Autosomal-dominant mutations within the gene FUS (fused in sarcoma) are responsible for 5% of familial cases of amyotrophic lateral sclerosis (ALS). The FUS protein is physiologically mainly located in the nucleus, while cytoplasmic FUS aggregates are pathological hallmarks of FUS-ALS. Data from non-neuronal cell models and/or models using heterologous expression of FUS mutants suggest cytoplasmic FUS translocation as a pivotal initial event which leads to neurodegeneration depending on a second hit. Here we present the first human model of FUS-ALS using patient-derived neurons carrying endogenous FUS mutations leading to a benign (R521C) or a more severe clinical phenotype (frameshift mutation R495QfsX527). We thereby showed that the severity of the underlying FUS mutation determines the amount of cytoplasmic FUS accumulation and cellular vulnerability to exogenous stress. Cytoplasmic FUS inclusions formed spontaneously depending on both, severity of FUS mutation and neuronal aging. These aggregates showed typical characteristics of FUS-ALS including methylated FUS. Finally, neurodegeneration was not specific to layer V cortical neurons perfectly in line with the current model of disease spreading in ALS. Our study highlights the value and usefulness of patient-derived cell models in FUS-ALS.
Keywords: Aggregate prone disease; Amyotrophic lateral sclerosis; Cortical neurodegeneration; Disease spreading; FUS-FTLD; Frontotemporal dementia; Fused in sarcoma; Human cell models; Motor neuron disease; RNA granules; Stress granules; Translated in sarcoma; iPSC.
Copyright © 2015 Elsevier Inc. All rights reserved.