The lncRNA Sweetheart regulates compensatory cardiac hypertrophy after myocardial injury in murine males

Abstract

After myocardial infarction in the adult heart the remaining, non-infarcted tissue adapts to compensate the loss of functional tissue. This adaptation requires changes in gene expression networks, which are mostly controlled by transcription regulating proteins. Long non-coding transcripts (lncRNAs) are taking part in fine-tuning such gene programs. We describe and characterize the cardiomyocyte specific lncRNA Sweetheart RNA (Swhtr), an approximately 10 kb long transcript divergently expressed from the cardiac core transcription factor coding gene Nkx2-5. We show that Swhtr is dispensable for normal heart development and function but becomes essential for the tissue adaptation process after myocardial infarction in murine males. Re-expressing Swhtr from an exogenous locus rescues the Swhtr null phenotype. Genes that depend on Swhtr after cardiac stress are significantly occupied and therefore most likely regulated by NKX2-5. The Swhtr transcript interacts with NKX2-5 and disperses upon hypoxic stress in cardiomyocytes, indicating an auxiliary role of Swhtr for NKX2-5 function in tissue adaptation after myocardial injury.

Fig. 1. Expression and localization of the Sweetheart lncRNA.

a Strand-specific RNA-seq from E9.5 heart tubes showing the Nkx2-5 region. (gray box = first heart field enhancer). Plus-strand track is 20× amplified over the minus-strand track. The chromatin state tracks are given for the forebrain and heart from ENCODE in ChromHMM colors (red = TSS, yellow/orange = enhancers, light gray = poised enhancer, purple = heterochromatin, dark gray = quiescent). The number above the genome bar denotes the mm10 coordinates and the vertical tip bars represent 10,000 bp steps. The coordinates for the Swhtr transcription unit are chr17:26,849,907–26,859,715 (mm10). b Whole mount in situ hybridization of Nkx2-5 and Swhtr in E8.25 embryos. White arrows show the early heart tube with two different heart fields and black arrowhead points to the inflow tract region. The white line represents 500 µm and the staining was repeated two times. c Lineage tracing of Swhtr expressing cells in E9.5 embryos. OFT outflow tract, RV right ventricle, LV left ventricle, PE pre-pericardium. The white line represents 500 µm. d Lineage tracing of Swhtr expressing cells in late gestation embryos and heart/lung explant. The white line represents 1 mm. e Lineage tracing of Swhtr expressing cells in a transversal section of an E12.5 heart. RV right ventricle, LV left ventricle. The black line represents 500 µm. c–e are representative images from one of two independent transgenic lines that showed the same staining pattern f Quantitative Real-Time PCR timeline of Swhtr and Nkx2-5 expression levels in the hearts of E10.5 embryos to 8-week adult mice. Embryo hearts were pooled from independent litters and the postnatal stages represent data from individual hearts (n = 3). Error bars are given as SEM. g Subcellular fractionation of E11.5 cardiomyocytes (CMs) of marker transcripts and Swhtr (n = 2). h SmFISH of Nkx2-5 and Swhtr in 24 h cultured neonatal cardiomyocytes. The white line represents 5 µm. i Quantification of smFISH Swhtr signal in 24 h cultured neonatal cardiomyocytes. Note that on average two dots were found in the nucleus (n = 86). j Analysis of coding potential of Swhtr by CPAT compared to known coding and non-coding RNAs.

Fig. 2. No overt phenotype in Swhtr^3xpA/3xpA mutant embryos or adult mice.

a ChIP-seq (H3K4me3) and RNA-seq (E9.5 heart tubes) from WT and Swhtr^3xpA/3xpA mutant mice. b qRT-PCR validation of loss of Swhtr in E9.5 embryonic hearts (WT n = 5; Swhtr^3xpA/3xpA n = 3). Values are given in mean ± SEM. Statistical significance was tested by Two-way ANOVA with Šídák’s multiple comparisons test. ns not significant, **0.0016. c Embryos of indicated age from WT and Swhtr mutants. The white line represents 1 mm. Images are representative of three independent litters. d Eight-week-old Swhtr^3xpA/3xpA founder mice. e Bodyweight of Swhtr^3xpA/3xpA eight times backcrossed (C57Bl6J) mice (WT n = 19; Swhtr^3xpA/3xpA n = 34). f Selected cardiac parameters determined by echocardiography in 8-week-old mice of the indicated genotype (WT n = 19; Swhtr^3xpA/3xpA n = 34). Statistical significance was tested by two-way ANOVA. No statistically significant differences were detected.

Fig. 3. Induced myocardial infarction by left ascending artery ligation (LAD).

a Schematic of the analysis setup for echocardiography and LAD ligation in mice of age 8 weeks. The schematic of the hearts was taken from BioRender.com. b Reduced survival of Swhtr null mice compared to WT mice after LAD ligation (WT n = 13; Swhtr^3xpA/3xpA n = 19). Statistical significance was tested by Kaplan–Meier simple survival analysis. *0.0387. c–f Selected heart-specific parameters in mice (n = 9 animals per genotype) before and after (1 and 2 weeks) LAD ligation. Values are given in mean ± SEM. Statistical significance was tested by Two-way ANOVA with Tukey correction for multiple comparisons. ns not significant, **0.0012, ***0.0002. g Verification of infarct presence in mice 2 weeks after LAD ligation by Sirius red (fibrotic tissue) staining. The black line represents 5 mm. Images are representative of three independent heart sections. h Schematic of the rescue transgene and the resulting mouse line. The rescue transgene (tg) is comprised of the BAC (RP23-466K9) that includes the Nkx2-5 (H2Bvenus inserted in Nkx2-5 ATG) and Swhtr loci, randomly inserted into the genome of wild type C57BL6J mice. The color key for the genotype was generated using BioRender.com. i Verification of tg presence and activity by heart-specific presence of H2BVENUS. Image shows organs prepared from one specimen from the first litter j Expression verification of Swhtr in from the tg in Swhtr null mutants, after crossing (WT n = 2; Swhtr^3xpA/3xpA n = 2; Swhtr^3xpA/3xpA;tg n = 3). Values are given in mean ± SEM. Statistical significance was tested by One-Way ANOVA with Tukey correction for multiple comparisons. WT/Swhtr^3xpA/3xpA p = 0.0136, WT/Swhtr^3xpA/3xpA;tg p = 0.025. k Reduced survival of Swhtr null mice compared to WT and Swhtr rescue mice after LAD ligation (WT n = 13; Swhtr^3xpA/3xpA n = 19; Swhtr^3xpA/3xpA;tg n = 15). Statistical significance was tested by Kaplan–Meier Simple Survival Analysis. *0.0387. l–n Selected heart-specific parameters in mice (n = 9 animals per genotype) before and after (1 and 2 weeks) LAD ligation. Values are given in mean ± SEM. Statistical significance was tested by Two-way ANOVA with Tukey correction for multiple comparisons. ns = not significant, *0.0248, ***0.0002.

Fig. 4. Hypertrophy in Swhtr mutant cardiomyocytes.

a Relative enrichment of Swhtr in the main cell-types represented in the heart of 8-week-old mice as compared to marker genes (n = 2). Note that the Swhtr expression pattern resembles that of Tnni1. b Wheat Germ Agglutinin stained IHC of representative sections of IVS tissue in the indicated genotypes 2 weeks post sham or MI. The white line represents 50 µm. c Automated quantifications of relative cell sizes in the interventricular septum of three representative animals of the indicated genotype 2 weeks post sham or MI. Plotted are 5–95 percentile and the middle line of the box plot represents the median. Statistical significance was tested by One-way ANOVA with Dunnett correction for multiple comparisons. ns not significant, ****p < 0.0001.

Fig. 5. Swhtr dependent genes under cardiac stress.

a Schematic overview of the experimental procedure with representative pictures of heart slices (slices derived from 4 male mice per genotype from two independent litters). The white line represents 5 mm. The schematic of the heart was taken from biorender.com. b Number of deregulated genes after 7 days of hypoxia treatment followed by 7 days of normoxic conditions of Swhtr^3xpA/3xpA and WT slices compared to WT heart slices (n = 4). c GO-term enrichment analysis of deregulated Swhtr-dependent genes. d NKX2-5 occupation on expressed genes that are not dysregulated (gray) and on the 464 Swhtr-dependent genes (red). Percental distribution was tested for significant differences between Swhtr-dependent and expressed genes by binomial test (p < 0.0001).

Fig. 6. Swhtr dispersion under hypoxic stress.

a RNA immunoprecipitation of NKX2-5. Western Blot detection of NKX2-5 enrichment over input after pulldown. qRT-PCR analysis of relative enrichment of Swhtr in NKX2-5 pulldown of HL-1 cardiomyocytes under normoxic (21% O₂) and hypoxic (1% O₂) conditions (24 h) normalized to IgG negative control (IgG n = 4, Normoxia n = 4, Hypoxia n = 7). Values are given in mean ± SEM. b smFISH of Swhtr under normoxic (21% O₂) and hypoxic (1% O₂) conditions (24 h) in WT, Swhtr^3xpA/3xpA;tg and Swhtr^3xpA/3xpA neonatal cardiomyocytes. The white line represents 10 µm. c Quantification of smFISH Swhtr signals in the nucleus counted from isolated neonatal cardiomyocytes of 4 animals of the indicated genotype (n > 50). Color legend applies to (c) and (d). d Quantification of smFISH Swhtr signals in the nucleus of neonatal cardiomyocytes isolated from 4 WT or Swhtr^3xpA/3xpA;tg animals under normoxic (21% O₂) and hypoxic (1% O₂, 24 h) conditions (n > 50). Statistical significance was tested by Two-way ANOVA with Tukey correction for multiple comparisons. **0.0031, ***0.0006, ****<0.0001.