Organocatalytic atroposelective heterocycloaddition to access axially chiral 2-arylquinolines

Axially chiral heterobiaryls play a vital role in asymmetric synthesis and drug discovery. However, there are few reports on the synthesis of atropisomeric heterobiaryls compared with axially chiral biaryls. Thus, the rapid enantioselective construction of optically active heterobiaryls and their analogues remains an attractive challenge. Here, we report a concise chiral amine-catalyzed atroposelective heterocycloaddition reaction of alkynes with ortho-aminoarylaldehydes, and obtain a new class of axially chiral 2-arylquinoline skeletons with high yields and excellent enantioselectivities. In addition, the axially chiral 2-arylquinoline framework with different substituents is expected to be widely used in enantioselective synthesis.

A xially chiral biaryl scaffold is one of the most important structural units, which is widely found in many natural products 1,2 , bioactive molecules [3][4][5] , and functional materials 6,7 . Therefore, the study of axially chiral compounds has attracted extensive attention, and plays an important role in the development of chiral ligands 8,9 and organic catalysts [10][11][12] . In particular, axially chiral 1,1′-bi-2-naphthol (BINOL) (Fig. 1a, R 1 = OH) and its derivatives, as the most successful catalysts and ligands in enantioselective synthesis, have achieved great progress [13][14][15] . Despite advance has been made in the study of thesis axially chiral biaryls, there are still some limitations in the preparation of axially chiral heterobiaryls [16][17][18][19][20][21][22] . Atropisomeric isoquinoline derivatives (1-(isoquinolin-1-yl)naphthalen-2-ol, Fig. 1a, R 2 = OH) has emerged as a unique backbone of several famous catalysts and ligands (e.g., N,O-ligand, QUINOX) in asymmetric catalysis 23,24 . Especially, Noyori's BINAP catalyst 25 is widely used in industrial and pharmaceutical production. In sharp contrast to axially chiral BINOLs, the atroposelective synthesis of axially chiral isoquinoline derivatives has not been greatly explored. To date, there are few methods for assembling axially chiral isoquinoline derivatives, most of the thesis reports are focus on transition-metal-catalyzed direct cross-coupling reaction 26,27 , [2 + 2 + 2] cycloaddition 28,29 and recently reported (dynamic) kinetic resolution/transformation strategy [30][31][32][33][34] . However, building upon the increasing demand for this type of catalysts and ligands, a versatile, practical, and scalable method for the synthesis of optically pure axial chiral isoquinoline and corresponding analogues is undoubtedly an urgent need in this field. In addition, from the perspective of structural diversity, nonclassical isoquinoline-type analogues will offer more possibilities for catalyst/ligand development and drug discovery.
Although quinoline skeleton widely exists in bioactive compounds 35 , in sharp contrast to atropisomeric 2-arylisoquinoline, the research on axial chiral quinoline skeleton is very limited 34 . Meanwhile, most of the reports focused on the atroposelective synthesis of aryl-C4- 36,37 , C5-38 , or C8quinoline 39 skeletons, in which the distant from the chiral axis of the decisive nitrogen atom for coordination with metal center leading to the difficulty in stereo-induction, which sometimes restrict their further applications in the area of asymmetric synthesis greatly. To face this issue, Tan and coworkers 40 uncovered a chiral phosphoric acid-catalyzed This is the only report on the atroposelective synthesis of 2-arylquinoline skeletons. Overall, the catalytic atroposelective synthesis of axially chiral 2-arylquinolines in a highly atroposelective manner remains an attractive challenge.
In the past two decades, chiral amine catalyst has attracted considerable attention in asymmetric catalysis due to its advantages of operational simplicity, low toxicity, and minimal impact on the environment [41][42][43][44][45][46] . However, most of the reports solely focus on the assembly of central chirality, and the aminecatalyzed enantioselective construction of axial chirality is still in its infancy [47][48][49][50] . The Sparr group constructed a series of axially chiral biaryl skeletons via chiral amine-catalyzed asymmetric aldol condensation [47][48][49] . Recently, Cheng 20 and Wang 21 reported the synthesis of axially chiral 4-arylquinoline skeletons via aminecatalyzed asymmetric heterocycloaddition of ynals with 2-(tosylamino)aryl ketones, followed by aromatization and central-toaxial conversion (Fig. 1b). As part of our group ongoing efforts on organocatalytic synthesis of axially chiral molecules 51-53 , we have successfully reported carbene-catalyzed atroposelective desymmetrization of biphenols, the [3 + 3] annulation of cyclic 1,3diones with ynals and the kinetic resolution of anilides, resulting in valuable axially chiral biaryl amino alcohols (NOBIN analogues), α-pyrone-aryls, and isoindolinones, respectively. Despite the aforementioned achievements, unsolved challenges and the continuing demands for atropoisomers continue to drive us to develop more revolutionary protocols. We herein firstly report a chiral amine-catalyzed atroposelective heterocycloaddition to offer a class of axially chiral 2-arylquinolines (nonclassical isoquinoline-type analogues) with high yields and excellent enantioselectivities.
The key results of reaction optimization are summarized in Table 1. First, the chiral secondary amine catalysts A-C derived from L-proline with different steric size on the silyl groups were tested. As a result, these catalysts provided the target product 3a in high yields but with very low enantioselectivities ( Substrate scope. After obtaining the optimal conditions, we turned our attention to the substrate scope of ynals. An array of naphthalen-based propiolaldehydes bearing various R 1 were tested (Fig. 2). The results show that the electronic and steric effects of the substituents at different positions on the aromatic rings have little impact on the reaction, and the corresponding products were produced in high yields and good to excellent enantioselectivities (3c-3k). Moreover, the quinoline-based ynal provided the coresponding product 3l in 84% yield with 99:1 er. Notably, phenylpropiolaldehyde 1m generated the corresponding biaryl (quinoline-phenyl) product 3m with 92% yield and 96:4 er. Meanwile, a good level of er (89:11) was achieved for quinolinepyridine-type biaryl product 3n when pyridine-based propiolaldehyde 1n was used.
Encouraged by these results, we next examined the generality of o-aminoarylaldehyde 2. As shown in Fig. 3, substrates bearing either electron-rich groups or electron-deficient groups reacted smoothly with 1c affording the corresponding axially chiral products (4c-4k) in excellent yields (90-96%) and excellent er (96:4->99:1). When steric hindrance was introduced at the ortho position of the substrate reaction site, the corresponding high yields (84-85%), and the high er (all 94:6) for product 4a, 4b, and 4l were observed. The absolute configuration of 3k was determined by X-ray crystallography (Fig. 2), and other products were assigned by analogy.
Gram-scale synthesis and synthetic transformations. To evaluate the practicality of this protocol, a gram-scale reaction of 1c with 2b was carried out under standard condition (Fig. 4, 89% yield, 96:4 er), which indicated that the large-scale synthesis of enantioenriched 2-arylquinolines can be achieved. Synthetic transformations of 3c were also illustrated in Fig. 4b. On the basis of methylation and hydrogenolysis, 3c was easily converted into a versatile intermediatex axially chiral isoquinoline analogue 5 with a yield of 71%, and the er value is completely maintained. After methylation and oxidation, the axially chiral QUINOX analogue Fig. 2 Scope of ynals. Reaction conditions: a mixture of 1c (0.11 mmol), 2 (0.1 mmol) and catalyst D (10 mol%) in CHCl 3 (1.0 mL) was stirred at room temperature for 24-36 h. After the reaction was complete, the reaction was cooled to 0°C, then MeOH (0.5 mL) and NaBH 4 (0.2 mmol) were added to the mixture and stirred for another 0.5 h at room temperature. At last, HOAc (5.0 equiv) was added to the mixture and stirred for another 2 h. 6 (a potential Lewis base organocatalyst) was obtained in good yield without loss of enantiopurity.
A postulated reaction pathway is proposed in Fig. 4c. Initially, chiral secondary amine catalyst D added to alkynaldehyde 3c and subsequently dehydrated to produce alkynylamine cationic intermediate I. Then 2b reacted with I via aza-Michael addition to give axially chiral allenamine intermediate II, which underwent an intramolecular aldol reaction to give the chiral styrene intermediate III. Catalyst D was then released from intermediate III to produce chiral compound IV. Finally, compound IV was reduced by NaBH 4 and then dehydrated in the present of acid in one pot to deliever the desired product 3c.

Discussion
In summary, we have successfully synthesized the axially chiral 2-arylquinoline analogues via catalytic heterocycloaddition of alkynaldehydes with N-protected o-aminoarylaldehyde via the formation of a critical intermediate (axially chiral styrene). In the presence of commercially available amine catalyst, this conversion can deliver a variety of axially chiral 1-aryl isoquinoline analogues with high yields and excellent er's. The synthetic utility of this methodology is illustrated by further conversion to axially chiral QUINOX and isoquinoline analogues. Further studies on the application of the axially chiral 2-arylquinoline skeletons in asymmetric synthesis are currently underway in our laboratory.

Methods
Procedure for enantioselective syntheses of compound 3c. To a flame-dried Schlenk reaction tube equipped with a magnetic stir bar, was added the catalyst D (2.2 mg, 0.01 mmol), 1c (39.4 mg, 0.12 mmol), and 2b (0.10 mmol). The Schlenk tube was closed with a septum, CHCl 3 (2.0 mL) was added. The mixture was then stirred at room temperature and monitored by TLC until 2b was full consumed. Then the mixture was cooled to 0°C and MeOH (1.0 mL) was added, subsequently, NaBH 4 (11.3 mg, 0.3 mmol) was added slowly to the mixture, and stirred for 2.0 h at this temperature. After the completion of the reaction, as monitored by TLC, HOAc (5.0 equiv) was added to the mixture and then the mixture was warmed to room temperature and stirred for another 6 h. After the reaction was completed, as monitored by TLC, saturated NaHCO 3 was added and stirred for another 0.5 h. Then the mixture was extracted with EtOAc. The combined organic layers were washed with water and brine, dried over anhydrous Na 2 SO 4 , filtered, and concentrated. The residue was purified by a silica gel flash chromatography (Hexane/EtOAc = 5:1) to afford the desired product 3c.
Procedure for enantioselective synthesis of compound rac-3c. To a flamedried Schlenk reaction tube equipped with a magnetic stir bar, was added the catalyst rac-A (2.2 mg, 0.01 mmol), 1c (39.4 mg, 0.12 mmol) and 2b (0.10 mmol). The Schlenk tube was closed with a septum, CHCl 3 (2.0 mL) was added. The mixture was then stirred at room temperature and monitored by TLC until 2b was full consumed. Then the mixture was cooled to 0°C and MeOH (1.0 mL) was added, subsequently, NaBH 4 (11.3 mg, 0.3 mmol) was added slowly to the mixture and stirred for 2.0 h at this temperature. After the completion of the reaction, as monitored by TLC, HOAc (5.0 equiv) was added to the mixture and then the mixture was warmed to room temperature and stirred for another 6 h. After the reaction was completed, as monitored by TLC, saturated NaHCO 3 was added and stirred for another 0.5 h. Then the mixture Fig. 3 Scope of N-(2-formylphenyl)naphthalene-1-sulfonamides. Reaction conditions: a mixture of 1c (0.11 mmol), 2 (0.1 mmol) and catalyst D (10 mol%) in CHCl 3 (1.0 mL) was stirred at room temperature for 24-36 h. After the reaction was completed, the reaction was cooled to 0°C, then MeOH (0.5 mL) and NaBH 4 (0.2 mmol) were added to the mixture and stirred for another 0.5 h at room temperature. At last, HOAc (5.0 equiv) was added to the mixture and stirred for another 2 h.
was extracted with EtOAc. The combined organic layers were washed with water and brine, dried over anhydrous Na 2 SO 4 , filtered, and concentrated. The residue was purified by a silica gel flash chromatography (Hexane/EtOAc = 5:1) to afford the desired product 3c.

Data availability
The authors declare that the data supporting the findings of this study are available within the article and Supplementary information file, or from the corresponding author upon reasonable request. The supplementary crystallographic data for this paper could be