ABSTRACT
A myriad of natural and/or biologically active products include nitrogen- and oxygen-containing saturated heterocycles, which are thus considered as attractive scaffolds in the drug discovery process. As a consequence, a wide range of reactions has been developed for the construction of these frameworks, much effort being specially devoted to the formation of substituted tetrahydropyrans and piperidines. Among the existing methods to form these heterocycles, the metal-catalyzed heterocyclization of amino- or hydroxy-allylic alcohol derivatives has emerged as a powerful and stereoselective strategy that is particularly interesting in terms of both atom-economy and ecocompatibility. For a long time, palladium catalysts have widely dominated this area either in Tsuji-Trost reactions [Pd(0)] or in an electrophilic activation process [Pd(II)]. More recently, gold-catalyzed formation of saturated N- and O-heterocycles has received growing attention because it generally exhibits high efficiency and diastereoselectivity. Despite their demonstrated utility, Pd- and Au-complexes suffer from high costs, toxicity, and limited natural abundance, which can be barriers to their widespread use in industrial processes. Thus, the replacement of precious metals with less expensive and more environmentally benign catalysts has become a challenging issue for organic chemists. In 2010, our group took advantage of the ability of the low-toxicity and inexpensive FeCl3 in activating allylic or benzylic alcohols to develop iron-catalyzed N- and O-heterocylizations. We first focused on N-heterocycles, and a variety of 2,6-disubstituted piperidines as well as pyrrolidines were synthesized in a highly diastereoselective fashion in favor of the cis-compounds. The reaction was further extended to the construction of substituted tetrahydropyrans. Besides triggering the formation of heterocycles, the iron salts were shown to induce a thermodynamic epimerization, which is the key to reach the high diastereoselectivities observed in favor of the most stable cis-isomers. It is worth noting that spiroketals could be prepared by using this method, which was successfully applied to a synthetic approach toward natural products belonging to the bistramide family. We then turned our attention to heterocycles incorporating two heteroatoms such as isoxazolidines. These frameworks can be found in biologically active natural products, and in addition, they can be transformed into 1,3-amino alcohols, which are of importance in organic chemistry. The use of FeCl3·6H2O allowed the access to a large variety of 3,5-disubstituted isoxazolidines from δ-hydroxylamino allylic alcohol derivatives with good yields and diastereoselectivities in favor of the cis-isomer. Recently, a Lewis acid-catalyzed synthesis of six- and five-membered ring carbonates starting from linear tert-butyl carbonates was reported. In some cases, the mild and chemoselective InCl3 was preferred over FeCl3·6H2O to avoid side-product formation. The resulting cyclic carbonates were easily transformed into 1,3- or 1,2-diols, and a total synthesis of (3S,5S)-alpinikatin was achieved.
ABSTRACT
The platform C14-C40, which can be used to prepare bistramide C and 39-oxobistramide K, was synthesized in 19 steps with an overall yield of 6.2%. Furthermore, the chemoselective reduction of the ketone at C-39 was performed giving an easy access to bistramides A, B, D, K, and L. Finally, the versatility of the synthesis of the C14-C40 fragment can allow the preparation of a large variety of stereoisomers to produce bistramide analogues.
