Enantioselective [4+2] Annulation to the Concise Synthesis of Chiral Dihydrocarbazoles

Summary A highly efficient phosphine-catalyzed enantioselective [4 + 2] annulation of allenoates with 3-nitroindoles or 3-nitrobenzothiophenes has been developed. The protocol represents a unique dearomatization–aromatization process to access functionalized dihydrocarbazoles or dihydrodibenzothiophenes with high optical purity (up to 97% ee) under mild reaction conditions. The synthetic utility of the highly enantioselective [4 + 2] annulation enables a concise synthesis of analgesic agent.


INTRODUCTION
Fused polycyclic indoles are common structural motifs found in a vast array of natural and biologically active molecules (Saxton, 1996;Knö lker and Reddy, 2002;Schmidt et al., 2012;Tan and Cheng, 2019), such as kopsihainanine A, isoelliptitoxin, and analgesic agents (Scheme 1A) (Madalengoitia and Macdonald, 1993;Carmosin et al., 2000;Chen et al., 2011). In this regard, the development of efficient methods for enantioselective construction of hydrocarbazole skeleton is still highly demanded (Sings et al., 2001;Lu et al., 2012;Zhou et al., 2015;Gu et al., 2016). The group of Jørgensen disclosed a novel [4 + 2] annulation by trienamine catalysis, thus obtaining dihydrocarbazoles in good yields and enantioselectivities (Li et al., 2016b). In this context, we hypothesized that the development of new methods through the enantioselective phosphine-catalyzed [4 + 2] dearomatization would provide practical and efficient approach to this class of enantioenriched heterocycles (Scheme 1B).
Phosphine catalysis has been recognized as a reliable tool for the development of unique transformations of allenoates, allowing for the discovery of novel asymmetric synthetic methodology (Lu et al., 2001;Methot and Roush, 2004;Ye et al., 2008;Cowen and Miller, 2009;Wei andShi, 2010, 2017;Fan and Kwon, 2013;Wang et al., 2014Wang et al., , 2016Ló pez and Mascareñ as, 2014;Xie and Huang, 2015;Li and Zhang, 2016a;Li and Lu, 2017;Ni et al., 2018;Guo et al., 2018). Considerable research efforts have been devoted to the development of new methods for the phosphine-catalyzed enantioselective reactions. The use of phosphine catalysts has introduced a set of elementary steps that operate via discrete reactive species, allowing access to natural products and pharmaceuticals (Tran and Kwon, 2005;Andrews and Kwon, 2012;Han et al., 2012;Wang and Krische, 2003;Cai et al., 2016). One particularly versatile and reactive species is the phosphine-mediated 1,4-dipole generated upon addition of the phosphine catalyst to an allenoate substrate, thus providing a concise approach for accessing enantioselective annulations. Specially, Kwon group reported the result of their pioneering studies toward the development of a novel [4 + 2] annulation reaction of allenoates and N-tosylimines in the presence of phosphine catalyst (Zhu et al., 2003). Later, Fu group reported the phosphine-catalyzed highly enantioselective [4 + 2] annulation of N-tosylimines with allenoates (Wurz and Fu, 2005). Although great achievements have been made, concise syntheses of useful heterocycles involving phosphine-catalyzed [4 + 2] annulations in asymmetric versions were still rare (Tran and Kwon, 2007;Wang and Ye, 2010;Tran et al., 2011;Xiao et al., 2011;Baskar et al., 2011;Yu and Ma, 2012;Zhong et al., 2012;Takizawa et al., 2014;Yu et al., 2014;Liu et al., 2016;Wang and Guo, 2019a). Moreover, the development of an enantioselective phosphine-catalyzed [4 + 2] dearomatization reaction would provide an attractive and complementary approach for construction of privileged motifs, which will be of great value for the synthesis of bioactive molecules (Scheme 1C).
Enantioselective dearomatization reactions of heteroaromatic compounds are very powerful transformations because they provide direct access to a wide variety of chiral heterocycles (You, 2016;Roche and Porco, 2011;Zhuo et al., 2012;Zhuo et al., 2014;Zheng and You, 2016;Sun et al., 2016;Wu et al., 2016). In recent years, 3-nitroindole was demonstrated to be a good substrate for various dearomatization processes, and a number of enantioselective approaches have been reported (Awata and Arai, 2014;Zhao et al., 2015aZhao et al., , 2015bZhao et al., , 2018Zhao et al., , 2019Gerten and Stanley, 2016;Trost et al., 2014;Cheng et al., 2018;Zhang et al., 2018;Sun et al., 2018;Yue et al., 2017;Yang et al., 2019). Importantly, Lu group (Li et al., 2019) and Zhang group (Wang et al., 2019b) independently reported the efficient phosphine-catalyzed enantioselective [3 + 2] annulation of 3-nitroindoles with allenoates to afford cyclopentaindoline products in high yields and excellent enantioselectivities. We envisaged that heteroaromatic systems bearing an electron-withdrawing group could react with phosphine-mediated zwitterionic intermediate in a process involving the [4 + 2] reaction to achieve the chiral dihydrocarbazole scaffold (Scheme 1C). With this objective in mind, a readily available 3-nitroindole derivative was selected as a model substrate to investigate the optimum reaction condition for the enantioselective [4 + 2] dearomatization reaction using a phosphine catalyst.

RESULTS AND DISCUSSION
Based on our previous work on phosphine chemistry (Wang and Guo, 2019a), we initiated the study by investigating the reaction between 1a and 2a in the presence of the phosphine 4a (Table 1, entry 1). Initially, diverse chiral phosphine catalysts were examined (entries [1][2][3][4][5]. However, the catalyst 4a to 4c did not work for this reaction (entries [1][2][3]. To our great delight, the desired dihydrocarbozole3a could be obtained when the chiral phosphine 4d was employed (entry 4). After surveying an array of additives, we determined that silica gel can promote elimination of HNO 2 for the aromatization process to afford the corresponding adduct in 92% yield with 94% ee (entry 4) (So and Mattson, 2012;Long et al., 2016). Other additives, such as Sc(OTf) 3 , Et 3 N, and SnCl 2 led to byproducts (for further details, see Table S1 in the Supplemental Information). Furthermore, the ee values of the 3a decreased to 80% with low yield in the presence of 4e as catalyst (entry 5). Varying the solvents led to no improvement in the reaction, and toluene was proven to be the best choice (entries 4 vs 6-8). Further optimization studies revealed that the protection group of the 3-nitroindole was also sensitive to the reaction, and the variation of the N-substituent of the 3-nitroindole 1a 0 or 1a'' generated no product at all (entries 9 and 10) (Rivinoja et al., 2017;Suo et al., 2018). With the optimal reaction conditions in hand, we set out to explore the substrate scope of the procedure. As shown in Scheme 2, various electron-withdrawing or donating groups on the indole ring were well tolerated and resulted in excellent levels of enantioselectivities ranging from 86% to 97% ee (3a-3j). The extension of the protocol to the 3-nitroindole with a variety of substitution patterns at the 5-position was successful to afford corresponding adducts with excellent enantioselectivities (3b-3f). To our delight, substrates bearing substituents on different positions of the indole ring also facilitate the annulation with high yields and ee values (3b, 3g, and 3j). The absolute configuration of the enantiopure 3i, recrystallized from ethyl acetate and petroleum ether, was assigned by single-crystal X-ray diffraction analysis.
The generality of the reaction with respect to the scope of the allenoates 2 was also investigated using 3-nitroindole 1a as the reaction partner under the optimized conditions. A diverse array of allenoates (2) with a variety of functional groups (methyl, fluoro, chloro, bromo, ester, trifluoromethyl, and cyano) performed well in this annulation reaction, and the corresponding products were isolated in good yields with high ee values (3k-3q). Remarkably, this method was compatible with alkyl allenoate, affording the desired products in good yields with good enantioselectivity (3v-3x). Additionally, all reactions with different esters attached to the allenoates proceeded smoothly, giving the corresponding products in good yields and excellent ee (3y and 3z). To test the synthetic utility of the current annulation, we performed the reaction on a 1 mmol scale with the formation of 3z in 56% yield and 92% ee. Encouraged by the excellent results with various 3-nitroindoles, we then investigated the [4 + 2] annulation reaction with a range of 3-nitrobenzothiophenes (5). Remarkably, process where the 3-nitrobenzothiophene as a reactive partner for asymmetric annulation has been much less studied (Tran and Kwon, 2007;Wang and Ye, 2010;Tran et al., 2011;Xiao et al., 2011;Baskar et al., 2011;Yu and Ma, 2012;Zhong et al., 2012;Takizawa et al., 2014;Yu et al., 2014;Liu et al., 2016;Wang and Guo, 2019a;Cheng et al., 2000;Cheng et al., 2017;Suo et al., 2018;Yue et al., 2018;Chen et al., 2019). Using phosphine 4d in toluene at 0 C, we were able to access dihydrodibenzothiophene products 6 (Scheme 3). Under the optimized reaction condition (for further details, see Table S2 in the Supplemental Information), a broad range of allenoates 2 and 3-nitrobenzothiophenes 5 were investigated. Allenoates with different substituents on the aromatic ring underwent this catalytic transformation smoothly in good yields with excellent ee (6a and 6b). Furthermore, various substitutions of 3-nitrobenzothiophenes 5 at the aromatic ring had little impact on the reactions (6c-6h, 91%-97% ee).
To highlight the synthetic potential of the present method, the dihydrocabazole 3v, which was obtained from the enantioselective [4 + 2] annulation, can be easily converted into analgesic agent 9 (Scheme 4). In 2000, Carmosin and co-workers obtained the racemic analgesic agent 9 via the Diels-Alder reaction,

Scheme 2. Substrate Scope of Enantioselective [4 + 2] Annulation
Unless indicated otherwise, the reactions were conducted with 1 (0.1 mmol), 2 (0.15 mmol), and catalyst 4d (0.01 mmol) in toluene at room temperature for 12-48 h. Then silica gel was added to the reaction mixture to complete elimination of HNO 2 . a Yield of the isolated product after purification by chromatography on silica gel. b Enantiomeric excess determined by HPLC analysis. c Aromatization process was performed at 50 C. d 20 mol% of 4d. e The reaction was performed on 1 mmol scale.

Scheme 3. Enantioselective [4 + 2] Annulation of 3-Nitrobenzothiophene 5
Unless indicated otherwise, the reactions was conducted with 5 (0.1 mmol), 2 (0.15 mmol), and catalyst 4d (0.01 mmol) in toluene (1.0 mL) at 0 C for 48-60 h. Then silica gel was added to the reaction mixture to complete elimination of HNO 2 . Yield of the isolated product after purification by chromatography on silica gel. Enantiomeric excess determined by HPLC analysis. a 0.02 mmol of 4d was used.
iScience 23, 100840, February 21, 2020 and the optical product was obtained by using preparative chromatography (Carmosin et al., 2000). Taking advantage of our current phosphine-catalyzed enantioselective [4 + 2] reaction, we can easily obtain the analgesic agent 9 with excellent enantioselectivity. Hydrogenation of 3v in the presence of a catalytic amount of Pd/C, followed by amidation with MeNH 2 gave rise to the desired amide 7 in 84% yield over two steps. The configuration of compound 7 was assigned by X-ray analysis. The subsequent chlorination of alcohol, deprotection of the N-Boc group and cyclization furnished 8 in good yield. Finally, the amide 8 was reduced to generate the corresponding analgesic agent 9 in 78% yield and 92% ee.
The proposed catalytic cycle for the enantioselective [4 + 2] annulation is illustrated in Figure 1. The addition of phosphine catalyst 4d to the allenoate 2 gives the intermediate A, which could react with the 3-nitroindole 1 or 3-nitrobenzothiophenes 5 to form the intermediate B. Following migration and intramolecular conjugate addition give rise to the intermediate D and regenerate the phosphine 4d. This species D then undergoes elimination of HNO 2 through the aromatization process to furnish the final dihydrocarbzole 3 or dihydrodibenzothiophene 6.
In summary, we have developed simple and efficient synthetic routes to highly enriched hydrocarbozoles through chiral phosphine-catalyzed [4 + 2] annulation utilizing 3-nitroindole and allenoates as starting materials. This phosphine-catalyzed enantioselective [4 + 2] annulation procedure involving tandem dearomatization and aromatization steps proceeds under mild conditions. This reaction displays a broad substrate scope with respect to the substituents. Additionally, the obtained dihydrocarbozole could be efficient transformed to an analgesic agent containing polycyclic indole frameworks.

Limitations of the Study
The synthesis of the products needs two steps in one pot. No product was formed with the initial addition of silica gel.

Supplemental figures and tables for X-Ray structures
Figure S157. X-Ray crystal data of 3i, related to Scheme 2.

General Information
Unless otherwise noted, all reagents were purchased from commercial suppliers and used without further purification. NMR spectra were recorded on a Brucker-400 MHz spectrometer.

General procedure
All the racemic products were obtained by use of Cy3P as catalyst.

Scheme S1. General procedure for phosphine-catalyzed enantioselective [4+2] annulation reaction of 3-nitroindoles 1 and allenoates 2, related to Scheme 2.
A dried tube with a magnetic stir bar was charged with 3-nitroindole derivative 1 (0.10 mmol), allenoate derivative 2 (0.15 mmol, 1.5 equiv.), catalyst 4d (10 mol%), followed by the addition of toluene (1.0 mL), and the reaction mixture was stirred at room temperature. When the reaction was finished (determined by TLC). The mixture were added silica gel and toluene (1.0 mL) continued stir at room temperature when the aromatization process was finished (determined by TLC). Then solvent was evaporated and the residue was purified by column chromatography on silica gel using hexane/ethyl acetate as the eluent to afford the products 3.

Scheme S2. General procedure for phosphine-catalyzed enantioselective [4+2] annulation reaction of 3-nitrobenzothiophene 5 and allenoate 2, related to Scheme 3.
A dried tube with a magnetic stir bar was charged with 3-nitrobenzothiophenes derivative 5 (0.10 mmol), allenoate derivative 2 (0.15 mmol, 1.5 equiv.), catalyst 4d (10 mol%), followed by the addition of toluene (1.0 mL), and the reaction mixture was stirred at 0 ℃. When the reaction was finished (determined by TLC). The mixture were added silica gel and toluene (1.0 mL) continued stir at room temperature when the aromatization process was finished (determined by TLC). Then solvent was evaporated and the residue was purified by column chromatography on silica gel using hexane/ethyl acetate as the eluent to afford the products 6.

Scheme S3. 1-mol scale reaction, related to Scheme 2.
A dried tube with a magnetic stir bar was charged with 3-nitroindole derivative 1a (1.0 mmol), allenoate derivative 2o (1.5 mmol, 1.5 equiv.), catalyst 4d (10 mol%), followed by the addition of toluene (10.0 mL), and the reaction mixture was stirred at room temperature. When the reaction was finished (determined by TLC). The mixture were added silica gel (2.0 g) and toluene (10.0 mL) continued stir at room temperature when the aromatization was finished (determined by TLC). Then solvent was evaporated and the residue was purified by column chromatography on silica gel using hexane/ethyl acetate as the eluent to afford the products 3x (251.4 mg, 56%, 92% ee).

Scheme S3. Synthesis procedure of derivatization reaction, related to Scheme 4.
A dried tube with a magnetic stir bar was charged with 3-nitroindole derivative 1a (1.00 mmol), allenoate derivative 2m (1.50 mmol, 1.5 equiv.), catalyst 4d (10 mol%), followed by the addition of toluene (10.0 mL), and the reaction mixture was stirred at room temperature. When the reaction was finished (determined by TLC). The mixture were added silica gel (2.0 g) and toluene (10.0 mL) continued stir at room temperature when the aromatization process was finished (determined by TLC). Then solvent was evaporated and the residue was purified by column chromatography on silica gel using hexane/ethyl acetate as the eluent to afford the products 3v (261.6 mg, 57%, 92% ee).
A suspension of 3v (261.6 mg, 0.57 mmol) and 10% palladium on carbon (130.0 mg) in MeOH (10.0 mL) was maintained under an atmosphere of hydrogen gas for 8 h at rt. The insoluble solids were removed by filtration and the filtrate was concentrated. The residue was dissolved in MeNH2 (30 wt. % in absolute EtOH 4.0 mL), and the resulting mixture was stirred 1 h. Then solvent was evaporated and the residue was purified by column chromatography on silica gel using hexane/ethyl acetate as the eluent to afford the product 7 (179.2 mg, 84%, 92% ee).
To a solution of 7 (179.2 mg, 0.48 mmol) in anhydrous DCM (5 mL) was slowly added SOCl2 (1 mol/L in DCM, 2.0 mL) at 0 ℃. The suspension was allowed to warm to room temperature and continues to stir 2 h. After that, the reaction mixture was reduced in vacuo. The residue was dissolved in EtOH (5.0 mL) and potassium tert-butoxide (336.0 mg, 3.0 mmol) was added and the reaction stirred at 40 ℃ for 36 h. The solvent was removed and the residue was purified by column chromatography using MeOH/DCM as the eluent to give 8 (91.2 mg, 79%, 92% ee).
To a solution of 7 (28.8 mg, 0.12 mmol) in anhydrous THF (5 mL) was added LiAlH4 (45.6 mg, 1.2 mmol) at 0 ℃. The suspension was allowed to warm to room temperature and continues to stir at 40 ℃ for 36 h. After that, saturated aqueous Na2SO4 (4 mL) was added. The solid formed was filtered and washed with DCM. The organic layers were combined and dried with MgSO4. The solvent was removed and the residue was purified by column chromatography (DCM/MeOH/Et3N = 100/5/1) to give 8 (21.2 mg, 78%, 92% ee).