Researchers from the University of Edinburgh and the Spanish National Research Council have harnessed the power of artificial intelligence (AI) to discover three potent molecules that could potentially slow down the ageing process. The study, led by Integrated Biosciences, a biotechnology company dedicated to ageing research, demonstrates the tremendous potential of AI in uncovering novel senolytic compounds capable of suppressing age-related processes such as fibrosis, inflammation, and cancer.
Senolytic drugs are designed to eliminate senescent cells, which are metabolically active but no longer capable of replication, hence often referred to as “zombie cells.” These cells can contribute to age-related diseases by releasing inflammatory molecules that adversely affect neighboring cells.
The researchers trained machine learning models using known examples of senolytic and non-senolytic compounds. These models successfully differentiated between the two types and predicted the senolytic potential of previously untrained molecules. Following rigorous analysis, the team identified 21 potential drug candidates, three of which—periplocin, ginkgetin, and oleandrin—were found to effectively eliminate senescent cells without harming healthy cells.
The team plans to proceed with further investigations and is currently conducting tests on human lung tissue to validate the efficacy of these senolytic compounds. Although the results of these experiments may take approximately two years to be released, the discovery of these highly efficient drug candidates marks a significant milestone in longevity research and drug development.
Senescent cells are implicated in various age-related diseases, including diabetes, cancer, Alzheimer’s disease, and cardiovascular disease. Senolytic compounds selectively induce apoptosis in these non-dividing cells, but previous compounds faced challenges such as limited bioavailability and undesirable side effects. The newly discovered compounds, however, exhibit favorable medicinal chemistry properties, making them more promising for successful clinical applications.
This significant milestone achieved by Integrated Biosciences and their AI-guided approach showcases the potential of AI in transforming the field of drug discovery. By leveraging the capabilities of AI to explore vast chemical spaces virtually, researchers are able to identify promising compounds for further development and potential clinical trials.
The research, conducted in collaboration with scientists from the Massachusetts Institute of Technology (MIT) and the Broad Institute of MIT and Harvard, utilised deep neural networks trained on experimental data to screen over 800,000 compounds. The resulting three compounds, demonstrating high selectivity and potency as senolytics, bind to Bcl-2, a protein involved in regulating apoptosis and a target for chemotherapy. Moreover, these compounds exhibited favorable toxicity profiles in additional experiments.
Co-lead author Felix Wong, co-founder of Integrated Biosciences, emphasises the significance of these findings and their potential impact on clinical interventions. He believes that the AI-driven discovery of multiple anti-ageing compounds with superior properties compared to existing senolytics holds tremendous promise for restoring health in ageing individuals.
