1521-3773/asset/olalertbanner.jpg?v=1&s=0c0a7bae38bad177497650f0bc0083504684f4fa)
The copper(I)-catalyzed azide–alkyne cycloaddition (CuAAC)1 forming 1,2,3-triazoles has become a prime example of click chemistry because of its reliability, specificity, and biocompatability. Click chemistry has been widely used in different areas of science, such as drug discovery, bioconjugation, polymer and supermolecular chemistry.2 The triazole products of this reaction are more than simply connecting linkers: they are very important pharmacophores, widely used in medicinal chemistry.3 Among them, the multisubstituted 5-heterofunctionalized 1,2,3-triazoles are a type of privileged triazoles, present in many synthetic molecules with a variety of biological activitites (Scheme 1). For instance, carboxylamidotriazole I (CAI) exhibits anticancer activity,4a the triazole II is an active potassium channel activator,4b the triazole III is a potent h-NK1 antagonist,3b the sulfur-containing triazole IV is a potential herbicide with antifungal activity,4c and the triazole V is an excellent chiral ligand used in asymmetric catalysis.4d However, the CuAAC reaction is limited to terminal alkynes and cannot produce these functionalized structures. Consequently, the development of an efficent strategy to access diverse hetero-functionailized triazoles is of great importance.
Scheme 1.
Important 5-functionalized triazoles.
In 2010, Fokin and co-workers reported a sequential halide exchange and subsequent SNAr reaction to produce various 5-functionalized triazoles.5 This strategy requires multiple reaction steps and a halide exchange step that involves high temperatures. The recently reported RuAAC6a and IrAAC6b reactions have partially addressed this challenge, and could afford trisubstituted triazoles with high regioselectivity. However, these approaches need to use noble metal catalysts and prepare the functionalized internal alkynes in advance. To date, no widely applicable direct synthesis of various heteroatom-functionalized triazoles from easily available terminal alkynes has been reported.7
The CuAAC reaction shown in Scheme 2 A is an effective approach to triazoles. The Ackermann group reported a copper-catalyzed one-pot reaction to introduce an aryl group onto the triazole ring at elevated temperature.8 Herein, we proposed another interrupted click reaction using a heteroatom electrophile to intercept the cuprate–triazole intermediate M1 forming 5-hetero-functionalized triazoles (Scheme 2 B). Previously reported successful interception of this intermediate with ICl or allyl iodide demonstrated the feasibility of this method.9–11 However, most reactions require stoichiometric amounts of a copper(I) catalyst. Realizing a catalytic reaction is highly desirable, but completion of this catalytic cycle is challenging because of two competing reactions: (1) protonation of the cuprate–triazole intermediate M1 producing the undesired 1,4-disubstituted triazole; and (2) reaction of copper(I) acetylide with the heteroatom electrophile generating a heteroatom-substituted internal alkyne, which is unreactive under click reaction conditions. Therefore, choosing an appropriate heteroatom electrophile with enough reactivity toward the vinyl copper intermediate, rather than the copper acetylide, is critically important.
Scheme 2.
Copper(I)-catalyzed interrupted click reaction by electrophiles to diverse 5-functionalized triazoles.
Previously, S-methyl benzenesulfonothioate 3 a was used successfully by Dai and Tang to introduce a sulfenyl group into D-camphor, building a chiral sulfide.12 Compound 3 a might be an excellent electrophilic sulfenylating reagent, because the cleaved benzenesulphinate is a good leaving group, with the pKa of phenylsulfinic acid being 2.76. To validate this concept, the phenylacetylene 1 a and the benzylazide 2 a were selected as model substrates with which to optimize the reaction conditions (Table 1; see the Supplementary Information for details). It was found that benzenethiosulfonate (3 a) serves as an efficient electrophile, affording the desired 5-thiotriazole (4 a) in 74 % yield, together with 8 % of the 5-H-triazole (5 a) in the presence of 20 mol % CuI (Table 1, Entry 1). Almost no thioalkyne (6) was observed in the reaction. Other electrophilic sulfenylating reagents (7, 8)13, 14 were also tested. The expected thiotriazole was not observed, but only the click product (5 a; Entries 2 and 3). Further screening of different bases indicated that LiOtBu was crucial for the success of this reaction (Entries 4–8). The highest isolated yield (90 %; Entry 10) was obtained by raising the temperature to 40 °C and increasing the amount of 3 a to two equivalents. Using an electrophilic amination reagent (9 a),15 5-amino triazole (10 a) could be obtained in 83 % yield by a similar CuI-catalyzed three-component reaction [Eq. (1)].
(1)| Entry | Base | Temp [°C] | Electrophile | Yield [%][b] | |
|---|---|---|---|---|---|
| 4 a | 5 a | ||||
| |||||
| 1 | LiOtBu | 25 | 3 a | 74 | 8 |
| 2 | LiOtBu | 25 |  | 0[c] | 66 |
| 3 | LiOtBu | 25 |  | 0 | 96 |
| 4 | KOtBu | 25 | 3 a | <5 | 64 |
| 5 | NaOMe | 25 | 3 a | 30 | 52 |
| 6 | NaH | 25 | 3 a | 25 | 56 |
| 7 | K2CO3 | 25 | 3 a | <5 | 92 |
| 8 | Et3N | 25 | 3 a | <5 | 94 |
| 9 | LiOtBu | 40 | 3 a | 81 | 5 |
| 10 [d] | LiOtBu | 40 | 3 a | 95 (90) | trace |
With the optimized conditions defined, we investigated the scope of various alkyne structures (Table 2). Both aromatic and aliphatic alkynes can participate in this three-component reaction to give the 5-thio- or 5-amino-triazoles in good to excellent yields. Electron-donating groups, such as methyl (4 e and 10 e) and methoxyl (4 f and 10 f), and electron-withdrawing groups, such as halogen (4 b,c and 10 b,c) and cyano (4 d and 10 d) moieties, are compatible with the reaction. Alkynes substituted by thienyl (4 g,h and 10 g,h), cyclopropyl (4 i and 10 i), and phenoxyl (4 k and 10 k) are also suitable for this transformation, giving good yields of the corresponding triazoles.
 |
|---|
|
 |
The reaction is also efficient with various alkyl and aromatic azides (Table 2). All of the aliphatic azides tested were effective substrates, giving the corresponding triazoles in good to excellent yields under standard conditions. Cinnamyl groups (4 o and 10 o), phthalimide-protected amines (4 q and 10 q), and indole skeletons (4 r and 10 r) are all tolerated under these mild conditions. Furthermore, the reaction is also efficient for aromatic azides, generating the 5-thio- and aminotriazoles in good yields (4 s and 10 s).
We next examined the scope of this transformation with respect to different heteroatom electrophiles (Table 3). For the sulfenylation reaction, a large variety of alkyl and aromatic thio groups could be easily introduced onto the triazole ring, giving 5-thiotriazoles (4 t-ab), mostly in very good yields, and functional groups such as allyl, halogen, and nitro are well tolerated. Significantly, 5-phenylselenyl triazole (11) could be produced in 71 % yield by this approach under very mild conditions. Various amines such as morpholine, piperidine, diethyl amine, and diallyl amine can all be installed on the triazole ring and give good yields (10 t–w). The structures of 4 aa and 10 n were unambiguously characterized by single-crystal X-ray crystallography. All of these heteroatom electrophiles can easily be prepared from readily available starting materials (Supporting Information), thus making this method a practical approach to various 5-hetero-functionalized triazoles.
 |
|---|
|
 |
The construction of medium and large rings is always a challenge in the organic synthesis,16 and we questioned whether this copper-catalyzed reaction can be applied to an intramolecular reaction to construct various ring systems. By linking the thiosulfonyl and alkyne groups in the same molecule, we synthesized the functionalized alkyne 12 (Supporting Information). The reaction of alkyne 12 with azide 2 a afforded the fused bicyclic triazole 13 in moderate to good yields (Table 4). Many 5- and 6-membered rings, and also 7- to 14-membered medium and large rings, can all be synthesized in good yields. Notably, the catechol-derived 12-membered ring 13 h and 14-membered ring 13 i can be formed in 80 % and 76 % yield, respectively. The structure of 13 h was confirmed by single-crystal X-ray crystallography.17 For formation of 8–14-membered ring systems, a slow injection of the alkyne substrate 12 into the reaction system was necessary.
 |
|---|
|
 |
The utility of this chemistry is demonstrated by its application to a two-step synthesis of the antifungal thiotriazole IV (Scheme 3).4d This biologically active compound was originally prepared from highly toxic thiophosgene and ethyl diazoacetate in multiple reaction steps with less than 15 % overall yield. However, it can be synthesized in 70 % yield by the standard three-component reaction of the azide 2 a, propiolic acid ethyl ester, and the sulfenylating reagent 14, which can be easily prepared by the reaction of PhSO2SNa with corresponding iodide.
Scheme 3.
Synthesis of antifungal active thiotriazole IV.
The significance of this chemistry is seen in the late-stage click/functionalization of bioactive natural compounds, saccharides, and amino acid derivatives shown in Scheme 4. Treatment of oleanane-type triterpene-derived alkyne 15 under standard reaction conditions afforded the corresponding thiotriazole 16 in 91 % yield. This reaction has great potential for drug discovery and development. Furthermore, the glucose derivative 17 and the amino acid derivative 19 were also amenable to this transformation giving desired products 18 and 20 with good yields.
Scheme 4.
Late-stage functionalization of biologically active molecules.
To gain an improved understanding of the reaction mechanism, controlled experiments were conducted. The reaction of phenylacetylene (1 a) with 3 a under standard conditions afforded the thioalkyne (6) in 90 % yield. However, the reaction of 6 with benzyl azide (2 a) failed to afford the target product (4 a) [Eq. (2)]. Disubstituted triazole (5), the major by-product of this reaction, failed to react with 3 a to form 4 a [Eq. (3)]. Addition of one equivalent of tetramethylpiperidine-1-oxyl (TEMPO), had almost no effect on the reaction, indicating that a radical process may not be involved [Eq. (4)]. Based on these experiments, a possible reaction mechanism was proposed (Scheme 5). The cycloaddition of copper(I) acetylide M0 with the azide 2 generates the cuprate–triazole M1. This intermediate reacts with the heteroatom electrophile E-LG to form the product 4, possibly through an oxidative addition and reductive elimination sequence. The recovered Cu-LG reacts with the alkyne in the presence of base to form copper(I) acetylide M0, completing the catalytic cycle.
(2)
Scheme 5.
Proposed reaction mechanism.
In summary, we have developed a copper(I)-catalyzed interrupted click reaction to access diverse 5-functionalized triazoles. The reaction proceeds under very mild conditions, with only catalytic amounts of an inexpensive copper catalyst and no required ligands. This practical method exhibits a very broad scope with respect to alkynes, azides, and also different heteroatom electrophiles. The intramolecular reaction can be used to construct bicyclic triazoles with various ring sizes. A notable feature of this reaction is the late-stage functionalization of bioactive compounds, thus providing efficient and practical synthetic routes for drug discovery and development, which the current click reaction (CuAAC) is unable to do. Further applications of this interrupted click reaction in medicinal chemistry are in progress in our laboratory.