Abstract 4370739: Integrative analysis of MoTrPAC and LINCS transcriptomes identifies cardiovascular-relevant exercise-mimetic drugs

Introduction: Exercise is a cornerstone of cardiovascular health, yet not all individuals can engage in sufficient physical activity. Identifying pharmacological agents that mimic exercise-induced molecular adaptations offers a promising strategy for disease prevention. Here, we integrate transcriptomic data from the Molecular Transducers of Physical Activity Consortium (MoTrPAC)—a multi-omics effort characterizing exercise responses—including human skeletal muscle after acute exercise (0-24h), and both acute and endurance training (1-8 weeks) in rat heart and skeletal muscle. These are combined with the Library of Integrated Network-Based Cellular Signatures (LINCS), which catalogs drug-induced transcriptional responses in 292 human cell lines exposed to 20,272 compounds. Through this integrative approach, we aim to identify compounds that replicate transcriptional effects of endurance exercise. Methods: To address transcriptomic differences between tissues and cell lines, we developed a pipeline combining functional and regulatory analyses. We assessed pathway enrichment via FGSEA and inferred transcription factor (TF) activity using VIPER and DoRothEA (Figure 1A). Shared pathways and upstream regulators were integrated into a mimetic score ranking candidate drugs by their functional and regulatory similarity to exercise. Results: We identified between 600–800 candidate mimetics for skeletal muscle in acute human and rat responses, as well as during endurance training. Notably, predicted mimetics aligned well between species (Figure 1B). Among top hits, midodrine, an α1-adrenergic agonist investigating for its cardioprotective properties, matched 24h post-exercise signatures in both species and activated TFs linked to mitochondrial function, vascular remodeling, and metabolism (FOXP1, CREB1). PD-0325901, a MEK inhibitor, showed strong similarity to early phases (15–45 min), inducing early-response TFs (FOS, HIF1A, RELA), and reflected endurance training upregulation of oxidative phosphorylation. Mimetic profiles from rat heart clustered with muscle at matched timepoints, highlighting the potential to extend heart-based predictions to humans via conserved signatures (Figure 1C). Conclusion: This integrative analysis identifies candidate compounds that mimic distinct phases of exercise-induced molecular remodeling. Ongoing validation in cardiac models aims to translate these findings into cardiovascular contexts.