Chem. Pharm. Bull. 53(2) 207—213 (2005)

widely in nature and have unique biological activities. Recently, Naganuma and his colleagues found that three natural triterpenes, betulin (1), uvaol (2), and soyasapogenol B (3), have reducing effects against the toxicity of cadmium chloride in HepG2 cells (Fig. 1). They also reported that betulin induced certain proteins, though not metallothionein, which is the representative protein in reducing heavy metal toxicity. To clarify the reduction mechanism of cadmium toxicity, we investigated the relationship between expression of activities and structures, with particular focus on the functional groups of betulin, using its analogues. The results showed that both the polar functional group, found at either the C3 or C28 positions, and the isopropenyl group play important roles in reducing cadmium toxicity and the cytotoxicity of betulin (1). However, it is difficult to carry out further investigation into the bioactivities of betulin because the only functional groups of this compound are hydroxyl and isopropenyl, both of which are crucial for its activity. Therefore, we decided to synthesize analogues of betulin, which could not be otherwise derived from natural products. Our synthetic strategy is summarized in Fig. 2, which shows the pentacyclic ring system being divided into two fragments: the AB fragment 4 and the DE fragment 5. Since our focus was on the angular substituents between the D and E rings, the starting material for 5 needed to be the optically pure bicyclo[4.4.0]decaline derivative 6b, and the starting material for 4 the Wieland–Miescher ketone 6a. The optically active Wieland–Miescher type bicyclic ketone 6a was effectively synthesized from the prochiral tricarbonyl compound 7a, using chiral proline as a chiral catalyst (Fig. 3). The optically pure 6a can be obtained by a single recrystallization, and can be utilized for subsequent syntheses of many natural products. However, the cyclization reaction of compounds with side chains other than a methyl group (e.g., allyl group 7b) was observed to have serious depreciation of the optical yield. In addition, proline did not work as a catalyst for this reaction, meaning a stoichiometric amount of proline is required to complete the reaction. If expensive synthetic proline has to be used, this presents a serious economic obstacle. Therefore, before starting the synthesis of the betulin analogues, it was decided to develop an efficient method for synthesizing optically pure bicyclo[4.4.0]decaline derivatives.