The transient-chelating-group-controlled stereoselective Rh(i)-catalyzed silylative aminocarbonylation of 2-alkynylanilines: access to (Z)-3-(silylmethylene)indolin-2-ones

A new method involving mild acryl transient-chelating-group-controlled stereoselective Rh(i)-catalyzed silylative aminocarbonylation of 2-alkynylanilines with CO and silanes is presented for producing (Z)-3-(silylmethylene)indolin-2-ones. Upon using an acryl transient chelating group, 2-alkynylanilines undergo an unprecedented alkyne cis-silylrhodation followed by aminocarbonylation to assemble (Z)-3-(silylmethylene)indolin-2-ones. Mechanistic studies show that acryl transient chelating effects result in the key alkyne cis-silylrhodation process.


Introduction
Oxindoles, including methylene oxindoles, are a class of importantly coveted scaffolds for organic and medicinal chemistry purposes due to their omnipresence in natural products and biologically active molecules, and their widely established utilization as versatile synthetic building blocks. 1,2 In particular, the use of 3-methylene-indolinone scaffolds has already been successfully established for VEGFR, Trk A, CDK, and GSK3 kinase inhibition, antitumor, antibacterial, anti-inammatory, analgesic, and antimalarial applications ( Fig. 1). 1 As a result, developing efficient methods, especially stereoselective ones, for the synthesis of a diverse range of 3methylene-indolinones is unarguably critical for continued progress in the area of drug development. [3][4][5][6][7][8][9] Despite this growing demand, the stereoselective construction of the substituted methylene moiety of 3-methylene-oxindoles has been a longstanding challenge and, for these reasons, highly stereoselective preparation methods remain rare to date. Classical approaches for the assembly of methylene oxindoles mainly involve the intermolecular condensation of oxindoles with aryl carbonyl compounds, including diaryl ketones and aromatic formaldehydes, but these transformations face serious stereoselective control issues and substrate scope limitations. 1,3 To overcome these issues, transition-metal-catalyzed tandem annulation reactions with unsaturated hydrocarbons, 4 such as cross-coupling-enabled annulation cascades of N- (2haloaryl)propiolamides, 5 N-arylpropiolamides, 6 or 2-(alkynyl) arylisocyanates; 7 the carbonylative annulation of 2-alkynylanilines or 2-alkenylanilines; 8 the chloroacylation of alkynetethered carbamoyl chlorides; 9 and the cross-dehydrogenation coupling (CDC) of 2,3-diarylacrylamides or N-cinnamoylanilines, 10 have been developed. Common transition-metal catalysts (such as those containing Pd, Rh, Co 2 Rh 2 , and Ni) are efficient for use in these transformations to access various functionalized 3-methylene-oxindoles; however, the careful control of stereoselectivity sometimes remains a problem, with most congurations being unknown before the conclusion of the reaction. Moreover, reports detailing the deliberate control of stereoselectivity are dominated by the introduction of halogen atoms to construct 3-(halogenated methylene) scaffolds; as a result, there is an urgent need to discover conceptually novel stereoselectivity-control strategies for building diverse functionalized scaffolds other than halogenated ones. For example, our group has reported the palladium-catalyzed carbonylative annulation of 2-alkynylanilines with CO for producing 3-(halomethylene)-indolin-2-ones using stoichiometric CuX 2 (X ¼ Br, Cl) as both the halogen source and oxidant (Scheme 1a). 8a The stereoselectivity mainly depended on the substrate choice, and the assembly of (E)-3-(halomethylene)indolin-2-ones is limited to 2-(alkylalkynyl)-anilines and sterically bulky 2-(3-substituted arylalkynyl)anilines. Lautens, Schoenebeck, and coworkers disclosed the Pd(0)-catalyzed trans-selective intramolecular chloroacylation of alkynetethered carbamoyl chlorides for assembling (E)-3-(halomethylene)indolin-2-ones, in which both sterically bulky silyl alkynyl substituents (such as TIPS and TBS) and bulky phosphorus ligands (such as phenyl phosphaadamantanes (PA-Ph)) are necessary to precisely direct the stereoselectivity toward (E)-isomers. 9a,b Very recently, Lautens and coworkers found that the use of hexauoroisopropanol at high temperature (about 100 C) allowed for the cycloisomerization of alkyne-tethered carbamoyl chlorides to forge only (E)-3-(chloromethylene)oxindoles. 9d This method has the advantage of simple operation and stereospecicity under metal-free conditions, but it is not applicable to sterically bulky TIPS alkynyl substituents. The same group developed a Pd(II)-catalysis-based method to shi the stereoselectivity of the intramolecular chloroacylation of alkyne-tethered carbamoyl chlorides mainly toward the corresponding (Z)-isomers, with Z/E ratios ranging from 3.8 : 1 to >99 : 1. 9c Subsequently, they employed a similar Pd(II) catalysis strategy to allow the domino cyclization of alkyne-tethered carbamoyl chlorides with 2-ethynylanilines through linked C(sp 2 )-C(sp 2 ) bond stereospecic formation to access (Z)-3-((1H-indol-3-yl)methylene)indolin-2-ones. 9e By comparing these ndings, 8,9 steric hindrance effects and, especially, cooperative ligand/substrate coordination with transition-metal catalysts unarguably play important roles in the stereoselectivity control.
On that basis, we envisioned that if a transient chelating group 11 was present to coordinate with transition-metal catalysts, it may be possible to carefully control the corresponding stereoselectivity. Herein, we report a new method involving the acryl-transient-chelating-group-controlled stereoselective [Rh I (cod)Cl] 2 -catalyzed silylative aminocarbonylation of 2-alkynylanilines with CO and silanes, 12 enabling the synthesis of (Z)-3-(silylmethylene)indolin-2-ones in moderate to good yields and with >99 : 1 Z/E stereoselectivity (Scheme 1b). The method utilizes an in situ generated acryl group on the nitrogen atom as the transient chelating group to coordinate with the active Rh I species, thus resulting in unprecedented alkyne cis-silylrhodation followed by aminocarbonylation, providing (Z)-3-(silylmethylene)indolin-2-ones.

Results and discussion
We began to test our hypothesis that a transient chelating group could control the stereoselectivity during the silylative aminocarbonylation reaction with the use of N-(4-bromobenzyl)-2ethynylaniline 1a, CO, triethylsilane 2a, and acryloyl chloride 3a as starting materials (Table 1). In the presence of 2 mol% [Rh I (cod)Cl] 2 , 2 equiv. of K 2 CO 3 , and 1 equiv. of chloride 3a, the silylative aminocarbonylation of substrate 1a with CO (1 atm) and triethylsilane 2a at room temperature aer 18 h was efficiently performed, giving the desired (Z)-1-methyl-3-((triethylsilyl)methylene)indolin-2-one 4aa with 58% yield and >99 : 1 Z/E stereoselectivity ( Table 1, entry 1). However, omitting the chloride 3a led to a lower yield (30%) and stereoselectivity inversion (1 : 5 Z/E) ( Table 1, entry 2). Gratifyingly, the reaction could be efficiently executed to deliver 75% yield of 4aa in the absence of both K 2 CO 3 and chloride 3a, but the stereoselectivity was shied to 1 : 5 Z/E (Table 1, entry 3). Decreasing (0.5 equiv.) or increasing (1.5 equiv.) the amount of chloride 3a resulted in diminished yields (Table 1, entries 4 and 5). Both K 2 CO 3 and [Rh I (cod)Cl] 2 are essential for this reaction, since the omission of either resulted in no detectable desired product 4aa (Table 1, entries 6 and 11). A brief assessment of the effects of the loading of K 2 CO 3 and the effects of the base (K 2 CO 3 , Na 2 CO 3 , Cs 2 CO 3 , NaHCO 3 , or Et 3 N) revealed that the reaction with 2 equiv. of K 2 CO 3 afforded the best results ( With the optimized conditions in hand, we set out to further investigate the feasibility of this transient-chelating-groupbased strategy (Scheme 2). Directly using N-(4-bromobenzyl)-N-(2-ethynylphenyl)acrylamide 1b in a reaction with CO, silane 2a, [Rh(cod)Cl] 2 , and K 2 CO 3 afforded (Z)-4aa with a lower yield (22%) (Scheme 2, eqn (1)), whereas the omission of K 2 CO 3 increased the yield of (Z)-4aa to 43%. The results show that the base can improve the acylation process via the removal of chloride ions, but it suppresses the silylative aminocarbonylation. Furthermore, the in-situ-generated transient chelating group process is more efficient than the process involving the direct use of substrate 1b, probably because coordination effects relating to the acryloyl chloride may improve the catalytic activity of the Rh catalyst. Similarly, the treatment of N-(2-ethynylphenyl)-N-methylacrylamide 1c with CO, silane 2a, and [Rh(cod)Cl] 2 also afforded the (Z)-isomer 4ca in 80% yield (Scheme 2, eqn (2)); meanwhile acryloyl chloride was found to be the optimal transient-chelating-group reagent and it could efficiently allow the silylative aminocarbonylation of substrate 1d, stereoselectively assembling (Z)-4ca exclusively with a slightly increased yield (84%; Scheme 2, eqn (3), run 1). Using cinnamoyl chloride 3b in a reaction with the N-benzylsubstituted substrate 1e decreased the yield of (Z)-4ea to 41%, with 50% yield of the alkyne silylformylation product (E)-5eab (Scheme 2, eqn (3), run 2). Both acetyl chloride 3c and the formyl system 3d 13 were less reactive (Scheme 2, eqn (3), runs 3 and 4). Notably, the treatment of the formyl system 3d with substrate 1d, CO, and silane 2a mainly resulted in the alkyne silylformylation product (E)-5dad in 62% yield, with a lower yield (12%) of (Z)-4ca (Scheme 2, eqn (3), run 4). These ndings suggest that the cis-silyl vinyl-Rh intermediate may be initially formed via the cis-silylrhodation of the alkyne moiety, followed by the insertion of CO. However, methyl iodide 3e was inert (Scheme 2, eqn (3), run 5).
Scheme 3 The utilization of (Z)-4ea and control experiments.
A plausible mechanism for the silylative aminocarbonylation protocol was proposed (Scheme 4).  11,12 Intermediate D may undergo two pathways for the insertion of CO: 5c-f,8,12 One is the direct insertion of CO into the vinyl-Rh bond with the simultaneous formation of a N-Rh bond via the reductive loss of the acryl group with the aid of the base (K 2 CO 3 ) 12j,k to generate the carbonyl-Rh III -N six-membered ring intermediate F; the other involves the formation of the vinyl-Rh III -N ve-membered ring intermediate E through the reductive decomposition of the acryl C(sp 2 )-N bond with the aid of the base, 12j,k followed by the insertion of CO to generate the intermediate F. The reductive elimination of intermediate F results in the desired product (Z)-4 and regenerates the active Rh I species.
Using cinnamoyl chloride 3b as the transient chelating group may afford the alkyne cis-addition intermediate D 0 and the alkyne trans-addition intermediate D 00 due to steric hindrance and electron effects from the cinnamoyl group. 11,12 The alkyne cis-addition intermediate D 0 undergoes CO insertion, C-N bond cleavage, and N-Rh bond formation to afford the intermediate F, whereas the alkyne classic trans-addition intermediate D 00 may undergo hydroformylation with CO to form (E)-5eab. This is because the in situ generated cinnamyl C-N bond in the intermediate D 00 involving conjugative effects is more stable than the acryl C-N bond, leading to no cleavage of the cinnamyl C-N bond.

Conclusions
In summary, we have developed a novel strategy involving a mild acryl transient chelating group for the stereoselective rhodium(I)-catalyzed silylative aminocarbonylation of 2-alkynylanilines with CO and silanes, enabling the formation of (Z)-3-(silylmethylene)indolin-2-ones. The method involves the use of an acryl transient chelating group to enable the unprecedented cis-silylrhodation of alkynes and aminocarbonylation cascades to produce (Z)-3-(silylmethylene)indolin-2-ones; the highlights include exquisite stereoselectivity, a wide substrate scope, and excellent functional group tolerance. This acryl-transientchelating-group-controlled stereoselectivity strategy provides a conceptually novel approach for stereoselective transformations of unsaturated hydrocarbons and it could inspire the further development of new and efficient methods for stereoselective synthesis.

Data availability
Experimental data is provided in the ESI. †

Conflicts of interest
The authors declare no competing nancial interests.