce a novel, linear approach for the total
10.1021/ol991100g CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/05/1999

5b5 was prepared from the known 4-benzyloxy-3-methoxy-
benzoic acid 66 in six steps as shown in Scheme 2. Methyl
nickel(II) chloride as catalyst to generate aniline 9, which
was converted to acid 10 by hydrolysis. The sodium salt of
10 was treated with a phosgene-toluene solution directly
to give 7-benzyloxy-6-methoxyisatoic anhydride 11 as a solid
after simple filtration. The desired N-allyl anhydride 5b was
then prepared by treatment of 11 with sodium hydride
followed by allyl bromide in N,N-dimethylacetamide.7
The N-allylisatoic anhydrides (5a and 5b) were reacted
with 3-bromopropylamine hydrobromide in the presence of
triethylamine as base (Scheme 3).5 The purified allylamino
Scheme 2. Synthesis of 5ba
Scheme 3
ester 7, which was obtained by esterification of 6, was
nitrated with fuming nitric acid to afford an 83% yield of
methyl 4-benzyloxy-5-methoxy-2-nitrobenzoate 8. The nitro
group in 8 was cleanly reduced by sodium borohydride using
(1) Thurston, D. E. In Molecular Aspects of Anticancer Drug-DNA
Interactions; Neidle, S., Waring, M. J., Eds.; The Macmillan Press Ltd.:
London, 1993; Vol. 1, pp 54-88.
(2) (a) Thurston, D. E.; Bose D. S. Chem. ReV. 1994, 94, 433-465 and
references cited therein. (b) Kamal, A.; Rao, M. V.; Reddy, B. S. Chem.
Heterocycl. Compd. 1999, 34, 1342.
(3) For recent syntheses of PBD’s, see: (a) Kraus, G. A.; Melekhov, A.
Tetrahedron 1998, 54, 11749. (b) Prabhu, K. R.; Sivanand, P. S.;
Chandrsekaran, S. Synlett 1998, 47. (c) Kamal, A.; Howard, P. W.; Reddy,
R. S. N.; Reddy, B. S. P.; Thurston, D. E. Tetrahedron 1997, 53, 3223. (d)
O′Neil, I. M.; Murray, C. L.; Hunter, R. C.; Kalindjian, S. B.; Jenkins, T.
C. Synlett 1997, 75. (e) Kamal, A.; Reddy, B. S. N.; Reddy, B. S. P. Bioorg.
Med. Chem. Lett. 1997, 7, 1825. (f) Kamal, A.; Damayanthi, Y.; Reddy, B.
S. N.; Lakminarayana, B.; Reddy, B. S. P. Chem. Commun. 1997, 1015.
(g) Kamal, A.; Reddy, B. S. P.; Reddy, B. S. N. Tetrahedron Lett. 1996,
37, 6803. (h) Kamal, A.; Reddy, B. S. P.; Reddy, B. S. N. Tetrahedron
Lett. 1996, 37, 2281. (i) Kamal, A.; Rao, N. V. Chem. Commun. 1996,
385. (j) Eguchi, S.; Yamashita, K.; Matsushita, Y.; Kakehi, A. J. Org. Chem.
1995, 60, 4006. (k) Molina, P.; Diaz, I.; Tarraga, A. Tetrahedron 1995, 51,
5617.
benzamides 12 slowly self-condensed, as shown by nonpolar
spot formation on TLC during chromatographic purification,
and thus were directly protected as the stable trifluoro-
(4) Purchased from Maybridge Inc.
(6) Thurston, D. E.; Murty, V. S.; Langley, D. R.; Jones, G. B. Synthesis
1990, 81.
(7) We followed the following literature for a similar reaction: Coppola,
G.; Damon, R. E.; Hardtmann, G. E. Synthesis 1981, 391.
(5) For reviews of isatoic anhydride chemistry, see: (a) Coppola, G. M.
Synthesis 1980, 505. (b) Kappe, T.; Stadlbauer, W. AdV. Heterocycl. Chem.
1981, 28, 127.
1836
Org. Lett., Vol. 1, No. 11, 1999

acetamides 13. The allylic double bond in 13 was subjected
to ozonolysis followed by quenching with dimethyl sulfide.
The resulting crude hemiaminal intermediates 14 were
dehydrated in refluxing toluene with camphorsulfonic acid
(CSA) as catalyst to afford the bicyclic enamides 15. The
pivotal step, the tributyltin hydride mediated radical cycliza-
tion of 15, afforded the ring-closure tricyclic compounds 16
in excellent yields. The trifluoroacetyl groups were removed
in methanolic aqueous potassium carbonate solution to yield
the amino compounds 3a and 17. The benzyl protecting
group in 17 was removed by palladium-carbon-catalyzed
hydrogenation using ammonium formate as the hydrogen
source to give 3b. Since 3b had already been converted to
DC-81 (2) by Kamal and Thurston in a recently developed
procedure,3c this synthesis constitutes a formal total synthesis
of (()-DC-81 (2).
synthesis and should provide entry to the synthesis of other
naturally occurring PBD’s and their analogues. This would
be especially true for the synthesis of PBD’s with additional
C-ring functionalities, many of which are important for
biological activity1,8 (by starting with C2-substituted 3-bro-
mopropylamine derivatives).
Acknowledgment. We thank Belle Abrera and Ann
Sjolander, formerly of Novartis, for help in MS determina-
tion.
Supporting Information Available: Full experimental
1
procedures for syntheses and spectra data (IR, H and 13C
NMR and MS) of 7-11, 5b, 13a, 13b, 15a, 15b, 16a, 16b,
17, 3a and 3b. This material is available free of charge via
the Internet at http://pubs.acs.org.
OL991100G
In conclusion, using bicyclic benzodiazepinone intermedi-
ates 15, we were able to generate a novel route to PBD’s
via radical cyclization to form the C1-C11a bond. This
synthesis complements the known methodologies2,3 for PBD
(8) Thurston, D. E.; Bose, D. S.; Howard, P. W.; Jenkins, T. C.; Leoni,
A.; Baraldi, P. G.; Guiotto, A.; Cacciari, B.; Kelland, L. R.; Foloppe, M.-
P., Rault, S. J. Med. Chem. 1999, 42, 1951.
Org. Lett., Vol. 1, No. 11, 1999
1837