UCSF

This thesis describes my odyssey with ternatin, a fungus-born poison that kills cancer cells. It follows the path of chemical biology––using tools from chemistry to interrogate biological systems. Inspired by the rich history of natural products as both clinical drugs and chemical probes, we set out to discover ternatin’s target and mechanism of action. We found that ternatin disrupts a central biological process, protein translation, by binding to the translation elongation factor 1A.

Chapter 1 introduces the reader to the deep field of translational elongation. eEF1A participates in a carefully orchestrated series of reactions with guanine nucleotides, transfer RNAs (tRNAs), a guanine nucleotide exchange factor, and the elongating ribosome. Thus, the possible points of inhibition and pharmacological implications for each possible inhibitory mechanism are myriad. Chapter 2 describes initial work synthesizing improved ternatin congeners. Ternatin-4, our most potent compound, killed cancer cells at doses ranging into the high picomolar. Furthermore, we synthesized a photo-affinity ternatin derivative to directly identify the cellular target. Photo-labeling experiments, along with biochemical reconstitution and a drug-resistance mutation identified the eEF1A ternary complex with GTP and aminoacyl tRNA (aa-tRNA) as the target of ternatin.

We next followed ternatin into the field and methods of translation biochemistry, working with collaborators to characterize the kinetic and structural basis of ternatin’s mode of action: trapping eEF1A on elongating ribosomes. Chapter 3 thus describes the mechanism of ternatin and compares it with an unrelated natural product, didemnin B. Didemnin B competes with ternatin for eEF1A binding, and traps eEF1A on the ribosome, but in a kinetically distinct state. Comparative pharmacological analysis of ternatin-4 and didemnin B in intact cells revealed that biochemical differences correlated with qualitative pharmacological differences; despite similar potencies under continuous treatment, the effects of ternatin-4 could be reversed upon washout, whereas didemnin B was irreversible. Intriguingly, ternatin-4 was also less toxic to cardiomyocytes than didemnin B, leading us to speculate that ternatin-4 might have an improved safety profile compared to didemnin B.

Finally, while musing on recent findings in translational quality control, specifically that the first known mechanistic step for rescuing aberrantly stalled ribosome ought to be inhibited on ternatin-trapped ribosomes, we unexpectedly found that ternatin induces degradation of eEF1A. Degradation appears to be specific to ternatin’s mechanism of action, as didemnin B did not induce similar degradation. Proteasome inhibitors prevented eEF1A degradation, suggesting the presence of an as-yet undiscovered ubiquitin pathway for targeted eEF1A degradation. We describe a fluorescent reporter system for studying ternatin-induced eEF1A degradation. Studying the ternatin-induced eEF1A degradation pathway may reveal new components of the ribosome-associated quality control pathway.