The synthesis of 3-amino-5-arylisothiazoles from propynenitriles

Highlights

• Novel synthesis of 3-amino-5-arylisothiazoles.

• Tolerant to a range of functional groups.

• Operationally simple.

Abstract

A new synthesis of 3-amino-5-arylisothiazoles is reported. The reaction is operationally simple, utilises readily synthesised propynenitriles as starting materials and is tolerant of a range of functional groups. The optimised reaction conditions can also be used with 3-chloropropenenitriles in place of propynenitriles.

Introduction

During a recent drug discovery programme a range of N-(5-phenyl-1,2-thiazol-3-yl)amides 3 were required (Fig. 1). These were accessed via acylation of the corresponding 3-amino-5-arylisothiazoles 1. 3-Amino-isothiazoles are important heterocyclic building blocks incorporated into MCH1R antagonists and a drug used to treat schizophrenia.¹ A survey of the literature indicated that the 3-amino-5-arylisothiazole core is not well precedented. The reported syntheses require multiple steps to obtain suitable starting materials,² or require the use of concentrated hydrazine at elevated temperatures.³ Hackler and co-workers reported the synthesis of 3-amino-5-t-butylisothiazole 6 from Z-3-chloro-4,4-dimethyl-2-pentenenitrile 4 albeit in low yield (Scheme 1, i).⁴ We presumed this route would also be applicable to 3-chloro-3-phenylprop-2-enenitrile 7 and were pleased to isolate the desired product using the same conditions (Scheme 1, ii). As we continued to synthesise a range of analogues we found that the synthesis of a range of substituted 3-chloro-3-phenylprop-2-enenitriles were low yielding and somewhat capricious and therefore sought a more reliable route.

We envisaged the addition of sodium sulfide to 3-phenylpropiolonitrile 9 would generate a thioketoenolate similar to 5. Propynenitriles are readily synthesised via a variety of methods, including cyanation of alkynes,⁵ reaction of aldehydes with trichloroacetonitrile and base,⁶ and dehydration of benzoylacetonitriles with Mukaiyama’s reagent.⁷ Gratifyingly, the initial reaction using Hackler’s conditions yielded aminoisothiazole 8 in comparable yield (Scheme 2).

With this result in hand, optimisation of the reaction conditions using readily available 3-phenylpropiolonitrile 9 was investigated (Table 1). Initially this optimisation focused on the addition of sodium sulfide to generate thioketoenolate 10. Intermediate 10 is unstable to isolation but could be detected by LCMS in the crude reaction mixture.⁸ Analysis of the reaction mixture after the addition of sodium sulfide showed two new products were formed; one was the desired thioketoenolate (10) and the other was a dimeric species (11).

Varying the order of addition (Table 1, Entries 1 and 2) and increasing the number of equivalents of sodium sulfide (Entries 3 to 6) resulted in a higher yield of 10, however significant amounts of 11 remained. Changing the solvent to MeOH and MeOH/water gave comparable yields to EtOH (Entries 7 and 8), while other solvents resulted in lower yields and multiple by-product formation (Entries 9 to 11). A significant discovery was that heating the reaction to 70 °C for 30 min improved the ratio of 10:11, presumably the elevated temperature allowed the excess sodium sulfide to react with 11 to generate 10 in good yield (Entry 12). Optimisation of the second step revealed that 7 equivalents of chloramine were necessary to convert the intermediate to the desired product in reasonable yield (Scheme 3).

The optimised conditions were used to explore the substrate scope of the reaction (Table 2). The reaction is tolerant of electron withdrawing and donating groups, halogens, esters and nitriles. Heterocycles were also tolerated albeit with a reduced yield.

Finally, 3-chloro-3-phenylprop-2-enenitrile 7 was re-examined as a starting material under the optimised reaction conditions. Pleasingly, the desired product 8 was isolated in an improved 61% yield (Scheme 4), demonstrating that 3-amino-5-arylisothiazoles can be accessed from either starting material depending on availability.

In summary, we have developed reaction conditions which enable the synthesis of 3-amino-5-arylisothiazoles from either propynenitriles or 3-chloro-3-phenylprop-2-enenitriles. The reaction is fast, operationally simple and compatible with a broad range of substrates.