Synthesis of [5,6]-Bicyclic Heterocycles with a Ring-Junction Nitrogen Atom: Rhodium(III)-Catalyzed C?H Functionalization of Alkenyl Azoles

Article abstract of DOI:10.1002/anie.201703967

The first syntheses of privileged [5,6]-bicyclic heterocycles, with ring-junction nitrogen atoms, by transition metal catalyzed C?H functionalization of C-alkenyl azoles is disclosed. Several reactions are applied to alkenyl imidazoles, pyrazoles, and triazoles to provide products with nitrogen incorporated at different sites. Alkyne and diazoketone coupling partners give azolopyridines with various substitution patterns. In addition, 1,4,2-dioxazolone coupling partners yield azolopyrimidines. Furthermore, the mechanisms for the reactions are discussed and the utility of the developed approach is demonstrated by iterative application of C?H functionalization for the rapid synthesis of a patented drug candidate.

Full text of DOI:10.1002/anie.201703967

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201703967
German Edition: DOI: 10.1002/ange.201703967
Nitrogen Heterocycles
Synthesis of [5,6]-Bicyclic Heterocycles with a Ring-Junction Nitrogen
À
Atom: Rhodium(III)-Catalyzed C H Functionalization of Alkenyl
Azoles
Kim Søholm Halskov, Howard S. Roth, and Jonathan A. Ellman*
Abstract: The first syntheses of privileged [5,6]-bicyclic
heterocycles, with ring-junction nitrogen atoms, by transition
À
metal catalyzed C H functionalization of C-alkenyl azoles is
disclosed. Several reactions are applied to alkenyl imidazoles,
pyrazoles, and triazoles to provide products with nitrogen
incorporated at different sites. Alkyne and diazoketone
coupling partners give azolopyridines with various substitution
patterns. In addition, 1,4,2-dioxazolone coupling partners yield
azolopyrimidines. Furthermore, the mechanisms for the reac-
tions are discussed and the utility of the developed approach is
À
demonstrated by iterative application of C H functionaliza-
tion for the rapid synthesis of a patented drug candidate.
P
erhaps as a result of their electronic and shape comple-
mentarity to the adenine and guanine nucleobases, [5,6]-
bicyclic heterocycles with a ring-junction nitrogen atom have
become increasingly prominent in medicinal chemistry, as
exemplified by clinical candidates such as filgotinib, volitinib,
and dinaciclib, as well as the FDA-approved drugs zolpidem,
trazodone, ibudilast, ponatinib, and zaleplon.[1] Transition
metal catalyzed chelation-assisted aromatic and, to a lesser
2
À
extent, alkenyl, C(sp ) H activation and annulation has
enabled the convergent synthesis of diverse heterocyclic
compounds.[2] However, despite their pharmaceutical impor-
tance, only a narrow subset of [5,6]-bicyclic heterocycles with
a ring-junction nitrogen atom have been assembled by this
approach, specifically with the rhodium(III)-catalyzed annu-
lation of N-vinyl imidazoles and alkynes (Figure 1a).[3,4]
Figure 1. Synthesis of fused [5,6]-bicyclic nitrogen heterocycles by
À
rhodium(III)-catalyzed C H functionalization.
Herein we describe general methods for the rhodium(III)-
2
À
catalyzed alkenyl C(sp ) H functionalization of C-alkenyl
obtained (Figure 1b), while dioxazolones give azolopyrimi-
dines (Figure 1c).
azoles for the synthesis of fused [5,6]-bicyclic heterocycles
which incorporate from two to four nitrogen atoms (Fig-
ure 1b,c). While previous reports have demonstrated C H
functionalization of C-aryl azoles for the synthesis of tricyclic
and higher-order heterocycles,[5] to the best of our knowledge,
this study represents the first investigation of C-alkenyl azole
substrates.[6] These transformations are effective for alkenyl
imidazoles, pyrazoles, and triazoles incorporating a variety of
substitution patterns on both the alkene and the azole. With
alkyne or diazoketone coupling partners, azolopyridines are
Upon identifying optimal reaction conditions for the
annulation of C-alkenyl azoles and alkynes (see Table S1 in
the Supporting Information), the reactivities of a variety of
alkenyl azole and alkyne substrates were evaluated (Table 1).
Both alkyl- and aryl-substituted alkynes coupled with a C2
trisubstituted alkenyl imidazole in good yields to give the
imidazopyridines 3a and 3b, respectively. In addition, reac-
tion of nonsymmetric 1-phenyl-1-propyne gave the product
3c with high regioselectivity. Imidazole substrates lacking a b-
substituent on the alkene (3d and 3e) and with a b-methyl
group (3 f) were all effective coupling partners, while a C-
vinyl imidazole coupled in a more modest yield (3g). As
illustrated with 3h, substitution on the imidazole ring was also
tolerated. The reaction of alternative alkenyl diazoles was
also investigated. Methyl urocanate, with an alkene at the 4-
position of the imidazole, reacted with 3-hexyne to give 3i,
and reaction with 1-phenyl-1-propyne yielded 3j with high
regioselectivity. Moreover, an alkenyl pyrazole was a compe-
À
[*] Dr. K. S. Halskov, Dr. H. S. Roth, Prof. Dr. J. A. Ellman
Department of Chemistry, Yale University
225 Prospect St., New Haven, CT 06520 (USA)
E-mail: jonathan.ellman@yale.edu
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.201703967.
Angew. Chem. Int. Ed. 2017, 56, 9183 –9187
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9183

Angewandte
Communications
Chemie
À
À
Table 1: Rhodium(III)-catalyzed C H functionalization of alkenyl azoles
Table 2: Rhodium(III)-catalyzed C H functionalization of alkenyl azoles
with alkynes.[a,b]
with diazoketones.[a,b]
[a] Reaction conditions: 1 (0.50 mmol), 4 (1.5 equiv), 0.1m, 16 h.
[b] Yield of isolated products after chromatography on silica. [c] 608C.
[d] 808C, MeOH as solvent. THF=tetrahydrofuran.
[a] Reaction conditions: 1 (0.50 mmol), 2 (1.5 equiv), 0.1m, 16 h.
[b] Yield of isolated products after chromatography on silica. [c] 3.0 equiv
2. [d] 1008C. Cp*=C5Me5.
(3y), sulfone (3z), and phosphonate (3aa) substituents were
also synthesized in good to excellent yields.
We next investigated the coupling of alternative alkenyl
azoles with diazoketones. Pyrazole and triazole substrates
tent substrate in the reaction, giving the pyrazolopyridine 3k
upon reaction with 3-hexyne. Notably, coupling an a-chloro-
alkenyl pyrazole provided the chloro-substituted pyrazolo-
À
pyridine 3l in good yield. To our knowledge, this represents
were competent C H partners, thus giving the products 3ab
2
À
the first example of group IX metal-catalyzed C(sp ) H
and 3ac, respectively (Table 2). The compound 3ac was
isolated as a single isomer, which was characterized by X-ray
crystallography to be distinct from that previously obtained
for alkyne coupling (3m–r; Table 1).[11]
haloalkene annulation,[7] and is of utility because chloro
substituents on fused [5,6]-bicyclic nitrogen heterocycle
frameworks are present in clinical candidates[8] and serve as
versatile handles for further elaboration. Finally, alkenyl
1,2,4-triazoles gave the triazolopyridines 3m–r with a greater
than 90:10 cyclization regioselectivity to provide the depicted
triazolo[4,3-a]pyridines, as confirmed by X-ray crystallogra-
phy of 3m.[9]
In addition to the diazoketones substituted with an
electron-withdrawing group, when CsOAc was employed as
a stoichiometric additive, the phenyl-substituted diazoketone
4 f also coupled to give the products 3ad and 3ae (Figure 2a).
It can be challenging to develop reactions in which both
acceptor/acceptor and donor/acceptor diazo compounds
react.[12] To the best of our knowledge, this report represents
We next investigated the reaction of alkenyl azoles with
the diazoketones
4 (Table 2). These compounds have
À
emerged in the past several years as competent coupling
partners for heterocycle synthesis by rhodium(III)-catalyzed
the first example of rhodium(III)-catalyzed C H addition
into both classes of diazo compounds. Moreover, it is
significant that the phenyl diazoketone exclusively provided
3ae, while the regioisomeric 3c preferentially formed in the
alkyne coupling, thus demonstrating the complementary
nature of these approaches (Figure 2b).
[10]
À
C H functionalization.
Unlike for alkyne coupling, the
reaction of alkenyl azoles with diazoketones is redox-neutral
and requires no stoichiometric oxidant. For C-alkenyl imida-
zoles, simple heating at 408C in THF was effective (see
Table S2 for optimization). The reaction of ethyl diazoace-
toacetate (4a) with alkenyl imidazoles gave the imidazopyr-
idines 3s–v with ester substituents. A substrate with a methyl
substituent on the imidazole ring underwent annulation with
high selectivity to give the methyl-substituted imidazopyr-
idine 3w. It was also possible to vary the identity of the
diazoketone substrate. A phenyl ketone coupled efficiently to
afford 3x, and imidazopyridine products containing ketone
It is noteworthy that divergent isomer formation with
alkyne and diazoketone coupling partners was also observed
for the alkenyl triazole substrates. Reaction with alkynes gave
the triazolo[4,3-a]pyridines 3m–r (Table 1), while the reaction
with a diazoketone yielded the triazolo[1,5-a]pyridine 3ac
(Table 2). The different product outcome can be understood
by the disparate mechanisms for product formation
(Scheme 1). Alkyne coupling begins with the concerted
9184
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 9183 –9187

Angewandte
Communications
Chemie
In contrast, reaction of a diazoketone with A results in
loss of nitrogen to give the intermediate C (Scheme 1b).
Protonolysis releases the intermediate D and regenerates the
catalyst. Cyclodehydration of D then provides the triazolo-
pyridine 3ac. The high regioselectivity for cyclization of D
likely occurs to avoid unfavorable steric interactions between
the triazole methyl group and the methyl ketone.
In addition to azolopyridines, we sought to develop
a method for the synthesis of azolopyrimidines, which contain
an additional nitrogen atom within the six-membered ring.
We investigated the reaction of alkenyl azoles with the
dioxazolones
5
(Table 3), which have previously been
À
Table 3: Rhodium(III)-catalyzed C H functionalization of alkenyl azoles
with dioxazolones.[a,b]
Figure 2. Reaction of donor/acceptor diazoketone allowing divergent
product formation. [a] After 16 h, added AcOH (5 mL) and stirred at
1008C for 24 h.
[a] Reaction conditions: 1 (0.50 mmol), 5 (1.5 equiv), 0.1m, 16 h.
[b] Yield of isolated products after chromatography on silica. [c] Second
step at 1008C. [d] Addition and cyclization performed at 808C with AcOH
as solvent.
[14]
À
employed in rhodium(III)-catalyzed C H amidation, and
found that heating at 808C in 1,4-dioxane for five hours was
optimal for the synthesis of the enamides 6. Treatment with
acetic acid facilitated cyclization of the enamides to the
desired azolopyrimidines 7. Reaction of alkenyl imidazoles
with a methyl-substituted dioxazolone gave the imidazopyr-
imidines 7a–d in moderate to high yields. Annulation of
a substrate with a methyl group on the imidazole ring
afforded the imidazopyrimidine 7e with complete selectivity
for the methyl group at the 2-position. Reactions with
a phenyl-substituted dioxazolone provided products 7 f and
7g. For these entries, a higher temperature was necessary in
the second step, as the intermediate enamide cyclizes less
efficiently. Finally, the alkenyl 1,2,4-triazole substrate coupled
to afford the triazolopyrimidine 7h, but required that both the
Scheme 1. a) Mechanism of rhodium(III)-catalyzed coupling of alkenyl
azoles with alkynes. b) Mechanism of rhodium(III)-catalyzed coupling
of alkenyl azoles with diazoketones.
metalation/deprotonation of alkenyl triazole to form the
rhodacycle A (Scheme 1a).[13] Migratory insertion of 3-
hexyne gives the intermediate B, and reductive elimination
yields 3m. Oxidation of rhodium(I) then regenerates the
rhodium(III) catalyst. The isomer formed is determined by
selective rhodium coordination at N4 of the 1,2,4-triazole
substrate, presumably because the electron density is greatest
at this site. Reductive elimination then gives the triazolo[4,3-
a]pyridine.
À
C H bond addition and cyclization be performed in acetic
Angew. Chem. Int. Ed. 2017, 56, 9183 –9187
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9185

Angewandte
Communications
Chemie
acid. This one-step protocol was successful for other substrate
Conflict of interest
pairings, but was lower yielding for the imidazopyrimidines.
Selective formation of 7h occurs during the cyclization step,
likely because of minimization of unfavorable steric inter-
actions in analogy to the synthesis of triazolopyridines from
diazoketones (Scheme 1b).
The authors declare no conflict of interest.
À
Keywords: C H activation ·cyclization ·
homogeneous catalysis ·nitrogen heterocycles ·rhodium
The utility of the developed methodology was demon-
How to cite: Angew. Chem. Int. Ed. 2017, 56, 9183–9187
Angew. Chem. 2017, 129, 9311–9315
À
strated by the modular and iterative application of C H
functionalization methods for the rapid synthesis of 12,
a potential drug candidate for central nervous system (CNS)
disorders (Scheme 2).[15] An initial C H alkylation of quin-
À
[1] Typing the names of these drugs and drug candidates into
PubChem provides the compound structure, bioactivity, full list
of literature, and access to ongoing clinical trials, applications,
and usage.
[2] For relevant reviews, see: a) T. Satoh, M. Miura, Chem. Eur. J.
2010, 16, 11212; b) G. Song, F. Wang, X. Li, Chem. Soc. Rev.
2012, 41, 3651; c) J. Yamaguchi, A. D. Yamaguchi, K. Itami,
Angew. Chem. Int. Ed. 2012, 51, 8960; Angew. Chem. 2012, 124,
9092; d) T. Mesganaw, J. A. Ellman, Org. Process Res. Dev. 2014,
18, 1097; e) M. Gulías, J. L. MascareÇas, Angew. Chem. Int. Ed.
2016, 55, 11000; Angew. Chem. 2016, 128, 11164.
[3] a) J.-R. Huang, Q.-R. Zhang, C.-H. Qu, X.-H. Sun, L. Dong, Y.-
C. Chen, Org. Lett. 2013, 15, 1878; b) L. Dong, J.-R. Huang, C.-
H. Qu, Q.-R. Zhang, W. Zhang, B. Han, C. Peng, Org. Biomol.
Chem. 2013, 11, 6142; c) For imidazo[1,2-a]pyridinium salts, see:
R. Thenarukandiyil, H. Thrikkykkal, J. Choudhury, Organo-
metallics 2016, 35, 3007.
À
[4] Methods have been developed for the direct C H arylation of
pre-assembled azolopyridines and azopyrimidines. For reviews
that provide leading references, see: a) I. V. Seregin, V.
Gevorgyan, Chem. Soc. Rev. 2007, 36, 1173; b) R. Rossi, M.
Lessi, C. Manzini, G. Marianetti, F. Bellina, Synthesis 2016, 3821.
[5] a) N. Umeda, H. Tsurugi, T. Satoh, M. Miura, Angew. Chem. Int.
Ed. 2008, 47, 4019; Angew. Chem. 2008, 120, 4083; b) K.
Morimoto, K. Hirano, T. Satoh, M. Miura, Org. Lett. 2010, 12,
2068; c) X. Li, M. Zhao, J. Org. Chem. 2011, 76, 8530; d) W. B.
Ma, K. Graczyk, L. Ackermann, Org. Lett. 2012, 14, 6318; e) N.
Kavitha, G. Sukumar, V. P. Kumar, P. S. Mainkar, S. Chandra-
sekhar, Tetrahedron Lett. 2013, 54, 4198; r) R. Wang, J. R. Falck,
J. Organomet. Chem. 2014, 759, 33; g) L. Zheng, R. Hua, J. Org.
Chem. 2014, 79, 3930; h) A. G. Algarra, W. B. Cross, D. L.
Davies, Q. Khamker, S. A. Macgregor, C. L. McMullin, K. Singh,
J. Org. Chem. 2014, 79, 1954.
Scheme 2. Synthesis of potential drug candidate 12 utilizing developed
methodology.
oline by rhodium(I) catalysis[16] provided 8, which upon
reaction with hydrazine gave the hydrazide 9 in high yield.
Condensation of 9 with the imidate 10 afforded the alkenyl-
substituted triazole 11. Final rhodium(III)-catalyzed annula-
tion with the dioxazolone 5a provided the target compound
12 in 78% yield. By the application of two rhodium-catalyzed
À
C H functionalization steps, this target could be synthesized
À
[6] For the synthesis of pyridinylamines by C H functionalization
in only four steps and in a good overall yield.
and ring fragmentation of alkenyl oxadiazolones, see: X. Yu, K.
Chen, Q. Wang, S. Guo, S. Zha, J. Zhu, Angew. Chem. Int. Ed.
2017, 56, 5222; Angew. Chem. 2017, 129, 5306.
In summary, we have developed novel methods for the
synthesis of fused [5,6]-bicyclic heterocycles with a ring-
À
junction nitrogen atom by rhodium(III)-catalyzed C H
À
[7] For C H halogenation of an alkenyl bromide, see: N. Kuhl, N.
functionalization of C-alkenyl azoles. The three different
classes of coupling partners that were employed allow the
construction of diverse heterocycles of biological relevance.
Moreover, the applicability to drug discovery was illustrated
Schrçder, F. Glorius, Org. Lett. 2013, 15, 3860.
[8] See PubChem Compound Identifier numbers 54897 (alpidem)
and 68234908.
[9] CCDC 1509748 (3m) contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre. See
SI for additional details.
[10] Selected examples: a) W.-W. Chan, S.-F. Lo, Z. Zhou, W.-Y. Yu,
J. Am. Chem. Soc. 2012, 134, 13565; b) Z. Shi, D. C. Koester, M.
Boultadakis-Arapnis, F. Glorius, J. Am. Chem. Soc. 2013, 135,
12204; c) L. Shi, K. Yu, B. Wang, Chem. Commun. 2015, 51,
17277; d) Y. Wu, P. Sun, K. Zhang, T. Yang, H. Yao, A. Lin, J.
Org. Chem. 2016, 81, 2166; e) X. G. Li, M. Sun, Q. Jin, K. Liu,
P. N. Liu, J. Org. Chem. 2016, 81, 3901; f) X. Chen, X. Hu, S. Bai,
Y. Deng, H. Jiang, W. Zeng, Org. Lett. 2016, 18, 192.
[11] CCDC 1509749 (3ac) contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre.
À
by modular and iterative application of C H functionaliza-
tion methods for the preparation of a patented compound
with CNS activity.
Acknowledgements
This work was supported by the NIH (R35GM122473). We
gratefully acknowledge Dr. Brandon Mercado (Yale Univer-
sity) for solving the crystal structures of 3m and 3ac. KSH
thanks the Villum Foundation for post-doctoral funding
(VKR023371).
9186
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 9183 –9187

Angewandte
Communications
Chemie
[12] T. K. Hyster, K. E. Ruhl, T. Rovis, J. Am. Chem. Soc. 2013, 135,
Org. Lett. 2016, 18, 2062; e) Q. Wang, F. Wang, X. Yang, X.
5364.
Zhou, X. Li, Org. Lett. 2016, 18, 6144.
[13] a) L. Li, W. W. Brennessel, W. D. Jones, J. Am. Chem. Soc. 2008,
130, 12414; b) A. P. Walsh, W. D. Jones, Organometallics 2015,
34, 3400.
[14] For a review, see: a) Y. Park, Y. Kim, S. Chang, Chem. Rev. 2017,
117, https://doi.org/10.1021/acs.chemrev.6b00644; For selected
examples: b) Y. Park, K. T. Park, J. G. Kim, S. Chang, J. Am.
Chem. Soc. 2015, 137, 4534; c) H. Wang, G. D. Tang, X. W. Li,
Angew. Chem. Int. Ed. 2015, 54, 13049; Angew. Chem. 2015, 127,
13241; d) J. Wang, S. Zha, K. Chen, F. Zhang, C. Song, J. Zhu,
[15] J. E. Campbell, M. C. Hewitt, P. Jones, L. Xie, PCT Int. Appl.
2011, 150156.
[16] G. Tran, K. D. Hesp, V. Mascitti, J. E. Ellman, Angew. Chem. Int.
Ed. 2017, 56, 5899; Angew. Chem. 2017, 129, 5993.
Manuscript received: April 17, 2017
Accepted manuscript online: June 6, 2017
Version of record online: July 5, 2017
Angew. Chem. Int. Ed. 2017, 56, 9183 –9187
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9187