Open Access Article
Wen-Run
Zhu‡
,
Qiong
Su‡
,
Xiao-Yi
Deng
,
Jia-Sheng
Liu
,
Tao
Zhong
,
Shan-Shui
Meng
,
Ji-Tao
Yi
,
Jiang
Weng
* and
Gui
Lu
*
Guangdong Provincial Key Laboratory of Chiral Molecule and Drug Discovery, School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou, 510006, P. R. China. E-mail: wengj2@mail.sysu.edu.cn; lugui@mail.sysu.edu.cn
First published on 23rd November 2021
A highly general and straightforward approach to access chiral bis(indolyl)methanes (BIMs) bearing quaternary stereocenters has been realized via enantioconvergent dehydrative nucleophilic substitution. A broad range of 3,3′-, 3,2′- and 3,1′-BIMs were obtained under mild conditions with excellent efficiency and enantioselectivity (80 examples, up to 98% yield and >99
:
1 er). By utilizing racemic 3-indolyl tertiary alcohols as precursors of alkyl electrophiles and indoles as C–H nucleophiles, this organocatalytic strategy avoids pre-activation of substrates and produces water as the only by-product. Mechanistic studies suggest a formal SN1-type pathway enabled by chiral phosphoric acid catalysis. The practicability of the obtained enantioenriched BIMs was further demonstrated by versatile transformation and high antimicrobial activities (3al, MIC: 1 μg mL−1).
Bis(indolyl)methanes (BIMs) are privileged structural motifs in natural products and bioactive molecules with significant antibacterial, anticancer, and anti-inflammatory activities (Scheme 1a).9 Therefore, the development of efficient approaches for BIMs have drawn much attention from the synthetic community. However, these established methods are mainly focused on the construction of symmetrical BIMs bearing tertiary stereocenters,10,11 and the asymmetric synthesis of enantioenriched BIMs bearing quaternary carbon centers still remain less explored.12,13 Moreover, existing synthetic methods are largely limited to symmetrical and unsymmetrical 3,3′-BIMs. In contrast, other types of BIMs (such as 3,1′-, 3,2′- and 3,n′-BIMs) received less attention and their asymmetric synthesis using the same strategy has remained unstudied, although these BIMs exhibit potential biological activities and deserve special attention as well. Therefore, a broadly applicable synthetic strategy for a wide variety of chiral BIMs would be of high importance.
Given the significance of both the dehydrative nucleophilic substitution and bis(indolyl)methanes (BIMs), herein we report the direct SN1-type dehydrative nucleophilic substitution of racemic 3′-indolyl tertiary alcohols with various indole nucleophiles catalyzed by chiral phosphoric acids, providing various BIMs bearing quaternary stereogenic carbon centers with good to excellent enantioselectivities; remarkably, we establish a unified strategy for the first time to enable the asymmetric synthesis of a wide range of enantioenriched 3,3′-, 3,2′- and 3,1′-BIMs under mild conditions (Scheme 1b). We further illustrate that the BIM products could be transformed into other valuable enantioenriched molecules bearing all-carbon quaternary centers such as chiral triarylmethanes that are difficult to access. Moreover, preliminary biological activities demonstrate that these chiral BIMs are potential antibacterial agents with MIC (minimum inhibitory concentration) of up to 1 μg mL−1 against Bacillus subtilis.
:
36 er (Table 1, entry 1). Next, a series of BINOL- and SPINOL-derived chiral phosphoric acid catalysts were examined to improve the enantioselectivities (Table 1, entries 1–11). We were pleased to find that catalyst (R)-A6 bearing a perfluorophenyl substituent at the 3,3′-position of the BINOL backbone showed a significant increase in enantioselectivity (91
:
9 er) with 90% yield (Table 1, entry 6). Interestingly, the catalyst loading was successfully reduced to 5 mol% without any loss in chemical efficiency and enantioselectivity (entry 12). Various solvents, reaction temperatures, concentrations and additives were further investigated in the presence of 5 mol% (R)-A6 (Table 1, entries 13–16. For details, see the ESI†). Finally, the highest yield (92%) and the best enantioselectivity (96
:
4 er) were achieved when the reaction was catalyzed by 5 mol% (R)-A6 in DCM (3 mL) at 10 °C with MgSO4 (35 mg) as the additive (Table 1, entry 16).
| Entry | Catalyst | Solvent (× mL) | Time | Yieldb (%) | er |
|---|---|---|---|---|---|
a Unless otherwise specified, all reactions were carried out with the catalyst (10 mol%), 1a (0.10 mmol) and 2a (0.12 mmol) in DCM (× mL) at room temperature.
b Isolated yield of 3a.
c Determined by chiral-phase HPLC analysis.
d Catalyst (5 mol%) was used.
e MgSO4 (35 mg) was used.
f Performed at 10 °C.
g
1a : 2a (1 : 1) was used.
|
|||||
| 1 | (R)-A1 | DCM (1) | 20 h | 90 | 64 : 36 |
| 2 | (R)-A2 | DCM (1) | 72 h | Trace | — |
| 3 | (R)-A3 | DCM (1) | 20 h | 90 | 56 : 44 |
| 4 | (R)-A4 | DCM (1) | 20 h | 88 | 61 : 39 |
| 5 | (R)-A5 | DCM (1) | 72 h | Trace | — |
| 6 | (R)-A6 | DCM (1) | 20 h | 90 | 91 : 9 |
| 7 | (R)-A7 | DCM (1) | 20 h | 68 | 53 : 47 |
| 8 | (R)-A8 | DCM (1) | 20 h | 65 | 65 : 35 |
| 9 | (R)-A9 | DCM (1) | 20 h | 53 | 64 : 36 |
| 10 | (R)-B1 | DCM (1) | 72 h | Trace | — |
| 11 | (R)-B2 | DCM (1) | 72 h | Trace | — |
| 12d | (R)-A6 | DCM (1) | 20 h | 90 | 91 : 9 |
| 13d | (R)-A6 | DCM (3) | 24 h | 89 | 93 : 7 |
| 14d,e | (R)-A6 | DCM (3) | 20 h | 90 | 94 : 6 |
| 15d,e,f | (R)-A6 | DCM (3) | 5 d | 92 | 95 : 5 |
| 16d,e,f,g | (R)-A6 | DCM (3) | 5 d | 92 | 96 : 4 |
Having established the optimal reaction conditions, we then proceed to examine the reaction scope (Tables 2–4). First, we evaluated the method for the synthesis of chiral 3,2′-BIMs bearing all-carbon quaternary stereocenters. As illustrated in Table 2, a broad range of 3-methylindoles 2 bearing either electron-donating or -withdrawing groups at different positions of the indole ring (C4, C5 and C6) participated in this reaction, furnishing corresponding 3,2′-BIMs (3a–3j) in 62–98% yields with 87
:
13–98
:
2 ers. Next, various α-indolyl propargylic alcohols 1 bearing substituents on the indolyl moiety were investigated, and excellent enantioselectivities (95
:
5–98
:
2 ers) were obtained for the corresponding 3,2′-BIMs (3k–3ab).
| a Unless otherwise specified, all reactions were carried out with the catalyst (5 mol%), 1 (0.10 mmol), 2 (0.10 mmol) and MgSO4 (35 mg) in DCM (3 mL) at 10 °C for 12–288 h. b Carried out with the catalyst (10 mol%), 1 (0.10 mmol), 2 (0.12 mmol) and MgSO4 (25 mg) in DCM (1 mL) at rt. c Carried out with the catalyst (10 mol%), 1 (0.10 mmol), 2 (0.10 mmol) and MgSO4 (35 mg) in DCM (3 mL) at rt. |
|---|
|
Notably, α-ethoxycarbonyl substituted propargylic alcohol was used to produce 3s in 85% yield and 95
:
5 er. In addition, a scale-up reaction for the synthesis of 3t was also successfully performed with similar results. To our delight, the R3 group of 1 could be expanded from alkyne to various alkenyl, alkyl and aryl functionalities. When racemic allyl alcohols were employed, alkenyl substituent products 3ac–3ak were afforded in up to 95% yield and 96
:
4 er, while moderate yields and enantioselectivities were observed for alkyl-substituted tertiary alcohols. Finally, triarylmethane 3an could also be obtained with high yield and excellent enantioselectivity (98
:
2 er). The absolute configuration of 3t was determined to be R based by X-ray crystallographic analysis after derivatization, and those of other chiral 3,2′-BIMs were assigned analogously.18
As a significant extension of this asymmetric catalytic method, we then turned our attention to the construction of more challenging chiral 3,3′-BIMs which have two similar indole rings. To our disappointment, the optimal catalyst (R)-A6 for chiral 3,2′-BIMs failed to deliver chiral 3,3′-BIM products with good enantioselectivity (Table S2 in the ESI†). After slight modification of the reaction conditions (see ESI for details†), the highest enantioselectivity of 97
:
3 er with 92% yield was achieved when the reaction was catalyzed by spirocyclic phosphoric acid (R)-B1 (10 mol%) in DCM at room temperature (5a, Table 3). Next, the scope of the enantioenriched 3,3′-BIMs was investigated. As shown in Table 3, various substituted groups on the indole 4 could be well tolerated (5a–5g, 80%–98% yields, 92
:
8–97
:
3 ers). Similarly, this process was applicable to a wide range of tertiary alcohols 1 bearing various R2/R3 substituents, which afforded products 5h–5r in generally high yields (62%–96%), and moderate to excellent enantioselectivities (70
:
30–97
:
3 ers). A single crystal X-ray analysis of 5i established the absolute configuration of the 3,3′-BIM products.18
| a Unless otherwise specified, all reactions were carried out with the catalyst (10 mol%), 1 (0.10 mmol) and 4 (0.12 mmol) in DCM (1 mL) at room temperature for 10–240 h. b Carried out with the catalyst (5 mol%), 1 (0.10 mmol) and 4 (0.12 mmol) in DCM (1 mL) at rt. c Carried out with catalyst (R)-A6 (10 mol%), 1 (0.10 mmol) and 4 (0.12 mmol) in toluene (1 mL) at 80 °C. |
|---|
|
During the investigation of the nucleophilic substitution reaction of 3-methylindole 2a with α-indolyl propargylic alcohol 1ae, an unexpected cyclic 3,1′-BIM 7a was isolated in 30% yield with 53
:
47 er under the standard conditions shown in Table 2, which was proposed to be formed through the intramolecular hydroamination of the allene intermediate 8a (Scheme 2). Encouraged by this interesting result, we next turned our attention to improve the enantioselectivity of this reaction. After optimization (see the ESI for details†), 3,1′-BIM 7a could be afforded in 86% yield and 97
:
3 er when the reaction was performed at room temperature with (R)-A7 as the catalyst (Table 4, 7a).
| a Unless otherwise specified, all reactions were carried out with the catalyst (10 mol%), 1 (0.10 mmol) and 2 (0.12 mmol) in DCM (1 mL) at room temperature for 10–96 h. |
|---|
|
Next, the scope of cyclic 3,1′-BIMs was explored. As shown in Table 4, various indoles 2 with different R1/R4 substituents produced the desired 3,1′-BIMs 7a–7j in 45%–96% yields with 95
:
5–98
:
2 ers. Different substituents including halogens, methyl and methoxyl on the phenyl ring of the indolyl moiety were all compatible, giving 7k–7w in 68%–98% yields with an excellent stereocontrol (92.5
:
7.5–>99
:
1 ers). Besides, when α–CHF2-substituted propargylic alcohol was subjected to this process, the corresponding 3,1′-BIM 7x was obtained in 76% yield with 96
:
4 er. The absolute configurations of these 3,1′-BIMs were assigned based on the X-ray crystal structure of compound 18 (derivated from 7a, Scheme 4).18
To gain some insights into the mechanism, a series of control experiments were performed. First, the products with opposite chirality can be isolated in a similar yield after subjecting racemic 1a to the chiral (R)- or (S)-phosphoric acid catalyst A6 (Scheme 3a). Using the same (R)-A6 catalyst, both (+)-1a and (−)-1a reacted with 2a to deliver product (+)-3a with the same configuration (Schemes 3b and 3c). Based on these results, an enantiospecific mechanism could be ruled out. Moreover, under the standard conditions, substrate 1k was observed to undergo a partial kinetic resolution process (Scheme 3d). However, variations in the initial concentrations of 2a (with the same initial concentration of 1a) produce good consistency in the reaction rate, which shows that the reaction is zeroth-order for 2a, and it has no rate dependence on the initial concentration of the nucleophile 2a (Scheme 3g). Meanwhile, reactions carried out with different initial concentrations of 2a produce good overlay in the enantiomeric excess data of product 3a, indicating no catalyst decomposition or product inhibition in this process (Scheme 3e). When taken together, these experimental results excluded the SN2 mechanism coupled with dynamic kinetic resolution.
To explore the origin of the excellent stereocontrol in the formation of chiral cyclic 3,1′-BIMs, control experiments were carried out using 1ae and 2a under the standard reaction conditions (Scheme 3f). After 2.5 hours, the allene intermediate 8a can be isolated in 22% yield with low enantioselectivity (55
:
45 er). Subjecting 8a to the standard conditions provided 7a with comparable yield and enantioselectivity (96
:
4 er) to those observed in the one-pot as shown process in Table 4 (see the ESI for details†), thus indicating that hydroamination of the alkene intermediate is the enantiodetermining step (Scheme 3f, right).
Next, ten randomly selected chiral BIMs (3l, 3o, 3u, 3y, 3ad, 3ah, 3al, 5m, 7a and 7g) were evaluated for their antimicrobial activities with the FDA-approved broad-spectrum antimicrobial drug ampicillin or vancomycin as the positive control. We first screened their antibacterial activities against escherichia coli and bacillus subtilis with ampicillin as the positive control. As shown in Table 5, 3,2′-BIMs 3o and 3u exhibited antibacterial effects towards escherichia coli. Meanwhile, all selected compounds showed remarkable antibacterial activities against bacillus subtilis (MIC: 1–8 μg mL−1). Notably, 3,2′-BIM 3al bearing a butyl group was 2 times more potent than ampicillin with a MIC value of 1 μg mL−1. We further examined their antibacterial potency against methicillin-resistant staphylococcus aureus (MRSA) using vancomycin as the reference. To our delight, most of these compounds showed impressive antibacterial effects against MRSA (MIC: 2–16 μg mL−1), especially compounds 3al and 5m (MIC: 2–4 μg mL−1). These preliminary antibacterial tests indicated that enantioenriched BIMs might be developed as novel antibacterial agents.
| Compound | MIC (μg mL−1) | ||
|---|---|---|---|
| Escherichia coli | Bacillus subtilis | Methicillin-resistant Staphylococcus aureus | |
| a MIC values were obtained by the microdilution method following CLSI standards and are given in μg mL−1. | |||
| 3l | 64 | 4 | 16 |
| 3o | 8 | 4 | 16 |
| 3u | 16 | 2 | 16 |
| 3y | >64 | 8 | >64 |
| 3ad | 64 | 2 | 16 |
| 3ah | 64 | 8 | >64 |
| 3al | >64 | 1 | 2 |
| 5m | >64 | 4 | 4 |
| 7a | 64 | 4 | 16 |
| 7g | 64 | 8 | 16 |
| Ampicillin | 2 | 2 | — |
| Vancomyci | — | — | 1 |
To illustrate the synthetic utilities of this new methodology, we have also tried to transform these 3,n′-BIM products into a variety of useful enantioenriched compounds with quaternary stereocenters. In the case of 3ac, its methoxyl group can be smoothly converted into triflate (9) in good yield without loss of enantiomeric purity, which provides more possibilities for further functionalization. For 3,2′-BIM 3t, regioselective protection of the indole NH moiety provided 10; hydrogenation of the alkyne afforded α-alkyl BIM 11; removal of the TMS group generated terminal alkyne 12, which underwent either a Cu-catalyzed azide–alkyne cycloaddition (CuAAC) to deliver triarylmethanes 13a–13c, or an intramolecular cyclization to give 14 with a pyrrolo[1,2-a]indole scaffold (Scheme 4a). Furthermore, 3,3′-BIM 5m with the TMS-protected alkyne moiety can be transformed into aryl alkyne 15 through desilylation followed by Sonogashira coupling (Scheme 4b). The synthetic utility of 3,1′-BIM 7a was also demonstrated by N–H protection, selective hydrogenation and bromination (Scheme 4c).
:
1 er). The rich chemistry of the indolyl, alkynyl, and other functional groups in the obtained BIMs allows for facile transformation into valuable chiral compounds bearing quaternary centers that are otherwise difficult to access. Mechanism investigations exclude the SN2 mechanism and suggest a formal SN1-type pathway enabled by CPA catalysts. Preliminary biological evaluation demonstrated that these enantioenriched BIMs exhibited impressive preliminary antibacterial activities, and notably 3al is 2 times more potent than ampicillin against bacillus subtilis with a MIC value of 1 μg mL−1. This strategy not only provides an efficient and unified method for accessing valuable chiral BIMs, but also opens opportunities for asymmetric dehydrative cross-coupling reactions employing tertiary alcohols as competent electrophiles without pre-activation.
Footnotes |
| † Electronic supplementary information (ESI) available. CCDC 2101620, 2101638 and 2101639. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/d1sc05174a |
| ‡ Authors with equal contribution. |
| This journal is © The Royal Society of Chemistry 2022 |