Asymmetric [4+2] annulation of N-sulfonyl ketimines: access to N-heterocycles bearing 1,3-nonadjacent stereocenters with a quaternary carbon center

INTRODUCTION

Chiral benzofused 5-membered sultam with a quaternary stereocenter as an important structural motif shows a broad spectrum of biological activities^[1-4], such as HIV-1 inhibitors, γ-secretase inhibitors and aldose reductase inhibitors [Figure 1]. It is also used as chiral auxiliaries in asymmetric chemistry^[5]. On the other hand, quinazoline skeleton widely exists in natural products^[6-9] and marketed drugs [Figure 1]. For instance, antiviral drug Letermovir^[10], with 3,4-dihydroquinazoline, has potent anti-cytomegaloviral activity both in vitro and in vivo, and it is used clinically to treat patients who are seropositive for cytomegalovirus after undergoing allogeneic hematopoietic stem cell transplantation. The serum high-density lipoprotein cholesterol raising agent (R)-SDZ 267-489^[11] can reduce the risk of atherosclerosis by raising cholesterol levels.

Figure 1. Bioactive compounds containing benzofused sultams and quinazolines.

The combination of two pharmacophores through covalent bonds to create new bioactive molecules is a traditional method of drug design. We envisioned that the N-heterocyclic compounds, combining the two chiral structures mentioned above containing 1,3-nonadjacent stereocenters, might possess potential biological activities. Meanwhile, N-heterocyclic compounds bearing 1,3-nonadjacent stereocenters also serve as the fundamental structure of biologically active natural products^[12-15], such as tetraponerines, which act as inhibitors of neuronal nicotinic acetylcholine receptors [Scheme 1A]. Moreover, in the total synthesis of this class of natural products, the construction of 1,3-nonadjacent stereocenters usually starts with chiral substrates^[16-18] or requires the participation of chiral auxiliaries^[19,20]. There is only one example of asymmetric catalytic approach, which requires the stepwise introduction of two chiral centers to construct the chiral hexahydropyrimidine skeleton^[21]. Therefore, the development of an efficient approach to the construction of benzenesulfonamide fused tetrahydroquinazoline compounds bearing 1,3-nonadjacent stereocenters in a direct, concise and stereoselective manner is of great importance. Currently, there are limited asymmetric catalytic synthetic methods available for constructing N-heterocyclic compounds with 1,3-nonadjacent stereocenters. As shown in Scheme 1, He et al. developed a cascade hydroamination/redox reaction for the synthesis of cyclic aminals under the combined catalysis of gold complexes and Brønsted acid^[22]. Sim et al. developed organocatalyzed stereoselective [4+2] annulations of cyclic N-sulfimines, affording benzosulfamidate fused tetrahydroquinazoline and hexahydropyrimidine derivatives, respectively^[23,24]. Mun et al. also developed a palladium-catalyzed decarboxylative [4+2] annulation of vinyl benzoxazinanones with cyclic N-sulfimines for the synthesis of benzosulfamidate fused tetrahydroquinazolines^[25].

Scheme 1. Asymmetric synthesis of N-heterocycles bearing 1,3-nonadjacent stereocenters.

It is widely recognized that quaternary carbon chiral centers represent a prominent research focus and challenge in the field of asymmetric catalysis due to their steric hindrance. Therefore, the development of 1,3-nonadjacent stereocenters with a quaternary carbon center is of great significance. Notably, there is lack of documented techniques for producing N-heterocyclic frameworks bearing 1,3-nonadjacent stereocenters with a quaternary carbon center, as illustrated in Scheme 1. Herein, we reported a catalytic enantioselective [4+2] annulation of N-sulfonyl ketimines with aminochalcones to access benzenesulfonamide fused tetrahydroquinazoline compounds bearing 1,3-nonadjacent stereocenters with a quaternary carbon center.

EXPERIMENTAL

To a stirred solution of cyclic N-sulfonyl ketimine 1 (0.10 mmol, 1.0 equiv) and aminochalcone 2(0.15 mmol, 1.5 equiv) in CHCl₃ (1.0 mL) was added catalyst (0.01 mmol, 10 mol%) at 25 °C. Then, the resulting mixture was stirred at the same temperature. After completion of the reaction as monitored by thin-layer chromatography (TLC), the reaction mixture was directly charged to column chromatography on silica gel (petroleum ether/ethyl acetate = 20/1 to 5/1) to give the product 3.

RESULTS AND DISCUSSION

To evaluate the practical feasibility of our proposal, the model reaction between N-sulfonyl ketimine 1a and aminochalcone 2a was carried out in MeCN at 25 °C. As indicated in Table 1, several chiral catalysts with hydrogen bond donors and acceptors were first examined. Cyclohexyl guanidine catalysts C1, C2, C5, C6 and benzocyclohexyl guanidine catalysts C3, C4 were tested first (Table 1, entries 1 to 6). It was found that catalyst C2 led to the best enantioselectivity, and the reaction proceeded well, producing the desired [4+2] cycloadducts in 54% yield with 89% ee (Table 1, entry 2). We then tested different types of cyclohexyl catalysts without guanidino C7-C9. To our delight, the enantioselectivity was further improved, and C8 exhibited the most effective catalytic effect (7:1 dr, 95% ee) (Table 1, entry 8), which is supposed to follow the catalytic pathway of the catalyst to form a ketimine intermediate with the aminochalcone to catalyze the reaction. Next, the reaction conditions were further optimized by varying solvents (Table 1, entries 10-15). When halogenated hydrocarbon solvents were tested, the enantioselectivity improved, and CHCl₃ as the solvent gave the best enantioselectivity (99% ee) and diastereoselectivity (10:1 dr) (Table 1, entry 14).

Table 1

Evaluation of catalysts and optimization of reaction conditions^a

Entry	Cat.	Solvent	Yield (%)b	Drc	Ee (%)d
1	C1	MeCN	42	2:1	0
2	C2	MeCN	54	2:1	89
3	C3	MeCN	39	1:1	11
4	C4	MeCN	< 5	-	-
5	C5	MeCN	51	2:1	76
6	C6	MeCN	58	2:1	81
7	C7	MeCN	56	5:1	92
8	C8	MeCN	52	7:1	95
9	C9	MeCN	55	5:1	91
10	C8	EtOAc	21	2:1	80
11	C8	Toluene	45	2:1	74
12	C8	THF	35	5:1	88
13	C8	Et2O	52	2:1	78
14	C8	CHCl3	60	10:1	99
15	C8	DCM	41	7:1	92

^aConditions: 1a (0.10 mmol), 2a (0.15 mmol), and catalyst (0.01 mmol) in solvent (1.0 mL) at 25 °C; ^bIsolated yield; ^cDetermined by ¹H NMR of the crude reaction mixture; ^dThe ee values were determined by chiral HPLC analysis. THF: Tetrahydrofuran; DCM: dichloromethane.

With the optimal reaction condition in hand, we then investigated the substrate scope of N-sulfonyl ketimines 1 and aminochalcones 2. As shown in Scheme 2, a range of substituted N-sulfonyl ketimines were evaluated to examine the generality of the method. The electron-withdrawing groups and electron-donating groups at different positions of the aromatic ring of 1 were tolerated well in this catalytic reaction, giving the corresponding chiral benzenesulfonamide fused tetrahydroquinazoline products 3b-3g with good diastereoselectivities (> 8:1) and excellent enantioselectivities (97%-99% ee). When replacing the benzene ring with naphthalene rings of N-sulfonyl ketimines 1, the transformation also maintained excellent enantioselectivities, affording 3h in 98% ee and 3i in 95% ee. Additionally, the replacement of ethyl ester with methyl ester in N-sulfonyl ketimines 1 also worked well to give the desired product 3j with 10:1 dr and 99% ee.

Scheme 2. Substrate scope of cyclic N-sulfonyl ketimines 1^a. ^aConditions: 1 (0.10 mmol), 2a (0.15 mmol), and catalyst (0.01 mmol) in CHCl₃(1.0 mL) at 25 °C; ^bIsolated yield; ^cDetermined by ¹H NMR of the crude reaction mixture; ^dThe ee values were determined by chiral HPLC analysis.

Next, a variety of aminochalcones 2 were investigated. Firstly, we tried different R³ substituents on the benzene ring. As shown in Scheme 3, electron-withdrawing R³ groups (4-F, 4-Cl, 4-Br, 5-F, 5-Cl) and electron-donating R³ groups (4-Me, 4-OMe) worked well, delivering products 3k-3q with good diastereoselectivities and excellent enantioselectivities. In addition, the R⁴ groups of 2, which involved a phenyl ring containing electron-withdrawing or donating substituents, were all well tolerated. The reactions proceeded smoothly with excellent efficiency, furnishing products 3r-3u with 10:1 dr and excellent enantioselectivities (96%-98% ee). Heterocyclic aromatic rings such as thienyl and furyl also worked well, and the reactions proceeded efficiently to afford the corresponding products 3v and 3w with excellent enantioselectivities. The absolute configuration of 3s was confirmed by X-ray single crystal diffraction (detailed data are in Supplementary Materials) and those of other products were assigned accordingly.

Scheme 3. Substrate scope of aminochalcones 2^a. ^aConditions: 1a (0.10 mmol), 2 (0.15 mmol), and catalyst (0.01 mmol) in CHCl₃ (1.0 mL) at 25 °C; ^bIsolated yield; ^cDetermined by ¹H NMR of the crude reaction mixture; ^dThe ee values were determined by chiral HPLC analysis.

To demonstrate the potential application of this protocol, the scale-up reaction of N-sulfonyl ketimines 1a with aminochalcone 2a was carried out under the standard condition [Scheme 4], and the corresponding product 3a could be obtained in 56% yield (1.04 g) with 99% ee. Then, the reduction of the carbonyl group of 3a using NaBH₄ and CeCl₃·7H₂O furnished the hydroxy-functionalized compound 4 in 60% yield. We also tried the Suzuki coupling of 3r, and the desired product 5 was obtained with a 55% yield.

Scheme 4. Scale-up reaction and further synthetic transformation.

CONCLUSIONS

In summary, we have developed a concise and efficient [4+2] annulation of N-sulfonyl ketimines and aminochalcones. The reaction proceeds with excellent enantioselectivity under mild conditions, providing a convenient approach to benzenesulfonamide fused tetrahydroquinazoline compounds with 1,3-nonadjacent stereocenters containing a quaternary carbon center. This straightforward synthetic protocol exhibited excellent yields with a wide substrate scope. Studies on the bioactivities of the benzenesulfonamide fused tetrahydroquinazoline compounds are in progress in our laboratory. Additionally, the GAP (Group-assisted purification) protection groups^[26-28] could be further considered for catalyst recycling/recovery purposes for this reaction, especially for large-scale synthesis with larger loading of catalysts.