Evaluating the Causal Effect of Mitochondrial Dysfunction on Alzheimer’s Disease and Parkinson’s Disease Using Polygenicrisk Scores and Mendelian Randomization (Paper May 26)

Key Takeaways

• LHC‑MR links higher blood mtDNAcn to lower AD and PD risk
• Standard MR and PRS analyses found no significant causal effect
• Reverse MR suggests AD genetic liability may increase mtDNAcn levels
• Platelet contamination and measurement method heavily influence mtDNAcn GWAS results
• Findings support mitochondria as a causal factor, not a proven therapy

Pulse Analysis

Mitochondrial health has long been implicated in Alzheimer’s and Parkinson’s pathology, but observational studies cannot untangle cause from consequence. By leveraging genome‑wide association data, the Chatterjee team applied a suite of genetic epidemiology tools to test whether the number of mitochondrial DNA copies circulating in blood—an indirect proxy for cellular energy capacity—plays a causal role in neurodegeneration. This approach moves beyond correlation, aiming to satisfy the three core MR assumptions of relevance, independence, and exclusion restriction.

The investigators combined four independent mtDNAcn GWASs, each differing in ancestry composition, assay technology, and adjustment for platelet contamination, to mitigate bias from measurement artefacts. Conventional univariable MR and multivariable MR that accounted for platelet counts yielded null results, suggesting no straightforward causal link. In contrast, latent‑heritable‑confounder MR, which models an unobserved genetic confounder, uncovered a protective effect of higher mtDNAcn on both AD/dementia and PD risk. A bidirectional analysis further revealed that AD genetic risk may elevate mtDNAcn, hinting at a compensatory mitochondrial response during early disease stages.

These findings reinforce mitochondria as a potential upstream driver of neurodegenerative risk, yet they also caution against simplistic therapeutic extrapolation. Blood‑derived mtDNAcn is an imperfect surrogate for neuronal mitochondrial function and is vulnerable to cell‑type composition and technical variation. Future work should validate these genetic signals in brain tissue, explore mechanistic pathways linking mtDNAcn to neuronal survival, and assess whether interventions that genuinely boost mitochondrial quality—not merely copy number—can modify disease trajectories. The study exemplifies how advanced MR techniques can refine causal inference in complex, polygenic diseases.