Ze-Min
Lai‡
,
Ying
Xie‡
,
Le-Le
Huang
,
Jing
Guo
* and
Gui
Lu
*
Guangdong Provincial Key Laboratory of Chiral Molecule and Drug Discovery, State Key Laboratory of Anti-Infective Drug Development, School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou 510006, P. R. China. E-mail: 1043397061@qq.com; lugui@mail.sysu.edu.cn
First published on 23rd January 2025
α-C chiral sulfones are privileged building blocks widely found in pharmaceuticals, agrochemicals, natural products, and ligands. Although many nucleophilic or electrophilic protocols have been developed for their construction, radical-based asymmetric catalysis, especially that involving photoactive electron donor–acceptor (EDA) complexes, remains a significant unmet challenge. Herein, we present the first catalytic asymmetric production of α-C chiral sulfones enabled by merging a photoactive EDA complex with a chiral Ni catalyst. With this cooperative asymmetric catalysis system, a wide range of α-C chiral sulfones are achieved in good yields with excellent enantioselectivities (53 examples, up to 99% yield, 99:
1 er). The synthetic utility of this protocol is further demonstrated by the first asymmetric synthesis of the selective MMP-3 (stromelysin-1) inhibitor. Detailed mechanistic and spectroscopic studies suggest that a newly identified type of EDA complex generated from sulfonyl chlorides and Hantzsch esters (HEs) is crucial to the success as a precursor of sulfonyl radicals.
In the past decade, the discovery of more generic synthetic paradigms has been a major goal in the field of catalytic asymmetric photocatalysis.42–45 As a result, the EDA complex activation strategy has emerged as a useful tool in synthetic chemistry.46–53 In many cases, this strategy is generally characterized by its mild reaction conditions, operational simplicity, and lower toxicity in the absence of external photocatalysts. Two Li groups respectively employed this approach to racemic sulfonylation reactions, using sulfonyl chlorides as acceptors for alkene sulfonylation.54,55 However, due to the high reactivity of radicals and the inherent significant racemic background reactions, catalytic asymmetric sulfonylation processes facilitated by a photoactive EDA complex remain a tough task. This not only limits the diversity of asymmetric sulfonylation reactions, but also highlights the importance of designing new asymmetric EDA platforms that can facilitate the formation and asymmetric transformations of radicals. Hantzsch esters (HEs) have recently been used as efficient bifunctional electron donors due to their high reduction potential and the hydrogen atom transfer (HAT) nature of the resulting dihydropyridine radical cation.56 Because of their high oxidation potential, commercially available and cheap sulfonyl chlorides (TsCl, Ered = −0.87 V vs. saturated calomel electrode) can produce sulfonyl radicals via single-electron transfer.57,58 Inspired by these observations, we hypothesized that a new class of EDA complexes providing access to sulfonyl radicals in a way that differs from the traditional generation method of photochemistry may be used to stimulate hitherto underexplored asymmetric sulfonylation reactions. Thus, we have designed a new catalytic system in which EDA complexes containing potential sulfonic radicals may be formed via ground-state molecular association between HEs and sulfonyl chlorides in the absence of an external photo-redox catalyst. Under visible light, the EDA complex triggers an intra-complex SET event, resulting in transient sulfonyl radical intermediates that add to alkenes and afford the desired α-C chiral sulfones. A suitable chiral Ni catalyst acts as a Lewis acid, not only activating α,β-unsaturated alkenes but also providing a suitable chiral environment for efficient stereoselective control. By taking advantage of this EDA complex and cooperation with chiral Ni catalysis, we have realized the asymmetric installation of alkyl and aryl sulfone moieties to afford optically enriched α-C chiral sulfones (Fig. 1C). However, this approach requires solving other issues, such as the instability of the bifunctional electron donor in the catalytic cycle, the incompatibility of asymmetric induction with EDA complex activation, and the limited substrate scope.
Entry | 2 | [Ni] | Ligand | Solvent | HE | Yieldb (%) | Erc (%) |
---|---|---|---|---|---|---|---|
a Reaction conditions: 1a (0.2 mmol), 2 (0.1 mmol), HE (0.2 mmol), Ni catalyst (0.01 mmol), ligand (0.012 mmol), DIPEA (0.2 mmol), solvent (2 mL) under two 24 W 400 nm LEDs at 30 °C, 24 h. b NMR yield using styrene as internal standard. c Determined by HPLC analysis on a chiral stationary phase. d Isolated yield. e 36 h. f DIPEA (0.05 mmol). g No light. | |||||||
1 | 2a | NiBr2 | L1 | DCM | HE-1 | 80 | 98![]() ![]() |
2 | 2a | NiBr2 | L2 | DCM | HE-1 | 80 | 93![]() ![]() |
3 | 2a | NiBr2 | L3 | DCM | HE-1 | 82 | 5![]() ![]() |
4 | 2a | Ni(acac)2 | L1 | DCM | HE-1 | 79 | 97![]() ![]() |
5 | 2a | Ni(OTf)2 | L1 | DCM | HE-1 | 70 | 96![]() ![]() |
6 | 2a | NiCl2 | L1 | DCM | HE-1 | 67 | 97![]() ![]() |
7 | 2a | NiBr2 | L1 | THF | HE-1 | 20 | 87![]() ![]() |
8 | 2a | NiBr2 | L1 | DCE | HE-1 | 81 | 96![]() ![]() |
9e | 2a | NiBr2 | L1 | DCM | HE-1 | 85 | 98![]() ![]() |
10e,f | 2a | NiBr2 | L1 | DCM | HE-1 | 93 (90d) | 98![]() ![]() |
11e,f | 2a | NiBr2 | L1 | DCM | HE-2 | 68 | 96![]() ![]() |
12e,f | 2a | NiBr2 | L1 | DCM | HE-3 | 83 | 97![]() ![]() |
13e,f | 2a1 | NiBr2 | L1 | DCM | HE-1 | 85 | 92![]() ![]() |
14e,f | 2a2 | NiBr2 | L1 | DCM | HE-1 | 40 | 50![]() ![]() |
15e,f | 2a3 | NiBr2 | L1 | DCM | HE-1 | NR | — |
16e,f | 2a | — | — | DCM | HE-1 | 17 | — |
17e,f | 2a | NiBr2 | L1 | DCM | — | — | — |
18e,f,g | 2a | NiBr2 | L1 | DCM | HE-1 | — | — |
Having identified the optimal reaction conditions, we next investigated the substrate scope of this asymmetric sulfonylation reaction with various sulfonyl chlorides and α,β-unsaturated N-acylpyrazoles. As shown in Table 2, a wide range of sulfonyl chlorides underwent this asymmetric radical addition reaction smoothly to achieve the desired products in good yields and excellent enantioselectivities. For the commercially available aryl sulfonyl chlorides, various functional groups (–Me, –OMe, –Et, –F, –Cl, –Br, –I) and even strong electron-withdrawing groups (–CF3) substituted at the para position of the aryl moiety were found to be well tolerated and give the corresponding products 3a–3i in good yields with 94:
6–99
:
1 er values. The absolute structure of 3a was unambiguously determined by X-ray single-crystal analysis.59 Other aryl sulfonyl chlorides bearing electron-withdrawing or electron-donating substituents at the meta or the ortho position of the aryl ring also showed good compatibility in this transformation, providing the desired products 3j–3n in 77–91% yields with 91
:
9–93
:
7 er values. Moreover, fused- and heteroaryl sulfonyl chlorides afforded the related α-C chiral sulfone products 3o–3p in satisfactory yields and enantioselectivities. Remarkably, the introduction of a linear, cyclic, or branched alkyl sulfonyl chloride to deliver the desired products (3q–3t) also proved successful.
Next, we turned our attention to further investigating the scope of various α,β-unsaturated N-acylpyrazoles bearing different substituents. In general, the position and electronic nature had limited effect on the enantioselectivity. α,β-Unsaturated N-acylpyrazoles bearing electron-neutral (–H), electron-withdrawing (–Br, –F, –Cl, –CN, –COOMe, –CF3), and electron-donating (–Me, –OMe) groups on the phenyl ring all worked well to afford the corresponding products 3u–3ab in good yields (77–90%) with high enantioselectivities (86.5:
13.5–97
:
3 er). For α,β-unsaturated N-acylpyrazoles with ortho- and meta-substituents on the phenyl ring (3ac–3ag), both electron-deficient and electron-rich arenes were also compatible in this transformation, providing good yields (80–93%) and high stereoselectivities (92
:
8–95
:
5 er). The sulfonyl of disubstituted α,β-unsaturated N-acylpyrazole also proceeded smoothly to give the desired products in 85% yield with 95
:
5 er. Moreover, a wide range of heteroaryl groups, such as furan (3ai) and thiophene (3aj) could be introduced into the α,β-unsaturated N-acylpyrazole, resulting in equally good results (78% yield, 95
:
5–95.5
:
4.5 er). Given the potential for wide applications, determining whether α,β-unsaturated N-acylpyrazoles with a variety of alkyl groups substituted at the β position would still perform well in this asymmetric C–S bond forming reaction is also important. To our delight, not only acyclic alkyl groups (3ak–3an), especially the branched alkyl group (3ao), but also cycloalkyl groups including three- (3ap), four- (3aq), and six- (3ar–3as) membered rings, were all compatible with regard to yields (70–99%) and enantioselectivities (92
:
8–95
:
5 er). Furthermore, the good functional group compatibility and excellent enantioselectivities of this method inspired us to look into its potential for the functionalization of (−)-10-camphorsulfonyl chloride and diverse pharmaceutical and natural products bearing a Michael acceptor. As shown in Table 2, both (−)-10-camphorsulfonyl chloride (3at) and a series of complex molecules derived from (−)-menthol (3au), diacetone-D-glucose (3av), diacetone-fructose (3aw), estrone (3ax), pregnenolone (3ay), and cholesterol (3az) could be effectively employed as Giese-addition substrates. This allows for the straightforward synthesis of complex drug and natural product derivatives containing chiral α-C sulfones with multiple stereocenters in good yields and excellent diastereoselectivities (>20
:
1 dr). These results together highlight the practicality and robustness of this transformation, as well as its potential for wide-ranging applications in synthetic and medicinal chemistry.
The efficiency and convenience of this asymmetric sulfonylation reaction were further demonstrated in a short synthesis of enantiomerically pure γ-trifluoromethyl γ-sulfone hydroxamate (5a) as a selective MMP-3 (stromelysin-1) inhibitor for the treatment of heart failure and cancer therapy, which was previously synthesized in six steps with 17% overall yield. In this synthesis, chromatographic separation was required to form the key enantiomerically pure intermediate.60,61 Our asymmetric sulfonylation approach employed a commercially available sulfonyl chloride (1b) to produce α-C chiral sulfone intermediate 3ba in 85% yield with 92:
8 er, and the er could be further improved to 96
:
4 after recrystallization. Treatment of 3ba with 6 M HCl then provided the corresponding (S)-4,4,4-trifluoro-3-((4-methoxyphenyl)sulfonyl) butanoic acid (4a) in 95% yield. Subsequent amidation and hydrogenation led to the MMP-3 inhibitor 5a in 60% yield with 95
:
5 er. This unique four-step route not only achieved the first asymmetric catalytic synthesis of a selective MMP-3 inhibitor, but also represented a significant simplification and improvement to 48% overall yield (Scheme 1A). In addition, a scaled-up experiment was performed using 1a and 2a (Scheme 1B). The reaction proceeded smoothly and generated 3a in a slightly decreased yield (81% yield) but with excellent er (97
:
3); 3a can be converted into the corresponding amide (6) or ester (7) by substitution of the pyrazole moiety with an amino or an ethoxy group, and an alcohol derivative (8) was obtained with the aid of NaBH4.
To shed light on the mechanism of this visible light-induced asymmetric sulfonylation reaction, we then performed control experiments. Firstly, a radical-trapping experiment was conducted in the presence of 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) under standard conditions. The desired product 3a was not observed, but the TEMPO-coupling product 9a was detected by HRMS, which implied possible involvement of sulfonyl radicals in this reaction (Scheme 1C). Next, we carried out control experiments to explain the source of the sulfonyl radicals. When 1a was irradiated by visible light, no reaction was observed, and a large amount of 1a was recycled. A similar outcome occurred when Ni/L was added to catalyze the reaction. In contrast, when 1a and HE-1 were subjected to the reaction conditions without the Ni/L system, both sulfonyl dimer and Ox-HE were detected (Scheme 1D), which unambiguously demonstrated that a new EDA complex might be generated between electron-deficient sulfonyl chloride and electron-rich HE-1, proving the formation of plausible sulfonyl radicals via SET events. Moreover, the reaction progress of 1a, 2a, and HE-1 was probed over time under the standard conditions. While the yield increased gradually over the course of the reaction, the er value remained constant at around 98:
2 (Scheme 1E1). An on/off visible light irradiation experiment showed that this reaction did not proceed during the “dark” period (Scheme 1E2). Furthermore, we analyzed the reaction components using UV/vis absorption spectroscopy. The individual absorption spectra of sulfonyl chloride (1a), DIPEA, and α,β-unsaturated N-acylpyrazoles (2a) in DCM were located in the ultraviolet region, whereas HE-1 displayed absorption in the visible light region (Scheme 1F2, orange line). In contrast, a significant bathochromic shift was observed with the mixture of sulfonyl chloride 1a and HE-1 in DCM (Scheme 1F2, red line), which demonstrated that the mixture had formed a new molecular aggregate, that is, a colored EDA complex. Preliminary studies on mixing different concentrations of HE-1 with 1a revealed an association constant of 5.52 M−1 for 1a/HE-1 through the Benesi–Hildebrand method,62 suggesting plausible EDA complex formation prior to homolytic fragmentation (Scheme 1F3 and ESI†). The analysis of this EDA complex via the Yoe and Jones method demonstrated that 1
:
1 stoichiometry was the most effective absorption ratio (Scheme 1F4).63–65
Based on the abovementioned experiments, we envisioned a possible mechanism for this visible-light-induced asymmetric sulfonylation (Scheme 2). The EDA complex was initially formed with the electron-deficient sulfonyl chloride and the electron-rich HE-1via π–π stacking. Under visible light irradiation, an intracomplex SET process from HE-1 to sulfonyl chloride was triggered to produce a dihydropyridine radical cation, chlorine anion, and sulfonyl radical. At the same time, the chiral nickel catalyst underwent ligand exchange with α,β-unsaturated N-acylpyrazole to obtain an intermediate complex, which reacted with the sulfonyl radical via asymmetric radical Giese addition. Subsequently, the resulting radical intermediate was quenched by the dihydropyridine radical cation through a hydrogen atom transfer (HAT) process to yield final product 3.
Footnotes |
† Electronic supplementary information (ESI) available: Procedures, optimization, characterization data, HPLC and NMR spectra. CCDC 2237566. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4sc07264b |
‡ Authors with equal contribution. |
This journal is © The Royal Society of Chemistry 2025 |