Pathway-Informed Stratification of Major Depression: Links to NMN-Rescued Aging Programs in Mitochondrial Metabolism and Trafficking
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background: Major depressive disorder is clinically heterogeneous, and current diagnostic categories often mix patients with different biological liabilities. We tested whether symptom-level genetic signals for major depression converge on cellular pathways previously nominated from nicotinamide mononucleotide, or NMN, aging biology. Methods: We analyzed transcriptome-wide association study outputs derived from the Adams et al. 2024 genome-wide meta-analysis of major depression symptom dimensions. The analysis used S-PrediXcan-style brain tissue models and a focused set of 10 KEGG pathways selected from a Cheung 2026 re-analysis of the Mills et al. 2016 NMN mouse study. These pathways were anchored in 35 cross-tissue robust genes whose age-associated expression shifts were reported to be rescued by NMN, including RAB11A-linked endocytosis, CPT2-linked fatty acid degradation, MAP2K2-linked MAPK/Ras signaling, and autophagy-related programs. We tested enrichment, symptom-pair divergence, profile proximity, leave-one-out gene influence, pathway–symptom network structure, and family-wise false discovery rate. Results: Fatty acid degradation was the broadest pathway signal, significant in 8 of 12 symptom dimensions and acting as the highest-degree pathway hub. The strongest individual association was endocytosis with hypersomnia, or sleepMore, with Stouffer Z = −6.42, bootstrap 95% CI = −9.67 to −3.02, permutation p = 0.0004. Autophagy was strongly associated with weight loss, Z = −5.93, permutation p = 0.0008. Weight loss and weight gain showed the clearest biological divergence, with mean Lin’s concordance correlation coefficient = −0.32 and low mean sign-concordance, supporting their treatment as distinct symptom dimensions rather than opposite ends of a single scale. Symptom clustering separated atypical sleep-weight features from core affective-cognitive-fatigue symptoms. Conclusions: These findings support a dimensional, pathway-informed model of major depression. The strongest signals implicate mitochondrial fatty acid handling, endosomal trafficking, and autophagy in symptom-specific genetic liability. The results are hypothesis-generating and do not establish therapeutic effects of NMN, but they nominate tractable pathways for replication and mechanism-targeted studies in biologically defined depression subgroups.