Altered intercellular communication and extracellular matrix signaling as a potential disease mechanism in human hypertrophic cardiomyopathy

Abstract

Hypertrophic cardiomyopathy (HCM) is considered a primary disorder of the sarcomere resulting in unexplained left ventricular hypertrophy but the paradoxical association of nonmyocyte phenotypes such as fibrosis, mitral valve anomalies and microvascular occlusion is unexplained. To understand the interplay between cardiomyocyte and nonmyocyte cell types in human HCM, single nuclei RNA-sequencing was performed on myectomy specimens from HCM patients with left ventricular outflow tract obstruction and control samples from donor hearts free of cardiovascular disease. Clustering analysis based on gene expression patterns identified a total of 34 distinct cell populations, which were classified into 10 different cell types based on marker gene expression. Differential gene expression analysis comparing HCM to Normal datasets revealed differences in sarcomere and extracellular matrix gene expression. Analysis of expressed ligand-receptor pairs across multiple cell types indicated profound alteration in HCM intercellular communication, particularly between cardiomyocytes and fibroblasts, fibroblasts and lymphocytes and involving integrin β1 and its multiple extracellular matrix (ECM) cognate ligands. These findings provide a paradigm for how sarcomere dysfunction is associated with reduced cardiomyocyte secretion of ECM ligands, altered fibroblast ligand-receptor interactions with other cell types and increased fibroblast to lymphocyte signaling, which can further alter the ECM composition and promote nonmyocyte phenotypes.

Isolating Nuclei From Frozen Human Heart Tissue for Single‐Nucleus RNA Sequencing

“…Read counts of mitochondrial genes from snRNA-seq data can serve as an assessment of sample quality. In comparison to sequencing data obtained using other protocols, we see a very low level of mitochondrial reads in sequencing data of tissue dissociated using this protocol (Figure 4B; Larson et al, 2022).…”

Single Nucleus RNA-sequencing Reveals Altered Intercellular Communication and Dendritic Cell Activation in Nonobstructive Hypertrophic Cardiomyopathy

“…To quantify potential cardiac cell-cell communication in Normal and nonobstructive HCM IVS tissue, we quantified the number of possible expressed ligand-receptor pairs among cell types as previously described [ 16 , 17 , 29 ]. We examined the expression of a curated list of 3627 unique human ligand-receptor (L-R) pairs derived by combining a collection of 2557 human L-R pairs [ 30 ] with another set of 3398 human L-R pairs [ 31 ] and eliminating duplicate pairs.…”

“…We examined the expression of a curated list of 3627 unique human ligand-receptor (L-R) pairs derived by combining a collection of 2557 human L-R pairs [ 30 ] with another set of 3398 human L-R pairs [ 31 ] and eliminating duplicate pairs. Ligands and receptors were considered expressed if their associated gene was detectable in ≥ 20% of cells in a cell type, a cutoff used commonly in previous studies [ 16 , 17 , 29 ]. Notably, quantification of cell-cell communication in this study represents potential communication as we account for only expressed ligand-receptor pairs and not the position or boundaries of cell types.…”

“…Trajectory analysis was performed using Monocle3 [ 26 ] to determine the relationship between subtypes of cells identified in our clustering analysis as previously described [ 16 , 17 ]. Since our data do not represent a developmental time course, we determined the root nodes for each subtype by hierarchical clustering prior to generating trajectories and assigning pseudotime to each cell.…”

“…Our cell-cell communication networks were derived as described previously [ 29 ], using a list of 2557 human ligand-receptor pairs [ 30 ] combined with another list of 3398 human ligand-receptor pairs [ 31 ], to give a total of 3627 unique human ligand-receptor pairs, largely as described previously [ 16 , 17 ]. A ligand or receptor for each cell type or cluster was considered expressed if the corresponding gene showed an above zero gene count in ≥ 20% of cells in our snRNAseq data, a threshold used commonly in other studies [ 16 , 17 , 29 ]. We initially analyzed the potential signaling interactions between the 8 cell types identified in our snRNAseq data.…”