A Cross-Tissue Transcriptome-Wide Association Study Identifies New Susceptibility Genes for Insomnia

Background

Insomnia is the second most prevalent psychiatric disorder, affecting nearly one-third of the global population. It not only diminishes quality of life but also increases the risk of cardiovascular diseases, metabolic disorders, mood disorders, and neurodegenerative diseases. Although insomnia has a significant heritability (estimated at 22%-25%), the understanding of its genetic mechanisms remains limited. Traditional genome-wide association studies (GWAS) have revealed some genetic loci associated with insomnia, but they rely solely on genotype data, potentially overlooking the impact of gene expression regulation on disease risk. Therefore, researchers aimed to identify new susceptibility genes for insomnia by integrating gene expression data with GWAS analysis using transcriptome-wide association study (TWAS) techniques, providing new insights for developing personalized treatment strategies.

Study Source

This study was conducted by Li Li, Dongjin Wu, Cuiping Zhang, and co-authors from the Department of Anesthesiology at the First Affiliated Hospital of Xiamen University and the School of Medicine at Xiamen University. The paper, titled “A Cross-Tissue Transcriptome-Wide Association Study Identifies New Susceptibility Genes for Insomnia,” was published on January 2, 2025, in the Journal of Neurophysiology.

Study Workflow

1. Data Sources and Preprocessing

The research team obtained GWAS summary data for insomnia from the UK Biobank, including 462,341 European participants. Additionally, gene expression data from the Genotype-Tissue Expression (GTEx) project were used to construct cross-tissue and single-tissue gene expression models.

2. Transcriptome-Wide Association Study (TWAS)

The study employed the Unified Test for Molecular Signatures (UTMOST) method, integrating GTEx data and GWAS summary statistics to conduct cross-tissue TWAS analysis and identify genes significantly associated with insomnia. To validate the reliability of the results, three validation methods were used: FUSION, FOCUS, and MAGMA.

• UTMOST: This method integrates gene expression data across multiple tissues by constructing single-tissue and cross-tissue covariance matrices to elucidate gene-trait associations.
• FUSION: Predicts individual gene expression levels to evaluate their associations with insomnia.
• FOCUS: Uses probabilistic fine-mapping to accurately identify individual genes associated with disease risk.
• MAGMA: Performs gene association analysis, gene set enrichment analysis, and tissue-specific analysis to reveal biological pathways and functional categories related to the disease.

3. Conditional and Joint Analysis

FUSION software was used for conditional and joint analyses to screen for genes independently associated with insomnia. Conditional analysis identifies genetic variants with independent effects, while joint analysis aggregates the effects of multiple SNPs to improve the detection of rare variants.

4. Tissue and Functional Enrichment Analysis

Tissue-specific enrichment analysis and gene set enrichment analysis were conducted using MAGMA. The study revealed enrichment of insomnia-related SNPs in specific brain regions such as the cerebellum, frontal cortex, hypothalamus, and hippocampus, as well as identified relevant biological pathways, including the SMAD2/3 signaling pathway, synaptic function, and oxidative stress.

5. Mendelian Randomization Analysis

The study used the “TwoSampleMR” and “MendelianRandomization” R software tools for two-sample Mendelian randomization analysis to verify causal relationships between significant genes and insomnia. By integrating the effects of multiple genetic instrumental variables, the study assessed the causal associations between gene expression and insomnia.

Results

1. TWAS Results

In cross-tissue TWAS analysis, 195 genes showed significant signals after FDR correction. In single-tissue validation, 332 genes were significantly associated with insomnia. Ultimately, the research team identified 15 candidate genes, with VRK2 and MMRN1 showing significant associations across all four methods.

2. Tissue and Functional Enrichment Analysis

MAGMA analysis indicated that insomnia-related SNPs were predominantly enriched in brain regions such as the cerebellum, frontal cortex, hypothalamus, and hippocampus. Functional enrichment analysis highlighted the roles of the SMAD2/3 signaling pathway, synaptic function, and oxidative stress in insomnia.

3. Conditional and Joint Analysis

Conditional and joint analyses identified two genomic regions independently associated with insomnia: 2p16.1 (VRK2) and 4q22.1 (MMRN1). These genes remained significant after conditional analysis, indicating their independent associations with insomnia.

4. Mendelian Randomization Analysis

Mendelian randomization analysis suggested a causal relationship between the VRK2 gene and insomnia, with individuals carrying the VRK2 gene having a 5% increased risk of developing insomnia.

Conclusion

This study successfully identified two new susceptibility genes for insomnia, VRK2 and MMRN1, by integrating GWAS and gene expression data using multiple TWAS methods. VRK2 may influence insomnia risk by regulating neuroinflammation or neuronal survival, while MMRN1 may affect sleep regulation through neurovascular function. The study also revealed the crucial roles of the SMAD2/3 signaling pathway, synaptic function, and oxidative stress in insomnia. These findings provide new insights into the genetic basis of insomnia and potential targets for personalized treatment strategies.

Highlights

1. Discovery of New Susceptibility Genes: Identified VRK2 and MMRN1 as associated with insomnia for the first time using TWAS methods.
2. Innovation in Cross-Tissue Analysis: The multi-tissue integration approach of UTMOST enhanced statistical power and reliability in gene discovery.
3. Revealing Functional Pathways: Functional enrichment analysis clarified the roles of the SMAD2/3 signaling pathway, synaptic function, and oxidative stress in insomnia.
4. Validation of Causal Relationships: Mendelian randomization analysis confirmed the causal relationship between the VRK2 gene and insomnia for the first time.

Additional Valuable Information

The research team noted that future studies should further integrate environmental factors and gene-environment interactions to comprehensively understand the pathogenesis of insomnia. Additionally, since this study primarily relied on GWAS data from European populations, the generalizability of the results may be limited, necessitating validation in broader ethnic groups.