Bioinformatic analysis of differentially expressed mitochondrion-related genes, immune cell infiltration, and diagnostic value in Alzheimer's disease.
TL;DR
A novel landscape of MRGs in AD is defined, their tight crosstalk with the immune microenvironment is deciphered, and a promising MRG-based signature for AD diagnosis is established.
Abstract
Background
Mitochondrial dysfunction and neuroinflammation are critically implicated in the pathogenesis of Alzheimer's disease (AD). However, a systematic exploration of key mitochondrion-related genes (MRGs) in AD, and their specific roles in reshaping the immune microenvironment and serving as diagnostic biomarkers, remains insufficient.
Methods
To address this, we conducted an integrative bioinformatics analysis. Differentially expressed MRGs were identified from public AD transcriptomic datasets. Their biological functions were elucidated through enrichment analyses. The correlations between core MRGs and ssGSEA-derived immune-cell signature enrichment scores were quantified using transcriptome-based computational analysis. Finally, machine learning models were constructed and validated to assess the diagnostic potential of identified MRG signatures.
Results
A robust set of dysregulated MRGs was identified in AD brains, showing predominant enrichment in pathways of oxidative phosphorylation and energy metabolism. Notably, the expression of key MRGs correlated significantly with altered infiltration abundances of specific immune cell types, including neutrophil-, eosinophil-, NK CD56bright cell-, and T follicular helper cell-related signatures. A diagnostic model constructed from a refined MRG signature exhibited promising predictive accuracy, with area under the curve (AUC) values reaching approximately 0.82 in the training cohort and around 0.74 in independent validation cohorts.
Conclusion
Our study defines a novel landscape of MRGs in AD, deciphers their tight crosstalk with the immune microenvironment, and establishes a promising MRG-based signature for AD diagnosis. These findings provide fresh insights into the potential molecular interplay between mitochondrial dysfunction and neuroinflammation in AD and nominate candidate mitochondrion-related biomarkers and regulatory mechanisms that warrant further experimental and clinical validation.