Understanding ALS: insights from genetics, genomics and functional biology

Abstract

Amyotrophic lateral sclerosis is a progressive neurological disorder. It is characterized by the selective degeneration of central and peripheral motor neurons, leading to muscle wasting and weakness, and subsequent paralysis. Patients die on average 3 to 5 years after disease onset, mostly due to respiratory failure. Progress in understanding the genetic background of ALS has recently begun to improve our understanding of pathogenic changes underlying the disease. However, both at the preclinical and clinical levels, many questions remain to be addressed. In this thesis, we expand the knowledge on ALS pathogenesis by combining different approaches from genetics, genomics and functional biology. First, we used gene expression profiling to study ALS pathogenesis in the SOD1-G93A mouse model of ALS. We show that blood gene expression profiles can be used to distinguish transgenic from wild-type mice by studying overlap and differences in gene expression of different tissues in this mouse model. Also, we used gene expression profiling to study the effect of the only currently available drug for ALS, riluzole, in SOD1-G93A mice. Several pathways were identified as influenced by riluzole treatment, including ubiquitin-mediated proteolysis, RNA-protein interaction and mitochondrial function. Changed RNA-protein interaction has recently emerged as important pathogenic mechanism in ALS. Therefore, we next investigated the occurrence of rare variants in previously ALS-associated genomic regions on human chromosome 9 and 19 and found that rare variants in the coding parts of these regions were not associated with ALS. We have confirmed the presence of FUS variants in a cohort of Dutch ALS patients, at equal frequencies as has been described in other studies. We then focused on further studying cellular mechanisms associated with FUS-caused ALS. ALS-linked mutations in FUS often lead to cytosolic mislocalization of the protein. We therefore performed a protein-interaction screen and determined numerous FUS-interacting proteins, including SMN and FMRP. Both SMN and FMRP are involved in neurological disease (SMA and FXS, respectively) and, in addition, genetic variation in SMN has been shown to be associated with ALS. Genetic variation in FMRP, however, is not associated with ALS, as described in this thesis. Subsequently, we showed that both SMN and FMRP are sequestered into FUS-associated cytosolic protein aggregates in primary neuron models. The sequestration of these proteins into cytosolic protein aggregates leads to their depletion from the axon, the endogenous expression site of the proteins. This impairs their normal function, as both SMN and FMRP have specific axonal functions. Depletion of SMN by ALS-mutant FUS aggregates leads to morphological defects, particularly at the growth cone, in primary cortical neurons. When the effect of protein sequestration of FMRP was studied in a zebrafish model of FUS, we observed a decrease in integrity of the neuromuscular junction. Importantly, each defect could be rescued by restoring expression of SMN and FMRP, respectively. These findings illustrate how protein aggregation in ALS leads to sequestration of proteins, depleting them from endogenous sites of expression and thereby leading to defects in axonal connectivity. Approaches to decrease protein aggregation toxicity and tackle connectivity defects are promising therapeutic strategies for ALS.