Modulation of miR-29 influences myocardial compliance likely through coordinated regulation of calcium handling and extracellular matrix

MicroRNAs (miRs) control the expression of diverse subsets of target mRNAs, and studies have found miR dysregulation in failing hearts. Expression of miR-29 is abundant in heart, increased with aging, and altered in cardiomyopathies. Prior studies demonstrate that miR-29 reduction via genetic knockout or pharmacologic blockade can blunt cardiac hypertrophy and fibrosis in mice. Surprisingly, this depended on specifically blunting miR-29 actions in cardiomyocytes versus fibroblasts. To begin developing more translationally-relevant vectors, we generated a novel transgene-encoded miR-29 inhibitor (TuD-29) that can be incorporated into a viral-mediated gene therapy for cardioprotection. Herein, we corroborate that miR-29 expression and activity is higher in cardiomyocytes versus fibroblasts and demonstrate that TuD-29 effectively blunts hypertrophic responses in cultured cardiomyocytes and mouse hearts. Furthermore, we found that adeno-associated viral (AAV)-mediated miR-29 overexpression in mouse hearts induces early diastolic dysfunction, whereas AAV:TuD-29 treatment improves cardiac output by increasing end-diastolic and stroke volumes. Integration of RNA-seq and miR-target interactomes reveals that miR-29 regulates genes involved in calcium handling, cell stress and hypertrophy, metabolism, ion transport, and extracellular matrix remodeling. These investigations support a likely versatile role for miR-29 in influencing myocardial compliance and relaxation, potentially providing a unique therapeutic avenue to improve diastolic function in heart failure patients. This experiment used RNA-seq to assess differential gene expression in mouse heart after acute miR-29 overexpression. C57BL6/J male mice aged 3-5 weeks old, were randomized and injected IV via jugular vein with cardiotropic AAV2/9 that co-expresses the primary miR-29a transcript and GFP or an AAV2/9:GFP control virus (n=4 each). After 3 weeks post-injection, mice were euthanized, hearts dissected, and snap frozen in liquid nitrogen. Total RNA was extracted using Trizol, DNAse digested, and libraries were prepared for RNA sequencing using Illumina TruSeq Stranded kit with 500 ng of input RNA. Libraries were multiplexed and sequenced on a NovaSeq 6000 with an entire SP flowcell at a read depth of 11M reads per sample (90% aligned). Pseudoalignments to mm10 (GRCm38) and quantification were performed using the Kallisto/Sleuth pipeline aggregating on Ensembl gene ID, and pairwise differential gene expressions were calculated between the AAV:GFP control samples and the AAV:miR29a samples. Gene expression data are expressed as "scaled_reads_per_base" where the average number of reads mapping to each base across the whole gene, and beta values are natural log fold change.