Abstract
Amyotrophic lateral sclerosis (ALS) is a rare neurodegenerative disease that causes progressive paralysis leading to death. In 90% of cases, the disease occurs sporadically, while the remaining 10% runs in families. Growing knowledge of the genetics of ALS and the development of gene therapies have led to an increase in genetic testing for the disease. The Biochemistry and Molecular Biology Laboratory at Nîmes University Hospital is one of the 3 reference medical biology laboratories (LBMR) that carry out molecular diagnosis of the disease. Molecular diagnosis is carried out for each patient, revealing variants of varying rarity that need to be classified. The impact of these variants on the disease is analyzed using an international system developed by the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP). This system offers 5 categories: Benign, Probably Benign, Variant of Uncertain Pathogenicity (VUS), Probably Pathogenic and Pathogenic. The real challenge in molecular diagnostics is the lack of key factors for interpreting variants, particularly in the context of initial analysis. Many of these variants are placed in the VUS class, due to a lack of information or contradictory information. One of the strong criteria of this classification system is the functional data, which is generally an important argument for classifying a variant in a pathogenic or benign class. The aim of this thesis is to set up a model to aid molecular diagnosis using zebrafish to create functional data for as many VUS variants as possible. We began by carrying out a functional study of this method on the SOD1 gene to ensure that it could work. The variant used for this first phase was SOD1 A5V, which is a variant known to be pathogenic in humans. Once it had been validated, we began injecting a larger number of variants and optimized the method to make it both more reliable and faster. The variants tested in this thesis project are: SOD1 D77V, SOD1 I113M, SOD1 H49P, SOD1 D91A, SOD1 V120L, SOD1 D77del, SOD1 G13A; V15G. The second objective is to apply these tests to another ALS gene. We have selected the FUS gene, which is a perfect candidate for this method. The results for the batch of SOD1 variants are very positive. They highlighted the impact on locomotor behavior and on the morphology of motor neuron axons in zebrafish larvae at 2 days post-fertilization. The results allow us to conclude on the pathogenicity of the V120L, I113M, H49P, D77del and G13A; V15G variants, which caused defects in the injected larvae. The D91A variant gave results that were like fish injected with SOD1 WT and to uninjected fish. Finally, the D77V variant showed less pathogenicity than the other variants tested, with non-significant results for locomotion but with affected motor neuron morphology. Finally, the preliminary results for the FUS variants are encouraging, with affected motor neuron axon morphology and reduced locomotion in the larvae. We have optimized a concentration at which injection of FUS WT does not cause problems for the fish. The results are not yet definitive but are promising for the implementation of this method for the FUS gene. The work in this thesis demonstrates the feasibility and reliability of these tests using zebrafish. These results provide a solid basis for developing this method on a larger scale. This will contribute to obtaining functional data very quickly for VUS variants to reclassify them. This will enable patients with these reclassified variants to have access to experimental protocols and their descendants to be able to carry out presymptomatic tests.