Figure 4. Enhancer-associated ncRNAs at Ryr2 and Atp2a2 are necessary for calcium-handling gene expression and cellular phenotype.

(A) Gene Ontology functional enrichment for gene targets within a 2-Mb window of the TBX5-activated enhancers. Odds Ratio (OR) and P-value of overlap with GO terms ‘Calcium Ion Transport Genes’ and ‘Calcium Ion Transmembrane Transport Genes’. (B) Representative 3 days post fertilization (Dpf) zebrafish embryos injected with a reporter construct containing the wildtype Ryr2 enhancer reporter, showing cardiac EGFP expression. One Representative embryos from two stable lines are also shown (Line 1 and Line 2, one embryo each). Stable lines display EGFP fluorescence in the heart in addition to other tissues (see Supplement). (C) Genomic view of the Ryr2 locus showing four candidate TBX5-dependent regulatory elements identified by TBX5 ChIP-seq (gray) and by RACER expression (Enh 4, red). (D) Relative luciferase activity in HL-1 cardiomyocytes of candidate Ryr2 enhancers, normalized to co-transfected Renilla vector and to a vector with a scrambled insert (n > 3 replicates). (E) Genomic view of the Ryr2 locus, showing identified noncoding RNA Ryr2 Associated Cis- Element RNA (RACER) in blue. A putative regulatory element associated with ncRNA is marked in red. Tracks show TBX5 ChIP (GSE21529), RNASeq from R26^CreERt2 and Tbx5^fl/fl;R26^CreERt2, ATAC-Seq in HL-1 cardiomyocytes, DNASe hypersensitivity (ENCODE), and H3K27 acetylation (GSE52123). (F) Gene expression of Ryr2 mRNA and the RACER transcript after knockdown of Ryr2 mRNA (top) and knockdown of RACER (below). Relative transcript expression (RTE) after mRNA knockdown: 0.52 ± 0.03, p = 5.3E-5, for mRNA; RTE 0.92 ± 0.09, p = 0.43, and 0.82 for RACER. For ncRNA knockdown: RTE 0.44 ± 0.03 for mRNA and 0.45 ± 0.02 for RACER, p = 5.64E-6 and 0.04, respectively. (G, H) Representative calcium transient line scans from control (left) and ASO knockdown (right) HL-1 cardiomyocytes in the presence or absence of isopropteranol (+ISO, –ISO resp) after knockdown of Ryr2 mRNA (top) or RACER (bottom) with antisense oligonucleotides (ASO). Quantification of rise velocity under control and isoproterenol treatments (right). (I) Relative transcript expression of chromatin enriched vs soluble nuclear fractions in HL1 cardiomyocyte cells of several ncRNAs and (as controls) Xist and the Gapdh-coding mRNA transcript. Expression was normalized to 18S ribosomal RNA. P values compared the relative expression in soluble vs nuclear fractions of the ncRNAs and Xist to that of GAPDH mRNA, and were less than Bonferroni corrected p<6e-3 for all ncRNAs. (J) Fold-enrichment of RNA Polymerase 2 (Pol2) occupancy by ChIP at housekeeping gene Btf3 promoter (prom.) and at the Ryr2 promoter in the control (left) and after antisense oligonucleotide knockdown of Ryr2 enhancer ncRNA (ASO, right). Normalized to a control locus near the Gapdh gene.

Figure 4—figure supplement 1. Knockdown of control Hprt locus does not change Ryr2 or RACER expression.

Gene expression of Ryr2 mRNA and RACER transcript after knockdown of Hprt mRNA. Relative transcript expression (RTE) after mRNA knockdown for RACER 1.23 ± 0.25, p = 0.77; for Ryr2 RTE 0.84 ± 0.09, p = 0. 21.

Figure 4—figure supplement 2. Ryr2 enhancer shows cardiac expression in 3 days post fertilization (dpf) zebrafish.

(A) Representative F0 (left) and F1 (right) 3 dpf zebrafish embryos surviving injection with a reporter construct containing the wildtype Ryr2 enhancer reporter, showing cardiac EGFP expression. Fluorescence was assessed at 3 dpf, after injection of constructs with either wildtype or T-box mutant enhancers. F0 and F1 transgenic zebrafish show cardiac expression of the reporter construct. (B) High-resolution images of stable transgenic zebrafish from two independent lines showing myocardial expression of the wildtype Ryr2 enhancer reporter at 3 dpf. * indicates atrial expression pattern in high-resolution image of line 1 (bottom).