Synthesis and cross-coupling reactions of 7-azaindoles via a new donor-acceptor cyclopropane

Article abstract of DOI:10.1021/ol061379i

A new type of donor-acceptor cyclopropane has been prepared from commercially available cyclopropane-1,1-diesters. This cyclopropane reacts with triflic anhydride to produce an isolable tristrifloxy intermediate which when treated with primary amines give

Full text of DOI:10.1021/ol061379i

ORGANIC
LETTERS
2006
Vol. 8, No. 17
3777-3779
Synthesis and Cross-Coupling
Reactions of 7-Azaindoles via a New
Donor−Acceptor Cyclopropane
Xiaomei Zheng and Michael A. Kerr*
Department of Chemistry, The UniVersity of Western Ontario,
London, Ontario, Canada N6A 5B7
makerr@uwo.ca
Received June 6, 2006
ABSTRACT
A new type of donor−acceptor cyclopropane has been prepared from commercially available cyclopropane-1,1-diesters. This cyclopropane
reacts with triflic anhydride to produce an isolable tristrifloxy intermediate which when treated with primary amines gives 6-trifloxy-7-azaindolines
which in turn can be dehydrogenated to the azaindoles. The 6-trifloxy substituent can be used to introduce diversity at this position via a
variety of cross-coupling reactions thus preparing potentially interesting compounds based on the important 7-azaindole pharmacophore.
The importance of the indole moiety in both natural products
and pharmaceutical chemistry is well understood, due in no
small part to the vast number of bioactive natural products
as well as medicinally important unnatural compounds based
on the benzopyrrole skeleton.1 Some of the drawbacks of
the indole as a pharmacophore, such as poor water solubility,
are being addressed by the investigation of azaindoles, where
one or more of the carbons of the indole ring system have
been replaced by a nitrogen atom.2 7-Azaindoles, in par-
ticular, are of interest because the proximity of the two
nitrogen atoms allows this ring system to mimic purines in
its role as a hydrogen-bonding partner. Although a relatively
few natural products contain the 7-azaindole motif (variolin
B is one of the few3), it is the subject of pharmaceutical
patent literature too extensive to reference here, although the
HIV attachment inhibitor BMS-3788064 and the kinase
inhibitors investigated by Johnson and Johnson5 stand as
examples of unnatural bioactive 7-azaindoles (Figure 1).
Figure 1. Representative 7-azaindoles.
In our continuing investigation into the chemistry of
donor-acceptor cyclopropanes, we had hypothesized that a
cyclopropane such as 1 would react with nitrones or imines
in a manner we have described previously,6 to afford
spiroadducts such as 2 or 3 (Scheme 1). Our interest lay in
the fact that this substructure is present in a target of interest
(1) (a) Gribble, G. W. In ComprehensiVe Heterocyclic Chemistry, 2nd
ed.; Pergammon Press: New York, 1996; Vol. 2, pp 203-257. (b) Snieckus
V. A. In The Alkaloids; Academic Press: New York, 1968; Vol. 11, Chapter
1.
(2) For a recent synthesis of 7-azaindoles with leading references
concerning both synthesis and pharmaceutical aspects, see: Schirok, H.
Synlett 2005, 1255-1258.
(3) Anderson, R. J.; Hill, J. B.; Morris, J. C. J. Org. Chem. 2005, 70,
6204-6212 and references therein.
(4) Yang, A.; Zadjura, L.; D’Arienzo, A. M.; Santone, K.; Llunk, L.;
Green, D.; Lin, P.-F.; Colonno, R.; Wang, T.; Neanwell, N.; Hansel, S.
Biopharm. Drug Dispos. 2005, 26, 387-402.
(5) O’Neill, D. J.; Shen, L.; Prouty, C.; Conway, B. R.; Westover, L.;
Xu, J. Z.; Zhang, H.-C.; Maryanoff, B. E.; Murray, W. V.; Demarest, K.
T.; Kuo, G.-H. Bioorg. Med. Chem. 2004, 12, 3167-3185.
10.1021/ol061379i CCC: $33.50
© 2006 American Chemical Society
Published on Web 07/29/2006

which had served us so well previously for the activation of
cyclopropane diesters, led to no reaction at ambient tem-
perature and decomposition at higher temperatures. Stronger
Lewis acids such as BF3 etherate or TiCl4 led to decomposi-
tion.
Scheme 1. Proposed Synthesis of a Nakadomarin A
Substructure
At this time, we turned to an alternative activation via
triflic anhydride.7 Treatment of 1 with Tf2O and pyridine in
CH2Cl2 led to relatively clean formation of the tristriflate
11 (Scheme 4). In the absence of pyridine, the reaction
Scheme 4. Reaction of 1 with Triflic Anhydride
to us, namely, nakadomarin A (4). We were disappointed to
discover that these cycloadditions which had proceeded so
smoothly with other substrates, failed with this cyclopropane.
We were, however, able to activate this cyclopropane toward
nucleophilic ring opening with various amines under the
influence of triflic anhydride. This communication describes
this chemistry and a novel, flexible, and efficient synthesis
of 7-azaindoles.
Cyclopropane 1 was prepared from the commercially
available 1,1-dicarbomethoxy cyclopropane 5 in the manner
shown in Scheme 2. Monoammonolysis of the diester moiety
produced in a near 1:1 mixture of 11 and the dihydrofura-
nopyridine 12.
The synthetic utility of 11 with its three potential elec-
trophilic sites was immediately obvious. It was felt that
treatment of 11 with a nucleophile capable of a double
displacement would first react with the 1° triflate via an SN2
process followed by an intramolecular SNAr reaction. To this
end, 11 was treated with a variety of 1° amines in CH2Cl2.
The result was the clean formation of 6-trifloxy-7-azaindo-
lines in 76-88% yields (Scheme 5). The adducts could be
Scheme 2. Preparation of Cyclopropane 1
followed by DIBAL reduction produces amido aldehyde 6
in 54% overall yield. Horner-Emmons olefination of the
aldehyde followed by a base-induced ring formation produces
the target cyclopropane 1.
Scheme 5. Synthesis of 7-Azaindolines and 7-Azaindoles
Our initial attempts to activate 1 toward ring-opening
reactions with Lewis acids were met with unequivocal failure.
We had envisioned an activated complex such as 7 or 8
(Scheme 3) which would, in the presence of a suitable
dehydrogenated with MnO2 in benzene to yield the 7-aza-
indoles in 79-91% yields. It is interesting, but not totally
unexpected, that a second SNAr reaction of the other aryl
triflate moiety did not occur. Clearly, the intramolecular SN-
Ar reaction was faster, and moreover, the replacement of
one trifloxy group with an amino group deactivated the ring
toward subsequent nucleophilic attack.
Scheme 3. Proposed Ring Opening of Cyclopropane 1
(6) (a) Wanapun, D.; Van Gorp, K. A.; Mosey, N. J.; Kerr, M. A.; Woo,
T. K. Can. J. Chem. 2005, 83, 1752-1767. (b) Carson, C. A.; Kerr, M. A.
J. Org. Chem. 2005, 70, 8242-8244. (c) Young, I. S.; Williams, J. L.;
Kerr, M. A. Org. Lett. 2005, 7, 953-955. (d) Ganton, M. D.; Kerr, M. A.
J. Org. Chem. 2004, 69, 8554-8557. (e) Young, I. S.; Kerr, M. A. Org.
Lett. 2004, 6, 139-141. (f) Young, I. S.; Kerr, M. A. Angew. Chem., Int.
Ed. 2003, 42, 3023-3026.
(7) The activation of amides with triflic anhydride of pyridine has been
well documented. See, for example: (a) DeRoy, P. L.; Charette, A. B. Org.
Lett. 2003, 5, 4163-4165 and references therein. (b) Mahuteau-Betzer, F.;
Ding, P.-Y.; Ghosez, L. HelV. Chim. Acta 2005, 88, 2022-2031 and
references therein.
nucleophile, open to form an aromatically stabilized anion
such as 9 or 10. Lanthanide triflates and magnesium iodide,
3778
Org. Lett., Vol. 8, No. 17, 2006

With the aim of preparing more substituted 7-azaindoles,
we investigated the preparation of the cyclopropane 17
(Scheme 6). Diazo transfer to 15 using 2-azido-3-ethylben-
Table 1. Cross-Coupling of 7-Azaindoles
Scheme 6. Preparation of Cyclopropane 1
zothiazolium tetrafluoroborate resulted in the preparation of
16 in 73% yield. A rhodium-catalyzed carbenoid insertion
to styrene resulted in the formation of the target compound
albeit in low yield.8 All attempts to employ 179 in ring-
opening chemistry met with failure.
With 7-azaindoles 14a-f in hand, we sought to utilize
the remaining trifloxy group toward the installation of
diversity via either SNAr or cross-coupling reactions. Treat-
ment of 6-trifloxy-7-azaindoles such as 14 with a variety of
nucleophiles under even forcing conditions failed to produce
substitution products. Cross-coupling reactions, however,
were far more productive. Table 1 shows a variety of cross-
coupling reactions of trifloxy-7-azaindoles 14a or 14c.
Suzuki, Sonogashira, Stille, Heck, and carbonylation reac-
tions were all successful in the installation of functionality
at position-6. This cross-coupling study is only a survey, and
one may imagine that, considering the advances in cross-
coupling chemistry, an enormous degree of molecular
diversity may be installed.
In conclusion, we have prepared a new donor-acceptor
cyclopropane and probed its reactivity in ring-opening
reactions. Triflic anhydride was effective in activating the
cyclopropane toward ring opening. The product of ring
opening was shown to be a useful synthetic intermediate for
the preparation of a variety of 7-azaindoles bearing a trifloxy
substituent at the 6-position. The triflate proved to be a useful
handle for building molecular complexity via a variety of
cross-coupling reactions.
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada and Boehringer
Ingelheim Canada for funding. We are grateful to Mr. Doug
Hairsine (University of Western Ontario) for performing MS
analyses.
Supporting Information Available: Full experimental
procedures and spectroscopic data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(8) One of the reviewers has suggested the use of a bulkier rhodium
catalyst to suppress suspected carbenoid dimerization side reactions. This
will be explored in future studies.
(9) The relative stereochemistry of 17 was not determined because the
stereochemistry would have been lost upon success in the intended reaction.
OL061379I
Org. Lett., Vol. 8, No. 17, 2006
3779