which provide considerable potential for further manipulation
of this heterocyclic ring system. In addition, we establish a
new illustration of the directed remote metalation (DreM)
concept3 on the Suzuki-derived heterobiaryl 7 leading to the
new heterocyclic 2,5,11c-triazadibenzo[cd,f]azulen-7(6H)-
one framework 8. These results constitute the first investiga-
tion of carbanionic chemistry on the imidazo[1,5-a]pyrazine
ring system which is of current interest in the development
of new antimicrobial, cardiovascular, and antitumor drugs4
and which also portends broader significance for synthetic
studies for this and related heterocyclic systems.
Table 1. Synthesis of 3-Substituted
8-Chloroimidazo[1,5-a]pyrazines 2a-2g
EX
R1
product (%)a
1
2
3
4
5
6
7
8
MeI
DMF
Me
CHO
4-MeOC6H4CHOH
MesCHOH
CO2Me
Et2NCO
H
I
2a (70)b
2b (72)
2c (74)
2d (83)
2e (86)c
2f (65)
n.r.d
4-MeOC6H4C(O)H
Mesitaldehyde
CO2
ClCONEt2
ClSiMe3
I2
2g (78%)
a Yield of isolated material. b Purified by supercritical fluid chroma-
tography to remove bis-Me byproduct. c Isolated as the methyl ester due to
purification difficulties. d No reaction as confirmed by 1H NMR of the
recovered material.
Figure 1. The direct Heck and Suzuki-Miyaura reaction sites (A)
and observed metalation sites (B, C) for imidazo[1,5-a]pyrazines.
introduction of various electrophiles. Compound 1 was
treated with 2.3 equiv of nBuLi (optimal conditions)
followed by quench with 2.5 equiv of several electrophiles
to give C3/C5-disubstituted imidazopyrazines 4a-4d in
moderate to excellent yields (Table 2). The formation of
product with high deuterium incorporation (entry 1)
constitutes evidence for the intermediacy of the dilithiated
species 3.7 Consistent with the observation of Table 1,
entry 7, quench with 2.5 equiv of TMSCl gave only the
C5-silylated product 4d in moderate yield, presumably
owing to the derived C3-TMS derivative instability.
Nevertheless, this result remains of value since it poten-
tially offers access to other C5 analogues, either directly
through ipso-desilylation8 or through Hiyama type cross-
coupling chemistries via C5 siloxanes or fluorosilanes.9
To initiate this study, the readily accessible5 8-chloroimi-
dazo[1,5-a]pyrazine 1 was subjected to reaction with nBuLi
(1 equiv) and quenched with various electrophiles to afford
exclusively C3-substitution products 2a-2g (Table 1). Reac-
tions with carbon electrophiles at three oxidation states
(entries 1-6) and iodine (entry 8) proceeded uneventfully
whereas that with TMSCl failed, an observation which may
be due to product instability, related to the known lability
of the imidazole C2-Si bond.6
Efforts then turned to investigating the potential for
dimetalation of this heterocyclic core and for the sequential
(3) Whisler, M. C.; MacNeil, S.; Snieckus, V.; Beak, P. Angew. Chem.,
Int. Ed. 2004, 43, 2206.
(4) Examples of such biological applications are: IGF-IR inhibitors, for
example: (a) Mulvihill, M. J.; Ji, Q.-S.; Coate, H. R.; Cooke, A.; Dong, H.;
Feng, L.; Foreman, K.; Rosenfeld-Franklin, M.; Honda, A.; Mak, G.;
Mulvihill, K. M.; Nigro, A. I.; O’Connor, M.; Pirrit, C.; Steinig, A. G.;
Siu, K.; Stolz, K. M.; Sun, Y.; Tavares, P. A. R.; Yao, Y.; Gibson, N. W.
Bioorg. Med. Chem. 2008, 16, 1359. c-SRC inhibition and treatment of acute
ischemic stroke: (b) Mukaiyama, H.; Nishimura, T.; Kobayashi, S.; Ozawa,
T.; Kamada, N.; Komatsu, Y.; Kikuchi, S.; Oonota, H.; Kusama, H. Bioorg.
Med. Chem. 2007, 15, 868. Corticotropin releasing hormone receptor
ligands: (c) Hartz, R. A.; Gilligan, P. J.; Nanda, K. K.; Tebben, A. J.;
Fitzgerald, L. W.; Miller, K. Bioorg. Med. Chem. Lett. 2002, 12, 291.
(5) Initial syntheses of imidazo[1,5-a]pyrazines: (a) Abushanab, E.;
Bindra, A. P.; Goodman, L.; Peterson, H., Jr. J. Org. Chem. 1973, 38, 2049.
(b) Abushanab, E.; Bindra, A. P.; Lee, D. Y.; Goodman, L. J. Heterocycl.
Chem. 1975, 12, 211. (c) Abushanab, E.; Bindra, A. P.; Goodman, L.
J. Heterocycl. Chem. 1975, 12, 207. (d) Abushanab, E.; Bindra, A. P.; Lee,
D. Y.; Goodman, L. J. Org. Chem. 1975, 40, 3373. (e) Abushanab, E.;
Lee, D. Y.; Goodman, L. J. Org. Chem. 1975, 40, 3376. (f) Abushanab,
E.; Lee, D. Y.; Goodman, L. J. Org. Chem. 1975, 40, 33792. Syntheses of
more highly substituted imidazo[1,5-a]pyrazines: (g) Albert, A.; Ohta, K.
J. Chem. Soc. (C) 1970, 11, 1540. (h) Davey, D. D. J. Org. Chem. 1987,
52, 4379. (I) Chattopadhyay, G.; Ray, S. J. Chem. Res. Synop. 1992, 5,
170. (j) Marchand, E.; Morel, G. Tetrahedron Lett. 1993, 34, 2319. (k)
Trcˇek, T.; Meden, A.; Vercˇek, B. Synlett 2000, 10, 1458. (l) Eltsov, O. S.;
Mokrushin, V. S.; Tkachev, A. V. Russ. Chem. Bull., Int. Ed. 2004, 53,
2293. (m) Trcˇek, T.; Vercˇek, B. Synthesis 2006, 20, 3437. (n) Mulvihill,
K. M.; Castelhano, A. L. US Patent 2007/0129547 A1.
In view of the selective C3 over C5 kinetic anion
reactivity established by the results of Table 1, we tested
the thereby expected greater nucleophilic reactivity at the
C5 site of the dianion 3.10 Thus, treatment of 1 with 2
equiv of nBuLi11 followed by 1 equiv of iodine or CD3OD
initially provided evidence of selectivity; however, sub-
sequent results were inexplicably inconsistent giving ratios
between 4:1 and 1:8 of C5:C3 products within the same
batch of reactions, and despite close control of the various
reaction variables.12
With the results of studies on compound 1 in hand, we
turned our attention to the metalation of the equally
accessible5 8-chloro-3-methylimidazo[1,5-a]pyrazine 2a.
In contrast to expectation,13 treatment of 2a with 1.1 equiv
(6) The N-protected imidazole C2-TMS system is moderately unstable
towards nucleophilic attack at silicon even when generated in situ, and its
isolation can be highly problematic due to hydrolysis: Carpenter, A. J.;
Chadwick, D. J. Tetrahedron 1986, 42, 2351.
Org. Lett., Vol. 11, No. 22, 2009
5119