Transcriptomic Analysis of Nuclear- Encoded Mitochondrial Genes in Amyotrophic Lateral Sclerosis (ALS) Across Two Independent Cohorts

Abstract

Mitochondrial dysfunction is a central hallmark of amyotrophic lateral sclerosis (ALS), yet the nuclear-encoded mitochondrial (NEM) transcriptomic landscape remains underexplored. This study conducts an integrative transcriptomic analysis of NEM genes using two independent post-mortem brain cohorts, BrainBank (BB) and TargetALS (TALS), to uncover convergent and divergent mechanisms of mitochondrial impairment in ALS. Differential expression analysis identified significant NEM gene dysregulation in both cohorts, with 35 genes overlapping. Over-representation analysis (ORA) of these shared differentially expressed genes revealed enrichment in mitochondrial transport, membrane organisation, and redox enzyme activity, implicating disrupted bioenergetics and structural integrity. Gene Set Enrichment Analysis (GSEA) in the BB dataset further identified downregulation of pathways associated with the mitochondrial inner membrane and oxidative stress response, supporting widespread mitochondrial destabilisation. Weighted Gene Co-expression Network Analysis (WGCNA) uncovered reproducible NEM hub genes across cohorts, including GFER, MCAT, MRPL28, and CHCHD6, which are involved in mitochondrial translation, redox regulation, and cristae maintenance. Notably, SOD1 emerged as a cross-validated marker of mitochondrial pathology through DEA, GSEA, and WGCNA. This multi-cohort, multi- method analysis highlights core transcriptional signatures of mitochondrial failure and identifies consistent molecular targets potentially relevant for therapeutic development in ALS.