Furo[2,3-b]furanones in Natural Products: Synthesis, Derivatization and Biological Evaluation of (+)-Paeonilide and Studies toward Dermatolactone

The furo[2,3-b]furanone moiety is present in more than 100 natural products. Two representatives of this class which are substituted in the C-4 position are the highly oxygenated monoterpenoid (+) paeonilide (49) and the sesquiterpene dermatolactone (199). In the present thesis, an enantioselective synthesis of (+)-paeonilide (49) and derivatives was developed and their biological activity against the PAF-induced human platelet aggregation was evaluated. Furthermore, studies toward the total synthesis of dermatolactone (199) were conducted. The first chapter describes the enantioselective synthesis of (+)-paeonilide (49) starting from achiral 3-furoic acid methyl ester (100). The key step in order to introduce chirality was the asymmetric Cu(I)-catalyzed cyclopropanation to obtain bicyclus 142. As this reaction only gave an enantiomeric excess of 83% and recrystallization led to great loss of product, a chiral resolution was carried out to achieve enantiomerically pure product. Therefore, 142 was first hydrogenated and then the methyl ester was selectively saponified. After the coupling of oxazolidin-2-one 152, it was possible to separate the two diastereomers and reductive cleavage of the chiral auxiliary led to enantiomerically pure alcohol 157. Transformation of the hydroxyl group to a bromide and subsequent acid-mediated ring-opening/lactonization resulted in the formation of furo[2,3-b]furanone 159 which could be obtained as single diastereomer after purification. Oxidative ring-opening then enabled access to the introduction of an allyl group in the acetal position. Finally, substitution of the bromide using potassium benzoate and oxidation of the allyl side chain applying an oxymercuration/oxidation protocol gave the desired (+)-paeonilide (49). In this way, (+)-paeonilide (49) was afforded in an 11 step, straightforward synthesis with an overall yield of 10.4%. Remarkably, this is the only enantioselective synthesis of (+) paeonilide (49) until today. Moreover, this synthetic route allowed several modifications and one suitable precursor 169 for further derivatizations could be synthesized in 9 steps with an overall yield of 20.6%. The second chapter describes the synthesis of a small library of (+)-paeonilide derivatives. In total, 9 derivatives with modifications either in the two side chains or in the α-position of the lactone were synthesized. These derivatives and the synthesized (+)-paeonilide (49) were tested for the inhibition of the PAF induced platelet aggregation and the results are described in the third chapter. The bioassay indicated that the functional groups in both side chains have a great impact on the biological activity of the compound. Especially the oxygen functionality seemed to play a crucial role, as all deoxygenated derivatives (182, 184 and 186) showed a big drop in inhibitory effect. Furthermore, the introduction of long alkyl groups in the side chains (184 and 189) had an adverse effect on the biological activity. Comparing the different functional groups in the C-4 position, the benzoate (49) proved to be the best followed by the bromide (187) while the alcohol (188) was revealed to be completely inactive. The introduction of an alkyl group in the α-position of the lactone (192) showed a beneficial effect on the biological activity. The last chapter deals with studies toward the total synthesis of dermatolactone (199). The key step in this synthesis should be a decarboxylative cross-coupling of carboxylic acid 178 which was already obtained during the synthesis of (+)-paeonilide (49). As acid 178 was shown to be insufficient for a photocatalyzed decarboxylative conjugate addition, it was converted to its corresponding active ester 240. This compound underwent visible light-mediated cross-coupling with Michael acceptor 211 to form the addition product 242, however, accompanied by undesired decarboxylation product 243. As this reaction did not give the desired result, it was looked for an alternative synthetic route. In order to directly introduce an olefin, a Ni-catalyzed decarboxylative alkenylation was envisaged. Therefore, a suitable model substrate was synthesized first. After preparation of the corresponding alkenylzinc reagent 261 in situ, it should be coupled with N acyloxyphthalimide 240. Unfortunately, only traces of the desired product 262 were obtained in a first test reaction. However, this strategy or a combination of photoredox and metal catalysis should be more intensively investigated in the future to enable access to the synthesis of dermatolactone (199)