Genome-wide association study of idiopathic hypersomnia in a Japanese population

Abstract

Idiopathic hypersomnia (IH) is a rare sleep disorder characterized by excessive daytime sleepiness, great difficulty upon awakening, and prolonged sleep time. In contrast to narcolepsy type 1, which is a well-recognized hypersomnia, the etiology of IH remains poorly understood. No susceptibility loci for IH have been identified, although familial aggregations have been observed among patients with IH. Narcolepsy type 1 is strongly associated with human leukocyte antigen (HLA)-DQB1*06:02; however, no significant associations between IH and HLA alleles have been reported. To identify genetic variants that affect susceptibility to IH, we performed a genome-wide association study (GWAS) and two replication studies involving a total of 414 Japanese patients with IH and 6587 healthy Japanese individuals. A meta-analysis of the three studies found no single-nucleotide polymorphisms (SNPs) that reached the genome-wide significance level. However, we identified several candidate SNPs for IH. For instance, a common genetic variant (rs2250870) within an intron of PDE9A was suggestively associated with IH. rs2250870 was significantly associated with expression levels of PDE9A in not only whole blood but also brain tissues. The leading SNP in the PDE9A region was the same in associations with both IH and PDE9A expression. PDE9A is a potential target in the treatment of several brain diseases, such as depression, schizophrenia, and Alzheimer's disease. It will be necessary to examine whether PDE9A inhibitors that have demonstrated effects on neurophysiologic and cognitive function can contribute to the development of new treatments for IH, as higher expression levels of PDE9A were observed with regard to the risk allele of rs2250870. The present study constitutes the first GWAS of genetic variants associated with IH. A larger replication study will be required to confirm these associations.

Supplementary information: The online version contains supplementary material available at 10.1007/s41105-021-00349-2.

Keywords: Genome-wide association study (GWAS); Idiopathic hypersomnia; Phosphodiesterase; Single-nucleotide polymorphisms (SNPs); Sleep.

Fig. 1

Manhattan plot of the GWAS of IH. A genome-wide Manhattan plot was generated using the chromosomal positions of individual SNPs (X-axis) and the negative logarithm of P-values (Y-axis) calculated based on chi-square test in an allelic model. Horizontal blue line in the Manhattan plot indicates a suggestive association P-value (P = 1.0E−04)

Fig. 2

Regional association plots for the top four IH-risk loci. Each of the markers is indicated by a purple diamond. Color intensity indicates the extent of the LD with the marker SNP, red (r² ≥ 0.8), orange (0.6 ≤ r² ≤ 0.8), green (0.4 ≤ r² ≤ 0.2), light blue (0.2 ≤ r² ≤ 0.4), and dark blue (r² ≤ 0.2). Light blue in the background indicates the local recombination rate. a rs2049916 is located more than 70 kbp downstream of the nearest gene, LINC01370. b rs3092351 is the imputed SNP within the same region (LINC01370) as rs2049916. c rs11113904 is located in an intron of LINC01498. d rs3742302 is a missense variant of USPL1. e rs2250870 is located in an intron of PDE9A. f rs13049942 is the imputed SNP within the same region (PDE9A) as rs2250870

Fig. 3

Violin plots of expression levels of the nearest gene according to the genotypes of the top four IH-risk SNPs. All eQTL analyses were performed based on the Genotype-Tissue Expression (GTEx) project. The number of individuals of each genotype is in brackets. a Correlation between LINC01370 expression and rs2049916 in the liver. b Correlation between LINC01498 expression and rs11113904 in the testis. c Correlation between USPL1 expression and rs3742302 in the cortex of the brain. d Correlation between PDE9A expression and rs2250870 in whole blood