Therapeutic Inhibition of LincRNA-p21 Protects Against Cardiac Hypertrophy

Abstract

Background: Cardiac hypertrophy is an adaptive response to pressure overload aimed at maintaining cardiac function. However, prolonged hypertrophy significantly increases the risk of maladaptive cardiac remodeling and heart failure. Recent studies have implicated long noncoding RNAs in cardiac hypertrophy and cardiomyopathy, but their significance and mechanism(s) of action are not well understood.

Methods: We measured lincRNA-p21 RNA and H3K27ac levels in the hearts of dilated cardiomyopathy patients. We assessed the functional role of lincRNA-p21 in basal and surgical pressure-overload conditions using loss-of-function mice. Genome-wide transcriptome analysis revealed dysregulated genes and pathways. We labeled proteins in proximity to full-length lincRNA-p21 using a novel BioID2-based system. We immunoprecipitated lincRNA-p21-interacting proteins and performed cell fractionation, ChIP-seq (chromatin immunoprecipitation followed by sequencing), and co-immunoprecipitation to investigate molecular interactions and underlying mechanisms. We used GapmeR antisense oligonucleotides to evaluate the therapeutic potential of lincRNA-p21 inhibition in cardiac hypertrophy and associated heart failure.

Results: lincRNA-p21 was induced in mice and humans with cardiomyopathy. Global and cardiac-specific lincRNA-p21 knockout significantly suppressed pressure overload-induced ventricular wall thickening, stress marker elevation, and deterioration of cardiac function. Genome-wide transcriptome analysis and transcriptional network analysis revealed that lincRNA-p21 acts in trans to stimulate the NFAT/MEF2 (nuclear factor of activated T cells/myocyte enhancer factor-2) pathway. Mechanistically, lincRNA-p21 is bound to the scaffold protein KAP1 (KRAB-associated protein-1). lincRNA-p21 cardiac-specific knockout suppressed stress-induced nuclear accumulation of KAP1, and KAP1 knockdown attenuated cardiac hypertrophy and NFAT activation. KAP1 positively regulates pathological hypertrophy by physically interacting with NFATC4 to promote the overactive status of NFAT/MEF2 signaling. GapmeR antisense oligonucleotide depletion of lincRNA-p21 similarly inhibited cardiac hypertrophy and adverse remodeling, highlighting the therapeutic potential of inhibiting lincRNA-p21.

Conclusions: These findings advance our understanding of the functional significance of stress-induced long noncoding RNA in cardiac hypertrophy and demonstrate the potential of lincRNA-p21 as a novel therapeutic target for cardiac hypertrophy and subsequent heart failure.

Keywords: RNA, long noncoding; gene expression profiling; heart failure; hypertrophy; ventricular remodeling.

Figure 1.. Muscle-enriched lincRNA-p21 is required for pressure overload-induced cardiac hypertrophy and failure.

A, Relative murine lincRNA-p21 expression, normalized to 18s. Cardiac lincRNA-p21 was set to 1. n=4. B, lincRNA-p21 expression in control and Myh6-Calcineurin (Myh6-CnA) mouse ventricles (GSE175711). n=3 and 6. C, Relative expression of human LINCRNA-P21 in dilated cardiomyopathy (DCM) and healthy donors’ ventricles, normalized to 18S. n=5 and 7. D&E, Relative expression levels in hiPSCs and hiPSC-CMs (differentiation day 38). n=3 and 6. F, Relative lincRNA-p21 expression levels in neonatal mouse cardiomyocytes (CMs) and cardiac fibroblasts (CFs), normalized to 18s. CM expression was set to 1. n=6. G, Nuclear to cytoplasmic RNA ratio in adult mouse ventricular cardiomyocytes. Lp21_ps1 and Lp21_ps2 represent two independent RT-qPCR primer pairs for lincRNA-p21. Nuclear plus cytoplasmic RNA was set to 100%. n=6. H&I, Echocardiographic analysis of ventricular size and function. n=9, 5, 4 and 6 for control and n=9, 10, 4 and 8 for KO mice. J, Timeline of experiment to test the effect of lincRNA-p21 global knockout on pressure overload-induced hypertrophy. K, Gross morphology, H&E, and Fast green and Sirius red staining of control and lincRNA-p21 KO ventricles. Scale bar =2 mm. L, Ventricular weight (VW) normalized to body weight (BW) or tibia length (TL). n=5, 6, 7 and 7 (left→right). M, Echocardiographic analysis of control and lincRNA-p21 global KO mouse ventricles 8 weeks after sham/TAC surgery. n=5, 4, 6 and 7 (left→right). N, Representative images of ventricular sections stained with wheat germ agglutinin (WGA). Scale bar =55 μm. n=4. Quantification of cell area relative to Sham_Ctrl. O, Relative mRNA levels of stress markers in control and lincRNA-p21 KO ventricles. n=5, 6, 6 and 6 (left→right). Data are presented as mean ± SEM. For two-group comparisons: Bootstrap Welch t-test. For L, M, N &O: two-way ANOVA with Tukey’s multiple comparisons test, except for M-LVID;d using Kruskal-Wallis with Dunn’s multiple comparisons test. L-O: significant interaction between surgery type and genotype by two-way ANOVA.

Figure 2.. Transcriptomic changes and cardiac-specific depletion of lincRNA-p21

A, Principal component analysis (PCA) of control and lincRNA-p21 global KO ventricle RNA-sequencing data under sham/TAC conditions. n=3. B&C, Volcano plots and Venn diagrams comparing differentially expressed genes (DEGs) in TAC_Ctrl vs. Sham_Ctrl and TAC_KO vs. TAC_Ctrl groups. Overlapping genes in the top 20 DEGs are underlined (B). D&E, Top dysregulated KEGG gene sets and heatmaps of lincRNA-p21 global knockout compared to the control ventricles. F&G, Gross morphology and H&E staining of control and lincRNA-p21 cKO ventricles. Scale bars represent 2 mm (F) and 1 mm (G). H, Normalized ventricular weights of control and lincRNA-p21 cKO mice. n=8, 6, 8 and 8 (left→right). I, Echocardiography analyses of lincRNA-p21 cKO and control mice 8 weeks after sham/TAC surgery. n=7, 5, 10 and 5 (left→right). J, Normalized lung weights of control and lincRNA-p21 cKO mice. n=8, 6, 8 and 8 (left→right). K, Representative WGA staining images of control and lincRNA-p21 cKO ventricles. Scale bar =50 μm. n=4. Quantification of cell area relative to Sham_Ctrl. L, Fast green and Sirius red staining of control and lincRNA-p21 cKO ventricles. Scale bar =1 mm. M, Relative mRNA levels of hypertrophic stress marker genes in control and lincRNA-p21 cKO ventricles. n=6, 7, 9 and 8 (left→right) for Nppa, Nppb and Acta1; n=6, 7, 9 and 7 (one outlier removed) for Myh7. Data are presented as mean ± SEM and were analyzed using two-way ANOVA with Tukey’s multiple comparisons test. H-K & M: significant interaction between surgery type and genotype.

Figure 3.. Immediate transcriptomic changes upon lincRNA-p21 inhibition under hypertrophic condition.

A, Relative mRNA levels of hypertrophic stress markers in neonatal mouse cardiomyocytes transfected with negative control (gr-NC) or lincRNA-p21 (gr-Lp21) GapmeR ASO under PBS and phenylephrine (PE) conditions. n=4. Representative of 3 independent experiments. B, Gene expression log2 fold change between gr-Lp21 and gr-NC against the mean normalized read counts across all six samples for each non-zero read gene. Red indicates Padj<0.005. C-F, Top differentially enriched KEGG gene sets, differentially expressed genes, and the most significantly dysregulated canonical pathways (BioCarta) for PE + gr-Lp21 vs. PE + gr-NC. F, Cytoscape gene regulatory network and BioCarta canonical pathways gene sets analysis using PE + gr-Lp21 vs. PE + gr-NC cardiomyocyte RNA-sequencing data. G&H, NFAT and MEF2D signaling pathways were enriched for genes downregulated between PE + gr-Lp21 and PE + gr-NC. n=3. I, Relative expression of Klf2 and Acta2, genes downstream of MEF2, in gr-NC and gr-Lp21 treated cardiomyocytes under PBS/PE treatments. n=3. J, Relative expression of NFAT signaling components in PE + gr-NC vs. PE + gr-Lp21 cardiomyocytes and TAC_Ctrl vs.TAC_KO mouse ventricles. Data are presented as mean ± SEM and were analyzed using two-way ANOVA with Tukey’s multiple comparisons test. A&I: significant interaction between hypertrophic stimulus and GapmeR treatment.

Figure 4.. lincRNA-p21 functions in trans under hypertrophic condition.

A&B, Gene expression fold-change (log2FC) in cardiomyocytes treated with PE + gr-Lp21 vs. PE + gr-NC, in the neighborhood (± 2Mb, A) of the lincRNA-p21 locus and in more remote regions (−2~−10Mb and +2~+10Mb, B). C, DEG/total gene ratios in the regions centered around lincRNA-p21 locus. lincRAN-p21 itself was not counted. D, Neighborhood effect analysis compared to 10,000 random genomic windows with the same genomic size. Data from Groff et al. (Ref. 17) was used as a positive control. E&F, Genomic views of normalized Cardiomyocyte RNA-sequencing reads at cardiomyocyte stress markers and lincRAN-p21 neighboring genes. n=3. G, MF20 fluorescence staining of neonatal rat ventricular cardiomyocytes (NRVMs) infected with negative control (NC) and lincRNA-p21 overexpression (OE) adenovirus-associated virus under PBS and 50 μM phenylephrine (PE) treatments. Scale bar =100 μm. n=4. Quantification of cell area relative to PBS_NC. H, Relative RNA levels of lincRNA-p21 and cardiac stress markers in neonatal rat ventricular cardiomyocytes. n=9. Data are presented as mean ± SEM. F (Cdkn1a&Srsf3) and H (lincRNA-p21): Bootstrap Welch t-test. Remaining panels were analyzed using two-way ANOVA followed by Tukey’s multiple comparisons test. F (Nppb&Acta1), G, and H (Nppa & Nppb): significant interaction between hypertrophic stimulus and GapmeR or overexpression treatment.

Figure 5.. lincRNA-p21 targets scaffold protein KAP1 during cardiac hypertrophy.

A, Schematic of BioID2-based full-length lncRNA-protein proximity labeling (BioID-FLPPL) system. B, Identification of proteins in proximity to lincRNA-p21 in C2C12 muscle cells. n=2. C, 145 enriched candidates were classified based on three key parameters: nuclear localization, RNA binding activity, and functional relevance (see Detailed Methods). D&E, Co-immunoprecipitation (co-IP) blotting to validate the proximity between lincRNA-p21 and KAP1 in 293T cells. Representative of 3 independent experiments. F, Co-IP assay to investigate the physical interaction between lincRNA-p21 and KAP1 in 293T cells. Representative of 3 independent experiments. G, Nuclear and cytoplasmic KAP1 protein levels in control and lincRNA-p21 cKO ventricles 8 weeks after sham/TAC surgery. Total Histone H3 and GAPDH: nuclear and cytoplasmic protein internal controls. n=3. H, Kap1 mRNA levels in control and lincRNA-p21 cKO ventricles. n=6, 7, 9 and 8 (left→right). I, Relative Kap1 mRNA levels in neonatal rat ventricular cardiomyocytes (NRVMs) transfected with NC and Kap1 siRNA combined with PBS/PE treatment. n=4. Bootstrap Welch t-test. J, MF20 fluorescence staining of NRVMs transfected with NC and Kap1 siRNA under PBS/PE treatments. Scale bar =100 μm. n=4. Quantification of cell area relative to PBS_si-NC. Representative of 3 independent experiments. K, Cardiac stress marker expression in NRVMs. n=4. Data are presented as mean ± SEM and were analyzed using two-way ANOVA with Tukey’s multiple comparisons test, if not indicated otherwise. G (nuclear KAP1), J, &K: significant interaction between hypertrophic stimulus and genotype or siRNA treatment.

Figure 6.. Cardiac-specific knockout of lincRNA-p21 protects against pathological cardiac hypertrophy via suppressing overactive NFAT/MEF2 signaling.

A, H3K9me3 and H3K27me3 immunoblotting in control and lincRNA-p21 cKO ventricles 8 weeks after sham/TAC surgery. Total Histone H3: nuclear protein internal control. n=3. For H3K9me3: Kruskal-Wallis with Dunn’s multiple comparisons test. B, H3K9me3 and H3K27me3 signals in the loci of the top 50 downregulated DEGs in neonatal mouse cardiomyocytes treated with PE + gr-Lp21 vs. PE + gr-NC. C&D, Co-immunoprecipitation to examine the association of KAP1 with either full-length (GFP-NFATC4) or constitutively active NFATC4 (ΔNFATC4). Representative of 3 independent experiments. E&F, Nuclear NFATC4 protein and downstream Rcan1 mRNA levels in control and lincRNA-p21 cKO ventricles. n=3 (E). n=6, 7, 9 and 8 (F, left→right). G, Nuclear and cytoplasmic MEF2A immunoblotting in control and lincRNA-p21 cKO ventricles. Total Histone H3 and GAPDH: nuclear and cytoplasmic protein internal controls. n=3. H, The expression of lincRNA-p21, Kap1, Nppa and Rcan1 in NMVMs treated with PBS/PE combined with NC or lincRNA-p21 GapmeR plus NC or Kap1 siRNA transfection, normalized to 18s. n=3. For lincRNA-p21 and Kap1: Bootstrap Welch t-test. Data are presented as mean ± SEM and were analyzed using two-way ANOVA with Tukey’s multiple comparisons test, if not indicated otherwise. E-G: significant interaction between surgery type and genotype. H(Nppa&Rcan1): significant interaction between GapmeR treatment and combination of hypertrophic stimulus and siRNA treatment.

Figure 7.. Therapeutic potential of GapmeR ASO-mediated lincRNA-p21 inhibition in pressure-overload induced cardiac hypertrophy and heart failure.

A, Schematic of sham/TAC procedures and GapmeR ASO administration in C57BL/6 wildtype mice. B&C, Gross morphology and H&E staining of NC and lincRNA-p21 GapmeR-treated ventricles under sham/TAC conditions at 6 weeks after surgery. Scale bars represent 2 mm (B) and 1 mm (C). D, Normalized ventricular weights of NC and lincRNA-p21 GapmeR ASO-treated ventricles. n=7, 7, 12 and 11 (left→right). E, Echocardiographic analyses of NC and lincRNA-p21 GapmeR-treated mouse ventricles. n=7, 7, 12 and 11 (left→right). F, WGA staining of NC and lincRNA-p21 GapmeR ASO-treated ventricles. Scale bar =100 μm. n=4. Quantification of cell area relative to Sham_gr-NC. G, mRNA levels of stress markers in the ventricles of NC and lincRNA-p21 GapmeR-treated mice. n=6 (one outlier removed), 7, 9, and 7 (two outliers removed) (left→right) for Nppa; n=7, 7, 9 and 9 (left→right) for Nppb and Myh7. H&I, Fast green & Sirius red staining of NC and lincRNA-p21 GapmeR ASO treated ventricles. Scale bar represents 1 mm (H) and 200 μm (I). n=4. Quantification of fibrosis area relative to Sham_gr-NC. J, Normalized fluid weight in lung of NC and lincRNA-p21 ASO-treated mice. n=7, 7, 12 and 11 (left→right). K, Relative mRNA levels of Calcineurin-NFAT signaling components in the ventricles of NC and lincRNA-p21 GapmeR-treated mice. n=5 (two outliers removed), 7, 9 and 9 (left→right) for Rcan1; n=7, 7, 9 and 9 (left→right) for Edn1. L, Working model. ASO targeting lincRNA-p21 protects against cardiac hypertrophy and subsequent adverse remodeling, without significantly impacting the physiological signaling. Data are presented as mean ± SEM and were analyzed using two-way ANOVA with Tukey’s multiple comparisons test, except for E (4-week) and J using Kruskal-Wallis with Dunn’s multiple comparisons test. D, E (6-week), F, G, I& K: significant interaction between surgery type and GapmeR treatment.