Myosin 7b is a regulatory long noncoding RNA (lncMYH7b) in the human heart

Abstract

Myosin heavy chain 7b (MYH7b) is an ancient member of the myosin heavy chain motor protein family that is expressed in striated muscles. In mammalian cardiac muscle, MYH7b RNA is expressed along with two other myosin heavy chains, β-myosin heavy chain (β-MyHC) and α-myosin heavy chain (α-MyHC). However, unlike β-MyHC and α-MyHC, which are maintained in a careful balance at the protein level, the MYH7b locus does not produce a full-length protein in the heart due to a posttranscriptional exon-skipping mechanism that occurs in a tissue-specific manner. Whether this locus has a role in the heart beyond producing its intronic microRNA, miR-499, was unclear. Using cardiomyocytes derived from human induced pluripotent stem cells as a model system, we found that the noncoding exon-skipped RNA (lncMYH7b) affects the transcriptional landscape of human cardiomyocytes, independent of miR-499. Specifically, lncMYH7b regulates the ratio of β-MyHC to α-MyHC, which is crucial for cardiac contractility. We also found that lncMYH7b regulates beat rate and sarcomere formation in cardiomyocytes. This regulation is likely achieved through control of a member of the TEA domain transcription factor family (TEAD3, which is known to regulate β-MyHC). Therefore, we conclude that this ancient gene has been repurposed by alternative splicing to produce a regulatory long-noncoding RNA in the human heart that affects cardiac myosin composition.

Keywords: alternative splicing; gene regulation; heart failure; long noncoding RNA (lncRNA); myosin heavy chain (MyHC); myosin heavy chain 7b (MYH7b).

Figure 1

MYH7b is alternatively spliced in the heart and correlates with β-MyHC in heart disease.A, MYH7b is expressed as a full-length sarcomeric myosin protein in the human brain and inner ear, but is alternatively spliced, skipping exon 8 (in green) in the heart and skeletal muscle to become noncoding. A putative short peptide (MYH7b_sp) could be produced from the exon-skipped transcript, and a microRNA (miR-499) is encoded regardless of splicing. B, the expression of exon 7-skipped MYH7b correlates with that of β-MyHC in nonfailing (NF), dilated cardiomyopathy (DCM), and ischemic cardiomyopathy (ICM). Data pulled from GSE116250. C, splicing analysis of MYH7b in 64 human hearts shows that exon 8 is almost nonexistent in the human heart. D, quantification of exon skipping in human hearts.

Figure 2

miR-499 does not affect the expression of either MYH7b or β-MyHC.A, miR-499 is effectively knocked down by the anti-miR. miR-499 is still produced upon skipping of exon 7 and could have a role in β-MyHC regulation. B, an anti-miR targeted to miR-499 does not change the expression levels of MYH7b or β-MyHC.

Figure 3

MYH7b_sp does not affect the β-/α-MyHC ratio and is undetectable in human hearts.A, miR-499 has been eliminated as the active molecule involved in MyHC regulation in the heart; however, MYH7b_sp could have a regulatory role. B, single-molecule FISH shows that MYH7b RNA is largely is localized in the cytoplasm, but not along the nuclear envelope where NMD largely occurs. C, quantification of MYH7b RNA localization using FISH. D, western blotting of failing human hearts cannot detect MYH7b_sp using a custom MYH7b antibody that does detect full-length MYH7b in mouse cerebellum. E, RNA-seq of MYH7b_sp overexpression shows no significant change in the β-MyHC/α-MyHC ratio.

Figure 4

lncMYH7b controls β-MyHC expression at both the RNA and protein level.A, MYH7b RNA decreases at least 48 h before there is a change in β-MyHC, RNA levels are reduced. B, ASOs targeted against MYH7b RNA do not affect myomiR levels. C, myosin-separating gels show a reduction in the β-/α-MyHC ratio at the protein level upon lncMYH7b KD in hiPS-CMs. D, representative myosin-separating gel.

Figure 5

lncMYH7b KD shows little overlap with MYH7b_sp overexpression.A, RNA-seq of lncMYH7b KD in iPS-CMs shows a reduction in the β-/α-MyHC ratio, confirming that lncMYH7b is affecting β-MyHC RNA. B, Venn diagram of the differentially expressed genes from the MYH7b_sp overexpression and lncMYH7b KD shows that there is little overlap, which further supports the conclusion that MYH7b_sp is nonconsequential. C, RNA-seq of lncMYH7b KD shows an enrichment of downregulated genes. D, KEGG analysis of the lncMYH7b KD RNA-seq shows enrichment of focal adhesion pathways, indicating possible role of FAK in lncMYH7b regulation.

Figure 6

lncMYH7b KD affects adrenergic signaling and beat rate in hiPS-CMs.A, KEGG pathway diagram showing the effects seen on adrenergic signaling genes upon lncMYH7b KD. B, hiPS-CMs treated with MYH7b ASO have a significantly slower spontaneous beat rate than those treated with control ASO. C, changes in RyR2 and PP2A expression shown by RNA-seq are recapitulated at the protein level.

Figure 7

lncMYH7b KD causes significant effects on sarcomere organization in hiPS-CMs.A, representative images showing the three phenotypes of α-actinin staining we scored in hiPS-CMs. The “organized” category was broken down into two subphenotypes: organized sarcomeres and presarcomeres that have been shown to transition to fully organized sarcomeres in hiPS-CMs (36). This is in comparison to the disorganized phenotype, where there is no discernible organization of alpha-actinin. B, quantification of the change in α-actinin phenotype upon lncMYH7b knockdown. There is a significant decrease in the percentage of cells with α-actinin organization with MYH7b ASO treatment. Cells from three wells from two independent differentiations were scored.