Bioactivity-based classification of metabolites retained in kidney disease

Uremic toxins lead to high risk of adverse pathophysiological outcomes upon chronic kidney dysfunction. However, the specific toxins and mechanisms that are most pathogenic remain unclear. In this study, we investigated bioactivity of 93 putative uremic toxins systematically with morphological profiling (cell painting). As a result, we identified twenty-four bioactive uremic toxins that significantly induced dose-response changes in cell morphology. These toxins were clustered into four groups based on activity fingerprints, which predict their mechanism of action. The first group is Nrf2-activator related toxins. These toxins were strongly correlated to CDDO compounds, which induce HO-1 by Nrf2 or BACH1 regulation. Label-free proteomics analysis further supports the mechanism of action as Nrf-2 activator and the toxins potentially disrupt the oxidative stress response. The second group and third group exerted non-specific toxicities. Considering the second group only included bile acids, their function as detergents to disrupt cellular membranes may have led to non-specific toxicity. The third group correlated various compounds that are non-related to each other such as antibiotics x-206, salinomycin, narasin, and nigericin. The fourth group contained p-cresol and 4-hydroxyindole that disturb DNA replication. Genotoxicity was observed by cell cycle arrest at S phase and phosphorylation of γH2aX at Ser139. Lastly, adapting activity-based proteomics, we found strong evidence that the group 4 toxins form adduct on reactive cysteines on proteins, which may lead to adverse effects to cells. All in all, we provided unbiased classification of uremic toxins by their bioactivity This abstract was presented at the American Physiology Summit 2026 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.