| Kidney disease and toxicity is increasingly being tied to mitochondrial dysfunction. Cheminformatic modeling and high-throughput phenotypic screens can discover discrete subsets of selective ligands that correlate with mitochondrial biogenesis (MB) and accelerate recovery from acute kidney injury (AKI). Serotonergic (5-HT) stimulation has been shown to cause MB. Subtype specific ligands for 5-HT1F and 5-HT2A/C receptors show increased uncoupled oxygen consumption rate (OCR). Our overall goal, in a screening assay, is to discover small molecules that increase MB to be used as drugs in the treatment of diseases that exhibit mitochondrial dysfunction. To achieve this goal we created pharmacophore models of these ligands to virtually explore new chemical space in large chemical libraries to narrow the list of compounds tested to a tractable number. 4394 compounds were identified for 5-HTIF from a 1.6 M compound library and 112 for 5-HT2A/C from a 50,000 compound library. These compounds were screened for mitochondrial biogenesis and identified with chemotypes generated by computational approaches including pharmacophore and ligand-receptor docking. Additionally, pharmacophores and ligand-receptor docking provided increased correlation with biological activity compared to chemotypes generated by the much more commonly used Tanimoto Coefficient (TC). These descriptive and predictive models were then used to probe vast amounts of virtual small molecule chemical space to discover novel scaffolds for mitochondria) biogenesis within virtual compound databases, and these compounds were then validated in renal proximal tubule cells (RPTCs) using a respirometry assay. Two compounds identified by the 5-HT1F pharmacophore and 5 compounds identified by the 5-HT2A/C pharmacophore were found to be biogenic in the RPTCs and result in MB. The compounds found with the 5HT2A/C pharmacophore were then evaluated for receptor internalization and calcium flux to show receptor specificity. Novel predictive models and experimental therapeutics identified here, will allow for continued identification of pharmacologic agents that cause MB which will bring the reality of new therapeutics for kidney disease one step closer to the clinic. |
