Scalable synthesis and properties of 7-methyl- 4-azaindole

Introduction

In medicinal chemistry, building an effective structure-activity relationship (SAR) study requires the possibility of introducing substituents into every position of the molecule identified as an initial hit [1]. Even the smallest structural modification, such as a methyl group, can sometimes dramatically influence the biological properties of the compound when placed in a proper site of the molecular scaffold [2]. Pyrrolopyridines (‘azaindoles’) are scaffolds which have been widely used in drug discovery; the recently registered anti-cancer drugs venetoclax (1) and vemurafenib (2) [3], an anti-HIV agent BMS-378806 (3) [4], an anti-mycobacterial compound 4 [5] and a potential agent for the treatment of the cognitive impairments 5 [6] can be mentioned among the most prominent examples (Figure 1).

Figure 1

Some biologically active azaindoles.

While the synthesis of various substituted azaindoles is generally well documented, some of the regioisomeric derivatives remain hardly accessible, in particular, 7-methyl-4-azaindole (6), a building block which was needed for our project on azaindole derivatives in multigram quantities. It was involved in the design of M₁ receptor positive allosteric modulators [6], muscarinic acetylcholine receptor modulators [7], decaprenylphosphoryl-β-D-ribose-2′-epimerase (DprE1) inhibitors [8] or HIV-1 attachment inhibitors [9]. However, all preparations of 6 reported so far (Scheme 1) were inefficient for the multigram preparation due to a dramatic yield drop at the key steps of Bartoli cyclization [10] and sp²–sp³ Suzuki coupling [8]) upon scale-up. In this work, we report an efficient method for the multigram synthesis of 6, which in our opinion may also be useful for the preparation of other 7-substituted 4-azaindole derivatives.

Scheme 1

The known syntheses of 7-methyl-4-azaindole (6).

Results and discussion

Our approach to the synthesis of compound 6 relied on the known reductive cyclization of 2-(3-nitropyridin-2-yl)acetonitriles of general formula 7 (Scheme 2) [11], [12], [13]. To the best of our knowledge, this method was not used for the preparation of 7-substituted 4-azaindole derivatives previously. The initial retrosynthetic approach to 6, according to the selected strategy, led us to compound 8. As there was no obvious way to assemble this 2,3,4-trisubstituted pyridine or any relevant synthetic precursor in a scalable manner (e.g., the related known compound 9 [14] is obtained via non-chemoselective nitration of amine 10), we switched to the tetrasubstituted derivative 11 as a possible synthetic equivalent of 8. The bromine atom was introduced into the molecule of 11 as a ‘place holding group’ for the sterically accessible C-5 position of the pyridine ring, which might be removed upon reductive conditions of the last step.

Scheme 2

Our synthesis of compound 11 commenced from 2-amino-4-methylpyridine (10), which is readily available from commercial sources on a kilogram scale (Scheme 3). Compound 10 was transformed into 2,3,4,5-tetrasubstituted pyridine 12 using reported procedures [15], [16], [17]. Diazotation of 12 in the presence of HBr and Br₂ gave dibromide 13 (65%), which was purified by crystallization from hexanes. Reaction of 13 with tert-butyl cyanoacetate proceeded in a regioselective manner and the product 14 was isolated in a 49% yield. It should be noted that 14 existed in equilibrium with its tautomer 14′. Acidic hydrolysis of 14 was accompanied by decarboxylation and led to the target compound 11 in an excellent yield of 92%. The key step of the synthesis – reductive cyclization of 11 – proceeded smoothly using inexpensive and common 10% Pd-C as a catalyst. As expected, heterocyclization was accompanied by debromination and gave the target 4-azaindole 6 in a good yield of 84%. Notably, the reaction sequence described above allowed the preparation of up to 30 g of 6 in a single run. In our opinion, the developed method allows the preparation of even larger quantities of the target compound.

Scheme 3

Some transformations were performed with azaindole 6 to demonstrate its chemical properties and potential for functionalization (Scheme 4). A partial reduction of the bicyclic system was achieved by treatment with BH₃·Me₂S, which furnished 2,3-dihydro-1H-pyrrolo[3,2-b]pyridine derivative 15 in a 71% yield. Further reduction of the pyridine ring was performed by catalytic hydrogenation; it occurred in a diastereoselective manner and gave (3aR*,7R*,7aR*)-isomer 16 in a 76% yield, which was isolated as a sulfate salt. The relative stereochemistry of 16·H₂SO₄ was established using the nuclear Overhauser effect spectroscopy (NOESY) experiment (Figure 2). It should be noted that direct hydrogenation of 6 to 16 was not successful. Compound 16 is a representative of bicyclic conformationally restricted diamines which are valuable building blocks in medicinal chemistry [18]. Finally, formylation of 6 with urotropine in aqueous AcOH gave 3-formyl derivative 17 in a 62% yield.

Scheme 4

Figure 2

Significant correlations in the NOESY spectrum of 16·H₂SO₄.

Conclusions

A convenient and scalable approach to 7-methyl-4-azaindole was developed. The method relies on using a bromine atom as a ‘place holding group’ for one of the carbon atoms of the pyridine ring throughout the reaction sequence, which is removed only upon the final reductive cyclization. The described reaction sequence can also be used for the synthesis of other substituted 4-azaindoles.

Experimental

The solvents were purified according to the standard procedures [19]. Compound 12 was prepared in a 85% yield from 10 using a reported method [15], [16], [17]. All other starting materials were purchased from commercial sources. Analytical thin-layer chromatography was performed using Polychrom SI F254 plates. Column chromatography was performed using Kieselgel Merck 60 (230–400 mesh) as the stationary phase. ¹H and ¹³C NMR spectra were recorded on a Varian Gemini 2000 spectrometer (at 400MHz for ¹H and 101 MHz for ¹³C). Mass spectra were recorded on an Agilent 1100 LCMSD SL instrument using electrospray ionization (APESI).

Supplementary material

¹H and ¹³C NMR spectra are available as online-only supplement.