Two Catalytic Annulation Modes via Cu-Allenylidenes with Sulfur Ylides that Are Dominated by the Presence or Absence of Trifluoromethyl Substituents

Summary We disclose the Cu-catalyzed enantioselective synthesis of 3-methyl-3-propargyl-indolines, which contain a quaternary stereogenic carbon center, via the decarboxylative [4 + 1] annulation of 4-methyl-4-propargyl-benzoxazinanones with variety of sulfur ylides. The reaction proceeds predominantly through a γ-attack at the Cu-allenylidene intermediates by sulfur ylides to provide the corresponding indolines in good yield and high enantioselectivity (up to 91% ee). In contrast, the reaction of 4-trifluoromethyl-4-propargyl-benzoxazinanones with sulfur ylides delivers 3-trifluoromethyl-2-functionalized indoles in good to high yield via an unexpected α-attack at the Cu-allenylidene intermediates. Control over the α/γ-attack at the Cu-allenylidene intermediates by the same interceptors was achieved for the first time by the use of trifluoromethyl substituents.

Herein, we disclose the first successful attempt to control the a/g chemo-selectivity at Cu-allenylidene zwitterionic intermediates via a fluorine effect. Specifically, the Cu-catalyzed decarboxylative annulation of non-fluorinated 4-methyl (Me)-4-propargylic benzoxazinanones 3 with sulfur yields 2 furnished chiral non-racemic 3-Me-3-propargyl-indolines 5 in a g-selective fashion in good to high yield with high enantioselectivity (up to 91% ee;Scheme 1C). As examples of the generation of all-carbon quaternary stereocenters at the propargylic position are rare (Tsuchida et al., 2016;Sanz-Marco et al., 2016;Shemet and Carreira, 2017;Wendlandt et al., 2018;Zhang et al., 2018a;Li et al., 2019;Xu and Hu, 2019), the obtained results might help to activate the corresponding area of research. On the other hand, the a-selective addition was predominantly observed for the Cu-catalyzed decarboxylative annulation of fluorinated variants such as 4-trifluoromethyl (CF 3 )-4-propargylic benzoxazinanones 4 with 2, which led to the formation of 3-CF 3 -2-functionalized indoles 6 in good to high yield with high E/Z-selectively via a rare a-attack at the Cu-allenylidene zwitterionic intermediates (Scheme 1D). Given that CF 3 -containing N-heterocycles have gained considerable attention in academic and industrial research on pharmaceutics and agrochemicals (Kawai and Shibata, 2014;Engl et al., 2015;Huang et al., 2015;Meyer, 2016;He et al., 2019), CF 3 -substituted indoles 6 that contain 2-functional groups should represent versatile building blocks for the preparation of drug candidates. To the best of our knowledge, this is the first example of controlling the a/g chemoselectivity at Cu-allenylidene zwitterionic intermediates that does not depend on the interceptor.
Encouraged by these unprecedented preliminary results, we initially studied the enantioselective [4 + 1] cycloaddition reaction of 4-Me-propargyl benzoxazinanone 3a with sulfur ylide 2a (Scheme 3, Table 1). First, the effect of (R)-BINAP on this transformation was examined at room temperature under a variety of conditions (entries 1-4). However, the enantioselectivity of 5aa was only moderate (up to 44%; entry 2). Subsequently, we focused on the use of Pybox ligands for the improvement of the enantioselectivity in this transformation. After a careful evaluation of chiral ligands, Lewis acids, solvents, and substituents on sulfur ylides 2a (2a 0 ) (entries 5-16; Tables S1-S7), we found that the commercially available iso-propyl-substituted Pybox ligand L3 exhibited the best performance, producing chiral indoline 5aa in 72% yield with 74% ee (entry 10). More details of the screening of other ligands such as L5 and L6 are shown in the Supplemental Information (Table S1). An investigation into the solvent effect (Table S3) revealed that dichloromethane (DCM) provided the best reaction efficiency with a slightly lower yield and improved enantiocontrol (entry 12, 69% yield, 78% ee). An evaluation of different bases showed that N-ethyl morpholine was superior to other bases (entry 13, 84% yield, 82% ee). Gratifyingly, a more favorable outcome (85% ee) was observed without a significant decrease in yield when the reaction was carried out with 1.5 equiv. of 2a' (entry 15, 83% yield, 85% ee). In all these cases, >95:5 diastereoselectivity was confirmed by a 1 H NMR analysis of the crude reaction mixture. While the amount of N-ethylmorpholine can be reduced to a catalytic amount, the corresponding yield decreased slightly (79% yield, 85% ee, entry 16). The absolute configuration of 5aa, induced by L3, was determined to be 2(S) and 3(R) by a single-crystal X-ray diffraction analysis (CCDC1971179). The 2(S), 3(R)-stereochemistry of 5aa is a surprise, as we expected the configuration of 5aa to be 2(R),3(R) or 2(S),3(S) based on a previous report (Scheme 1A) (Wang et al., 2016). Ts group on 3a is important since the reaction of Boc-protected variant of 3a with 2a 0 under the same conditions resulted in a complex mixture.
To demonstrate the synthetic utility of the 3-propargyl indoline products 5, we carried out two subsequent transformations (Scheme 5). Optically active indoline 5aa was smoothly converted into triazole 7 via a 1,3dipolar cycloaddition with tosyl azide in the presence of CuTc. As expected, 7 was formed in 99% yield without any loss of enantiopurity (85% ee). Furthermore, a Sonogashira coupling of 5aa with iodobenzene afforded the disubstituted alkyne 8 in 70% yield under retention of its enantiopurity.
Furthermore, we examined the reaction conditions to generate the indole product 6 with major E isomer. As mentioned in Scheme 8, the formation of the indole product 6 (standard reaction condition) and E/Z isomerization were achieved in concerted manner (Scheme 8).

Proposed Reaction Mechanisms
Based on the observed experimental results and previous reports (Wang et al., 2016(Wang et al., , 2018a(Wang et al., , 2018b(Wang et al., , 2018cLi et al., 2016Li et al., , 2017Li et al., , 2018Song et al., 2017;Lu et al., 2017Lu et al., , 2018aShao and You, 2017;Chen et al., 2018;Jiang et al., 2018;Zhang et al., 2018aZhang et al., , 2018bZhang et al., , 2019Ji et al., 2018;Simlandy et al., 2019;Sun et al., 2019), we would like to propose a feasible mechanism to rationalize the chemo/stereoselective formation of indolines/indoles from 4-substituted 4-propargyl benzoxazinanones (3,4) with sulfur ylides 2 (2 0 ) ( Figure 1A). As described in Figure 1A, the Cu complex initially activates the propargyl benzoxazinanone (3a or 4a) in the presence of a base to generate CuÀacetylide A. Then, the Cu-allenylidene zwitterionic intermediate B, which is stabilized by its resonance form, is generated via an extrusion of CO 2 . Depending on the substitution pattern at the propargylic position of the Cu-stabilized allenylidene zwitterionic intermediate B, the sulfur ylide 2 attacks at the g-(X = Me) or a-position (X = CF 3 ). The Me-substitution at the propargylic position of transient species B allows sulfur ylide 2a to attack at the g-position (propargylic position) to generate intermediate C, which further converts into copper-containing cycloadduct D via an intramolecular SN 2 reaction. Finally, 3-Me-3-propargyl indoline 5aa is produced through a proton transfer under concomitant regeneration of the copper catalyst to close the catalytic cycle. The 2,3-cis-selectivity of alkyne and benzoyl groups in 5aa could be explained by the bulkiness of 4-methyl group (C sp3 group) rather than Although the reasons for the noticeable a/g-selectivity depend on the 4-substitution in 4-propargyl benzoxazinanones 3 (Me) and 4 (CF 3 ) remain obscure at present, the a/g-selectivity could potentially be rationalized in terms of stabilization and steric effects of the reactive intermediates. Specifically, the Cu-stabilized allenylidene zwitterionic intermediate B, which contains a Me group, has a resonance structure B-I, in which the carbocation is stabilized by the positive inductive (+I) effect of the Me group. Thus, nucleophilic 2 approaches the g-position of Cu-allenylidene intermediate B ( Figure 1B). In the case of 4a, however, the similar intermediate carbocation B-II, generated from the Cu-stabilized allenylidene zwitterionic intermediate B with a CF 3 group, is not stabilized by the strong electronwithdrawing effect of the CF 3 group, whereas the vinyl cation in intermediate B-III is stabilized by the additional resonance structure B-IV induced by the electron-withdrawing effect of the CF 3 substituent. Moreover, the g-attack should also be unfavorable owing to the steric demand of the bulky CF 3 group. All of the aforementioned aspects should favor the unprecedented a-attack ( Figure 1C).

Conclusion
In conclusion, we have constructed optically active indolines 5, which contain an all-carbon quaternary stereocenter, in good yield with high enantioselectivity from the decarboxylative [4 + 1] annulation of Me-propargyl benzoxazinanones 3 and sulfur ylides 2. Irrespective of the substituents on 3 and 2, the reaction yielded the corresponding indoline derivatives 5 with excellent enantioselectivity (up to 91% ee) via a g-attack on a Cu-allenylidene zwitterionic intermediate. Interestingly, the reaction between CF 3 -propargyl benzoxazinanones 4 and 2 delivered indole derivatives 6 in good yield via an unprecedented a-attack on the Cu-allenylidene zwitterionic intermediate. In their entirety, these results represent the first example of controlling two modes (a-versus g-attack) of decarboxylative annulation of propargyl benzoxazinanones via Cu-allenylidenes with the same interceptors. With respect to the importance for research in the area of N-containing heterocycles, enantio-enriched indolines with all-carbon quaternary propargyl stereogenic center and CF 3 -substituted indoles with a 2-functional group are both extremely useful precursors in medicinal chemistry. Further investigations into unique reaction patterns that are dominated by fluorine-containing groups and non-fluorinated groups are currently in progress in our laboratories.   Ji, D., Wang, C., and Sun, J. (2018). Asymmetric [4 + 2]-cycloaddition of copper-allenylidenes with hexahydro-1,3,5-triazines: access to chiral tetrahydroquinazolines. Org. Lett. 20, 3710-3713.
Figure S162. 13 C NMR spectrum of (E)-6aa, related to Scheme 7. Figure S163. 1 H NMR spectrum of (E)-6ca, related to Scheme 7. Figure S164. 1 H NMR spectrum of (E)-6ga, related to Scheme 7. Supplemental Table   Table S1. Ligand screening a , related to  (R)-DTBM-SEGPHOS ND <10 −87 10 (R)-SEGPHOS ND 23 −32 a Reactions were carried out with 3a (0.1 mmol), 2a' (0.2 mmol), Cu(OTf)2 (10 mol %), ligand (12 mol %), i-Pr2NEt (DIPEA, 1.2 equiv.) in THF at room temperature. b Determined by 1 H NMR analysis of the reaction mixture. c Determined by 1 H NMR analysis of the crude reaction mixture using 1,3,5-trimethoxybenzene as an internal standard. d The ee was determined by chiral HPLC analysis.  HFIP -NR a Reactions were carried out with 3a (0.1 mmol), 2a' (0.2 mmol), Cu(OTf)2 (10 mol %), ligand (12 mol %), i-Pr2NEt (1.2 equiv.) in THF at room temperature. b Determined by 1 H NMR analysis of the reaction mixture. c Determined by 1 H NMR analysis of the crude reaction mixture using 1,3,5-trimethoxybenzene as an internal standard. d The ee was determined by chiral HPLC analysis. N-Ethylmorpholine 3.0 >95:5 62 83 a Reactions were carried out with 3a (0.1 mmol), 2a' (0.2 mmol), Cu(OTf)2 (10 mol %), ligand (12 mol %), base (1.2 equiv.) in THF at room temperature. b Determined by 1 H NMR analysis of the reaction mixture. c Determined by 1 H NMR analysis of the crude reaction mixture using 1,3,5-trimethoxybenzene as an internal standard. d The ee was determined by chiral HPLC analysis. e without base. Reactions were carried out with 3a (0.1 mmol), 2 (0.2 mmol), Cu(OTf)2 (10 mol %), ligand (12 mol %), N-Ethylmorpholine (1.2 equiv.) in THF at room temperature. b Determined by 1 H NMR analysis of the reaction mixture. c Determined by 1 H NMR analysis of the crude reaction mixture using 1,3,5-trimethoxybenzene as an internal standard. d The ee was determined by chiral HPLC analysis.      a Fellow the general method J, the crude product 6ax was then filtered through a short pad of silica, the filtrate was concentrated for the next run. Fellow the literature procedure (Makarov et al., 2018), an oven-dried tube was charged with 6ax, Iodine (10 mol%) and AcOH. The tube was sealed, and the resulting solution was stirred at 100 o C for 3 h. The resulting solution were then taken 19 F NMR to give the corresponding isomer rate. The 19 F NMR spectrum were attached below.

General Information
All reactions were performed in oven-dried glassware under a positive pressure of nitrogen or argon. Solvents were transferred via syringe and were introduced into the reaction vessels through a rubber septum. All solvents were dried by standard method. All the reactions were monitored by thin-layer chromatography (TLC) carried out on 0.25 mm Merck silica gel (60-F254). The TLC plates were visualized with UV light. All the reaction products were purified by column chromatography and was carried out on a column packed with silica gel 60N spherical neutral size 50-63 mm. The 1 H NMR (300 MHz and 500 MHz) and 19 F NMR (282 MHz) spectra as for solution in CDCl3 and DMSO were recorded on a Varian Mercury 300 and BRUKER 500 Ultra Shield TR. 13 C NMR (125.8 MHz) spectra for solution in CDCl3 was recorded on a BRUKER 500 Ultra Shield TR. The chemical shifts (δ) are expressed in ppm downfield from internal TMS (δ = 0.00) and coupling constants (J) are reported in hertz (Hz). The hexafluorobenzene (C6F6) [δ = −162.2 (CDCl3)] was used as internal standard for 19 F NMR. The following abbreviations were used to explain the multiplicities: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, br = broad. Mass spectra were recorded on a SHIMADZU GCMS-QP5050A (EI-MS) and SHIMADZU LCMS-2020 (ESI-MS). High resolution mass spectrometry (HRMS) was carried out on an electron impact ionization mass spectrometer with a micro-TOF analyzer and recorded on a Waters, GCT Premier (EI-MS) with a TOF analyzer. Infrared spectra were recorded on a JASCO FT/IR-4100 spectrometer. Melting points were recorded on a BUCHI M-565. Optical rotations were measured on a SEPA-300 instrument (HORIBA Ltd, Kyoto, Japan). HPLC analyses were performed on a JASCOLC-2000 Plus series using 4.6 x 250 mm CHIRALPAK series.
Commercially available chemicals were obtained from Aldrich Chemical Co., Alfa Aesar, TCI and used as received unless otherwise noted. Solvents acetonitrile, ethyl acetate, ethanol, Dioxane, DMF, DCM and THF were dried and distilled before use.

3-(2-Aminophenyl)pent-1-yn-3-ol (S2g):
Following the general method A, compound S2g was obtained as a red oil (0.75 g, Yield: 86% reduced pressure. To the residue, water was added slowly and followed by extraction with ethyl acetate (3 X 30 mL). Combined organic layers were finally washed with brine solution, dried over anhydrous Na2SO4 and then solvent was removed under reduced pressure. The crude product was purified by flash column chromatography ((Hexane/Ethyl Acetate = 9:1)) to obtain the pure product S3. The characterization data of S3 are summarized below.

4-Ethynyl-4-isopropyl-1-tosyl-1,4-dihydro-2H-benzo[d][1,3]oxazin-2-one (3h):
Following the general method C, compound 3h was obtained as a white solid (  General procedure for the synthesis of substituted 1-(2-aminophenyl)-2,2,2-trifluoroethanones (Method D), related to Scheme 6. Route 1: The substituted 1-(2-aminophenyl)-2,2,2-trifluoroethanones (S8) were prepared according to the reported literature procedures with slight modification from the starting materials 2-nitrobenzaldehydes (S4) (Cheng et al., 2013;Punna et al., 2019;Kim et al., 2013 In a flame dried 100 mL round bottom flask, aldehyde S4 (20 mmol, 1.0 equiv.) and dry K2CO3 (0.552 g, 0.2 equiv.) was suspended in anhydrous DMF (25 mL). To this solution TMSCF3 (5.68 g, 2.0 equiv.) in 5 mL was added and the mixture was stirred vigorously at room temperature under N2 atmosphere. Completion of the reaction was monitored by TLC. To this reaction mixture, aqueous HCl solution (2 M, 4 mL) was added and stirred for 30 min at room temperature. The reaction mixture was then extracted with ethyl acetate. Combined organic layers were finally washed with brine solution, dried and concentrated under reduced pressure. Then purification by chromatography on a short silica gel column (Hexane/Ethyl Acetate = 9:1) to afford compound S6 as pure product. In a flame dried 100 mL round bottom flask, PDC (9.4 g, 2.5 equiv.) was suspended in anhydrous DCM (25 mL). To this solution Alcohol S6 (10 mmol, 1.0 equiv.) in 25 mL DCM was added and the mixture was stirred reflux under N2 atmosphere. Completion of the reaction was monitored by TLC. Filtered through a pad of celite to remove the solid, and then concentrated under reduced pressure. Purification by chromatography on a short silica gel column (DCM) to afford compound S7 as pure product.
In a 100 mL round bottom flask, ketone S7 (9.1 mmol, 1.0 equiv.), Iron powder (1.55 g, 3.0 equiv.) and NH4Cl (2.95 g, 6 equiv.) was added subsequently into 30mL H2O/EtOH (v/v=1:5). The mixture was stirred at 80°C for 2h. Completion of the reaction was monitored by TLC. Filtered through a pad of celite to remove the solid, and then extracted with DCM, dried and concentrated under reduced pressure. Purification by chromatography on a short silica gel column (DCM) to afford compound S8 as pure product.

Route 2:
The substituted 1-(2-aminophenyl)-2,2,2-trifluoroethanones (S8) were prepared according to the reported literature procedures with slight modification from o-amino benzoic acids as starting materials (S5) (Allendörfer et al., 2012 The Substituted o-amino benzoic acid S5 (10 mmol, 1.0 equiv.) was dissolved in toluene (50 mL), then Ac2O (2.84 mL, 3.0 equiv.) and NEt3 (4.18 mL, 3.0 equiv.) were added. The mixture was stirred for 15 h at 110 °C. The solvent was removed under reduced pressure after complete consumption of starting material. The residue was taken up with water and ethyl acetate (3:1) and phases were separated. The organic layer was dried over Na2SO4 and the solvent removed under reduced pressure. The product S9 was used immediately without further purification. Under argon atmosphere benzoxazinone S9 (9.17 mmol, 1.0 equiv.) was dissolved in dry DMSO. Trifluoromethylation reagent (4.0 mL, 3.00 equiv.) and TBAF (0.10 equiv., 1 M in THF) were added into the solution, and the mixture was stirred at rt for 15 h. After complete consumption of the starting material, the reaction mixture was quenched with 6 M HCl and stirred for an additional 1 h. Then, water was added, and the mixture was extracted with DCM. The organic layer was washed with saturated aq NH4Cl and brine, dried and the solvent was removed under reduced pressure. Column chromatography (DCM) of the crude product yielded the trifluoromethylated ketones S8. Fellow the general literature procedure with slight modification (Yasuhara et al., 1999), to a solution of trifluoromethylated ketones S8 (5 mmol, 1.0 equiv.) in 10mL pyridine was added slowly p-toluenesulfonyl chloride (2.39 g, 2.5 equiv.). The resulting mixture was stirred at 50 °C under N2 atmosphere. The mixture was evaporated to remove pyridine, quenched with water and extracted with DCM. The combined organic layer was washed with brine, then dried and concentrated. The crude residue was then dissolved in 15 mL dry THF, then TBAF (1.0 equiv., 1 M in THF) were added into the solution and keep the reaction at room temperature for 2 h under N2 atmosphere. Completion of the reaction was monitored by TLC. The mixture was quenched with water and extracted with DCM. The combined organic layer was washed with brine, dried over anhydrous Na2SO4, filtered, and concentrated. The crude residue was purified by silica gel column chromatography to give S10.

General procedure for the synthesis of perfluoroalkyl substituted 4-ethynyl-1-tosyl-1H-benzo[d][1,3]oxazin-2(4H)-ones (Method F), related to Scheme 6.
Overall reaction steps for the synthesis of trifluoromethyl substituted 4-ethynyl-1-tosyl-1H-benzo[d] [1,3]oxazin-2(4H)-ones 4a to 4h is showing below . Under a dry nitrogen atmosphere, 30 mL of dry THF was added to a 100 mL round bottom flask, followed by the ethynyltrimethylsilane (2.2 mL, 16 mmol). The solution was then cooled at −78 °C and 1.6 M n-butyllithium solution in THF (10.0 mL, 16 mmol) was then added dropwise by syringe. After stirring for 20 min, 4-methyl-N- (2-(2,2,2trifluoroacetyl)phenyl)benzenesulfon-amide (S10) (2.49 g, 7.24 mmol) in THF was added slowly to the reaction mixture for 30 min. The mixture was then keep stirring for 1 h, and then checked for conversion of sulfonamide by TLC. After the complete conversion of sulfonamide, triphosgene (2.6 g, 9.4 mmol) in 5 mL dry THF was added dropwise. The reaction mixture was then stirred for 2 h. Once full conversion of the intermediate was verified by TLC, the reaction was quenched with water slowly. The solution was then concentrated to remove THF, then extracted with DCM, and the combined organic layers dried with sodium sulfate then concentrated to afford a dark brown crude solid. The residue was undergoing a short silica pad then directly used for next step. Under a nitrogen atmosphere, the crude solid was added into a 100 mL round bottom flask and dissolved in 30 mL of dry THF and cooled at −78 °C. Tetrabutylammonium fluoride solution (1.0 M) in THF (8.5 mL, 6.9mmol) was then added dropwise, and reaction was then stirred for 30 min. After the reaction completed as checked by TLC, the reaction was quenched with water dropwise and warm to room temperature. The solution was then concentrated to remove THF, then extracted with DCM, and the combined organic layers dried, concentrated to afford a dark brown crude solid. Purification by column chromatography (hexane/ethyl acetate = 5:1) afforded the pure trifluoromethylated propargyl benzoxazinanones.

General experimental procedure for the preparation of sulfur ylides and sulfonium salts (Method H), related to Scheme 4 and Scheme 6.
Sulfur ylides 2 were prepared according to known methods. A typical experimental procedure for the preparation of sulfur ylides were described below.
2) TEA, EtOH, 0 °C, 2h S13 ( R = aryl. alkyl, heteroaryl.) 2) 10% NaOH 4-Methylthiophenol (1.0 equiv., 10.00 mmol, 1.24 g) was charged into a dry 100 mL flask along with ethanol (20 mL), magnetic stir bar and K2CO3 (1.0 equiv., 10.0 mmol, 1.38 g). The α-bromo ketone (1.0 equiv., 10.0 mmol) was added in one portion. The resulting suspension was stirred for 2h at room temperature. The crude reaction mixture was filtered through a pad of celite and washed with EtOH. The solvent was removed in vacuo. The residue was purified by flash silica gel chromatography (using 95:5 hexane/ethyl acetate). The resulting sulfide was transferred into a vial. In a glove box, Me2SO4 (1.0 equiv.) was added and the vial was sealed. The vial was stirred for 1 h at 100 °C and allowed to cool to room temperature. The resulting semi-solid was transferred to a flask, EtOH (99.9%, 1.0 M) added and the mixture cooled to 0 °C. Triethylamine (1.1 equiv.) was added and the reaction stirred 2 hours at 0 °C. The reaction mixture was transferred to a separatory funnel containing water and DCM. The phases were separated and the aqueous was extracted twice with DCM. The combined organic phases were washed with water and then dried over MgSO4. All solvent was removed in vacuo yielding a solid which further recrystallized from DCM and hexane. The characterization data of 2a'-2g' are summarized below, and sulfur ylides 2a-2i were prepared according to the known procedure, the characterization data are match with the previous data (Søren et al., 2012;Anderson et al., 1984;Ratts et al., 1966;Payne et al., 1967;Quintana et al., 1973).