Direct synthesis of pterocarpans via aldol condensation

English: Pterocarpans, representing the second largest group of natural isoflavonoids, have received considerable interest on account of their medicinal properties over the last few years. These phytoalexins not only serve as antitoxins but also display antifungal, antiviral and antibacterial properties. Despite this, the study of these metabolites are restricted by their limited availability from natural sources. Furthermore, synthetic protocols allowing ready access to these compounds are restricted by the lack of availability of suitable starting materials and the potential introduction of stereoselectivity. Owing to the demand for enantiopure pterocarpans a direct stereoselective synthetic approach, based on the aldol condensation between appropriate phenylacetates and benzaldehydes, was developed. 2-Hydroxybenzaldehydes, protected as 2-0-methoxymethyl ethers, and 2- hydroxyphenylacetates, protected as TBDMS ethers, were subjected to aldol condensation employing lithium diisopropylamide, to afford the 2,3-diphenyl-3- hydroxypropanoates (40-76%, de = 22-100%). Subsequent reduction (LiAIH4), followed by Lewis acid (SnCI4, BnSH) deprotection of the 2'-O-MOM ethers, yielded the 3- benzylsulfanyl-2,3-diphenylpropanols (29-56%). Improved yields of these propanols were obtained by simply reversing the order of reactions (54-81%). B-ring formation using Mitsunobu conditions (TPP-DEAD) afforded the isoflavan silyl ethers in good yields (80-97%). The 2'-O-TBDMS derivatives were smoothly deprotected (TBAF) to yield the 2'-hydroxyisoflavans in excellent yields (96-99%). Finally, thiophilic Lewis acid (AgBF4, AgOTf or DMTSF) cyclisation produced the cis-pterocarpans in moderate to good yields (39-82%). Initial C-ring cyclisation (AgBF4) of the methyl 3-benzylsulfanyl-2,3-di(2- hydroxyphenyl)propanoates, followed by reduction (LiAIH4) and Mitsunobu (TPPDEAD) B-ring formation, afforded for the first time a trans-pterocarpan in a moderate overall yield of 12%. In order to address the issue of stereo control, we first attempted to introduce stereoselectivity during the aldol condensation. Stereoselective aldolisation employing diisopropylethylamine and chiral boron triflates, was evaluated utilizing achiral dibutylborontriflate. This system, though capable of effecting aldolisation, was ineffective to incorporate a broad range of substrates. Secondly, we converted the methyl propanoates to chiral derivatives of imidazolidin-2-one, bornane-l0,2-sultam and (lR,2S)-p-tol-N-norephedrine. Steric shielding of the enolates generated from these derivatives, prevented aldol condensation. Thirdly, using (-)-sparteine as chiral base afforded achiral products. Finally, in an effort to employ stereoselective epoxidation, attempts were made to synthesize 2-propenoates. All attempts to introduce the double bond gave disappointing yields. Although our attempts to introduce chirality failed, several alternatives still needs to be investigated in future endeavours. We have thus developed a highly efficient synthesis of cis-pterocarpans and succeeded in modifying this protocol to the novel synthesis of Irans-pterocarpanoids. Also, this synthetic protocol was modified to permit the stereoselective synthesis of 6ahydroxypterocarpans in high overall yields. The ease with which these protocols accommodate highly oxygenated substrates, featured by most natural pterocarpans, should contribute substantially to assess the chemical and physiological characteristics that may promote application of this class of phenolics as pharmaceutical or agricultural chemicals.