Xiaohua Xie,
Linhai Jing*,
Dabin Qin,
Wujun He,
Song Wu,
Lunqiang Jin and
Gan Luo
Chemical Synthesis and Pollution Control Key Laboratory of Sichuan Province, China West Normal University, Nanchong, 637002, China. E-mail: jlhhxg@cwnu.edu.cn
First published on 11th February 2014
Enantioselective O-alkylation of α-nitrophosphonates was reported for the first time. A series of optically active 3,3-disubstituted oxindoles was prepared via a less exploited organocatalyzed stereoablative reaction of 3-bromooxindoles.
Recently, a new enantioselective stereoablative strategy was presented by Stoltz group for the first time.5 Followed the tactic, the same research group disclosed the first enantioselective stereoablative alkylation of 3-halooxindoles to access oxindoles with C3 all-carbon quaternary stereocenters via the base-promoted in situ generation of a putative o-azaxylylene.6a Subsequently, Yuan and co-workers utilized key o-azaxylylene intermediates to construct chiral 3,3-disubstituted oxindoles bearing a keto-carbonyl group or hydroxyl group.6b,c Very recently, Lu6d and Wang6e applied the strategy to synthesize 3-spirocyclopropyl-2-oxindoles and (+)-perophoramidine, respectively. In spite of these remarkable advances, the enantioselective stereoablative strategy, especially using 3-halooxindoles as electrophilic partners, is still in its infancy. Given that 3,3-disubstituted oxindole structural skeletons are widely found in a large number of alkaloid natural products and pharmaceutically relevant compounds,7 and it has been documented that the modification of oxindole at C3 position is very crucial for the enhancement of the biological activity,8 we envisioned that the combination of oxindole motifs with oxygen atoms of α-nitrophosphonates for the formation of structurally diverse new oxindoles should be highly promising. As our continuous efforts for the asymmetric stereoablative tactic,9 herein we wish to demonstrate, to the best of our knowledge, the first highly region- and enantioselective stereoablative O-alkylation reaction of α-nitrophosphonate 1 with putative o-azaxylylene, generated in situ from 3-halooxindoles under chiral tertiary amine–squaramide catalysts (Scheme 1b).
To commence our study, a series of bifunctional catalysts 3 were used to evaluate their abilities to promote stereoablative O-alkylation of α-nitrophosphonate 1a with 3-benzyl-3-bromooxindole 2a in CHCl3 at 30 °C in the presence of K2CO3. As expected, all tested catalysts could smoothly afford the desired O-alkylation product 4a (Table 1). Surprisingly, the widely used tertiary amine–thiourea catalysts 3a–3c (ref. 10) gave almost racemic 4a albeit with good yields and moderate Z/E ratios (entries 1–3). Triggered by the recent advance of some organocatalysts, our attention turned to bifunctional tertiary amine–squaramide catalysts.11 Several cinchona alkaloid or cyclohexanediamine and 3,5-dis(trifluoromethyl)aniline-derived squaramide catalysts 3d–3g were first screened (entries 4–7). To our delight, quinidine-based 3g afforded the desired product with 55% ee, 83:
17 Z/E ratio and 87% yield (entry 7). Encouraged by this result, another amine component of catalyst was surveyed (entries 8–12). Gratifyingly, the catalyst 3j derived from quinidine and 3,5-bis(trifluoromethyl)benzylamine could give the best result in terms of enantioselectivity (entry 10). Further studies showed that the base played a significant effect on the reaction result (entries 13–17), and Na3PO4·12H2O was identified as the best base source in terms of Z/E ration and enantioselectivity (entry 16). After screening a set of solvents, no better results were attained compared with CHCl3 (entries 18–21). Lowering reaction temperature or adding molecular sieves did not also offer further improvement (entries 22–25). To improve the reaction results, several new cinchona alkaloid and dehydroabietylamine-derived12 squaramide catalysts were synthesized and applied to the current reaction in the presence of the above optimized base, solvent and temperature. However, to our disappointment, more inferior results were observed (Fig. 1).
Entry | Cat. | Base | Solvent | Yieldb (%) | Z/Ec | eed (%) |
---|---|---|---|---|---|---|
a Unless otherwise noted, the reactions were performed using 1a (0.11 mmol), 2a (0.1 mmol), 3 (10 mol%) and base (0.11 mmol) in solvent (1 mL) at 30 °C for 4 h.b Total yield of two isomers.c Determined by 1H NMR of the crude mixture.d The ee of Z isomer was determined by chiral HPLC.e At 0 °C for 12 h.f At −20 °C.g 3 Å MS was used.h 4 Å MS was used. | ||||||
1 | 3a | K2CO3 | CHCl3 | 74 | 58![]() ![]() |
<5 |
2 | 3b | K2CO3 | CHCl3 | 77 | 66![]() ![]() |
<5 |
3 | 3c | K2CO3 | CHCl3 | 76 | 70![]() ![]() |
<5 |
4 | 3d | K2CO3 | CHCl3 | 75 | 72![]() ![]() |
−21 |
5 | 3e | K2CO3 | CHCl3 | 81 | 76![]() ![]() |
−38 |
6 | 3f | K2CO3 | CHCl3 | 73 | 75![]() ![]() |
34 |
7 | 3g | K2CO3 | CHCl3 | 87 | 83![]() ![]() |
55 |
8 | 3h | K2CO3 | CHCl3 | 66 | 71![]() ![]() |
55 |
9 | 3i | K2CO3 | CHCl3 | 70 | 81![]() ![]() |
20 |
10 | 3j | K2CO3 | CHCl3 | 86 | 76![]() ![]() |
69 |
11 | 3k | K2CO3 | CHCl3 | 77 | 86![]() ![]() |
22 |
12 | 3l | K2CO3 | CHCl3 | 85 | 88![]() ![]() |
<5 |
13 | 3j | Na2CO3 | CHCl3 | 64 | 78![]() ![]() |
84 |
14 | 3j | NaHCO3 | CHCl3 | — | — | — |
15 | 3j | K3PO4·3H2O | CHCl3 | 94 | 78![]() ![]() |
71 |
16 | 3j | Na3PO4·12H2O | CHCl3 | 64 | 81![]() ![]() |
84 |
17 | 3j | DBU | CHCl3 | 72 | 64![]() ![]() |
14 |
18 | 3j | Na3PO4·12H2O | DCM | 87 | 76![]() ![]() |
26 |
19 | 3j | Na3PO4·12H2O | DCE | 86 | 84![]() ![]() |
27 |
20 | 3j | Na3PO4·12H2O | THF | 80 | 76![]() ![]() |
30 |
21 | 3j | Na3PO4·12H2O | Toluene | 70 | 87![]() ![]() |
47 |
22e | 3j | Na3PO4·12H2O | CHCl3 | 59 | 81![]() ![]() |
84 |
23f | 3j | Na3PO4·12H2O | CHCl3 | — | — | — |
24g | 3j | Na3PO4·12H2O | CHCl3 | 80 | 81![]() ![]() |
65 |
25h | 3j | Na3PO4·12H2O | CHCl3 | 92 | 81![]() ![]() |
52 |
With the optimal reaction conditions in hand, the generality of the asymmetric stereoablative O-alkylation reaction was investigated by using a series of α-nitrophosphonates 1 and 3-bromooxindoles 2, and the results are summarized in Table 2. It was clearly observed that all tested substrates affored the desired O-alkylation products 4 in high yields with good to excellent enantioselectivities. Moreover, aryl-substituted α-nitrophosphonates 1b–1f gave higher Z/E selectivities than alkyl-substituted 1a, this is perhaps because of the π–π interaction between reaction substrates. Based on the result, phenyl-substituted 1b was used to evaluate the effect of structural change of 2 on the reaction. As seen from Table 2, the position of the substituent on the phenyl ring of R1 has an important influence on the enantioselectivity of the reaction, and ortho-substituent seems to be more beneficial than para- or meta-substituent. For example, 2-OMe-substituted 2d gave 91% ee, whereas only 81% and 82% ee values were obtained in the cases of 4-OMe-substituted 2b and 3-OMe-substituted 2c, respectively (entry 5 vs. entries 3 and 4). Similar phenomena were also observed in the reactions of 2h compared with 2f and 2g (entry 9 vs. entries 7 and 8), and 2l compared with 2j and 2k (entry 13 vs. entries 11 and 12). The electronic property of the substituent on the phenyl ring of R1 has also been found to show a significant effect on the enantioselectivity, and electron-donating group gave better enatioselectivity than electron-withdrawing group (entries 3, 7 and 10 vs. entries 11 and 14; entries 4 and 8 vs. entry 12; entries 5 and 9 vs. entry 13). Particularly, heterocycle substituted 2n, alkyl-substituted 2o and vinyl-substituted 2p were also suitable substrates, 84%, 90% and 70% ee values could be obtained, respectively (entries 15–17). In addition, substrates 2q and 2r with different substituents at different positions on aromatic ring of 3-bromooxindoles could also be tolerated (entries 18 and 19). To our delight, several newly synthesized aryl-substituted α-nitrophosphonates 1c–1f could also be applied to this asymmetric process, and good to excellent enantioselectivities could be attained (entries 20–24). It is noteworthy that the reaction reported herein could be performed in a relatively large-scale without obvious loss of the stereoselectivities and reactivity (Scheme 2).
Entry | R | R1 | R2 | 4 | Yieldb (%) | Z/Ec | eed |
---|---|---|---|---|---|---|---|
a Unless otherwise noted, the reactions were performed using 1 (0.11 mmol), 2 (0.1 mmol), 3j (10 mol%) and Na3PO4·12H2O (0.11 mmol) in CHCl3 (1 mL) at 30 °C for 4 h.b Total yield of two isomers.c Determined by HPLC.d The ee of Z isomer was determined by chiral HPLC. | |||||||
1 | Me (1a) | Ph (2a) | H | 4a | 64 | 81![]() ![]() |
84 |
2 | Ph (1b) | Ph (2a) | H | 4b | 75 | 92![]() ![]() |
75 |
3 | Ph (1b) | 4-OMe–Ph (2b) | H | 4c | 68 | 90![]() ![]() |
81 |
4 | Ph (1b) | 3-OMe–Ph (2c) | H | 4d | 73 | 92![]() ![]() |
82 |
5 | Ph (1b) | 2-OMe–Ph (2d) | H | 4e | 73 | 91![]() ![]() |
91 |
6 | Ph (1b) | 3,4-DiOMe–Ph (2e) | H | 4f | 61 | 91![]() ![]() |
81 |
7 | Ph (1b) | 4-Me–Ph (2f) | H | 4g | 80 | 92![]() ![]() |
83 |
8 | Ph (1b) | 3-Me–Ph (2g) | H | 4h | 65 | 91![]() ![]() |
88 |
9 | Ph (1b) | 2-Me–Ph (2h) | H | 4i | 80 | 94![]() ![]() |
92 |
10 | Ph (1b) | 4-tBu–Ph (2i) | H | 4j | 84 | 93![]() ![]() |
80 |
11 | Ph (1b) | 4-Cl–Ph (2j) | H | 4k | 71 | 91![]() ![]() |
73 |
12 | Ph (1b) | 3-Cl–Ph (2k) | H | 4l | 68 | 91![]() ![]() |
73 |
13 | Ph (1b) | 2-Cl–Ph (2l) | H | 4m | 75 | 92![]() ![]() |
84 |
14 | Ph (1b) | 4-F–Ph (2m) | H | 4n | 68 | 92![]() ![]() |
71 |
15 | Ph (1b) | 2-Thienyl (2n) | H | 4o | 72 | 93![]() ![]() |
84 |
16 | Ph (1b) | tBu (2o) | H | 4p | 71 | 92![]() ![]() |
90 |
17 | Ph (1b) | CH2![]() |
H | 4q | 80 | 81![]() ![]() |
70 |
18 | Ph (1b) | 2-Me–Ph (2q) | 6-Cl | 4r | 70 | 92![]() ![]() |
75 |
19 | Ph (1b) | 2-Me–Ph (2r) | 5-OMe | 4s | 64 | 91![]() ![]() |
83 |
20 | 4-Cl–Ph (1c) | 2-Me–Ph (2h) | H | 4t | 80 | 93![]() ![]() |
81 |
21 | 4-Cl–Ph (1c) | 4-tBu–Ph (2i) | H | 4u | 75 | 92![]() ![]() |
70 |
22 | 3-Cl–Ph (1d) | 2-Me–Ph (2h) | H | 4v | 68 | 93![]() ![]() |
71 |
23 | 4-Me–Ph (1e) | 2-Me–Ph (2h) | H | 4w | 68 | 92![]() ![]() |
82 |
24 | 3-Me–Ph (1f) | 2-Me–Ph (2h) | H | 4x | 72 | 92![]() ![]() |
90 |
The absolute configuration of the O-alkylation product 4u was established to be R by utilizing single-crystal X-ray diffraction (Fig. 2),13 and the absolute configurations of other products were assigned by analogy.
Based on these experimental results, a plausible bifunctional transition state was proposed. The aliphatic tertiary amine unit of the catalyst acts as a Brønsted base to deprotonate and activate α-nitrophosphonate via hydrogen bonding. Meanwhile, the squaramide moiety of the catalyst serves as a Brønsted acid to interact with the prochiral o-azaxylylene intermediate by double hydrogen bonds, as shown in Scheme 3.
In conclusion, the first unprecedented regioselective asymmetric O-alkylation reaction of α-nitrophosphonate has been realized. By means of a less exploited stereoablative strategy, a series of optically active 3,3-disubstituted oxindoles have been prepared in the presence of chiral bifunctional tertiary amine–squaramide catalysts via o-azaxylylene intermediates generated in situ from 3-bromooxindoles. Mechanistic studies and more catalytic asymmetric O-alkylation reactions of α-nitrophosphonate are underway.
Footnote |
† Electronic supplementary information (ESI) available: Experimental procedure, characterisation data. CCDC 965127. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c3ra47100d |
This journal is © The Royal Society of Chemistry 2014 |