Synthetic pathways to fern antheridiogens from Gibberellins

Following the discovery of an antheridium-inducing substance in gametophytes of the bracken fern, Pteridium aquilinum,1,2 several discrete compounds (for which the term antheridiogen has been coined) have been isolated from a number of other fern species.3 These compounds appear to play an important role in promoting cross-fertilization and thereby maintaining genetic diversity.4 The very small quantities that have been obtained (nanograms to micrograms in most cases) have rendered structural analysis extremely difficult, however. An important clue to the constitution of these intriguing substances in the case of the family Schizaeaceae was the observation that they possessed gibberellin (GA)-like reactivity5 and, conversely, that GAs had antheridium-inducing properties.6,7 Nakanishi et al. were eventually able to arrive at formula I for antheridic acid,8 the major antheridiogen from Anemia phyllitidis, a member of the family Schizaeaceae. The structure of antheridic acid was later refined to II,9 following total syntheses of the racemates of I and II.10 Antheridic acid (II) has also been shown to be a natural antheridiogen in further members of the Anemia genus, i.e., A. hirsuta,11 A. rotundifolia, and A. flexuosa,12 but could not be detected in A. mexicana.13 However, a new GA-like substance was obtained from this last species, for which structure III was initially suggested, but which was ultimately shown to be IV.13,14