Multi-omics analyses of HFpEF progression and regression reveals multi-organ-specific metabolic dysregulation and impaired inflammation resolution

BACKGROUND: Heart failure with preserved ejection fraction (HFpEF) is a complex cardiometabolic syndrome with limited effective therapies. Given its heterogeneity, whether withdrawal of cardiometabolic stress reverses HFpEF-associated inflammatory, metabolic, and molecular remodeling remains unclear. OBJECTIVES: To characterize multi-organ metabolic, molecular, and bioactive-lipid remodeling in an obesity-hypertension HFpEF model and determine the extent of recovery during an 8-week combined high-fat diet withdrawal and L-NAME cessation protocol using parallel multi-omics profiling. METHODS: HFpEF was induced in C57BL/6 mice using a high-fat diet plus nitric oxide synthase inhibition(H + L). A regression cohort(R-H + L) underwent simultaneous return to standard chow and L-NAME cessation for 8 weeks. Multi-omics profiling included targeted and untargeted lipidomics, cardiac transcriptomics, metabolomics, fecal microbiome, and plasma short-chain fatty acid analyses, with detailed assessment of the left ventricle. RESULTS: H + L mice developed an obesity-hypertension HFpEF phenotype characterized by predominant diastolic dysfunction, concentric remodeling, hypertension, and increased fat mass. The combined regression of H + L protocol attenuated these phenotypic abnormalities but did not restore all measures to control levels. H + L was associated with multi-organ remodeling marked by altered arachidonic acid-derived oxylipid pathways, depletion of triacylglycerols and phospholipids, and enrichment of diacylglycerols and lysophospholipids, consistent with altered resolution-related lipid signaling. Cardiac transcriptomics revealed changes in immune, circadian, vascular, and mitochondrial gene programs, whereas metabolomics revealed altered energy, arginine, and redox metabolism. H + L was also associated with reduced microbial diversity and altered short-chain fatty acids, particularly acetate and propionate. Within the 8-week recovery window, partial reprogramming occurred, but differences remained in lipid signaling, metabolic stress responses, and microbiome-associated measures. CONCLUSIONS: The HFD/L-NAME model is accompanied by extensive multi-organ metabolic and molecular remodeling. Combined withdrawal of HFD/L-NAME attenuates the HFpEF phenotype however did not fully normalize molecular profiles within the 8-week recovery period. Persistent lipid-remodeling and metabolic dysregulation may represent potential therapeutic targets thus warrants pathway specific investigation.