Gold-catalyzed [4+3]- and [4+2]-annulations of 3-en-1-ynamides with isoxazoles via novel 6π-electrocyclizations of 3-azahepta trienyl cations

Abstract

New gold-catalyzed [4+3]-annulations of 3-en-1-ynamides with isoxazoles afford 4H-azepines efficiently; this process involves 6π electrocyclizations of gold-stabilized 3-azaheptatrienyl cations. In the presence of Zn(OTf)₂, the resulting 4H-azepines undergo skeletal rearrangement to furnish substituted pyridine derivatives. We subsequently develop new catalytic [4+2]-annulations between the same 3-en-1-ynamides and isoxazoles to deliver substituted pyridine products using Au(I)/Zn(II) catalysts. This work reports the first success of the 6π electrocyclizations of heptatrienyl cations that are unprecedented in literature reports.

---

Introduction

Electrocyclizations of acyclic conjugated π-motifs are powerful tools to access five-, six- and seven-membered carbocycles;¹ prominent examples include Nazarov cyclizations of pentadienyl cations² and 6π electrocyclizations of trienes,³ which have found widespread applications in organic synthesis.

In the context of seven-carbon π-motifs, heptatrienyl anions I undergo facile 8π electrocyclizations via rapid interconversions among various anion configurations (Scheme 1).⁴ In contrast, heptatrienyl cations III⁵ exclusively undergo Nazarov reactions because of the difficulties of forming all σ-cis configured cations V that have a high energy state.^5b 1-Aza- and 1-oxaheptatrienyl cations⁶ were also reported to follow Nazarov cyclizations. The realization of a 6π electrocyclization of conjugated seven-membered cations is formidable but challenging. This work reveals the first success of such seven-membered cyclizations of gold-stabilized 3-azaheptatrienyl cations V′ to form azacyclic products 3–4via a new C–C bond formation.

Scheme 1 Electrocyclizations of conjugated π-motifs.

The advent of gold catalysis has inspired new annulations between alkynes and poor nucleophiles.⁷ N–O containing nucleophiles serve as useful building blocks to construct valuable azacyclic frameworks.⁷ Ye and Hashmi reported interesting [3+2]-annulations of isoxazoles or benzisoxazoles with electron-rich ynamides, yielding substituted pyrrole derivatives through aza-Nazarov cyclizations of the key intermediate [eqn (1)].^7,8 These [3+2]-annulations were extensively expanded to other N–O heterocycles including benzisoxazoles, 1,2,4-oxadiazoles, 1,4,2-dioxazoles and 4,5-dihydro-1,2,4-oxadiazoles, yielding additional five-membered azacycles as depicted in [eqn (1)].⁹ Here, we report two distinct [4+3]- and [4+2]-annulations between 3-en-1-ynamides and isoxazoles using varied catalysts. An Au(I) catalyst alone delivers 4H-azepines 3–4 through 6π electrocyclizations of intermediates V′ [eqn (2)] whereas a combined action of Au(I)/Zn(II) on the same reactants furnishes highly functionalized pyridines 5 [eqn (3)]. With our convenient synthesis, the synthetic utility of new 4H-azepines 3–4 is also reported.¹⁰

Results and discussion

We examined the reactions of 3-methyl-3-en-1-ynamide 1a with 3,5-dimethylisoxazole 2a using various gold catalysts. Heating this mixture (1a/2a = 1 : 2 ratio) in hot DCE with 5 mol% LAuCl/AgNTf₂ [L = p(t-Bu)₂(o-biphenyl) and IPr] afforded a [4+3]-annulation product, 4H-azepine 3a, in 64% and 75% yields respectively (Table 1, entries 1–2). Under these conditions, a low loading (1.2 equiv.) of 3,5-dimethylisoxazole 2a gave 3a in a decreased yield, ca. 62% (entry 3). With a 10 mol% catalyst, IPrAuCl/AgNTf₂ gave a clean reaction, yielding desired 3a up to 91% (entry 4). We tested other phosphine ligands such as PPh₃ and P(OPh)₃, yielding desired 3a in satisfactory yields (78–81%, entries 5–6). Other counter anions such as OTf⁻ and SbF₆⁻ were also effective in producing 3a in 85–88% yields (entries 7–8). AgNTf₂ alone was not active at all (entry 9).

Table 1 [4+3]-Annulations over various gold catalysts

Entry	Catalyst [mol%]	x	Time [h]	Yield^b [%]
				1a	3a	1a–H′/1a–H′′
a [1a] = 0.15 M. b Product yields are reported after separation from a silica column. c L = p(t-Bu)2(o-biphenyl). d IPr = 1,3-bis(diisopropylphenyl)-imidazol-2-ylidene. Ms = methanesulfonyl, DCE = 1,2-dichloroethane, and Tf = trifluoromethanesulfonyl.
1^c	LAuCl/AgNTf2 [5]	2	3	20	64	—
2^d	IPrAuCl/AgNTf2 [5]	2	7	12	75	7 [2.5 : 1]
3	IPrAuCl/AgNTf2 [5]	1.2	7	23	62	5 [1 : 1]
4	IPrAuCl/AgNTf 2 [10]	2	3	—	91	Trace
5	PPh3AuCl/AgNTf2 [10]	2	3.5	—	81	5 [1.25 : 1]
6	[PhO]3PAuCl/AgNTf2 [10]	2	3.5	—	78	13 [1.1 : 1]
7	IPrAuCl/AgSbF6 [10]	2	2.5	—	85	6 [1.4 : 1]
8	IPrAuCl/AgOTf [10]	2	2	—	88	Trace
9	AgNTf2 [10]	2	15	33	—	11

---

Suitable substituents of 3-en-1-ynamides 1 are crucial to achieve 6π cyclizations of 3-azaheptatrienyl cations V′ [eqn (2)]. We tested the reactions on 3-en-1-ynes 1b–1m bearing a C(3)-substituent to circumvent aza-Nazarov cyclizations as reported in Ye's work.⁷ Herein, only entries 9 and 10 showed the presence of 3-azanorcaradienes 3′. We examined these [4+3]-annulations on 3-methyl-3-en-1-ynamides 1b–1e bearing various sulfonamides NTsR⁴ (R⁴ = Me, cyclopropyl, benzyl and N(n-C₄H₉)(−SO₂Bu)), affording the desired 4H-azepines 3b–3e in high yields (84–90%, Table 2, entries 1–4). Nevertheless, this new annulation becomes less efficient for 3-en-1-ynamide 1f bearing an oxazolidin-2-one to yield product 3f in 64% yield (entry 5).

Table 2 [4+3]-Annulations with various 3-en-1-ynamides

a [1] = 0.15 M. b Product yields are reported after separation from a silica column. EWG = electron withdrawing group.

---

We altered the C(3)-substituents as in substrates 1g–1i; their resulting products 3g–3h (R¹ = isopropyl and cyclopropyl) were obtained in 74–79%, and 3i (R¹ = Ph) with only 58% yield (entries 6–8). Notably, when a long n-butyl group was present as in species 1j and 1k, their corresponding reactions afforded compounds 3j/3j′ = 5/1 and 3k/3k′ = 11.1 : 1, respectively, in 55% and 68% yields (entries 9–10). For E-configured trisubstituted 3-en-1-yne 1l (R¹ = Me, R² = Ph and R³ = H), 4H-azepine 3l and pyrrole 6l were obtained in equal proportions (entry 11). When a cyclohexenyl group was present for alkene as in species 1m, pyrrole product 6m was dominant over azepine 3m (entry 12). Accordingly, preferable 3-en-1-ynes comprise a small R² or R³ substituent whereas R¹ must be substituted. Herein, the structures of 4H-azepines 3b and 3l, and pyrrole species 6m were confirmed with X-ray diffraction.¹¹

Isoxazoles of a wide scope are compatible with these [4+3]-annulations, as depicted in Table 3. The reaction of unsubstituted isoxazole 2b with model 3-en-1-ynamide 1b afforded the desired 4H-azepine 4a in 84% yield, together with pyrrole 7a′ in only 8% yield (entry 1). Mono-substituted 3-methyl or 5-methyl isoxazoles 2c and 2d are also suitable for these annulations to afford compounds 4b and 4c in 75% and 87% yields, respectively (entries 2–3). We prepared additional 3,5-disubstituted isoxazoles 2e–2i with R¹ = alkyl and phenyl, and R² = alkyl; their annulations proceed smoothly to produce desired 4d–4h in 69–85% yields (entries 4–8). For di-substituted isoxazoles 2j and 2k bearing R² = Ph, 4H-azepines 4i and 4j were obtained in 61% and 71% yields respectively, together with their rearrangement products 5i and 5j in 15–30% yields (entries 9–10). Compounds 4a and 5i were characterized by X-ray diffraction.¹¹

Table 3 [4+3]-Annulations with various isoxazoles

Entry	(R^1, R^2)	2	Time [h]	Yield [%]	4
a [1b] = 0.15 M. b Product yields are reported after separation from a silica column.
(1)	H, H	2b	4	84	4a (X-ray)
				8	7a′
(2)	H, Me	2d	3	75	4b
(3)	Me, H	2c	3	87	4c
(4)	Et, Et	2e	6	85	4d
(5)	n-Bu, n-Bu	2f	7	81	4e
(6)	Me, n-Bu	2g	3	82	4f
(7)	n-Bu, c-Pr	2h	2	77	4g
(8)	Ph, n-Bu	2i	4	69	4h
(9)	Ph, Ph	2j	6.5	61	4i
				30	5i (X-ray)
(10)	Me, Ph	2k	4	71	4j
				15	5j

---

Our convenient synthesis of 4H-azepines provides new synthetic utilities; several new functionalizations are depicted in Scheme 2. NaBH₄-reduction of species 3b delivered an alcohol derivative 7a in 84% yield. Selective hydrogenation of the same species afforded 2-aza-1,3-dien-5-one 7b in 71% yield. A final treatment of 4H-azepine 3b with NBS in acetone afforded compound 7c, of which the molecular structure was determined by ¹H NOE spectra.

Scheme 2 New functionalization of 4H-azepines.

The Lewis-catalyzed rearrangement of 4H-azepines 3–4 to substituted pyridines 5 [eqn (3)] is unprecedented in 4H-azepine chemistry.¹⁰ We undertook such novel [4+2]-annulations between 3-en-1-ynamides 1 and isoxazoles 2 using Au(I)/Zn(II) in a relay series, as depicted in Table 4. In the reactions of various 3-substituted 3-en-1-ynamides 1 (R¹ = methyl, n-butyl, cyclopropyl and isopropyl) with 3,5-dimethylisoxazole 2a, substituted pyridines 5a–5d were obtained in satisfactory yields (51–73%, entries 1–4). In entry 1, if the reaction was performed with combined Au(I)/Zn(II) catalysts in a non-relay operation, compounds 5a and 3b were isolated in 35% and 28% yields respectively. For 3-en-1-ynamide 1a bearing a NMs(n-butyl), the corresponding product 5e was obtained in 63% yield (entry 5). We tested the reactions on 3,5-disubstituted isoxazoles 2e–2f & 2h bearing all alkyl substituents, producing desired 5f–5h in good yields (69–78%, entries 6–8). For such disubstituted isoxazoles bearing R⁴ = Ph, the reactions afforded the desired pyridine derivatives 5i and 5j in 75–80% yields (entries 9–10). The molecular structures of compounds 5a and 5i were characterized by X-ray diffraction.¹¹

Table 4 [4+2]-Annulations between 3-en-1-ynamides and isoxazoles

Entry	(R^1, R^2, EWG)	1	(R^3, R^4)	2	Time [h]	Yield [%]	5
a [1] = 0.15 M. b Product yields are reported after separation from a silica column. c The value in parentheses is reported using a mixture of IPrAuCl/AgNTf2 (10 mol%) and Zn(OTf)2 (20 mol%) in hot DCE (70 °C, 48 h); 3b was also isolated in 28% yield.
(1)	Me, Me, Ts	1b	Me, Me	2a	19	73 (35)^c	5a (X-ray)
(2)	n-Bu, Me, Ts	1k	Me, Me	2a	33	64	5b
(3)	c-Pr, Me, Ts	1h	Me, Me	2a	20	56	5c
(4)	i-Pr, Me, Ts	1g	Me, Me	2a	15	51	5d
(5)	Me, n-Bu, Ms	1a	Me, Me	2a	28	63	5e
(6)	Me, Me, Ts	1b	n-Bu, n-Bu	2f	19	78	5f
(7)	Me, Me, Ts	1b	Et, Et	2e	16	69	5g
(8)	Me, Me, Ts	1b	nBu, c-Pr	2h	20	75	5h
(9)	Me, Me, Ts	1b	Ph, Ph	2j	24	80	5i (X-ray)
(10)	Me, Me, Ts	1b	Me, Ph	2k	30	75	5j

---

Scheme 3 rationalizes the crucial roles of substituents of 3-en-1-ynamides in the chemoselectivity that relies on two conformational structures DversusD′. The N-attack of isoxazole at gold-π-ynamide A is expected to form a gold-carbene D′, which can be visualized as a gold-stabilized cycloheptatrienyl cation. Conformation D is favorable with R = H, which prefers aza-Nazarov reactions.¹² When a C(3)-substituent is present (R = alkyl and aryl), all σ-cis configured species D′ are the preferable geometry to induce novel 6π electrocyclizations. This ring closure is expected to proceed through an attack of enamide at the alkenylgold moiety that is also visualized as a gold-stabilized cation. Additional C(4)-substituents render the formation of cations D′ difficult, thus yielding pyrrole 6 as byproducts. A loss of an acidic proton from seven-membered cations E is expected to yield azepines 3–4. 4H-Azepines 3–4 bear an enone conjugated with a triene; this extensive conjugation is very stable to impede a 6π electrocyclization of their triene moieties unless a Lewis acid is present. Zn(OTf)₂ likely coordinates with the carbonyl of 4H-azepine 3 to generate a 2-azapentadienyl cation F bearing a zinc enolate, further enabling an intramolecular cyclization to generate species G. A 1,2-acyl shift¹⁴ of species G delivers the observed product 5.¹³

Scheme 3 A plausible reaction mechanism.

Conclusions

In summary, this work describes new gold-catalyzed [4+3] annulations¹⁵ of 3-substituted 3-en-1-ynamides with isoxazoles to form 4H-azepines. A relay catalysis is also developed with Au(I)/Zn(II) catalysts to achieve [4+2] annulations from the same reactants. The mechanisms of gold-catalyzed [4+3] annulations involve unprecedented 6π electrocyclizations of 3-azacycloheptatrienyl cations to form 4H-azepines 3–4 efficiently. Control experiments confirm that 4H-azepines 3–4 are catalyzed by Zn(OTf)₂ to undergo new rearrangement reactions to form substituted pyridine derivatives.

Conflicts of interest

The authors declare no conflict of interest.

Acknowledgements

The authors thank the Ministry of Science and Technology and the Ministry of Education, Taiwan, for supporting this work.

Notes and references

1. (a) S. Sankararaman, in Pericyclic Reactions-A Textbook. Reactions, Applications and Theory, Wiley-VCH, New York, 2005, ch. 6, pp. 393–398; (b) C. M. Beaudry, J. P. Malerich and D. Trauner, Chem. Rev., 2005, 105, 4757; (c) P. V. R. Schleyer, J. I. Wu, F. P. Cossio and I. Fernandez, Chem. Soc. Rev., 2014, 43, 4909; (d) M. Bian, L. Li and H. Ding, Synthesis, 2017, 49, 4383; (e) E. C. Taylor and I. J. Turchi, Chem. Rev., 1979, 79, 181; (f) F. D. Proft, P. K. Chattaraj, P. W. Ayers, M. T. Sucarrat, M. Elango, V. Subramanian, S. Giri and P. Geerlings, J. Chem. Theory Comput., 2008, 4, 595; (g) N. Jana and G. Driver, Org. Biomol. Chem., 2015, 13, 9720.
2. (a) W. T. Spencer III, T. Vaidya and A. J. Frontier, Eur. J. Org. Chem., 2013, 3621; (b) R. L. Davis and D. L. Tantillo, Curr. Org. Chem., 2010, 14, 1561; (c) T. N. Grant, C. J. Rieder and F. G. West, Chem. Commun., 2009, 5676; (d) M. J. Riveira, L. A. Marsili and M. P. Mischne, Org. Biomol. Chem., 2017, 15, 9255; (e) N. Shimada, C. Stewart and M. A. Tius, Tetrahedron, 2011, 67, 5851.
3. (a) V. A. Guner, K. N. Houk and I. W. Davis, J. Org. Chem., 2004, 69, 8024; (b) R. V. Essen, D. Frank, H. W. Sunnemann, D. Vidovic, J. Magull and A. D. Meijere, Chem.–Eur. J., 2005, 11, 6583; (c) N. A. Magomedov, P. L. Ruggiero and Y. Tang, J. Am. Chem. Soc., 2004, 126, 1624; (d) E. N. Marvell, G. Caple, B. Schatz and W. Pippin, Tetrahedron, 1973, 29, 3781; (e) J. R. Otero, J. Org. Chem., 1999, 64, 6842; (f) P. E. Tessier, N. Nguyen, M. D. Clay and A. G. Fallis, Org. Lett., 2005, 7, 767; (g) C. L. Benson and F. G. West, Org. Lett., 2007, 9, 2545; (h) G. A. Barcan, A. Patel, K. N. Houk and O. Kwon, Org. Lett., 2012, 14, 5388.
4. (a) R. B. Bates, W. H. Delnes, D. A. McCombs and D. E. Potter, J. Am. Chem. Soc., 1969, 91, 4608; (b) K. Marx and W. Eberbach, Chem.–Eur. J., 2000, 6, 2063; (c) M. Reisser and G. Mass, J. Org. Chem., 2004, 69, 4913; (d) T. Hübner, Dissertation, University of Freiberg, 1987; (e) A. Arany, D. Bendell, P. W. Groudwater, I. Garnett and M. Nyerges, J. Chem. Soc., Perkin Trans. 1, 1999, 2605.
5. (a) N. C. Deno, C. U. Pittman and J. O. Turner, J. Am. Chem. Soc., 1965, 87, 2153; (b) O. N. Faza, C. S. López, R. Alvarez and A. R. de Lera, Chem.–Eur. J., 2009, 15, 1944.
6. D. Alickmann, R. Fröhlich, A. H. Maulitz and E. U. Wurthwein, Eur. J. Org. Chem., 2002, 1523.
7. (a) A. H. Zhou, Q. He, C. Shu, Y. F. Yu, S. Liu, T. Zhao, W. Zhang, X. Lu and L. W. Ye, Chem. Sci., 2015, 6, 1265; (b) X. Y. Xiao, A. H. Zhou, C. Shu, F. Pan, T. Li and L. W. Ye, Chem.–Asian J., 2015, 10, 1854; (c) W. B. Shen, X. Y. Xiao, Q. Sun, B. Zhou, X. Q. Zhu, J. Z. Yan, X. Lu and L. W. Ye, Angew. Chem., Int. Ed., 2017, 56, 605; (d) L. Li, T. D. Tan, Y. Q. Zhang, X. Liu and L. W. Ye, Org. Biomol. Chem., 2017, 15, 8483.
8. (a) H. Jin, L. Huang, J. Xie, M. Rudolph, F. Rominger and A. S. K. Hashmi, Angew. Chem., Int. Ed., 2016, 55, 794; (b) H. Jin, B. Tian, X. Song, J. Xie, M. Rudolph, F. Rominger and A. S. K. Hashmi, Angew. Chem., Int. Ed., 2016, 55, 12688.
9. (a) Z. Zeng, H. Jin, J. Xie, B. Tian, M. Rudolph, F. Rominger and A. S. K. Hashmi, Org. Lett., 2017, 19, 1020; (b) M. Chen, N. Sun, H. Chen and Y. Liu, Chem. Commun., 2016, 52, 6324; (c) W. Xu, G. Wang, N. Sun and Y. Liu, Org. Lett., 2017, 19, 3307.
10. For synthesis of nH-azepines (n = 2 or 4), see selected examples: (a) U. Gockel, U. Hartmannsgruber, A. Steigel and J. Sauer, Tetrahedron Lett., 1980, 21, 599; (b) I. R. Dunkin, A. E. Ayeb and M. A. Lynch, J. Chem. Soc., Chem. Commun., 1994, 1695; (c) K. Satake, R. Okuda, M. Hashimoto, Y. Fujiwara, H. Okamoto, M. Kimura and S. Morosawa, J. Chem. Soc. Perkin Trans 1, 1994, 1753; (d) Y. Luo and J. Wu, Chem. Commun., 2011, 47, 11137.
11. Crystallographic data of compounds 3b, 3l, 4a, 5a, 5i, and 6m were deposited in Cambridge Crystallographic Data Center: 3b: CCDC 1589549, 3l: CCDC 1589562, 4a: CCDC 1589561, 5a: CCDC 1589558, 5i: CCDC 1589559 and 6m CCDC 1589560.†.
12. For the aza-Nazarov cyclizations, see: (a) D. A. Klumpp, Y. Zhang, M. J. O'Connor, P. M. Esteves and L. S. de Almeida, Org. Lett., 2007, 9, 3085; (b) Z. X. Ma, S. He, W. Song and R. P. Hsung, Org. Lett., 2012, 14, 5736; (c) R. L. Sahani and R. S. Liu, Angew. Chem., Int. Ed., 2017, 56, 12736; (d) S. K. Pawar, R. L. Sahani and R. S. Liu, Chem.–Eur. J., 2015, 21, 10843; (e) C. Shu, Y. H. Wang, C. H. Shen, P. P. Ruan, X. Lu and L. W. Ye, Org. Lett., 2016, 18, 3254.
13. As suggested by one reviewer, an alternative mechanism is also possible for the Zn(II)-catalyzed rearrangement; this process involves an isomerization of initial species 3 to an unconjugated iminoyl ketone H, followed by a 6π-cyclization to generate species I. A subsequent Zn(II)-catalyzed aromatization of species I is expected to yield the final product 5. In this process, species H is relatively higher than 3 in energy, but its feasibility is not excluded.
14. (a) S. M. Wang and L. Zhang, Org. Lett., 2006, 8, 4585; (b) G. Li, G. Zhang and L. Zhang, J. Am. Chem. Soc., 2008, 130, 3704; (c) S. B. Wagh and R. S. Liu, Chem. Commun., 2015, 51, 15462; (d) R. Chaudhuri, A. Das, H. Y. Liao and R. S. Liu, Chem. Commun., 2010, 46, 4601.
15. For reviews of gold-catalyzed cycloaddition or annulation reactions of alkynes, see (a) D. J. Gorin, B. D. Sherry and F. D. Toste, Chem. Rev., 2008, 108, 3351; (b) A. S. K. Hashmi, Chem. Rev., 2007, 107, 3180; (c) F. López and J. L. Mascareñas, Beilstein J. Org. Chem., 2011, 7, 1075; (d) C. Aubert, L. Fensterbank, P. Garcia, M. Malacria and A. Simonneau, Chem. Rev., 2011, 111, 1954; (e) D. Garayalde and C. Nevado, ACS Catal., 2012, 2, 1462; (f) M. E. Muratore, A. Homes, C. Obradors and A. M. Echavarren, Chem.–Asian J., 2014, 9, 3066.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 1589549, 1589562, 1589561, 1589558, 1589559 and 1589560. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c8sc00232k

---