CRISPR-mediated gene correction links the ATP7A M1311V mutations with amyotrophic lateral sclerosis pathogenesis in one individual

Abstract

Amyotrophic lateral sclerosis (ALS) is a severe disease causing motor neuron death, but a complete cure has not been developed and related genes have not been defined in more than 80% of cases. Here we compared whole genome sequencing results from a male ALS patient and his healthy parents to identify relevant variants, and chose one variant in the X-linked ATP7A gene, M1311V, as a strong disease-linked candidate after profound examination. Although this variant is not rare in the Ashkenazi Jewish population according to results in the genome aggregation database (gnomAD), CRISPR-mediated gene correction of this mutation in patient-derived and re-differentiated motor neurons drastically rescued neuronal activities and functions. These results suggest that the ATP7A M1311V mutation has a potential responsibility for ALS in this patient and might be a potential therapeutic target, revealed here by a personalized medicine strategy.

Fig. 1. Filtering strategy to identify pathogenic mutation.

Trio whole genome sequencing was performed for the patient and his healthy parents to annotate inherited and de novo mutations. Candidate variants were prioritized according to measures of functional importance including allele frequency, high impact consequences, or in silico analysis. Variants were further prioritized using network-based prioritization or pathogenity scores. Nine variants were determined to be high priority; this list was narrowed down by review of related diseases, predicted genomic function, and selection of candidates with zygosity analysis. The final candidate mutation, ATP7A M1311V, was selected after a review of literature for neurodegenerative disease.

Fig. 2. CRISPR-mediated gene correction of the ATP7A M1311V mutation and construction of patient-derived isogenic iPS cell lines.

a Overview of our personalized medicine approach. b Strategy for correction of the ATP7A M1311V mutation in patient-derived iPS cells using CRISPR-Cas9. The ATP7A point mutation (A to G) is shown in red. The 20-nucleotide single guide RNA (sgRNA) target sequence and the protospacer adjacent motif (PAM) are shown in blue and yellow, respectively. The single-stranded oligodeoxynucleotides (ssODNs), which has 50-bp homology arms, was designed to change the disease-causing mutation to the normal sequence and disrupt the PAM sequence to prevent further Cas9-mediated cleavage after the correction. c Efficiency of homology-directed repair (HDR) for generating the gene-corrected isogenic iPS cell line. d Karyotype analysis of parental (ATP7A-M1311V) and gene-corrected (ATP7A-Cor1 and ATP7A-Cor2) iPS cell lines. e Sanger sequencing to confirm the mutant ATP7A M1311V and normal sequence in iPS cells, NPCs, and MNs.

Fig. 3. Gene correction of the ATP7A M1311V mutation results in improvement of pathophysiology.

a Time table showing differentiation protocol used for NPC from patient-derive iPS and MN from NPC. DMSB, Dorsomorphin and SB431542; EB, Embryoid body; RA, Retinoic acid; SAG, Smoothened Agonist; NTFs, Neurotrophic factors; EP, Electrophysiology. b Immunostaining with the anti-Nestin antibody in NPCs and anti-Tuj1 antibody after 3 weeks differentiation. Scale bars, 50 µm. c The Nestin-positive cells normalized to that of the 4′, 6-diamidino-2-phenylindole (DAPI)-positive area. n = 9. d The Isl1/2-positive cells normalized to that of the 4′, 6-diamidino-2-phenylindole (DAPI)-positive area. n = 14. e Immunostaining with the anti-isl1/2 antibody and anti-MAP2 antibody after 3 weeks differentiation. Scale bars, 50 µm. f The microtubule associated protein 2 (MAP2)-positive area normalized to that of the 4′, 6-diamidino-2-phenylindole (DAPI)-positive area. n = 13. g Immunostaining with the anti-MAP2 antibody (red). DAPI (blue) was used for nuclear staining. Scale bars, 50 µm. h LDH activity in motor neurons (MNs). n = 24 (Control-MN), n = 26 (ATP7A-M1311V-MN), and n = 20 (ATP7A-Cor1-MN). i, Quantification of the reactive oxygen species (ROS) level. n = 11 (Control-MN), n = 22 (ATP7A-M1311V-MN), and n = 14 (ATP7A-Cor1-MN). j, Quantification of the mitochondrial membrane potential using tetramethylrhodamine ethyl ester (TMRE) assay. n = 16 (Control-MN), n = 22 (ATP7A-M1311V-MN), and n = 12 (ATP7A-Cor1-MN). Data are presented as mean ± s.e.m,; unpaired two-tailed t test with Welch’s correction. Source data used for the graphs in (c, d, f) and (h–j) can be found in Supplementary Data 1.

Fig. 4. Action potential and current traces of voltage-gated sodium and potassium channels in gene-corrected MNs.

a Morphology of neural precursor cells (NPCs) after differentiation for 8 weeks for electrophysiology experiments. BF, bright field. Scale bars, 20 µm. b Representative traces of different types of action potentials: none, single, adaptive, repetitive. Repetitive firing was defined as action potentials that lasted for the duration of the current injection (1 s), whereas adaptive firing was defined as multiple action potentials that ceased before the end of the current injection. Comparisons between ATP7A-M1311V MNs and the other groups of cells were performed using data from recordings of firing cells. c Proportion of cells firing in each category as a measure of functional maturation. n = 23 (Control-MN), n = 25 (ATP7A-M1311V-MN), and n = 25 (ATP7A-Cor1-MN). The differences between groups were assessed using Fisher’s exact test with Bonferroni correction. d Sodium currents of Control-MN (n = 23), ATP7A-M1311V-MN (n = 25), and ATP7A-Cor1-MN (n = 30). Every data point was compared with the peak of the sodium current for ATP7A-M1311V-MN. e Potassium currents of Control-MN (n = 23), ATP7A-M1311V-MN (n = 25), and ATP7A-Cor1-MN (n = 30). Every data point was compared with the peak of the potassium current for ATP7A-M1311V. The error bars indicate ± s.e.m.; statistical analysis was performed using two tailed t-test. *p < 0.05, **p < 0.01. Source data used for the graphs in (c–e) can be found in Supplementary Data 2.