Abstract
Background: Majority of ALS cases are sporadic (sALS), as they lack defined genetic causes. Metabolic alterations shared between the nervous system and skin fibroblasts have emerged in ALS. Recently, we found that a subgroup of sALS fibroblasts (sALS1) is characterized by metabolic profiles (metabotype) distinct from other sALS cases (sALS2) and controls, suggesting that metabolic therapies could be effective in sALS. The metabolic modulators nicotinamide riboside and pterostilbene (EH301) are under clinical development for the treatment of ALS. Here, we studied the metabolome and transcriptome of sALS cells to understand the molecular bases of sALS metabotypes and the impact of EH301.Methods: Six fibroblast cell lines (3 male and 3 female subjects of similar ages) were used for each group (sALS1, sALS2, and controls). Metabolomics and transcriptomics were investigated at baseline and after EH301 treatment. Differential gene expression (DEGs) and metabolite abundance were assessed by a Wald Test and ANOVA, respectively, with FDR correction, and pathway analyses were performed. EH301 protection against metabolic stress was tested by thiol depletion. Weighted gene co-expression network analysis (WGCNA) was used to investigate the association of metabolic and clinical features and was also performed on the Answer ALS dataset from induced motor neurons (iMN). A machine learning model based on DEGs was tested as a sALS disease progression predictor. Results: We found that the sALS1 transcriptome is distinct from sALS2 and that EH301 modifies gene expression differently in sALS1, sALS2, and controls. Furthermore, EH301 had strong protective effects against metabolic stress, which is linked to anti-inflammatory and antioxidant pathways. WGCNA revealed that ALS functional rating scale and metabotypes are associated with gene modules enriched for cell cycle, immunity, autophagy, and metabolism terms, which are modified by EH301. Meta-analysis of publicly available transcriptomics data from iMNs confirmed functional associations of genes correlated with disease traits. A small subset of genes differentially expressed in sALS fibroblasts could be used in a machine learning model to predict disease progression.Conclusions: Multi-omics analyses of patient-derived fibroblasts highlighted differential metabolic and transcriptomic profiles in sALS metabotypes, which translate into differential responses to the investigational drug EH301.