A new research project at University College London aims to use artificial intelligence (AI) to develop treatments that attack diseased cells while sparing healthy ones in MND patients. This will enable safer delivery of gene therapies.

The project, led by Dr Oscar Wilkins, follows recent research on the nuclear protein, TDP-43, that in healthy people prevents unnecessary pieces of genetic material, called cryptic exons, from being mistakenly included in the final instructions (mRNA) used by cells to make proteins. In most cases of MND, however, TDP-43 moves out of the nucleus and accumulates in the cytoplasm of nerve cells. This leads to defective production of proteins that have an important role in neuronal function. Such defects are thought to impair neuronal function and contribute to neurodegeneration.

Dr Wilkins has previously developed a method, called TDP-REG, that uses this faulty system to sneak into the cell and activate potentially therapeutic genetic fragments that are hidden in cryptic exons. The elegance of this method is that the therapies are only activated in diseased cells, which reduces the risk of side effects, creating new treatment options for MND that might otherwise be too risky.

This project will use AI along with cell and animal models of MND to improve the safety of the TDP-REG technology to target diseased motor neurones while minimising the risk of off-target activation in other parts of the brain and body.

The £330,000 project, called ‘Improving the safety and specificity of the TDP-REG system for MND research and gene therapies’, is being jointly funded by Rosetrees, though a Lady Edith Wolfson Rosetrees Non-Clinical Fellowship, and the MND Association.