Elucidation of the mycosamine dependent sterol binding and antifungal activities of amphotericin B

Amphotericin B (AmB) is a clinically vital, yet highly toxic antifungal agent that is dependent on the presence of membrane sterols to exert its biological activity. This natural product has been shown to form ion channels in model membrane systems and therefore represents a molecule with the capacity to perform protein-like function, i.e. the formation of a transmembrane ion channel. Efforts to improve the therapeutic index of AmB or harness its potential to act as a molecular prosthetic to treat diseases that arise from a lack of protein function would benefit from an atomistic understanding of the mechanism of action. To elucidate this poorly understood mechanism, we developed a strategy focused upon the synthesis-enabled deletion of functional groups from the macrolide skeleton followed by determination of the biological and biophysical consequences. The first functional groups we targeted for deletion were the carboxylic acid and the mycosamine sugar appendages. Both of these functional groups had been predicted to be critical for the following roles: 1) stabilizing ion channel formation via an intermolecular salt-bridge interaction, 2) anchoring AmB to phospholipid bilayers and 3) binding to sterols. However, an alternative hypothesis states that AmB does not directly bind sterols but rather that sterol-induced global membrane properties are the source of the sterol dependency of AmB. This debate has been ongoing since the early 1970’s and had yet to be resolved at the time this work began. The functional group deletion strategy proved to be remarkably effective. Ultimately, we found that polar interactions between the acid and mycosamine are not required for ion channel formation or antifungal activity. In addition, we found that electrostatic interactions between the acid and/or mycosamine and the zwitterionic phospholipid headgroup are not required for AmB to partition into phospholipid bilayers. Finally, it was discovered that AmB directly binds ergosterol and cholesterol and that the mycosamine appendage is strictly required for this binding interaction. Furthermore, this binding interaction was shown to be absolutely required for forming ion channels and killing yeast cells. Consequently, the long contentious theory that AmB and membrane sterols participate in a functionally vital small molecule-small molecule interaction was finally confirmed. Based upon these results, and related studies with another natural product, natamycin, we proposed a novel, potentially general, two mechanism model to account for AmB’s potent antifungal activity. These discoveries provide a foundation for the more effective utilization of AmB and demonstrate the capacity of synthetic organic chemistry to illuminate even the most elusive aspects of small molecule function