CHCHD2 mutations in autosomal dominant late-onset Parkinson's disease: a genome-wide linkage and sequencing study

Background Identification of causative genes in mendelian forms of Parkinson's disease is valuable for understanding the cause of the disease. We did genetic studies in a Japanese family with autosomal dominant Parkinson's disease to identify novel causative genes. Methods We did a genome-wide linkage analysis on eight affected and five unaffected individuals from a family with autosomal dominant Parkinson's disease (family A). Subsequently, we did exome sequencing on three patients and whole-genome sequencing on one patient in family A. Variants were validated by Sanger sequencing in samples from patients with autosomal dominant Parkinson's disease, patients with sporadic Parkinson's disease, and controls. Participants were identified from the DNA bank of the Comprehensive Genetic Study on Parkinson's Disease and Related Disorders (Juntendo University School of Medicine, Tokyo, Japan) and were classified according to clinical information obtained by neurologists. Splicing abnormalities of CHCHD2 mutants were analysed in SH-SY5Y cells. We used the Fisher's exact test to calculate the significance of allele frequencies between patients with sporadic Parkinson's disease and unaffected controls, and we calculated odds ratios and 95% CIs of minor alleles. Findings We identified a missense mutation (CHCHD2, 182C>T, Thr61Ile) in family A by next-generation sequencing. We obtained samples from a further 340 index patients with autosomal dominant Parkinson's disease, 517 patients with sporadic Parkinson's disease, and 559 controls. Three CHCHD2 mutations in four of 341 index cases from independent families with autosomal dominant Parkinson's disease were detected by CHCHD2 mutation screening: 182C>T (Thr61Ile), 434G>A (Arg145Gln), and 300+5G>A. Two single nudeotide variants (-9T>G and 5C>T) in CHCHD2 were confirmed to have different frequencies between sporadic Parkinson's disease and controls, with odds ratios of 2.51 (95% CI 1.48-4.24; p=0.0004) and 4.69 (1.59-13.83, p=0.0025), respectively. One single nucleotide polymorphism (rs816411) was found in CHCHD2 from a previously reported genome-wide association study; however, there was no significant difference in its frequency between patients with Parkinson's disease and controls in a previously reported genome-wide association study (odds ratio 1-17,95% CI 0.96-1.19; p=0.22). In SH-SY5Y cells, the 300+5G>A mutation but not the other two mutations caused exon 2 skipping. Interpretation CHCHD2 mutations are associated with, and might be a cause of, autosomal dominant Parkinson's disease. Further genetic studies in other populations are needed to confirm the pathogenicity of CHCHD2 mutations in autosomal dominant Parkinson's disease and susceptibility for sporadic Parkinson's disease, and further functional studies are needed to understand how mutant CHCHD2 might play a part in the pathophysiology of Parkinson's disease.