Abstract P1023: Single-Cell Metabolic Reconstruction Predicts Regenerative Capacity In Cardiomyocytes

Abstract

Background: Adult mammalian hearts have limited regenerative capacity following ischemic injury, whereas neonatal hearts can fully regenerate. Uncovering the mechanisms underlying this ability could identify potential avenues for treatment following cardiac injuries. There has been evidence that changes in the tricarboxylic acid (TCA) cycle are involved in transition of cells out of a pluripotent state. Our hypothesis is that a specific metabolic profile will be associated with regenerative capacity in cardiomyocytes (CMs), and our subsequent goal is to characterize its activity. Methods: Previously published single-cell RNA-Seq data (GSE130699, GSE153481) from mouse hearts subjected to ischemic or sham injury at neonatal (P1) or adult (P8) stages (10987 CMs) were used to calculate metabolic reaction activity scores (RAS), which assign activity to each reaction based on gene expression (Damiani et al, 2020). We then used enrichment analysis to compare metabolic changes between experimental conditions. CMs were clustered into metabolic clusters based on RAS and compared to clusters based on total gene expression. We identified the metabolic reactions that are the best predictors of regenerative ability using LASSO regularization logistic regression. Results: Significant (p ≤ .05) differences in metabolism were identified between neonatal and adult CMs. Clustering based on RAS revealed four distinct metabolic profiles in neonatal CMs, but only two (regenerative and non-regenerative) in adult CMs. Compared to other CM clusters, regenerative CMs exhibited significantly upregulated TCA, nucleic acid, and folate metabolism. This subpopulation was enriched in the neonatal condition, resulting in the significant differences seen in overall CM metabolism between conditions. Further, regression using metabolic activity allowed us to predict the regenerative subpopulation of CMs with greater than 85% accuracy. Conclusion: In conclusion, we characterize the unique metabolic profiles of CMs that can be used to identify the regenerative subpopulation of CMs. This suggests that the different metabolic profile seen in regenerative CMs may play a significant role in their regenerative capacity and their differentiation to non-regenerative CMs.