The application of CRISPR-mediated genome editing to study the pathophysiological role of proteins related to neurological disorders

Abstract

Recently CRISPR-Cas9 has become a key technology in constructing cellular and animal models to study the correlation between genetic mutations, protein properties, and the pathologies of various diseases. The combination of CRISPR-based genome editing and neural differentiation from human induced pluripotent stem cells (hiPSCs) further offers the opportunities to extend the studies of neurological diseases on human neurons. To obtain hiPSCs with the desired genomic modification, it is necessary to isolate numerous isogenic clones for verification by DNA sequencing, which are tedious and time consuming. Besides, to enhance neuronal maturity the hiPSC-derived neurons are often co-cultured with a feeder layer of glial cells or transplanted into murine hosts. As a result, it is difficult to examine the specific localization of the protein of interest in hiPSC-derived neurons because the surrounding cells in the feeder layer or host brain also express the same protein. Here we report a CRISPR-based knock-in approach to add an epitope tag to the protein of interest in hiPSCs, which not only facilitates clonal selection but can also be utilized to observe the subcellular localization of the tagged protein in human neuron. We first validate a modified protocol for the differentiation from hiPSCs towards a mixed population of mature cortical neurons that form synaptic connections. In the proof-of-concept study, we labelled endogenous β-Actin with a N-terminal hemagglutinin (HA) tag through genome editing. After differentiation of the genome-edited hiPSCs into cortical neurons, the enrichment of β-actin in the growth cones and dendritic spines were visualized by immunostaining through anti-HA antibody. We extended the application of this knock-in approach to study the pathophysiological role of fused in sarcoma (FUS), a protein linked to the moto-neurodegenerative disorder amyotrophic lateral sclerosis (ALS). A C-terminal V5 tag was introduced simultaneously with three ALS-related FUS missense mutations (R521C, R524S, P525L) of different clinical severity. The presence of V5 tag facilitated the selection of isogenic clones with heterozygous or homozygous modifications on the FUS gene. Through V5 immunostaining, we found that the subcellular localization of FUS protein was affected to varying degrees by these FUS mutations in hiPSCs and the differentiated cortical neurons. The mis-localization of FUS was coupled with distinct effects on dendritic growth and arborization of the hiPSC-derived cortical neurons. Interesting, the missense mutations P525L and R521C significantly increased the density of excitatory synapses in the hiPSC-derived cortical neurons. The subcellular localization of the V5-tagged FUS protein in human neurons in vivo was further demonstrated after transplantation of hiPSC-derived neurons into the mouse brain. Collectively, our findings suggested the feasibility of CRISPR-mediated knock-in strategy to generate disease-related mutations in human hiPSC-derived cortical neurons and study the subcellular mis-localization of the protein. The effects of FUS mutations have been mostly characterized previously in spinal motor neurons. Our study has unravelled specific impacts of FUS mutations on cortical neurons, which may have important implications as increasing evidence shows that ALS patients not only suffer motor deficits but also alterations in cognitive functions.