CIC bioGUNE is making progress on new therapeutic strategies for Leigh syndrome using computational biology and artificial intelligence

Two international studies published in Nature Communications and Cell demonstrate how the integration of computational models and experimental validation accelerates our understanding of drug mechanisms of action and the discovery of new therapies for rare neurological diseases

What if we could understand in detail the effect of drugs on cells and turn that knowledge into new therapeutic options for rare diseases? This is the central question addressed by two recent studies involving international researchers from Germany, Italy, Luxembourg and Spain, published in the journals Nature Communications and Cell.

Both studies highlight the strategic value of the synergy between computational biology and experimentation in advancing our understanding of the molecular mechanisms of drugs and, in particular, in discovering new therapeutic opportunities in the context of Leigh syndrome, a rare neurological disease associated with mitochondrial dysfunction.

Leigh syndrome is a severe neurodegenerative disorder, usually with an early-onset, for which there are very limited treatment options. Understanding its molecular basis and identifying effective pharmacological strategies represents a major scientific and clinical challenge. In this context, studies involving the Computational Biology group led by Professor Antonio Del Sol, an Ikerbasque researcher at CIC bioGUNE, a member of BRTA, and the LCSB, demonstrate how approaches based on systems biology, artificial intelligence and human cellular models can accelerate translational research.

Computational biology to understand drug mechanisms of action

The research team developed advanced computational approaches to analyse gene regulatory networks and altered cellular pathways in Leigh syndrome, with the aim of gaining a deeper mechanistic understanding of a drug’s mode of action and accelerating the development of new therapeutic strategies.

In the study published in Cell, computational methods enabled researchers to elucidate the mechanism of action of sildenafil, which is proposed in this work for the first time as a potential treatment for Leigh syndrome. The analysis revealed how this drug modulates key processes related to neurodevelopment and neuronal function, providing a mechanistic basis that could guide the future design of more effective therapies.

Complementarily, the study published in Nature Communications focused on the discovery of new therapeutic candidates using a deep learning-based algorithm. Leveraging this approach to accelerate the process, the research team carried out an additional drug repositioning screen in a second study, which led to the identification of Talarozole—originally developed for acne and psoriasis—as a promising candidate for Leigh syndrome. Its therapeutic potential was subsequently validated experimentally by collaborating research groups in relevant cellular models, and a patent application was filed for its use in mitochondrial diseases.

Experimental validation and translational potential

The computational predictions were evaluated in patient-derived cellular models obtained from induced pluripotent stem cells (iPSCs) that had been differentiated into neurons and cerebral endothelial cells. The results confirmed functional and molecular effects consistent with an improvement in the pathological cellular state.

Furthermore, the compassionate use of sildenafil in six patients showed preliminary clinical and motor improvements, although larger clinical trials will be necessary to confirm its efficacy and safety.

Taken together, both studies demonstrate that the integration of computational biology, including artificial intelligence and experimental validation, enables both an understanding of drug mechanisms of action and the discovery of new therapeutic options, accelerating the development of treatments for rare neurological diseases such as Leigh syndrome.