A Continuum of Atrial Peristalsis Initiates the Bicuspid to Quadricuspid Valve Transition

Figures

• 

  Figure 1.. A 2- to 4-cusp AV valve transition develops between 14 and 45 days post-fertilization (dpf). (a) A zebrafish larva (left panel) at 12 dpf is shown, with its heart highlighted. The right panel illustrates the two-chambered cardiac anatomy and the quadricuspid atrioventricular (AV) valve. (b) From 14 to 45 dpf, confocal imaging of hearts in Tg(TP1:EGFP; flk:mCherry) larvae during the transition from 2-cusp to 4-cusp. Dashed lines indicate the developing lateral cusps extending from the commissures (asterisks). (c) At 15 dpf, confocal imaging of EdU-labeled hearts (24-hour incubation) from Tg(TP1:EGFP) larvae (n = 7) shows that the commissure remodels through endothelial proliferation (highlighted in the blue box) before VIC differentiation, after which VICs proliferate to initiate lateral cusp formation. (d) From 14 to 45 dpf, the average area of the left/right lateral cusps (LL/RL) increases more rapidly than that of the pre-existing superior/inferior cusps (SUP/INF), as indicated by the increasing percentage over time. The number (n) of hearts analyzed at each time point is displayed. Source data are available in a Source Data file. (e) From 14 to 45 days post-fertilization…

• 

  Figure 2.. Laser microdissection-assisted RNA sequencing reveals valvular maturation programs following the 2- to 4-cusp transition. (a) Laser microdissection-assisted RNA sequencing is performed on AV valve tissue from 28-, 35-, and 45-dpf WT zebrafish. Four samples from each time point are collected for sequencing, each containing 30~60 tissue pieces (depending on age) cut from ~10 hearts. One outlier from each time point is discarded based on the principal component (PC) analysis plot of normalized gene counts. Percentages of variance explained by the first two PCs are indicated on each axis. Source data are provided as a Source Data file. (b - c) Volcano plots show the log-fold changes and Benjamini & Hochberg (1995) adjusted p-values for upregulated genes (labeled with lines) at 35 or 45 dpf versus 28 dpf. The genes shown here are validated in the spatial transcriptomic atlas, involved in chondrogenesis (snorc, mmp2, prg4b, col2a1b, krt4, krt91, pcolce2a), protein synthesis (eef1a1a), retinoic acid signaling (rbp4), and glycogen metabolism (gyg1b, pygl). Original p-values are computed using the Wald test in DESeq2. Dashed lines mark the selected thresholds for log-fold changes (-1/1)…

• 

  Figure 3.. Peristaltic atrioventricular contraction aligns shear stress with the inferior-superior (S-I) axes during AV valve initiation. (a) At 25 hours post-fertilization (hpf), brightfield imaging of Tg(TP1:EGFP) embryos along two orthogonal planes of the heart tube (green and blue) throughout a cardiac cycle. Dashed lines indicate the corresponding valvular axes at the atrioventricular canal (AVC): superior-inferior (S-I) and left-right (L-R). See Supplementary Video 1. (b) At 33 and 48 hpf, brightfield imaging of Tg(TP1:EGFP) embryos along the AVC cross-section (gray plane) demonstrates the opening of the AVC along the superior-inferior (S-I) axis during diastole. By 48 hpf, the differentiated AVC becomes flattened along the S-I axis prior to the SUP/INF cusps. Colored lines delineate the AVC endocardial contours at various time points. See Supplementary Video 2. (c) Extracted atrioventricular canal (AVC) geometry from the brightfield images, and a 2D shear-stress simulation demonstrates the spatial variation in shear-stress along the L-R axis (green arrowheads) and the S-I axis (red arrowheads) at 48 hpf. The color scale indicates the normalized shear-stress for a steady,…

• 

  Figure 4.. Increased endocardial wall shear stress occurs along the left-right lateral (L-R) axis during the 2- to 4-cusp transition. (a) At 14 dpf, whole-mount in situ hybridization for klf2a mRNA in the Tg(TP1:EGFP) heart reveals high klf2a expression around the remodeled commissures. The right panels show 3D reconstructions of Notch (TP1) and klf2a fluorescent signals. (b) Average valvular klf2a fluorescence intensity (TP1-normalized, black line) along the L-R axis peaks at the remodeled commissures (n = 7 hearts). The light-brown shading indicates the Standard Error of the Mean (SEM). (c) Schematics showing variations in the shear stress along the superior and inferior cusps and the L-R commissures during the 2-cusp to 4-cusp transition. (d) At 10 dpf, 4D light-sheet imaging of a Tg(TP1:EGFP) heart reveals the opening and closing of AV leaflets and the remodeling of the left lateral (LL) commissure over a cardiac cycle (arrowheads). See Supplementary Video 5. (e) AVC geometry extracted from the light-sheet images and 2D computational fluid dynamics (CFD) reveal a decrease in the shear stress along the remodeled LL commissure (arrowheads) but not along the right lateral (RL) commissure.…

• 

  Figure 5.. Ventricular contraction-generated shear stress across AVC is required to initiate the SUP/INF cusps. (a) At 3 dpf, brightfield imaging of haf (“half-hearted”, myh7−/−) mutants and their wild-type (WT) siblings. In WT hearts, the atrial blood (red) completely empties into the ventricle, whereas in half hearts, atrial inflow encounters resistance from the blood remaining in the non-contracting ventricle (purple), resulting in incomplete emptying and reduced flow across the AVC. See Supplementary Video 6. (b) At 6 dpf, whole-mount in situ hybridization for notch1b, klf2a, and snai1b mRNA in the WT heart (n = 7) shows clear colocalization in the SUP/INF cusps. In the haf hearts (n = 7), notch1b expression is no longer localized to the AVC, and snai1b expression is absent, suggesting the absence of ventricular interstitial cells (VICs). klf2a enrichment at the AVC occurs in four of seven hearts (4/7) but not in the remaining three (3/7). (c) 3D reconstruction of notch1b fluorescent signals reveals that haf mutants fail to develop peristaltic contractions along the left-right and inferior-superior axes. Anatomic labels: Ven, ventricle; Atr, atrium; AVC, atrioventricular canal; SUP,…

• 

  Figure 6.. Atrial peristaltic contraction is required for 2- to 4-cusp transition. (a) At 2 dpf, brightfield imaging of wea (“weak-atrium”, myh6−/−) mutants and their wild-type (WT) siblings. The kymograph on the right shows blood flow dynamics across three cardiac cycles (Sys, systole; Dia, diastole). In WT hearts, blood flows across the AVC (top to bottom) during atrial contraction. In contrast, in wea hearts, blood experiences prolonged stalling (ST) or reversal/regurgitation (R) in the absence of atrial contraction. See Supplementary Video 7. (b) On average, wea mutants have significantly shorter atrial filling (AF) but longer stalling (ST) or regurgitation (R) periods than their WT siblings. Ventricular filling (VF) duration is similar between WT and wea. All values are reported as mean ± SD. Number of hearts analyzed: WT = 8, wea = 6. An ordinary two-way ANOVA followed by Sidak’s multiple-comparisons test was used to assess statistical significance. Source data are provided as a Source Data file. (c) At 14 dpf, whole-mount in situ hybridization for notch1b, klf2a, and ccm1 mRNA in the WT heart (n = 15) shows enrichment of klf2a at the remodeled commissure (arrowheads) before (upper…

• 

  Figure 7.. Shear stress at the remodeled commissure activates VIC differentiation. (a) The endocardial cell (solid outlines with arrowheads) undergoing differentiation loses Notch1b and Klf2a expression but retains Ccm1 expression. (b - c) The Notch-negative (TP1) differentiating endocardial cells (solid outlines with arrowheads) express snai1b (mRNA), a mesenchymal gene involved in EndoMT. snai1b becomes highly enriched in VICs after the initiation of 4-cusps. (d) Endothelial identity (fli1a:LifeAct-GFP signals) is also lost during VIC differentiation (solid outlines with stars). Arrowheads mark endothelial filopodia. A dashed line outlines the free edge of the SUP/INF cusps and the remodeled commissure. (e - f) Compared with WT, wea mutants have significantly fewer snai1b+ cells at the remodeled commissure (arrowheads). The dashed line outlines the free edge of the SUP/INF cusps and the remodeled commissure. All values are reported as mean ± standard deviation (SD). Each data point represents a single commissure (n = 21 from 12 WT hearts, n = 17 from 10 wea hearts). A Mann-Whitney test was applied to the means to assess statistical significance. Source data are provided as a Source Data…

• 

  Figure 8.. Spatial distributions of KLF4 and SNAI1 during the 2- to 4-cushion transition in the mouse AVC/OFT. (a) Schematic illustrating the 2- to 4-cushion transition during mouse AVC and OFT development. The SUP/INF cushions extend toward the center of the AVC/OFT and begin to fuse at E11.5. Meanwhile, the lateral (LC) and intercalated (IC) cushions develop along the orthogonal axis. The atrioventricular and OFT septa (Sep) arise from the fused cushions. The lateral and intercalated cushions become leaflets (blue shading) of the mitral/tricuspid (Mi/Tri) and aortic/pulmonary (Ao/Pu) valves, respectively. Partially created in BioRender. Wang, J. (2025) https://BioRender.com/jzz0scy (b) At E10.5, KLF4 and SNAI1 antibody staining shows colocalization at the initiating lateral cushion (solid outlines in the magnified lower panels). SNAI1+ cells (arrowheads) express weak KLF4 in the endocardium but not within the cushion, suggesting mesenchymal differentiation. The shape of KLF4+ endocardium resembles the dumbbell geometry, similar to the zebrafish AVC during the 2- to 4-cusp transition. (c) The intercalated cushion of OFT (solid outlines in the bottom panels) harbors very few SNAI1+ cells…