Work

A cross-platform Mendelian randomization study of mitochondrial DNA copy number across psychiatric disorders

Year: 2026

Type: preprint

Abstract: Mitochondrial DNA copy number (mtDNA-CN) is an accessible peripheral proxy of mitochondrial biology, but its relationship to psychiatric liability remains difficult to interpret because blood mtDNA-CN is influenced by assay platform, tissue context and blood-cell composition. We used two-sample Mendelian randomization to evaluate genetically proxied mtDNA-CN across psychiatric phenotypes using four partially overlapping European-ancestry GWAS exposure panels spanning qPCR, SNP-array intensity and whole-genome-sequencing-derived measures with and without blood-cell-composition adjustment. Because the exposure panels substantially overlapped through UK Biobank, panel-level estimates were treated as correlated and synthesized with a design-effect correction rather than as independent replications. Among the five primary disorders, genetically proxied higher mtDNA-CN was associated with lower liability to autism spectrum disorder (ASD; β = -0.182, design-effect-corrected p = 5.97 × 10^-3; BH-FDR q = 0.030). ADHD showed a weaker directionally concordant estimate that did not survive correction, and the ASD-ADHD neurodevelopmental factor showed a similar nominal pattern but should not be interpreted as independent replication. No robust associations were observed for schizophrenia, bipolar disorder, major depressive disorder, general psychopathology or other disorder-class factors. Reverse-direction MR provided no evidence that psychiatric liability influenced mtDNA-CN, although instrument power was limited for several outcomes. Gene-set enrichment supported the mitochondrial relevance of the mtDNA-CN instruments, whereas brain eQTL colocalization nominated hypothesis-generating candidate genes without establishing mediation. These findings support an ASD-led inverse association between genetically proxied mtDNA-CN and neurodevelopmental liability, while broader mechanistic interpretations require validation.

Source: Research Square

DOI: 10.21203/rs.3.rs-10308549/v1

Authors: Halil Velioğlu, Yusuf Çiçek, Murat Çelik, Abdulahad Bayraktar

Institutions: Istanbul Medipol University, Istanbul University-Cerrahpaşa, Istanbul University

Language: English

Topic: Genetic Associations and Epidemiology

Subfield: Genetics

Field: Biochemistry, Genetics and Molecular Biology

Domain: Life Sciences

SDG: Good health and well-being

Open Access status: green

Funder: Türkiye Bilimsel ve Teknolojik Araştırma Kurumu