|
Fig. 1. CASZ1 interacts with CHD5. (A,B) Full-length CASZ1 (bait) interacts with a partial CHD5 protein (prey) in a yeast two-hybrid assay. Cells placed under –His/–Ade selection. (C) Schematic of the structure of CHD5 depicting predicted domains. (D) RT-PCR on cDNA derived from Xenopus whole-embryo lysate. Ef1a was used as a loading control. (E) Expression of Chd5 RNA by in situ analysis in early tadpole Xenopus embryos (stage 37). h, heart; pr, pronephros; e, eye; pa, pharyngeal arches. (F-M) Colocalization of CHD5 and CASZ1 protein in the nuclei of the developing Xenopus (stage 40) myocardium (m). White boxes in F-I correspond to the magnified regions in J-M. Scale bars: 500 μm in E; 50 μm in F-I; 5 μm in J-M.
|
|
Fig. 2. CHD5 is required for cardiac morphogenesis. (A,B) CHD5-depleted embryos (stage 37) exhibit smaller, improperly looped hearts that fail to undergo chamber formation. Lateral views of whole-mount antibody staining with anti-tropomyosin (Tmy) antibody. (C,D) CHD5-depleted hearts fail to develop fully formed chambers and have a thicker myocardial layer. Transverse sections of cardiac actin-EGFP transgenic (CA-TG) Xenopus embryos (stage 37). (E-M) 3D reconstructions of early tadpole Xenopus embryos (stage 37) stained with anti-tropomyosin antibody viewed laterally and posteriorly. CHD5-depleted hearts (H-M) fail to complete migration and fusion (H-J, 54%), or fail to undergo proper looping and chamber formation (K-M, 31%). Only 15% display normal looping. n>25 per condition, two biological replicates. v, ventricle; o, outflow tract; i, inflow tract. D, dorsal; A, anterior; V, ventral; P, posterior. Scale bars: 100 μm in A-D; 50 μm in E-M.
|
|
Fig. 3. CHD5 is required for cardiac morphogenesis. (A-D) Scanning electron microcopy (SEM) of Xenopus embryos at early tadpole stage (stage 37) in (A) control and (B-D) CHD5-depleted embryos. CHD5-depleted embryos fail to complete looping and begin chamber formation, and fail to undergo normal cell shape changes associated with development and maturation of the linear heart tube. o, outflow tract; v, ventricle. Arrowhead indicates failure of cardiomyocyte fusion at ventral midline. (E,F) CASZ1-depleted embryos exhibit similar defects in cardiac looping and cardiac cell shape changes. n=10 per condition, two biological replicates. Scale bar: 50 μm.
|
|
Fig. 4. CHD5 and CASZ1 are required for myocardial tissue adhesion. (A-L) Transverse sections of CHD5 and CASZ1-depleted CA-GFP embryos (stage 37) display diffuse or reduced expression of ZO-1, indicating defects in tight junction formation. Phenotypes represented across multiple sampled embryos (n≥4, from two biological replicates). Boxes in A-F outline magnified images in G-L. Merge represents ZO-1 (red), DAPI (blue) and CA-GFP (green). (M-O) TEM imaging of transverse sections of control, CHD5-depleted and CASZ1-depleted myocardium (stage 37) reveal large intercellular gaps between cardiomyocytes (arrowheads) adjacent to the inner myocardial chamber (ic) in both CHD5-depleted and CASZ1-depleted embryos compared with control embryos (n≥2, from two biological replicates). Scale bars: 100 μm in A-F; 20 μm in G-L; 5 μm in M-O.
|
|
Fig. 5. CHD5 and CASZ1 are required for deposition of myocardial basement membrane. (A-C) The basement membrane (bm) forms as a discontinuous membrane on the basal surface of the myocardium in CHD5- and CASZ1-depleted embryos, as shown by TEM imaging (stage 37) (n=2). (D-I) CHD5- and CASZ1-depleted CA-GFP hearts (stage 37) exhibit areas of depleted or improperly localized laminin (yellow arrows). Phenotypes represented across multiple sampled embryos (n≥4, from two biological replicates). Merge represents laminin (red), DAPI (blue) and CA-GFP (green). Scale bars: 0.2 μm in A-C; 100 μm in D-I.
|
|
Fig. 6. CASZ1 and CHD5 are required for myocardial tissue adhesion prior to cardiac looping and chamber formation. (A-X) Control, CHD5-depleted and CASZ1-depleted CA-GFP embryos were examined for expression of the tight junction markers ZO-1 (A-L) and claudin-5 (M-X). Expression of both ZO-1 and claudin-5 is reduced and appears diffuse and unorganized in transverse sections of CHD5- and CASZ1-depleted hearts (stage 33). Phenotypes represented across multiple sampled embryos (n≥4). White boxes in A,B,E,F,I,J,M,N,Q,R,U,V correspond to magnified images in C,D,G,H,K,L,O,P,S,T,W,X, respectively. Merge represents ZO-1 or claudin-5 (red), DAPI (blue) and CA-GFP (green). Scale bars: 100 µm in A,B,E,F,I,J,M,N,Q,R,U,V; 20 µm in C,D,G,H,K,L,O,P,S,T,W,X.
|
|
Fig. 7. CHD5 is crucial for CASZ1 function. (A,B) The region of CASZ1 containing the minimal CHD5-interacting domain amino acid residues 785-998 was identified using a yeast two-hybrid assay. (C-N) Control (C-E) and CASZ1-depleted (F-H) Xenopus embryos compared with CASZ1-depleted embryos co-injected with mRNA encoding either full-length CASZ1 (I-K) or CASZ1δCID (L-N) at early tadpole stage (stage 37). Full-length CASZ1 partially rescued cardiac looping and chamber formation (v), and outflow tract (o) and inflow tract (i) formation, whereas CASZ1δCID failed to rescue any aspect of cardiogenesis. (O-Q) Three independent experiments (n>20 per condition) of embryos were scored for cardia bifida (O), degree of proper cardiac looping (P) and chamber formation (Q) using whole-mount staining with anti-tropomyosin antibody. Fisher's exact test was performed for significance: NS, not significant; *P=0.003; ***P<0.0001. Error bars represent calculated s.e.m. Scale bars: 500 μm in C,D, F,G,I,J,L,M; 100 μm in E,H,K,N.
|
|
Fig. 8. The CASZ1 CID domain disrupts heart development. (A-D) Control embryos at early tadpole stage (stage 37) exhibit looping and formation of the cardiac chambers. v, ventricle; o, outflow tract; i, inflow tract. (E-L) CASZ1 and CASZ1 CID misexpressing embryos display incomplete convergence of the outflow and inflow tract, as seen in dorsal views (I,K) (arrowheads), and a partially closed inflow tract, as seen in posterior views (H,L) (arrowheads). n=23 for CASZ1 mis; n=29 for CASZ1 CID mis; two biological replicates. Scale bars: 100 μm.
|
