Cardiac contraction activates endocardial Notch signaling to modulate chamber maturation in zebrafish

Abstract

Congenital heart disease often features structural abnormalities that emerge during development. Accumulating evidence indicates a crucial role for cardiac contraction and the resulting fluid forces in shaping the heart, yet the molecular basis of this function is largely unknown. Using the zebrafish as a model of early heart development, we investigated the role of cardiac contraction in chamber maturation, focusing on the formation of muscular protrusions called trabeculae. By genetic and pharmacological ablation of cardiac contraction, we showed that cardiac contraction is required for trabeculation through its role in regulating notch1b transcription in the ventricular endocardium. We also showed that Notch1 activation induces expression of ephrin b2a (efnb2a) and neuregulin 1 (nrg1) in the endocardium to promote trabeculation and that forced Notch activation in the absence of cardiac contraction rescues efnb2a and nrg1 expression. Using in vitro and in vivo systems, we showed that primary cilia are important mediators of fluid flow to stimulate Notch expression. Together, our findings describe an essential role for cardiac contraction-responsive transcriptional changes in endocardial cells to regulate cardiac chamber maturation.

Keywords: Blood flow; Cardiac chamber maturation; Cardiac contraction; Neuregulin 1; Notch; Primary cilia; Trabeculation; Zebrafish.

Fig. 1.

Cardiac contraction is required for myocardial trabeculation. (A,D) Schematic diagrams of (A) morpholino (MO) injection at one-cell stage or (D) pharmacological inhibition of cardiac contraction (6 μM blebbistatin) from 22 hpf. The morpholino-injected or chemical-treated Tg(kdrl:EGFP); Tg(myl7:dsRed) double transgenic embryos were allowed to develop further and examined for cardiac trabecular phenotypes at 3 dpf. (B,C,E) Maximal projection of confocal z-stacks reveals the overall shape of the heart. (B′,C′,E′) Mid-chamber confocal optical section of the same hearts shown in B,C,E. (B″,C″,E″) Magnified high-resolution images of the cardiac regions marked by dotted lines in B′,C′,E′. White arrows point to trabeculae. a, atrium; v, ventricle. Scale bars: 50 μm in C′,E′; 10 μm in C″,E″.

Fig. 2.

Cardiac contraction is required for endocardial Notch activation and notch1b transcription. (A) Schematic diagram of morpholino (MO) gene knockdown experiment, in which double transgenic Tg(Tp1:EGFP); Tg(myl7:dsRed) embryos were injected with (B,C) control, (D,E) tnnt2a or (F,G) notch1b morpholinos and imaged at 48-50 hpf. (B,D,F) Representative whole-mount images of Notch reporter with cardiac regions highlighted by circles. (C,E,G) Confocal maximal intensity projections of the hearts shown in (B,D,F) with cardiomyocytes labeled in red, (C′,E′,G′) Notch reporters in green and (C″,E″,G″) colocalized signal in yellow. Minimal colocalization indicates that Notch activation is in endocardial cells. (H,I) Whole-mount notch1b riboprobe hybridization in (H) control and (I) tnnt2a^−/− embryos, with the heart outlined in red. Scale bar: 50 μm. a, atrium; v, ventricle.

Fig. 3.

Notch activation in the ventricular endocardium. (A-F) Confocal z-stack maximal intensity projection of hearts from double transgenic Tg(Tp1:VenusPest); Tg(myl7:dsRed) embryos at designated time points, with Tp1:VenusPest expression in green and cardiomyocytes marked in red. Scale bar: 50 μm. a, atrium; v, ventricle.

Fig. 4.

Cardiac contraction promotes trabeculation through notch1b-efnb2a-nrg1 epistasis. (A,B,C) Mid-chamber confocal optical section of Tg(myl7:dsRED); Tg(kdrl:EGFP) double transgenic (A) control, (B) notch1b and (C) efnb2a morphant (MO) hearts, showing cardiomyocytes in red and endocardial cells in green. (A′,B′,C′) Magnified high-resolution images of the cardiac regions highlighted by dotted lines in A,B,C. (D-F) Expression of notch1b, efnb2a and nrg1 in hearts isolated from (D) tnnt2a, (E) notch1b and (F) efnb2a morphants compared, normalized to expression in control morphant hearts (dashed line). *P≤0.05-0.01, **P≤0.01-0.001, ***P<0.001 compared with control morphants (one-sample t-test compared with control morpholino fold change=1). Error bars are s.e.m. White arrows point to trabeculae. Scale bars: 50 μm in C; 10 μm in C′.

Fig. 5.

Notch1 activation rescues efnb2a and nrg1 expression in non-contractile hearts. (A) Experimental schematic diagram of morpholino injection and heat-shock overexpression of NICD. qRT-PCR and imaging were performed to examine gene expression and trabecular phenotype at 2-3 and 4 dpf, respectively. (B) Representative whole-mount images of cardiomyocytes (red) and Notch reporter (green) in (a,b) control and (c,d) tnnt2a morphants at 48 hpf (a,c) without or (b,d) with NICD overexpression. (C,D) Expression of efnb2a and nrg1 in hearts isolated from (C) control morphants and (D) tnnt2a morphants, comparing gene expression in embryos with or without NICD overexpression. (E-H) Confocal mid-chamber optical section of 4 dpf hearts, with dotted lines marking the inset magnified in E′-H′. ^#P≤0.075-0.05, *P≤0.05-0.01, **P≤0.01-0.001, ***P<0.001 compared with control morphants (one-sample t-test compared with control morpholino fold change=1). Error bars are s.e.m. White arrows highlight trabeculae. a, atrium; HS−, heat shock control without NICD overexpression; HS+, heat shock control with NICD overexpression; NICD, Notch intracellular domain; v, ventricle. Scale bars: 50 μm in H; 10 μm in H′.

Fig. 6.

Shear stress promotes notch1 expression in a primary cilia-dependent manner. (A-B′) The hearts of Tg(actb2:Arl13b-GFP) embryos were examined by confocal microscopy to assess localization of primary cilia in the endocardium. (A) Tg(actb2:Arl13b-GFP) reporter expression in the heart. The schematic diagram indicates the orientation of the heart and confocal section relative to the whole embryo at 30 hpf. (B) High-resolution view of a Tg(actb2:Arl13b-GFP) embryo merged with (B′) a bright field image demonstrating colocalization of the primary cilium base with an endocardial cell. (C,D) Whole-mount Tg(myl7:dsRed); Tg(Tp1:VenusPest) double transgenic (C) control and (D) ift88 morphants (MO) at 48 hpf. The hearts are marked with dashed circles. (C′,D′) Confocal maximal projection of the heart from the individual embryos shown in C,D, overlaying cardiomyocytes (red) and Notch reporter (green). (E,F) Confocal optical section of the (E) control and (F) ift88 morphant embryo cardiomyocytes (red) at 80 hpf. The insets marked by the dotted rectangle were magnified in E′,F′. (G) Quantification of ventricular Notch reporter (EGFP) MFI from whole-mount embryos at 48 hpf. (H,I) Relative expression of (H) notch1b and (I) nrg1 in control and ift88 morphant hearts. (J-L) Expression of (J) Notch1, (K) Efnb2 and (L) Nrg1 in DMSO- or 50 μM CBD-treated MLECs that were exposed to 2 dyn/cm² shear stress for 4 h compared with static DMSO- and CBD-treated controls. Red arrows highlight Arl13b-GFP in the endocardium. Blue arrows highlight Notch reporter signal in neural tissue. White arrows highlight trabeculae. *P≤0.05-0.01, **P≤0.01-0.001, ***P<0.001 compared with control morphants (one-sample t-test compared with 1.0-fold change or Student's t-test). Error bars are s.e.m. Scale bars: 50 μm in A; 10 μm in B,B′,F′; 100 μm in F. a, atrium; CBD, ciliobrevin D; lu, lumen; MFI, mean fluorescence intensity; v, ventricle.

Fig. 7.

Cardiac contraction activates endocardial Notch signaling in a primary cilia-dependent manner to regulate trabeculation. Schematic diagram. (Blue boxes) Cardiac contraction activates a regulatory Notch-Ephrin B2a-Neuregulin 1 pathway in endocardial cells to activate ErbB2 signaling in cardiomyocytes to promote trabeculation. (Red box) Given that Notch activation in non-contractile hearts is not sufficient for trabeculation, cardiac contraction also stimulates parallel pathways to promote trabeculation. (Purple boxes) Cardiac contraction causes blood flow, which is likely to be detected by primary cilia to activate Notch1.