DMAPO/Boc₂O-Mediated One-Pot Direct N1-Acylation of Indazole with Carboxylic Acids: A Practical Synthesis of N1-Functionalized Alkyl Indazoles

Graphical Abstract

We describe the one-pot direct N-acylation of indazole with carboxylic acids using DMAPO/Boc₂O system. This system is simple and provides N1-acyl indazoles in high yield with high N1 selectivities. This new method exhibits a wide substrate scope (>40 examples). In addition, a new synthesis of N1-functionalized alkyl indazoles utilizing N1-acyl indazoles as starting materials was achieved.

Abstract

This report describes the one-pot direct N-acylation of indazole with carboxylic acids using our previously developed 4-(N,N-dimethylamino)pyridine N-oxide (DMAPO)/di-tert-butyl dicarbonate (Boc₂O) system. This simple system provides N1-acyl indazoles in high yield with high N1 selectivities and does not require the use of activated derivatives of carboxylic acids or high temperatures. This new method exhibits a wide substrate scope (>40 examples). In addition, a new synthesis of N1-functionalized alkyl indazoles utilizing N1-acyl indazoles as starting materials was achieved. This stepwise protocol is useful for the selective synthesis of structurally diverse N1-functionalized alkyl indazoles, which are difficult to synthesize by other methods such as the Mitsunobu reaction and classical S_N2 alkylation of indazole.

Introduction

Nitrogen-containing heterocycles are important building blocks for many bioactive natural products, commercially available pharmaceuticals, and drug candidates. As pharmacologically important scaffolds, they have attracted considerable attention from chemists.¹ Among several N-heterocycles, indazoles are one of the most important classes of heterocyclic compounds that serve as indole bioisosteres in medicinal research.² Indazole derivatives are rare in nature and have been isolated from extracts of a limited number of plant species.³ However, this particular skeleton in a variety of synthetic compounds possesses a wide range of pharmacological activities,⁴ such as anti-inflammatory,⁵ anti-HIV,⁶ anti-cancer,⁷ anti-fungal,⁸ anti-bacterial,⁹ and anti-hypertensive activities.¹⁰ Selected examples of indazole-based drugs and potential candidates are shown in Scheme 1A. The selective 5-HT₃ receptor antagonist Granisetron has been used clinically to prevent nausea and emesis induced by cancer chemotherapeutic agents.¹¹ Benzydamine is a widely applied locally acting nonsteroidal anti-inflammatory drug with local anaesthetic and analgesic properties.¹² Indazol-3-carboxylic acid derivatives such as Lonidamine comprise non-hormonal and non-steroidal, anti-spermatogenic agents and have found interest in male contraception.¹³ Merestinib, which was developed as a kinase inhibitor, possesses significant anti-cancer activity in vitro and in vivo.¹⁴ Finally, N-acyl indazole 1 can specifically inhibit human neutrophil elastase (HNE), an attractive target for the treatment of chronic and acute inflammatory lung diseases, with an IC₅₀ value of 7 nM.¹⁵ Therefore, current interest is directed toward the synthesis of novel indazole derivatives to identify potential pharmacophores.

Scheme 1

Current methods for N1-acylation of indazoles and this work. Part A shows selected examples of bioactive N-1 substituted indazoles. Part B shows stepwise approaches from preactivated carboxylic acids. Part C shows oxidative approaches from aldehydes. Part D shows this work. rt=room temperature. Cbz=benzyloxycarbonyl. p=para. TBHP=tert-butyl hydroperoxide. DCE=1,2-dichloroethane. DCM=dichloromethane.

Given their importance as privileged scaffolds in medicinal chemistry, methodologies for the synthesis and functionalization of indazoles have received significant attention.¹⁶ Among them, N-alkylation of indazoles is the most efficient and well-studied method for the synthesis of indazole derivatives. However, the direct N-alkylation of indazoles normally produces a mixture of N1 and N2-alkylated products.¹⁷ Thus, controlled alkylation of indazoles in a regioselective manner is still a highly challenging topic, specifically for C3-unsubstituted indazoles. In 2016, the Masson group reported a visible-light mediated photoredox-catalyzed coupling reaction of azoles including indazole with carbamoyl sulfides.^18a In 2018, significant breakthroughs were made by the MacMillan group to realize regioselective N1-alkylation of indazoles using carboxylic acids controlled by dual copper and visible-light mediated photoredox catalysis.^18b Subsequently, the successful discovery of visible-light mediated photoredox catalysis for the selective N1-alkylation of azoles including indazoles was reported by the Larionov group,^18c Yang and Ma group,^18d Nagao and Ohmiya group,^18e and Lei and Cai group,^18f respectively. Around the same time, electrochemical methods for N1-selective alkylation of azoles including indazole were reported by various research groups.¹⁹ Synthetic efforts have also been developed for the N2-selective alkylation of indazoles.²⁰ Among these approaches, one of the most impressive is the alkylation of indazole reported by the Chen group, in which N1/N2 selectivity can be reversed by the choice of ligands under Pd-hydride catalysis.²¹ An elegant selective N-alkylation method as described above would provide a powerful tool for the selective synthesis of N-alkyl indazole derivatives.

In contrast to the numerous methods of N-alkylation of indazoles that have been studied, methods for N-acylation of indazoles have not been explored much. The most general and classical method known is the reaction of indazole with carboxylic acid chlorides, which usually gives N-acyl indazoles in moderate yields (Scheme 1B, 1).^{15, 22} It is known that N1-acyl indazole is obtained preferentially over N2-acyl indazole in the N-acylation of indazoles. It has been suggested that this selectivity is due to isomerization of the N2-acyl indazoles to the more thermodynamically stable N1-acyl indazoles.²³ Similarly, Conrow and coworkers discovered the reaction of indazole with the succinimidyl ester of Cbz-L-alanine (Scheme 1B, 2).²⁴ This reaction proceeds under mild conditions to provide N1-acyl indazoles in high yields. Song and coworkers reported that the transamination reaction of N-acyl benzimidazole with indazole proceeds under high temperature conditions to give N-acyl indazole in high yield (Scheme 1B, 3).²⁵ Vannucci and coworkers have developed a selective N1-acylation reaction of indazoles using acid anhydrides by an electrochemical method (Scheme 1B, 4).²⁶ All of these methods are stepwise approaches using activated derivatives of carboxylic acids. Although the direct use of a carboxylic acid is the most ideal method, to the best of our knowledge, surprisingly few one-pot direct N-acylation reactions of indazoles with carboxylic acids have been reported in the literature. On the other hand, oxidative methods have been developed using aldehydes as acyl donor precursors. Li and coworkers reported the oxidative N-acylation of indazole using TBHP as an oxidant. The reaction of 4-bromobenzaldehyde with indazole proceeded under heating conditions to give N1-acyl indazole in 44 % yield (Scheme 1C, 5).²⁷ Leadbeater and coworkers developed an oxidative N-acylation of indazoles using oxoammonium salt X1. The reaction of 4-methylbenzaldehyde with indazole proceeded under mild conditions to afford N1-acyl indazole in 81 % yield (Scheme 1C, 6).²⁸ In 2018, Sunden and coworkers developed an oxidative N-acylation of indazoles using N-heterocyclic carbene catalyst X2. The reaction of 6-aminoindazole with 4-methoxycinnamaldehyde produced the corresponding N1-acyl indazole in 69 % yield (Scheme 1C, 7).²⁹ These sophisticated reports significantly advanced the research area, but a large amount of substrate generally was required for achieving high product yields. Furthermore, the scope of the substrates was relatively narrow (or not well explored).

To date, enormous efforts have been devoted to the development of amide bond formation reactions (i. e., N-acylation). Of those methods, the N-acylation reaction utilizing 4-(N,N-dimethylamino)pyridine N-oxide (DMAPO) as a nucleophilic catalyst is particularly rare but highly valuable. For example, Shiina and coworkers discovered in 2008 that DMAPO is an effective catalyst for amide bond formation reactions.³⁰ Furthermore, Shiina and coworkers developed an efficient protocol for macrolactamization and reported that DMAPO was the best catalyst for the process.³¹ Ishihara and coworkers reported boronic acid−DMAPO cooperative catalysis for dehydrative condensation between carboxylic acids and amines.³²

In this context, we originally focused on the development of one-pot direct N-acylation of less nucleophilic N-heterocycles with carboxylic acids.³³ We recently developed the DMAPO/di-tert-butyl dicarbonate (Boc₂O)-mediated one-pot direct N-acylation of less nucleophilic nitrogen compounds with carboxylic acids.³⁴ A wide variety of N-heterocycles was applicable to this reaction, such as indoles, pyrroles, carbazoles, pyrazoles, and lactams, and the scope of carboxylic acids was also quite wide. Furthermore, the high potential of this reaction was demonstrated in the reaction of less nucleophilic nitrogen compounds with bulky carboxylic acids.³⁵ Encouraged by this success, we thought to develop the one-pot direct N-acylation of indazole with carboxylic acids utilizing the DMAPO/Boc₂O system (Scheme 1D, this work). There are several challenges that need to be addressed: 1) direct use of carboxylic acids without prior derivatization; 2) achieving high yields with a 1 : 1 ratio of substrates; 3) high functional group tolerance and wide range of substrate scopes; and 4) high chemoselectivity between N1-acylation and N2-acylation. We describe herein the development of the one-pot direct N-acylation of indazole with carboxylic acids.

Results and Discussion

We screened conditions for the direct N-acylation of indazole with carboxylic acids (Scheme 2). We chose the challenging reaction substrate 2,2-diphenylpropionic acid 2 in order to discover conditions with a wider substrate scope. Initially, the reaction was carried out under the standard conditions for our method (i. e., 1 : 1 substrate ratio, 2 mol% of DMAPO, 1.1 equivalents of Boc₂O, 2 equivalents of Et₃N, MeCN as reaction solvent, 22 h at room temperature).^{35, 36} As a result, an inseparable mixture of N1-acyl indazole 3 and N2-acyl indazole 4 was obtained in 71 % yield with a 9/1 ratio of 3/4 (entry 1). N-tert-Butoxycarbonyl (Boc) indazole 5 was also obtained as a byproduct in 29 % yield. When the amount of Et₃N was reduced, a higher chemical yield was obtained, in 76 % yield at 1.0 equivalent of Et₃N and 85 % yield at 0.5 equivalent of Et₃N (entries 2 and 3). Interestingly, a further decrease in the amount of Et₃N (0.2 equivalent) reduced the N1/N2 selectivity (entry 4). Notably, no product formation was observed in the absence of Et₃N (entry 5). Although the effect of Et₃N stoichiometry on chemoselectivity of the reaction was unclear, this careful fine-tuning revealed that 0.5 equivalents of Et₃N was optimal. Finally, the reaction was carried out at 40 °C to afford N1-acyl indazole 3 in 79 % isolated yield with satisfactory N1 selectivity (>20/1) (entry 6). For structure determination of N1-acyl indazoles 3, see Supporting Information.

Scheme 2

Screening of conditions for selective one-pot direct N-acylation of indazole with carboxylic acid 2. 0.4 mmol scale. rt=room temperature. n.d.=Not determined. DCM=dichloromethane. Unless otherwise noted, reactions were performed with amine (1 equiv) at rt. [a] Inseparable mixture of 3 and 4. Yields are for purified material of isomers 3 and 4. [b] Ratios of isomers 3 and 4 were determined by ¹H NMR analysis. [c] 2.0 Equivalents of Et3 N were used. [d] 0.5 Equivalents of Et3 N were used. [e] 0.2 Equivalents of Et3 N were used. [f] Reaction was carried out at 40 °C. [g] 100 mol% of HOBt was used. [h] 100 mol% of HOAt was used.

To evaluate the above results, a variety of control experiments were conducted as follows. No product formation was observed in the reaction catalyzed by 4-methoxypyridine N-oxide (MOPO) (entry 7). Reactions catalyzed by 4-(N,N-dimethylamino)pyridine (DMAP), 4-pyrrolidinopyridine (PPY), 9-azajulolidine, and N-methyl imidazole (NMI) gave significant byproduct 5 in 95 %, 85 %, 90 %, and 63 % yields, respectively, and no formation of the desired N-acyl indazole was observed (entries 8, 9, 10, and 11). When dimethyl dicarbonate (DMDC) was used as the activating reagent instead of Boc₂O, byproduct 6 was obtained in 36 % yield and only a trace amount of the desired product 3 was obtained (entry 12). Similarly, the use of diethyl dicarbonate (DEDC) gave byproduct 7 in 36 % yield, with only trace amounts of N-acyl indazole 3 (entry 13). In the case of reactions using pivalic anhydride (Piv₂O) or pivaloyl chloride (PivCl) as activating reagents, no N-acyl indazoles were obtained, giving a complex mixture of unidentified byproducts (entries 14 and 15).

Under conditions using ethyl chloroformate (ECF) or isobutyl chloroformate (IBCF), both commonly used peptide coupling reagents,³⁶ no N-acylation of indazole proceeded at all (entries 16 and 17). The reaction of indazole and 2,2-diphenylpropionic acid 2 with CDI, a very attractive reagent for amide coupling, especially for large scale reactions,³⁶ produced a mixture of N-acyl indazoles 3 and 4 in 5 % yield, with lack of N1/N2 selectivity (entry 18). Similarly, N-acylation of indazole did not proceed under the reaction conditions using EDCI, which is frequently used in peptide coupling,³⁶ even in the presence of effective additives such as DMAP, HOBt, and HOAt (entries 19, 20, and 21). On the other hand, the reactions subjected to PyBOP and HBTU, which are known as efficient peptide coupling reagents containing HOBt,³⁶ yielded the active ester 8 as the byproduct in 95 % and 88 % yields, respectively (entries 22 and 23). In these reactions, N-acyl indazoles were not formed even after increasing the reaction temperature. This result suggested that the less nucleophilic indazole was not involved in the N-acylation reaction with the active ester 8. Among a number of inefficient conditions, the reaction with HATU, an excellent peptide coupling reagent,³⁶ gave byproduct 9 in 32 % yield and a mixture of the desired products 3 and 4 in 39 % yield, but the N1/N2 selectivity was poor (1.4/1) (entry 24). Finally, T3P, which has recently become a popular choice for amide coupling,^{36, 37} was also examined, with disappointing results (entry 25). From the above condition screening, we concluded that the optimal reaction conditions consisted of 2 mol% of DMAPO, 1.1 equivalents of Boc₂O, and 0.5 equivalents of Et₃N in MeCN (0.2 M).

With the optimal conditions in hand, the scope of direct N-acylations of indazole was tested using structurally diverse carboxylic acids (Scheme 3). Initially, the reaction with benzoic acid proceeded smoothly at room temperature to provide N1-acyl indazole 10 in 88 % yield. The reaction with 4-methylbenzoic acid and 4-bromobenzoic acid also proceeded under the same conditions to give N1-acyl indazoles 11 and 12 in 84 % and 85 % yields, respectively. Notably, no formation of N2-acyl indazoles was observed from these reactions. Furthermore, N1-acyl indazoles 10, 11 and 12 are reported compounds in the literature,^25-28 and the various spectral data were in perfect agreement. Therefore, although it is a well-known fact that N-acylation of indazoles normally proceeds in an N1-selective manner, our method was found to exhibit extremely high N1-selectivity. 4-Substituted and 3-substituted aromatic carboxylic acids were good substrates and the reaction proceeded smoothly at room temperature to give the corresponding N1-acyl indazoles (13–17 and 20–22) in good to excellent yields. In the case of N1-acyl indazoles 18 and 19, elevated reaction temperatures were necessary to achieve high yields and high N1 selectivities. Both electron-withdrawing (e. g., 13–18) and electron-donating (e. g., 19 and 20) groups were found to be compatible with this reaction condition. These results also demonstrated that diverse functionalities including ketone, ester, cyano, nitro, trifluoromethyl, sulfonyl, methoxy, aryl iodide, and boronic ester groups were well tolerated. It is worth noting that those functionalities can serve as handles for various functionalizations, including transition metal-catalyzed reactions. In contrast, the reaction with 2-substituted aromatic carboxylic acids gave the desired products (23 and 24) in lower but still synthetically useful yields. This direct transformation of indazole also proceeded smoothly with naphthalene-containing carboxylic acids to provide the corresponding N1-acyl indazoles in good yields (25 and 26). Furthermore, various heteroaromatic carboxylic acids, such as furans, pyridines, pyrazine and quinoline, were found to be excellent substrates, and the corresponding N1-acyl indazoles (27–32) were obtained in satisfactory yields.

Scheme 3

Scope of one-pot direct N-acylation of indazole with carboxylic acids. 0.4 mmol scale. Isolated yields are shown. [a] Known compound, see Refs. [25] and [26]. [b] Known compound, see Ref. [28]. [c] Known compound, see Ref. [27]. [d] The reaction was carried out using 1.0 equivalent of Boc2O. PMB=para-methoxybenzyl. Boc=tert-butoxycarbonyl.

The reaction with sterically less demanding but functionalized carboxylic acids proceeded smoothly to yield the desired acylated products (33–41) in excellent yields. Olefin and ester groups were tolerated under the conditions (34 and 35). To our delight, an alkyl bromide, which is reactive under basic conditions, was also tolerated (36). Acid-labile protecting groups such as PMB ether, ethylene acetal, and Boc carbamate were also well tolerated (38–41). Although reactions with these sterically less demanding carboxylic acids could also be carried out at 0 °C (35 and 38), in the specific case of the synthesis of 39, poor N1/N2 selectivity was observed when the reaction was performed at room temperature. When the reaction was carried out at 40 °C, a drastic improvement of N1 selectivity was observed, and compound 39 was successfully obtained with satisfactory purity.

α-Cyclic aliphatic carboxylic acids were also nicely converted to the desired N1-acyl products (42–44) in excellent yields. Subsequently, the generality of the reaction of indazole with sterically hindered α-fully substituted carboxylic acids was investigated. To our delight, this challenging reaction proceeded smoothly to afford the corresponding N1-acyl indazoles (45–49) in high yield, although an elevated reaction temperature was required. It is of note that both indazole and sterically hindered α-fully substituted carboxylic acids have been challenging substrates for one-pot direct N-acylation. Indeed, in Scheme 2, the synthesis of compound 3 failed when using common amide coupling reagents, and thus the current results in Scheme 3 demonstrate the impact of this method.

To further illustrate the synthetic utility of this method, gram-scale reactions were conducted for the synthesis of 37 in 92 % yield without a significant loss of reaction efficiency. See Experimental Section for details.

Scheme 4 illustrates that N1-acyl indazoles can be synthetically valuable starting materials for the selective synthesis of N1-functionalized alkyl indazoles (Scheme 4). First, reduction of N-acyl indazole 37 with DIBAL−H, followed by one-pot acetylation, afforded N,O-acetal 50 as the pivotal intermediate in 95 % yield (Scheme 4A). Treatment of this N,O-acetal 50 with allyltributyltin in the presence of TMSOTf gave N-alkyl indazole 51 in 77 % yield. A similar nucleophilic substitution reaction of N,O-acetal 50 with methallyltrimethylsilane provided N-alkyl indazole 52 in 65 % yield. Moreover, cyanation of N,O-acetal 50 with trimethylsilylcyanide proceeded smoothly in the presence of TMSOTf to provide nitrile 53 in excellent yield, which can serve as a precursor to numerous important functional groups in organic synthesis, including N-heterocycles, carbonyl compounds, and amines.³⁸ Additionally, we were pleased to find that this nucleophilic substitution reaction could be extended to silyl enol ethers derived from acetophenone, and ketone 54 was successfully synthesized in excellent yield.

Scheme 4

Synthesis of N-1 functionalized alkyl indazoles. Part A shows synthesis of N1-functionalized alkyl indazoles starting from N1-acyl indazole. Part B shows conventional approaches toward synthesis of 51. [a] A complex mixture containing dehydrated compound 56 was obtained. DCM=dichloromethane. DMF=N,N-dimethylformamide. Ac₂O=acetic anhydride. DMAP=4-(N,N-dimethylamino)pyridine. DIBAL-H=diisobutylaluminum hydride. TMSOTf=trimethylsilyl trifluoromethanesulfonate. DIAD=diisopropyl azodicarboxylate.

Finally, to demonstrate the potential usefulness of this stepwise but straightforward protocol for the selective synthesis of N1-functionalized alkyl indazoles, several control experiments were performed (Scheme 4B). For example, upon treatment of indazole with alcohol 55 under Mitsunobu reaction conditions,^17i-17k a complex mixture containing dehydrated byproduct 56 was produced, and the desired N-alkyl indazole 51 was not obtained at all. Similarly, under the conditions of N-alkylation by a classical nucleophilic substitution reaction of indazole^{17d, 17j, 17k} with methanesulfonate 57, a complex mixture containing dehydrated byproduct 56 was obtained, but no desired N-alkyl indazole 51. These results demonstrate that our newly developed protocol shown in Scheme 4A would be useful for the selective synthesis of N1-substituted alkyl indazoles. Thus, we have successfully developed a stepwise but selective and practical synthesis of N1-functionalized alkyl indazoles. Given the drug discovery importance of N1-substituted indazoles and the increasing interest in them, the method developed here may be attractive.

Conclusions

We described a DMAPO/Boc₂O-mediated one-pot direct N-acylation of indazole with carboxylic acids. Our developed method is amenable to a variety of functional groups, shows a wide range of substrate scopes, and does not require the use of activated derivatives of carboxylic acids. Furthermore, it is expected that our method would be a powerful tool for the selective synthesis of N1-acyl indazoles due to its simple experimental manipulation, scalability, efficiency, high selectivity, and high yield. In addition, we have successfully demonstrated that N1-acyl indazoles can be useful starting materials for the selective synthesis of various N1-functionalized alkyl indazoles, which are difficult to synthesize by other methods such as the Mitsunobu reaction and classical S_N2 alkylation. As such, we anticipate that this approach will have meaningful impacts within the drug discovery and synthetic chemistry communities.

Experimental section

Synthesis of N-acyl indazole 10

A flame-dried two-necked test tube (10 mL) equipped with a magnetic stirring bar and a rubber septum was charged with indazole (0.424 mmol, 50.1 mg), benzoic acid (0.425 mmol, 51.9 mg), DMAPO (8.6×10⁻³ mmol, 1.2 mg), and freshly distilled Et₃N (0.21 mmol, 0.03 mL) in 2.1 mL of dry MeCN under an argon atmosphere. Then the mixture was stirred at room temperature. Boc₂O (0.478 mmol, 110 μL, neat) was added via syringe at room temperature. After stirring for 15 hours, the resulting mixture was concentrated under reduced pressure to give a crude material. Purification of the crude material by flash column chromatography (silica gel, hexanes/Benzene) gave N-acyl indazole 10 (0.375 mmol, 83.5 mg, 88 %) as white solid.

Gram-scale synthesis of N-acyl indazole 37

A flame-dried two-necked flask (200 mL) equipped with a magnetic stirring bar and a rubber septum was charged with indazole (8.54 mmol, 1.01 g), 3-(Phenylmethoxy) propanoic acid (8.65 mmol, 1.56 g), DMAPO (0.18 mmol, 25.0 mg), and freshly distilled Et₃N (4.32 mmol, 0.60 mL) in 35 mL of dry MeCN under an argon atmosphere. Then the mixture was stirred at room temperature. Boc₂O (8.70 mmol, 2.00 mL, neat) was added via syringe at room temperature. After stirring for 18 hours, the resulting mixture was concentrated under reduced pressure to give a crude material. Purification of the crude material by flash column chromatography (silica gel, hexanes/Benzene) gave N-acyl indazole 37 (8.02 mmol, 2.25 g, 92 %) as colorless solid.

For details and synthesis of other compounds, see Supporting Information.

Acknowledgments

This work was financially supported by JSPS KAKENHI Grant Number JP23K14314.

Conflict of interests

The authors declare no competing financial interest.

Open Research

Data Availability Statement

The data that support the findings of this study are available in the supplementary material of this article.