Xiuwen Jia1, Pinyi Li2, Xiaoyan Liu3, Jiafu Lin4, Yiwen Chu5, Jinhai Yu6, Jiang Wang7,8, Hong Liu9,10, Fei Zhao11. 1. Antibiotics Research and Re-evaluation Key Laboratory of Sichuan Province, Sichuan Industrial Institute of Antibiotics, Chengdu University, Chengdu 610052, China. jiaxiuwen2018@126.com. 2. Antibiotics Research and Re-evaluation Key Laboratory of Sichuan Province, Sichuan Industrial Institute of Antibiotics, Chengdu University, Chengdu 610052, China. pinyiLi19950206@126.com. 3. Antibiotics Research and Re-evaluation Key Laboratory of Sichuan Province, Sichuan Industrial Institute of Antibiotics, Chengdu University, Chengdu 610052, China. 19940826097@163.com. 4. Antibiotics Research and Re-evaluation Key Laboratory of Sichuan Province, Sichuan Industrial Institute of Antibiotics, Chengdu University, Chengdu 610052, China. linjiafu@cdu.edu.cn. 5. Antibiotics Research and Re-evaluation Key Laboratory of Sichuan Province, Sichuan Industrial Institute of Antibiotics, Chengdu University, Chengdu 610052, China. siiakyb@139.com. 6. School of Biological Science and Technology, University of Jinan, Jinan 250022, China. bio_yujh@ujn.edu.cn. 7. State Key Laboratory of Drug Research and CAS Key Laboratory of Receptor Research, Shanghai Institute of Materia Medical, Chinese Academy of Sciences, Shanghai 201203, China. jwang@simm.ac.cn. 8. University of Chinese Academy of Sciences, Beijing 100049, China. jwang@simm.ac.cn. 9. State Key Laboratory of Drug Research and CAS Key Laboratory of Receptor Research, Shanghai Institute of Materia Medical, Chinese Academy of Sciences, Shanghai 201203, China. hliu@simm.ac.cn. 10. University of Chinese Academy of Sciences, Beijing 100049, China. hliu@simm.ac.cn. 11. Antibiotics Research and Re-evaluation Key Laboratory of Sichuan Province, Sichuan Industrial Institute of Antibiotics, Chengdu University, Chengdu 610052, China. zhaofei@cdu.edu.cn.
Abstract
The present study describes an AuPPh₃Cl/AgSbF₆-catalyzed cascade reaction between amine nucleophiles and alkynoic acids in water. This process proceeds in high step economy with water as the sole coproduct, and leads to the generation of two rings, together with the formation of three new bonds in a single operation. This green cascade process exhibits valuable features such as low catalyst loading, good to excellent yields, high efficiency in bond formation, excellent selectivity, great tolerance of functional groups, and extraordinarily broad substrate scope. In addition, this is the first example of the generation of an indole/thiophene/pyrrole/pyridine/naphthalene/benzene-fused N-heterocycle library through gold catalysis in water from readily available materials. Notably, the discovery of antibacterial molecules from this library demonstrates its high quality and potential for the identification of active pharmaceutical ingredients.
The present study describes an AuPPh₃Cl/AgSbF₆-catalyzed cascade reaction between amine nucleophiles and alkynoic acids in water. This process proceeds in high step economy with water as the sole coproduct, and leads to the generation of two rings, together with the formation of three new bonds in a single operation. This green cascade process exhibits valuable features such as low catalyst loading, good to excellent yields, high efficiency in bond formation, excellent selectivity, great tolerance of functional groups, and extraordinarily broad substrate scope. In addition, this is the first example of the generation of an indole/thiophene/pyrrole/pyridine/naphthalene/benzene-fused N-heterocycle library through gold catalysis in water from readily available materials. Notably, the discovery of antibacterial molecules from this library demonstrates its high quality and potential for the identification of active pharmaceutical ingredients.
Rapid advances in genomics and proteomics have resulted in the identification of an increasing number of novel therapeutic targets [1,2,3,4,5], and the existing compound libraries can no longer well meet the needs of drug screening. Therefore, it is highly demanding to develop robust synthetic methods to construct new compound libraries for drug discovery aimed at these targets [6,7,8]. Considering the structurally diverse targets in a wide “biological space”, high-throughput screening (HTS) of skeletally diverse compounds, which occupy a broader “chemical space”, can apparently enhance the hit rates [9,10]. In addition to skeletal diversity, drug-like properties of the compounds are equally important for the generation of high-quality compound libraries [11,12,13,14], which can increase the possibility of identifying drug-like hit compounds. As a result, privileged structures have received wide attention in drug discovery because they are widely found in natural and pharmaceutical products [15,16,17,18,19]. Although privileged substructure-based diversity-oriented synthesis (pDOS) provides a useful access to assemble compound libraries with high-quality [20,21,22,23,24,25,26], it is still challenging to develop efficient and practical approaches to generate a variety of molecular frameworks embedded with privileged structures, especially in a green and sustainable manner. With our interests to develop green and efficient protocols to synthesize valuable N-heterocycles [27,28,29,30,31], we herein construct a library of privileged substructure-based N-heterocycles with diverse scaffolds using gold-catalyzed cascade reactions in water. To the best of our knowledge, this is the first example of the generation of pDOS compound library encompassing skeletal diversity, molecular complexity, and drug-like properties through gold catalysis in water.Alkynoic acids are extensively used to react with amine nucleophiles to furnish heterocyclic compounds because of the efficient cascade reaction developed by Dixon’s group [32], in which, an activated cyclic enol lactone species, which derives from alkynoic acids, is involved as the key intermediate (Scheme 1a). Dixon and co-workers disclosed linear aliphatic terminal alkynoic acids reacted smoothly with amine nucleophiles bearing a nucleophilic carbon atom in toluene or xylene to produce pyrrole- or indole-based heterocyclic frameworks catalyzed by AuPPh3Cl/AgOTf. It should be noted that, based on Dixon’s pioneering work, the reactions of alkynoic acids with amine nucleophiles in aprotic solvents such as toluene, xylene, 1,2-dichloroethane, dichloromethane, etc., have been well studied by Patil’s group as well as our group [33,34,35,36,37,38,39,40,41,42,43]. However, the reactivity of alkynoic acids and amine nucleophiles in the environmentally friendly, abundant, and cheap solvent—water—was seldom explored [44,45], mainly because the ring opening reaction of the activated enol lactone intermediate in water at elevated temperature may lead to the failure of the cascade reaction. This reason prevented researchers’ steps from investigating the nature of the cascade reaction in water. To date, only two successful examples in water were reported by our group [44,45], although the alkynoic acids were only limited to linear aliphatic terminal alkynoic acids, amine nucleophiles were only limited to amine nucleophiles carrying a nucleophilic carbon atom, and complicated Au catalysts were required (Scheme 1b). However, considering water often displays unique reactivity and selectivity which can’t be obtained in common organic solvents [46,47,48,49,50,51] and it is an environmentally benign solvent, we aim to develop a more general cascade process between various alkynoic acids and amine nucleophiles in water with great interest. Despite the possibility of hydrolysis of enol lactone intermediate in water at high temperature which may result in the failure of the cascade reaction, we hypothesize that a more general cascade process could also be achieved in water but with a proper catalytic system. In the present study, we develop a greener, and more general and efficient catalytic system (AuPPh3Cl/AgSbF6/CF3CO2H) in water, which tolerates a broader substrate scope of alkynoic acids as well as amine nucleophiles. As shown in Scheme 1c, not only linear aliphatic terminal and internal alkynoic acids but also cyclic aromatic terminal and internal alkynoic acids are well tolerated. In addition, extraordinarily broad amine nucleophiles bearing a nucleophilic carbon/nitrogen/oxygen atom turned out to be suitable substrates. Besides, the reaction mechanism in water was carefully checked and studied for the first time. Interestingly, when D2O was used as the reaction solvent, we observed the highly deuterated products generated from the H–D exchange between the reaction substrates/intermediates and D2O. Herein, we also present the library construction of skeletally diverse N-heterocycles embedded with privileged structures employing different alkynoic acids and various amine nucleophiles as the building blocks. To our delight, five antimicrobial compounds were identified from this library after biological evaluation. The production of N-heterocycles with diverse scaffolds and the discovery of active pharmaceutical ingredients (APIs) demonstrate the power of this method in both organic synthesis and medicinal chemistry.
Scheme 1
Gold-catalyzed tandem reactions between alkynoic acids and amine nucleophiles. (a). Dixon’s work. (b). Our previous work. (c). This work.
2. Results and Discussion
4-pentynoic acid (1a) and tryptamine (2a) were employed as the model substrates to optimize the cascade reaction conditions (Table 1). Pleasingly, treatment of starting materials 1a and 2a in water without any catalyst at 100 °C for 24 h gave the desired product SF1a, albeit with a low yield (entry 1). Then various metal catalysts were screened to improve the reaction yield. A screening of Pd, Cu, Ni, and Mn complexes (entries 2–11), disclosed, at first, that Pd(II) sources, such as PdCl2(CH3CN)2 and Pd(PPh3)2Cl2, NiOAc and Mn(OAc)2, had little enhancement on the yield; while Pd(0) sources such as Pd(PPh3)4 and Pd2(dba)3; copper catalysts, such as Cu(OAc)2 and Cu(OH)2, and CuCl; and Ni(PPh3)2Cl inhibited this transformation. A subsequent survey of Ag, Ru, and Co catalysts (entries 12–16) revealed that Ag2CO3, AgOAc, [RuCl2(p-cym)]2, CoCl2, and Co(acac)2 could obviously increase the yield, affording the product SF1a in moderate yields (38–57%). Notably, a good yield (70%) was achieved with AuPPh3Cl (entry 17), and increasing the temperature to 120 °C was much more efficient in improving the yield as compared with extending the reaction time to 48 h (entries 18 and 19). Encouraged by this result, an investigation of other Au catalysts at 120 °C was carried out (entries 20–24). Among them, none displayed higher catalytic reactivity than AuPPh3Cl. To further improve the yield of the product, a series of silver salts, which were proved to be able to increase the catalytic reactivity of gold catalysts [52,53,54,55,56,57], were screened as the additives (entries 25–28). To our delight, AgSbF6 was found to be the best choice, with which product SF1a was obtained in 91% yield (entry 28). In addition, other typical organic solvents were also tested as the reaction solvents, and toluene, xylene, and DCE also turned out to be suitable solvents, in which similar high isolated yields were observed (See Supplementary Materials for details). However, considering water is more environmentally benign, we then decided to use water as the solvent to explore the substrate scope of this method. In this way, the optimal reaction conditions were identified using a catalytic system consisting of AuPPh3Cl/AgSbF6 in water at 120 °C for 24 h.
Table 1
Reaction condition optimization for the tandem synthesis of compound SF1a
a.
Entry
Catalyst/Additive
Solvent
T (°C)
Yield (%) b
1
-
H2O
100
17
2
PdCl2(CH3CN)2
H2O
100
26
3
Pd(PPh3)2Cl2
H2O
100
31
4
Pd(PPh3)4
H2O
100
0
5
Pd2(dba)3
H2O
100
0
6
Cu(OAc)2
H2O
100
trace
7
Cu(OH)2
H2O
100
5
8
CuCl
H2O
100
trace
9
NiOAc
H2O
100
33
10
Ni(PPh3)2Cl
H2O
100
11
11
Mn(OAc)2
H2O
100
25
12
Ag2CO3
H2O
100
51
13
AgOAc
H2O
100
57
14
[RuCl2(p-cym)]2
H2O
100
48
15
CoCl2
H2O
100
38
16
Co(acac)2
H2O
100
50
17
AuPPh3Cl
H2O
100
70
18 c
AuPPh3Cl
H2O
100
76
19
AuPPh3Cl
H2O
120
83
20
AuBr3
H2O
120
52
21
AuI
H2O
120
46
22
Au1 catalyst d
H2O
120
81
23
Au2 catalyst e
H2O
120
78
24
Au3 catalyst f
H2O
120
80
25
AuPPh3Cl/Ag2CO3
H2O
120
85
26
AuPPh3Cl/AgOAc
H2O
120
86
27
AuPPh3Cl/AgOTf
H2O
120
89
28
AuPPh3Cl/AgSbF6
H2O
120
91
a Reaction conditions: 4-pentynoic acid 1a (0.6 mmol), tryptamine 2a (0.5 mmol), catalyst/additive (0.005 mmol), and solvent (4.0 mL). b Yield refers to isolated yield. c The reaction was performed for 48 h. d Au1 catalyst = Chloro[(1,1′-biphenyl-2-yl)di-tert-butylphosphine]gold(I). e Au2 catalyst = Chloro[1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene]gold(I). f Au3 catalyst = (Acetonitrile)[(2-biphenyl)di-tert-butylphosphine]gold(I) hexafluoroantimonate.
After determining the optimal reaction conditions, we then began to construct a high-quality library of privileged substructure-based N-heterocycles with diverse scaffolds. We first examined the generality of the process with various amine nucleophiles 2 containing a nucleophilic carbon on a heteroaromatic or aromatic ring. In general, this process tolerated a variety of amine nucleophiles 2 and alkynoic acids 1, and 26 scaffolds embedded with privileged structures were furnished in good to high yields in water under optimal or modified reaction conditions (Scheme 2). For example, tryptamines reacted smoothly with terminal alkynoic acids such as 4-pentynoic acid under standard conditions to give products SF1a and SF1b in high yields. The reactions of tryptamines with other terminal alkynoic acids, such as 5-hexynoic acid, 2-ethynylbenzoic acid, and 2-(2-ethynylphenyl)acetic acid, also afforded the corresponding products (SF2a, SF2b, SF3a, SF3b, and SF4a) in good to high yields under a modified two-step one-pot process, in which CF3CO2H (TFA) was added to promote the iminium ion formation. In addition, internal alkynoic acids, such as 5-phenylpent-4-ynoic acid, 6-phenylhex-5-ynoic acid, 2-(phenylethynyl)benzoic acid, and 2-(2-(phenylethynyl)phenyl)acetic acid, were also tested as the substrates. Unfortunately, they all failed to react with tryptamines to give the desired products (SF1c, SF2c, SF3c, and SF4b) even under further improved conditions. Interestingly, 2-(1H-indol-2-yl)ethylamines could undergo this transformation with terminal alkynoic acids as well as internal alkynoic acids to yield the desired products SF5–SF8 in 35–96% yields. It should be noted that no N1 ring closure products were observed when 2-(1H-indol-2-yl)ethylamines were used as the amine nucleophiles, suggesting excellent selectivity of this cascade process. This may be because the C3 nucleophilicity is stronger than N1 nucleophilicity [58]. Likewise, the reactivity of 2-(1H-indol-1-yl)ethanamines in this cascade reaction was very similar to that of tryptamines. They reacted well with terminal alkynoic acids while their reactions with internal alkynoic acids failed to give the desired products (SF9–SF12). Interestingly, the protocol was also compatible with 3-(1H-indol-1-yl)propan-1-amines, which furnished products SF13 carrying a seven-membered ring in 54–80% yields. Similarly, products SF14–SF16 were obtained in 59–86% yields when 2-(1H-indol-1-yl)anilines and alkynoic acids were subjected to the modified conditions. Notably, the indole-containing polycyclic frameworks, represented by compounds SF1–SF16, are regarded as valuable N-heterocycles considering their ubiquitous presence in biologically active molecules [59,60,61,62,63]. Subsequently, 2-(1H-pyrrol-2-yl)ethanamines, 2-(1H-pyrrol-1-yl)ethanamines, 2-(thiophen-2-yl)ethanamines, and 2-(thiophen-3-yl)ethanamines were employed as amine nucleophiles. Their reactions with diverse alkynoic acids took place successfully to provide pyrrole- or thiophene-fused compounds SF17–SF23 in moderate to high yields, despite the fact that stronger conditions were required. In particular, 2-phenylethanamines with electron-donating substituents on the benzene ring were also well tolerated, leading to the formation of benzene-fused heterocyclic products SF24–SF26 in yields ranging from 47 to 88%. It is also worth noting that excellent selectivity was achieved in the reactions of 2-(1H-pyrrol-2-yl)ethanamine, 2-(thiophen-3-yl)ethanamine or 2-phenylethanamines, even though two potential cyclization sites existed in the final step. It should be noted that compound SF26 is the analog of tetrahydroberberines, which were extracted from the Chinese herb Corydalis ambigua and exhibited a broad range of biological activities [64,65].
Scheme 2
Generation of scaffold diversity. a Reaction conditions: alkynoic acids 1 (0.6 mmol), amine nucleophiles 2 (0.5 mmol), AuPPh3Cl/AgSbF6 (0.005 mmol), H2O (4.0 mL), and 120 °C, 24 h. b Reaction conditions: (i) alkynoic acids 1 (0.6 mmol), amine nucleophiles 2 (0.5 mmol), AuPPh3Cl/AgSbF6 (0.01 mmol), H2O (4.0 mL), 120 °C, 20 h; (ii) TFA (0.5 mmol) was added, and then 120 °C, 4 h. c Reaction conditions: (i) alkynoic acids 1 (0.6 mmol), amine nucleophiles 2 (0.5 mmol), AuPPh3Cl/AgSbF6 (0.015 mmol), H2O (4.0 mL), 140 °C, 20 h; (ii) TFA (0.5 mmol) was added, and then 140 °C, 4 h. ND = Not detected.
To further broaden the substrate scope of this approach, amine nucleophiles 3 containing a nucleophilic heteroatom (Z = amide/aniline N, acid/alcohol O) were tested as substrates. Overall, this protocol was also applicable to diverse amine nucleophiles 3, and 17 scaffolds were constructed with high efficiency (Scheme 3). For instance, the reactions of 2-aminobenzamides with various alkynoic acids worked successfully, affording benzene-fused polycyclic products SF27–SF30 in moderate to high yields. Gratifyingly, this approach was compatible with 2-(aminomethyl)anilines and benzene-1,2-diamines, although the desired benzene-based heterocyclic products SF31–SF34 were obtained in lower yields. Remarkably, in the case of substrates such as 2-aminobenzamides and 2-(aminomethyl)anilines, which contain two nitrogen atoms as the nucleophiles, the nitrogen atom with stronger nucleophilicity tended to attack the enol lactone intermediate and therefore the other nitrogen atom with weaker nucleophilicity attacked the iminium ion intermediate to selectively provide the corresponding products, while not in the reverse way. Besides, 2-aminobenzoic acids, 3-amino-2-naphthoic acids, or 2-aminonicotinic acids reacted with various alkynoic acids smoothly, producing the corresponding benzene-, naphthalene-, or pyridine-fused heterocyclic products SF35–SF41 in yields of 50–96%. Surprisingly, 2-aminobenzyl alcohols were also found to be suitable substrates, which could undergo the cascade reaction with alkynoic acids to give the desired benzene-fused polycyclic products SF42 and SF43, albeit with lower yields. In addition, we also tried to synthesize the indole-fused compounds SF44–SF46 embedded with an eight- or nine-membered ring; unfortunately, we failed. This may be attributed to the low reactivities of the amine nucleophiles and the instability of the large rings in energetics.
Scheme 3
Generation of scaffold diversity. a Reactions conditions: (i) Alkynoic acids 1 (0.6 mmol), amine nucleophiles 3 (0.5 mmol), AuPPh3Cl/AgSbF6 (0.01 mmol), H2O (4.0 mL), 120 °C, 20 h; (ii) TFA (0.5 mmol) was added, and then 120 °C, 4 h. b The reaction was carried out at 140 °C. c The reaction was carried out with 5 mol% AuPPh3Cl/AgSbF6. d The reaction was performed with 5 mol% AuPPh3Cl/AgSbF6 at 140 °C. ND = Not detected.
The diversity of the library can be further expanded by the derivatization of the target compounds. We herein introduce the derivatization of the target compounds based on the simple reduction of the carbonyl group, and the selected results are shown in Scheme 4. Nine different scaffolds represented by indole-, pyrrole-, or thiophene-based polycyclic compoundsSF47–SF55 containing a tertiary amine were produced conveniently through an easy reduction of the corresponding precursors with LiAlH4/AlCl3. Notably, these scaffolds are very similar to those found in natural and pharmaceutical agents [65]. We expect the screening of these compounds towards specific biological targets might lead to the identification of bioactive molecules.
Scheme 4
Derivatization of the target compounds in the library.
Thus, using various amine nucleophiles and alkynoic acids as the building blocks, a library of privileged substructure-based N-heterocycles with diverse scaffolds was constructed through gold catalysis in a green and efficient manner. It is worth noting this cascade process constructs three new bonds together with two rings in one chemical process, suggesting the high efficiency of this cascade reaction in synthesizing nitrogen-containing heterocyclic compounds. Regarding the large occurrence of nitrogen-containing heterocyclic compounds in APIs [66,67,68], the method presented in this paper is prospective since it could provide an environmentally benign and useful platform for the preparation of diverse nitrogen-containing heterocyclic compounds.To verify our expectation that this approach could provide useful scaffolds with attractive bioactivities, a bioactivity study of this library was carried out. An initial pharmacological study of these nitrogen-containing heterocyclic compounds led to the discovery of five antimicrobial compounds—SF9d, SF29b, SF33, SF36, and SF41. The minimal inhibitory concentration (MIC) results revealed that compound SF36 displayed the most potent antibacterial activity against S. aureus strain, with a MIC90 value of 10–25 μg/mL (Table 2) (Time-kill assays and colony-forming unit studies of compounds SF9d, SF29b, SF33, SF36 and SF41 could be found in Supplementary Materials).
Table 2
MIC90 of compounds SF9d, SF29b, SF33, SF36, and SF41 against the S. aureus strain.
Compound
MIC90 (μg/mL)
SF9d
100–200
SF29b
50
SF33
100–200
SF36
10–25
SF41
100
Mechanistic studies were carried out with deuterium-labeling experiments, and the hydrogens of the products were assigned at first by the analysis of the 1H-NMR, 13C-NMR, HSQC, HMBC, and 1H-1H COSY spectrum to confirm the deuterated positions (See Supplementary Materials for details). Interestingly, the reaction of 4-pentynoic acid 1a with tryptamine 2a in D2O under the standard conditions afforded the deuterated product [D]n-SF1a, not only at the methyl position but also at the β-position of the carbonyl (Scheme 5a). Specifically, a 96% deuteration at the methyl position and a similar deuteration (90%) of the two unequal hydrogens at the β-position of the carbonyl were observed. The reactions of 2-(1H-indol-2-yl)ethylamines with 4-pentynoic acid 1a in D2O gave the similar results (Scheme 5b,c).
Scheme 5
Deuterium-labeling experiments.; (a) the deuterium-labeling experiment of substrates 1a and 2a in D2O; (b) the deuterium-labeling experiment of substrates 1a and 2c in D2O; (c) the deuterium-labeling experiment of substrates 1a and 2d in D2O.
According to the results of deuterium-labeling experiments, we hypothesize two possible reaction pathways (Scheme 6 and Scheme 7). The reaction of with 4-pentynoic acid 1a with tryptamine 2a in the presence of Au catalyst in D2O is taken as the example to illustrate the reaction pathway. The first hypothetic reaction pathway (Scheme 6) may involve the gold-catalyzed hydration of carbon–carbon triple bond, which was observed in our previous work [69]. The H–D exchange between the carboxyl group of 4-pentynoic acid 1a and D2O leads to the formation of intermediate A1. Gold-catalyzed addition of D2O to the carbon–carbon triple bond of intermediate A1 produces intermediate A2, which undergoes two keto–enol tautomerizations to give intermediate A4. The subsequent H–D exchange between the hydroxyl group of intermediate A4 and D2O affords intermediate A5, which undergoes keto–enol tautomerization again to give intermediate A6. The two acidic protons at the α-position of the carbonyl in intermediate A6 undergo H–D exchange with D2O via keto–enol tautomerizations to yield intermediate A7. The following condensation between intermediate A7 and tryptamine 2a, the subsequent iminium ion formation, and the final cyclization achieve the product [D]n-SF1a. The second hypothetic reaction pathway is shown in Scheme 7; the H–D exchange between the carboxyl group of 4-pentynoic acid 1a and D2O produces intermediate A1. Gold-catalyzed intramolecular cyclization of intermediate A1 produces enol lactone intermediate B1, which is attacked by tryptamine 2a to give intermediate B2. The subsequent H–D exchange between the hydroxyl group of intermediate B2 and D2O yields intermediate B3, which undergoes enol–keto tautomerization to provide intermediate B4. Similarly, the three acidic protons at the α-position of the carbonyl in intermediate B4 undergo H–D exchange with D2O via keto–enol tautomerizations to yield intermediate C1, which is converted into the product [D]n-SF1a via an iminium ion formation/cyclization sequence.
Scheme 6
Hypothetic reaction pathway 1.
Scheme 7
Hypothetic reaction pathway 2.
To further verify the reaction mechanism, the reaction of 4-pentynoic acid 1a and tryptamine 2a under gold catalysis in O18-labeled water was carried out first. This reaction in H2O18 was stopped after 0.5 h to track the reaction intermediates. As shown in Scheme 8a, apart from the remaining starting materials, intermediate 2a″ and the product SF1a were obtained in 21% and 15% yield, respectively. While the O18-labeled intermediate 2a′ was not observed. This result clearly indicates the hydration of alkyne moiety, which is proposed in Scheme 6, is not involved. By contrast, the commercially purchased enol lactone D reacted smoothly with tryptamine 2a without the gold catalyst (Scheme 8b). This result shows that the enol lactone species B1, which is proposed in Scheme 7, is likely to be the key intermediate.
Scheme 8
Mechanistic study experiments. (a) the mechanistic experiment of substrates 1a and 2a in H2O18; (b) the mechanistic experiment of substrates D and 2a without catalyst/additive.
On the basis of the above results of mechanistic experiments, a final proposed reaction mechanism is outlined in Scheme 9. The proposed mechanism commences with the coordination of the gold catalyst to the carbon–carbon triple bond of alkynoic acids to produce intermediate I1. The subsequent intramolecular exo cyclization of I1 yields intermediate I2. The following protodemetalation of intermediate I2 takes place to produce the enol lactone species I3 with the regeneration of the catalyst. Then intermediate I3 undergoes aminolysis by amine nucleophiles to give intermediate I4, which tautomerizes to produce intermediate I5. Intermediate I5 is converted into the iminium ion I8 under the catalysis of the gold catalyst. The final nucleophilic cyclization of intermediate I8 affords the desired products with the release of the gold catalyst. The stronger nucleophilicity of amine nucleophiles compared to that of H2O results in the aminolysis of enol lactone I3 by amine nucleophiles instead of hydrolysis by H2O. It should be noted that the reaction solvent H2O participates in this cascade reaction via the H–H exchange with the carboxyl group of alkynoic acids, the hydroxyl group of intermediate I4, and the α hydrogen atoms of the carbonyl group of intermediate I5, as demonstrated by deuterium-labeling experiments. Besides, TFA could promote this reaction by accelerating the formation of iminium ion I8.
Scheme 9
A final proposed reaction mechanism.
3. Materials and Methods
3.1. General Information
If not otherwise specified, the starting materials were obtained from commercial sources and used directly without purification. Analytical thin-layer chromatography (TLC): HSGF 254 (0.15–0.2 mm thickness). Detection under UV light at 254 nm. Column chromatography: Separations were carried out on silica gel FCP 200–300. Yields refer to isolated compounds. Melting point apparatus: a micro melting point apparatus, values are uncorrected. Nuclear magnetic resonance (NMR) apparatus: a Brucker instrument. Chemical shifts (δ) are given in ppm. Proton coupling patterns were recorded as singlet (s), doublet (d), triplet (t), quartet (q), and multiplet (m). LRMS (low-resolution mass) and HRMS (high-resolution mass) were measured on a spectrometer with an electrospray ionization (ESI) source.
3.2. General Procedure for the Preparation of Compounds SF1a, SF1b, SF5a, SF5b, and SF5c
A suspension of alkynoic acids 1 (0.6 mmol), amine nucleophiles 2 (0.5 mmol), and AuPPh3Cl/AgSbF6 (0.005 mmol) in H2O (4.0 mL) was stirred at 120 °C for 24 h. At ambient temperature, saturated Na2CO3 solution (25.0 mL) was added to the reaction mixture. The resulting mixture was then extracted with ethyl acetate (3 × 15 mL). The combined organic layers were washed with brine, and dried over Na2SO4. After filtration and removal of the solvents in vacuo, the crude product was purified by flash chromatography on silica gel to provide the desired product.11b-Methyl-5,6,11,11b-tetrahydro-1H-indolizino[8,7-b]indol-3(2H)-one (SF1a): white solid (109.7 mg, yield 91%), mp 260–261 °C. 1H-NMR (500 MHz, DMSO-d6) δ 1.54 (s, 3H), 2.07–1.99 (m, 1H), 2.32–2.20 (m, 2H), 2.66–2.55 (m, 2H), 2.75–2.67 (m, 1H), 3.11–3.00 (m, 1H), 4.25–4.16 (m, 1H), 7.01–6.94 (m, 1H), 7.10-7.02 (m, 1H), 7.32 (d, J = 8.0 Hz, 1H), 7.39 (d, J = 7.8 Hz, 1H), 11.06 (s, 1H); 13C-NMR (100 MHz, DMSO-d6) δ 171.9 (CO), 139.0 (C, Ar), 135.9 (C, Ar), 126.3 (C, Ar), 121.0 (CH, Ar), 118.6 (CH, Ar), 118.0 (CH, Ar), 111.1 (CH, Ar), 104.7 (C, Ar), 58.9 (C), 34.3 (CH2), 32.6 (CH2), 30.1 (CH2), 25.0 (CH3), 20.9 (CH2); ESI-LRMS m/z: 241 [M + H]+; ESI-HRMS m/z calcd for M + H+ 241.1335, found: 241.1331. The characterization data is in accordance with that reported in [32].8-Methoxy-11b-methyl-5,6,11,11b-tetrahydro-1H-indolizino[8,7-b]indol-3(2H)-one (SF1b): yellow oil (120.2 mg, yield 89%). 1H-NMR (500 MHz, DMSO-d6) δ 1.53 (s, 3H), 2.07–1.97 (m, 1H), 2.31–2.20 (m, 2H), 2.64–2.52 (m, 2H), 2.73–2.64 (m, 1H), 3.10–2.98 (m, 1H), 3.74 (s, 3H), 4.24–4.15 (m, 1H), 6.70 (dd, J = 8.7, 2.4 Hz, 1H), 6.89 (d, J = 2.3 Hz, 1H), 7.20 (d, J = 8.7 Hz, 1H), 10.88 (s, 1H); 13C-NMR (125 MHz, DMSO-d6) δ 171.9 (CO), 153.2 (C, Ar), 139.7 (C, Ar), 130.9 (C, Ar), 126.6 (C, Ar), 111.7 (CH, Ar), 110.8 (CH, Ar), 104.6 (C, Ar), 100.2 (CH, Ar), 58.9 (C), 55.4 (OCH3), 34.3 (CH2), 32.6 (CH2), 30.1 (CH2), 25.0 (CH3), 21.0 (CH2); ESI-LRMS m/z: 271 [M + H]+; ESI-HRMS m/z calcd for M + H+ 271.1441, found: 271.1437. The characterization data is in accordance with that reported in [43].11c-Methyl-5,6,7,11c-tetrahydro-1H-indolizino[7,8-b]indol-3(2H)-one (SF5a): yellow solid (114.2 mg, yield 95%), mp 96–97 °C. 1H-NMR (500 MHz, DMSO-d6) δ 1.54 (s, 3H), 2.02–1.92 (m, 1H), 2.27–2.17 (m, 1H), 2.54–2.47 (m, 1H), 2.65–2.54 (m, 1H), 2.81–2.67 (m, 2H), 3.16–3.04 (m, 1H), 4.27–4.18 (m, 1H), 7.00–6.92 (m, 1H), 7.08–7.00 (m, 1H), 7.30 (d, J = 8.0 Hz, 1H), 7.47 (d, J = 7.8 Hz, 1H), 10.90 (s, 1H); 13C-NMR (100 MHz, DMSO-d6) δ 171.5 (CO), 135.9 (C, Ar), 130.5 (C, Ar), 123.9 (C, Ar), 120.5 (CH, Ar), 118.6 (CH, Ar), 117.9 (CH, Ar), 115.3 (C, Ar), 111.1 (CH, Ar), 59.2 (C), 33.2 (CH2), 33.2 (CH2), 30.2 (CH2), 25.0 (CH3), 22.8 (CH2); ESI-LRMS m/z: 241 [M + H]+; ESI-HRMS m/z calcd for M + H+ 241.1335, found: 241.1332. The characterization data is in accordance with that reported in [43].10-Methoxy-11c-methyl-5,6,7,11c-tetrahydro-1H-indolizino[7,8-b]indol-3(2H)-one (SF5b): pale yellow solid (129.5 mg, yield 96%), mp 193–194 °C. 1H-NMR (500 MHz, DMSO-d6) δ 1.54 (s, 3H), 2.01–1.91 (m, 1H), 2.27–2.18 (m, 1H), 2.64–2.52 (m, 2H), 2.79–2.65 (m, 2H), 3.14–3.03 (m, 1H), 3.77 (s, 3H), 4.25–4.15 (m, 1H), 6.69 (dd, J = 8.7, 2.4 Hz, 1H), 6.93 (d, J = 2.3 Hz, 1H), 7.19 (d, J = 8.7 Hz, 1H), 10.72 (s, 1H); 13C-NMR (100 MHz, DMSO-d6) δ 171.5 (CO), 153.1 (C, Ar), 131.3 (C, Ar), 131.0 (C, Ar), 124.2 (C, Ar), 115.1 (C, Ar), 111.7 (CH, Ar), 109.8 (CH, Ar), 100.6 (CH, Ar), 59.2 (C), 55.5 (OCH3), 33.2 (CH2), 33.0 (CH2), 30.1 (CH2), 24.8 (CH3), 22.9 (CH2); ESI-LRMS m/z: 271 [M + H]+; ESI-HRMS m/z calcd for M + H+ 271.1441, found: 271.1437. The characterization data is in accordance with that reported in [43].10,11c-Dimethyl-5,6,7,11c-tetrahydro-1H-indolizino[7,8-b]indol-3(2H)-one (SF5c): pale yellow oil (118.4 mg, yield 93%). 1H-NMR (500 MHz, DMSO-d6) δ 1.53 (s, 3H), 2.02–1.91 (m, 1H), 2.29–2.17 (m, 1H), 2.38 (s, 3H), 2.50–2.46 (m, 1H), 2.64–2.54 (m, 1H), 2.80–2.65 (m, 2H), 3.15–3.03 (m, 1H), 4.26–4.15 (m, 1H), 6.86 (dd, J = 8.2, 1.0 Hz, 1H), 7.18 (d, J = 8.2 Hz, 1H), 7.25 (s, 1H), 10.75 (s, 1H); 13C-NMR (125 MHz, DMSO-d6) δ 171.5 (CO), 134.2 (C, Ar), 130.5 (C, Ar), 127.1 (C, Ar), 124.1 (C, Ar), 122.0 (CH, Ar), 117.6 (CH, Ar), 114.8 (C, Ar), 110.8 (CH, Ar), 59.2 (C), 33.3 (CH2), 33.2 (CH2), 30.2 (CH2), 25.0 (CH3), 22.8 (CH2), 21.3 (CH3); ESI-LRMS m/z: 255 [M + H]+; ESI-HRMS m/z calcd for M + H+ 255.1492, found: 255.1489. The characterization data is in accordance with that reported in [43].
3.3. General Procedure for the Preparation of Compounds and
In conclusion, a green and general tandem reaction between alkynoic acids and amine nucleophiles through gold catalysis in water has been developed. This process proceeds with high efficiency leading to the formation of two rings and three new bonds in a single operation. This approach features low catalyst loading, good to excellent yields, high efficiency in bond formation, high step economy, excellent selectivity, great functional group tolerance, and extraordinarily broad substrate scope, and has been successfully employed to construct a high-quality library of indole/thiophene/pyrrole/pyridine/naphthalene/benzene-fused N-heterocycles. In addition, five antimicrobial compounds were discovered from the library, suggesting the value of our strategy to identify APIs. This is the first example of the generation of pDOS compound library encompassing skeletal diversity, molecular complexity, and drug-like properties from readily available materials through gold catalysis in water. We anticipate that these valuable N-heterocycles will find more pharmaceutical applications after our further investigations.
Authors: Alain Daugan; Pascal Grondin; Cécile Ruault; Anne-Charlotte Le Monnier de Gouville; Hervé Coste; Jean Michel Linget; Jorge Kirilovsky; François Hyafil; Richard Labaudinière Journal: J Med Chem Date: 2003-10-09 Impact factor: 7.446
Authors: Alla I Vaskevych; Nataliia O Savinchuk; Ruslan I Vaskevych; Eduard B Rusanov; Oleksandr O Grygorenko; Mykhailo V Vovk Journal: Beilstein J Org Chem Date: 2021-11-25 Impact factor: 2.883