Evaluating the Cell-Type Specificity and Cellular Targets of Senotherapuetic Compounds with Unknown Mechanisms

National Institute On Aging

Investigators

Abstract

We initially screened eight flavonoids for senolytic activity (i.e., fisetin, myricetin, quercetin, apigenin, epicatechin, cyanidin c1, gingerenone, and procyanidin c1). Of these, fisetin and quercetin showed senescence specific senolytic activity in a dose-response study in both monocytes and renal epithelial cells. Myricetin, a structurally similar flavonoid, showed no senolytic activity, even at a maximal dose of 100 µM. In an analysis of proteomic changes associated with senescence in renal cells, senescence-associated proteins were enriched with plasma membrane ion transporters , including OAT transporters, known transporters of quercetin. Reasoning that elevated OAT transporters might lead to selective vulnerability to senolytic flavonoids via increased uptake of the compounds, we used targeted MS to measure intracellular levels of fisetin and quercetin to show elevated intake of senolytics into senescence cells as a possible mechanism of selective vulnerability. Using an OAT inhibitor, we could block the senolytic activity of Fisetin, suggesting that uptake through the OAT receptor mediates the mechanism. We are working to validate this further by genetically modulating OAT receptors with siRNA. To identify further protein interactions required for senolysis in monocytes, we utilized a chemoproteomic approach termed the cellular thermal shift assay with mass spectrometry (CETSA-MS), a method to screen for drug-protein interactions in living cells. To identify senolytic interactions while controlling for many potential non-senolytic interactions of flavonoids, we identified proteins that were thermally stabilized by fisetin and quercetin with respect to myricetin, a structurally similar non-senolytic. In the process of analyzing the data, we developed a novel computational pipeline, a software tool called SenoCETSA, to rigorously statistically assess potential drug-binding proteins, improving on published approaches. Applying our method, we identified 40 top protein candidate protein mediators of the senolytic mechanism of Fisetin or Quercetin. To validate our CETSA results, we performed an siRNA screen to knockdown the genes associated with the top 40 protein candidates. Notably, 27 of these were experimentally validated, suggesting novel and previously unknown mechanisms to target and kill senescent monocytes. In future work, we plan to identify drugs that modulate these proteins and test whether these mechanisms occur in vivo in mouse models. In summary, we identified several candidate mechanisms by which senescent monocytes and renal epithelial cells are vulnerable to fisetin and quercetin. One potential vulnerability mechanism in senescent renal cells is the increased uptake of flavonoids due to increased levels of OAT transporters. Direct interactions with one of several pro-death or pro-survival proteins identified by CETSA-MS may also underlie senolytic activity. We are pursuing both avenues in ongoing work.