AI Screening Identifies Two Drug Candidates for Leigh Syndrome
Researchers from Heinrich Heine University Düsseldorf, University Hospital Düsseldorf, and the University of Luxembourg have successfully utilized artificial intelligence and brain organoids to identify two potential drug candidates for Leigh syndrome, a rare and fatal mitochondrial disease. The findings, published in Nature Communications, offer new hope for a condition with no currently approved therapies. Leigh syndrome, affecting approximately one in 36,000 live births, is a progressive disorder that disrupts cellular energy metabolism, typically causing severe neurological symptoms in childhood and leading to early death. Research has historically been hampered by the scarcity of patients and a lack of accurate animal or cellular models that mimic the human disease process. To overcome these barriers, the international team developed human brain organoids, which are three-dimensional lab-grown structures that closely replicate the organization and function of the human brain. By using stem cells derived from patients, the researchers engineered organoids containing the specific genetic mutations responsible for Leigh syndrome. This innovation provided a reliable platform to study disease mechanisms and test treatments in a human-relevant context. The team employed a drug repurposing strategy, screening existing approved medications for potential use against this specific disease. To optimize this process, a deep learning algorithm developed by the Luxembourg team was applied to analyze how various drugs might affect the organoids. This AI-driven approach significantly accelerated the identification of promising candidates. The screening process pinpointed two drugs: talarozole, originally developed for acne treatment, and sertaconazole, a topical antifungal used for skin infections like athlete's foot. Laboratory results demonstrated that both drugs could sustain brain cell development, improve cell growth, and reduce lactate release within the organoids. Reduced lactate is a key indicator, as its accumulation is a hallmark of the mitochondrial dysfunction seen in Leigh syndrome. These improvements suggest that the medications could potentially slow disease progression and alleviate symptoms in patients, though further clinical trials are necessary to confirm efficacy in humans. Professor Alessandro Prigione, who led the study at the Düsseldorf team, emphasized the significance of this achievement, noting that the ability to model rare diseases using organoids is a major breakthrough. He expressed optimism regarding the potential of the identified drugs while acknowledging the need for further validation in patient studies. Professor Antonio Del Sol from the University of Luxembourg highlighted the broader implications of their work. He stated that the AI algorithm developed for this project could be adapted to screen for treatments for other rare diseases, making the drug repurposing process more efficient and accessible. The study marks a significant step forward in addressing the urgent unmet medical needs of patients with Leigh syndrome and demonstrates the power of combining advanced AI tools with cutting-edge biological models.
