Strategies for the synthesis of benzyltetrahydroisoquinoline alkaloids.

Abstract

The objectives of this project were to investigate the application of new methodologies for the preparation of benzyltetrahydroisoquinoline monomers and secondly, to synthesise the bisbenzyltetrahydroisoquinoline neferine and its analogues. Neferine was isolated from the roots of Nelumbo nucifera. This compound has been reported to exhibit important biological activities, which include anti-arrhymia, anti-platelet aggregation, anti-thrombosis, anti-cancer as well as anti-HIV activities. Moreover, neferine showed lower cytotoxicity compared to other isoquinolines. However, the total synthesis of this compound has not been reported. Two methodolologies based on the intramolecular hydroamination of aminostilbenes and aminoalkynes were investigated for the preparation of benzyltetrahydroisoquinolines with different oxygenation patterns. In these strategies, the aminostilbene and aminoalkyne precursors were successfully synthesised by the Heck and Sonogashira coupling reactions, respectively. The attempts to cyclise the aminostilbenes into the corresponding tetrahydroisoquinolines under base-catalysed, metal-catalysed and acid-catalysed conditions were unsuccessful. On the other hand, cyclisation of aminoalkynes into dihydroisoquinolines was achieved with the aid of titanium catalysts. Different titanium catalysts were tested for this hydroaminationation reaction. Optimum results were obtained with bis-(cyclopentadienyl)dimethyl titanium(IV) catalyst, albeit the yields were inconsistent when the reaction was performed on a larger scale. Induction of the desired stereochemistry on the dihydroisoquinolines prepared by the hydroamination of aminoalkynes was attempted with the chiral BINOL phosphoric acid catalyst without success. The catalyst was prepared in good yields and high enantiomeric excess from cheap and readily-available starting materials. Had this reaction been successful, this would have been a breakthrough in the stereoselective reduction of dihydroisoquinolines as most chiral catalysts, which are currently employed are expensive, difficult to prepare and some are air and moisture-sensitive. Although the first objectives of this project are not fully met, the results obtained in the synthesis of benzyltetrahydroisoquinolines by the hydroamination of aminostilbenes and aminoalkynes contribute greatly to the prevailing literature on the synthesis of benzyltetrahydroisoquinolines by these reactions. Presently, there is limited literature on the synthesis of benzyltetrahydroisoquinolines by these methods. Moreover, there is a need for the development of new synthetic strategies that would render benzyltetrahydroisoquinolines in minimum steps and good yields. It was planned that, upon successful synthesis of benzyltetrahydroisoquinolines from aminostilbene and aminoalkyne precursors, these modern methodologies would be applied in the synthesis of the two benzyltetrahydroisoquinoline scaffolds of neferine. However, these routes could not be pursued due to failure to ringclose the aminostilbenes and irreproducibility of results in the preparation of dihydroisoquinolines from aminoalkynes. Therefore, classical procedures were employed for the preparation of benzylisoquinoline nuclei of neferine. Three different synthetic routes were followed for the synthesis of neferine and its analogues. The first two methods were based on the Ullmann coupling reaction for the formation of the diaryl ether bond. The first method entailed an early construction of the ether link and late construction of the two isoquinoline rings on the ether bridge. The second method involved synthesis of the two isoquinoline nuclei, and coupling of the two units by the Ullmann reaction in the late stages of the synthesis. In the last synthetic strategy, the diaryl ether bridge was constructed by the nucleophilic aromatic substitution reaction. In all the three methods, the two isoquinoline rings were formed by the Bischler-Napieralski cyclisation reaction. In the first route, we succeeded in preparing the two major building blocks, which were N-(3,4-dimethoxyphenylethyl)-4-benzyloxy-3-iodophenylacetamide and [2- (4’-hydroxy-3’-methoxyphenyl)ethyl]carbamic acid tert-butyl ester. The Ullmann coupling of the two compounds afforded the diphenyl ether N-(3,4- dimethoxyphenylethyl)-4-benzyloxy-3-(4-(3-methoxyphenoxy)ethyl-tertbutylcarbamate) phenylacetamide, albeit in low yields. Although N-(3,4- dimethoxyphenylethyl)-4-benzyloxy-3-(4-(3-methoxyphenoxy)ethyl-tertbutylcarbamate) phenylacetamide was obtained in low yields, the successful formation of the diaryl ether bond from electron-rich haloacetamide and hydroxyphenethylamine is a great advancement in the synthesis of bisbenzyltetrahydroisoquinolines. In the second approach, the two benzyltetrahydroisoquinoline precursors for the Ullmann coupling reaction were successfully synthesised. These were the 7-hydroxybenzyltetrahydroisoquinoline and the 3’-iodobenzyltetrahydroisoquinoline. The Ullmann coupling reaction of the two isoquinolines did not give any fruitful results. In the last synthetic strategy, the formation of the diaryl ether bridge was based on the nucleophilic aromatic substitution reaction. In this route, we managed to synthesise the two coupling partners for the nucleophilic aromatic substitution reaction leading to Omethylneferine. One of the building blocks was the natural benzyltetrahydroisoquinoline, hydroxylaudanidine, and its coupling partner was N- [2-(4-fluoro-3-nitrophenyl)ethyl]-2-(4-methoxyphenyl)-N-methylacetamide. The major challenges in this route were encountered in the preparation the fluoroacetamide, which involved several low-yielding synthetic steps and tedious chromatographic purifications. The nucleophilic aromatic substitution reaction of the two precursors was attempted in vain. Even though the total synthesis of neferine could not be accomplished, it is strongly believed that the developed synthetic routes will enable us to complete the synthesis of the targeted compound and other naturally-occurring bisbenzyltetrahydroisoquinolines. The results obtained herein represent a significant advance considering the importance of the bisbenzyltetrahydroisoquinolines as biologically active compounds.