Transcriptomal Insights of Heart Failure from Normality to Recovery

Abstract

Current management of heart failure (HF) is centred on modulating the progression of symptoms and severity of left ventricular dysfunction. However, specific understandings of genetic and molecular targets are needed for more precise treatments. To attain a clearer picture of this, we studied transcriptome changes in a chronic progressive HF model. Fifteen sheep (Ovis aries) underwent supracoronary aortic banding using an inflatable cuff. Controlled and progressive induction of pressure overload in the LV was monitored by echocardiography. Endomyocardial biopsies were collected throughout the development of LV failure (LVF) and during the stage of recovery. RNA-seq data were analysed using the PANTHER database, Metascape, and DisGeNET to annotate the gene expression for functional ontologies. Echocardiography revealed distinct clinical differences between the progressive stages of hypertrophy, dilatation, and failure. A unique set of transcript expressions in each stage was identified, despite an overlap of gene expression. The removal of pressure overload allowed the LV to recover functionally. Compared to the control stage, there were a total of 256 genes significantly changed in their expression in failure, 210 genes in hypertrophy, and 73 genes in dilatation. Gene expression in the recovery stage was comparable with the control stage with a well-noted improvement in LV function. RNA-seq revealed the expression of genes in each stage that are not reported in cardiovascular pathology. We identified genes that may be potentially involved in the aetiology of progressive stages of HF, and that may provide future targets for its management.

Keywords: RNA-Seq; cardiac recovery; dilatation; heart failure; hypertrophy.

Figure 1

Hierarchical clustering and principal components analysis of gene expression. (A) The dendrogram shows hierarchical clustering of all samples according to their gene expression. The closeness of any two samples in the dendrogram shows that samples are similar in their gene expression. (B) The principal component analysis shows the samples plotted according to differential gene expression. The distance between points approximates differences in gene expression among samples.

Figure 2

Characteristics of differential gene expression. (A–D) The MA plots of the hypertrophy through recovery compared to control show the total expression of genes. The X-axis represents value A (log2-transformed mean expression level of each gene). The Y-axis represents value M (log2-transformed fold change of each gene). Red dots represent upregulated DEGs. Blue dots represent downregulated DEGs. Gray points represent genes that are not differentially expressed. (E) Heat map showing the differentially expressed genes in each group. (F) The Venn diagram shows the distribution of DEGs between the groups. (G–N) Top up-/downregulated genes that are either unique to a group or common to two or more groups are shown.

Figure 3

Validation of gene expression by real-time qRT-PCR analysis. The expression of four genes was analysed by qRT-PCR as GAPDH as the housekeeping control gene. The inset table shows the fold change values in the RNA-seq data. NS; non significant.

Figure 4

The PANTHER analysis of DEGs for functional annotation. The DEGs of each pathological stage were analysed for molecular function (MF), biological process (BP), cellular component (CC), protein classification (PC), and PANTHER pathways. Representation of the enrichment of each GO term by DEGs of (A) hypertrophy stage, (B) dilated stage, and (C) failure stage. (D) The significant PANTHER pathways enriched by DEGs of hypertrophy, dilated, or failure stages.

Figure 4

The PANTHER analysis of DEGs for functional annotation. The DEGs of each pathological stage were analysed for molecular function (MF), biological process (BP), cellular component (CC), protein classification (PC), and PANTHER pathways. Representation of the enrichment of each GO term by DEGs of (A) hypertrophy stage, (B) dilated stage, and (C) failure stage. (D) The significant PANTHER pathways enriched by DEGs of hypertrophy, dilated, or failure stages.

Figure 4

The PANTHER analysis of DEGs for functional annotation. The DEGs of each pathological stage were analysed for molecular function (MF), biological process (BP), cellular component (CC), protein classification (PC), and PANTHER pathways. Representation of the enrichment of each GO term by DEGs of (A) hypertrophy stage, (B) dilated stage, and (C) failure stage. (D) The significant PANTHER pathways enriched by DEGs of hypertrophy, dilated, or failure stages.

Figure 4

The PANTHER analysis of DEGs for functional annotation. The DEGs of each pathological stage were analysed for molecular function (MF), biological process (BP), cellular component (CC), protein classification (PC), and PANTHER pathways. Representation of the enrichment of each GO term by DEGs of (A) hypertrophy stage, (B) dilated stage, and (C) failure stage. (D) The significant PANTHER pathways enriched by DEGs of hypertrophy, dilated, or failure stages.

Figure 5

DisGeNET analysis for DEG–disease network identification. The DEGs in each stage were analysed for diseases enriched by them. (A–C) The graph denotes the significance log10(q) value, enrichment score, z-score, and the count of genes in the stage enriching the disease. (A–C) (i–iv)) Subsets of graphs show the DEG-enriched selected diseases. (D) The potential candidate genes for cardiovascular diseases were filtered and depicted as a heatmap according to their fold change expression compared to the control.