Sarcomeric signaling complexes are important to sustain proper sarcomere structure and function, however, the mechanisms underlying these processes are not fully elucidated. In a gene trap experiment, we found that vascular cell adhesion protein 1 isoform X2 (VCAP1X2) mutant embryos displayed a dilated cardiomyopathy phenotype, including reduced cardiac contractility, enlarged ventricular chamber and thinned ventricular compact layer. Cardiomyocyte and epicardial cell proliferation was decreased in the mutant heart ventricle, as was the expression of pAKT and pERK. Contractile dysfunction in the mutant was caused by sarcomeric disorganization, including sparse myofilament, blurred Z-disc, and decreased gene expression for sarcomere modulators (smyd1b, mypn and fhl2a), sarcomeric proteins (myh6, myh7, vmhcl and tnnt2a) and calcium regulators (ryr2b and slc8a1a). Treatment of PI3K activator restored Z-disc alignment while injection of smyd1b mRNA restored Z-disc alignment, contractile function and cardiomyocyte proliferation in ventricles of VCAP1X2 mutant embryos. Furthermore, injection of VCAP1X2 variant mRNA rescued all phenotypes, so long as two cytosolic tyrosines were left intact. Our results reveal two tyrosine residues located in the VCAP1X2 cytoplasmic domain are essential to regulate cardiac contractility and the proliferation of ventricular cardiomyocytes and epicardial cells through modulating pAKT and pERK expression levels.
The sarcomere is the basic contractile unit of cardiac and skeletal striated muscles1, consisting of central bipolar myosin thick filaments surrounded by parallel actin thin filaments, which are anchored to the Z-disc. Mutations in thin or thick filament components may cause hypertrophic cardiomyopathy or dilated cardiomyopathy (DCM)2,3. In addition to well-known calcium homeostasis and cell signaling events, cardiac contractility is regulated by sarcomere maintenance. The sarcomere is a dynamic structure under constant turnover, and the homeostatic quality control system involves chaperones, the calpain and ubiquitin proteasome system, and autophagy4. In addition to protein quality control, other modulators participate in the maintenance of sarcomere homeostasis in cardiac and skeletal muscle. For instance, Smyd1, a MYND and SET-domain-containing lysine methyltransferase, is specifically expressed in muscle tissues. Smyd1 knockout mice display defects in ventricular cardiomyocyte maturation and formation of the right ventricle5, and the protein localizes to the sarcomere of murine cardiomyocytes where it interacts with myosin at the sarcomeric M-line6. Furthermore, fla/smyd1 mutant zebrafish exhibited disrupted sarcomere assembly in both heart and skeletal muscle, demonstrating the importance of the Smyd1-myosin interaction in the assembly of thick filaments6.
Z-discs contain α-actinin as a core protein7, which interacts with F-actin and titin. These interactions are critical for mechanical function. In addition to its mechanical role, Z-disc proteins are important for cell signaling, gene expression and cell survival8, with mutations in α-actinin, calsarin, telethonin or MLP all causing DCM or hypertrophic cardiomyopathy3. The Z-disc is linked to the sarcolemma via desmin intermediate filaments, which connect to costameres7. Two major protein complexes, including the dystrophin-glycoprotein complex (DGC) and the integrin-vinculin-talin complex, exist in the costamere9. Mutations in several proteins that comprise the DGC may cause cardiomyopathy or muscular dystrophy10. The integrin heterodimer transduces signals through focal adhesion kinase or integrin-linked kinase (ILK) to regulate cardiac growth, contractility and repair11. Cardiac deletion of vinculin, β1 integrin or ILK in mice led to dilated cardiomyopathy3. Therefore, the costamere-Z-disc axis is important for force and signal transmission between the sarcomere, sarcolemma and extracellular matrix. Further characterization of proteins within this axis will serve to advance our knowledge of mechanotransduction in cardiac muscle.
Recently, the zebrafish has emerged as a model organism for cardiovascular research, partially due to the availability of the zebrafish genome sequence and conserved function of many human and zebrafish genes. Zebrafish heart development and function resemble those of mammals, allowing researchers to use genetic tools for studies on cardiogenesis or cardiac diseases. Intriguingly, zebrafish also have the ability to regenerate heart muscle, suggesting that this model may hold clues to uncovering therapeutic treatments for heart failure12,13. In order to discover proteins that modulate cardiac development, we conducted a Tol2 transposon-mediated gene trap study and identified a vascular cell adhesion protein 1 isoform X2 (VCAP1X2) mutant that displayed a DCM phenotype. A BLAST search revealed that the mammalian homologue to VCAP1X2 is vascular cell adhesion molecule 1 (VCAM1). Mammalian VCAM1 has been implicated in leukocyte transendothelial migration via engagement of the cell-surface ligand, very late antigen 4 (VLA4/α4β1-integrin)14,15. Moreover, VCAM1-deficient mouse embryonic hearts displayed thinned compact layer, reduction of intraventricular septum and epicardium formation defects, and the interaction between VCAM1 and integrin α4 was shown to be important for heart development16,17. VCAM1 is expressed in cardiomyocytes derived from human embryonic stem cells (hESCs) or human induced pluripotent stem cells (hiPSCs). These VCAM1-positive cells were found to express high levels of cardiac genes and display clear sarcomere structure, self-beating and action potentials similar to ventricular and pacemaker cells18.
In this report, we show that the DCM phenotype in the VCAP1X2 mutant is attributed to sarcomere disorganization, decreased pAKT and pERK expression, and reduced gene expression for sarcomere modulators, sarcomeric proteins and calcium regulators. We conducted rescue experiments by injecting embryos with variant mRNAs for either VCAP1X2 or smyd1b (a downstream effector), and probed the underlying signaling mechanisms by treating with PI3K and pMEK inhibitors or activators. Together, our results reveal two tyrosine residues located in the VCAP1X2 cytoplasmic domain are essential to regulate cardiac contractility and the proliferation of ventricular cardiomyocytes and epicardial cells through modulation of pAKT and pERK levels.
VCAP1X2 is expressed in the heart and localized to myocardium plasma membrane
To discover genes involved in cardiac development, we injected Tol2-Gal4-VP16; UAS: EGFP-Tol2 plasmid19 and produced a gene trap transgenic zebrafish with EGFP expression in the heart (Fig. 1A). The insertion was located in intron 1 of the si:ch211-74m13.3 gene (ZFIN ID: ZDB-GENE-060503-219) on chromosome 3. Two splice variants (si:ch211-74m13.3-201, si:ch11-74m13.3-202) consisting of seven or five exons were produced. The variants encode vascular cell adhesion protein 1 isoform X1 (VCAP1X1) (XP_002661307.2, 551 amino acids) and vascular cell adhesion protein 1 isoform X2 (VCAP1X2) (XP_005164393.1, 360 amino acids). VCAP1X2 protein contains a signal peptide, one intercellular adhesion molecule (ICAM) domain, two immunoglobulin (Ig) superfamily domains, a transmembrane domain, and a cytosolic domain, while VCAP1X1 possesses two extra Ig domains (Supplementary Fig. S1A). Sequence comparison indicated that VCAP1X1 and VCAP1X2 are identical at amino acids 1-303 (Supplementary Fig. S1B). However, strong expression of VCAP1X2, but not VCAP1X1, was identified at 72 and 96 hours post fertilization (hpf) by RT-PCR using specific reverse primers for each isoform, indicating that only VCAP1X2 was expressed during embryonic stages (Supplementary Fig. S1C,D).
RT-PCR analysis indicated splicing of exon 1 and exon 2 was blocked by the insertion, and only exon 1 of VCAP1X2 was synthesized in si:ch211-74m13.3 gene trap homozygous transgenic fish (Fig. 1B). Normal global embryonic morphology was observed in si:ch211-74m13.3 gene trap homozygous embryos at 48 hpf and intense EGFP expression was detected in the heart, primordial hindbrain channel and posterior cardinal vein (Fig. 1C). We then generated a polyclonal antibody using amino acids number 31–308 as an antigen. With this antibody, we detected VCAP1X2 in the heart ventricle at 96 hpf in Tg(myl7:EGFP; myl7:H2AFZ mCherry) transgenic fish but not homozygous mutant embryos, confirming that this gene trap line is indeed a VCAP1X2 mutant (Fig. 1D).
VCAP1X2 was expressed ubiquitously at shield and somite stages (5 s and 18 s) (Fig. 1E,a–c). Expression of VCAP1X2 was identified in the heart beginning from the 18 s stage and showed strong ventricular and weak atrial expression at 48 and 72 hpf (Fig. 1E,d,f,i,l,m). VCAP1X2 was co-expressed with myl7 in the myocardium at 24 and 30 hpf (Fig. 1E,g,j), and was also expressed in the primordial hindbrain channel and posterior cardinal vein during different developmental stages (Fig. 1E,e,h,k,m). VCAP1X2 immunofluoresence signal co-localized with GFP in the primordial hindbrain channel and posterior cardinal vein of Tg(fli1:GFP) embryos at 48 hpf (Fig. 1E,n–q).
To further clarify the localization of VCAP1X2 in the heart, we conducted immunofluorescence staining (Fig. 1F). VCAP1X2 was observed at the plasma membrane of EGFP-expressing ventricular myocardium cells in Tg (myl7: EGFP; myl7: H2AFZ mcherry) embryos at 96 hpf (Fig. 1F,a). No VCAP1X2 was detected in GFP-expressing outer epicardial cells of Tg(tcf21:NTR; tcf21:nucEGFP) embryos or in GFP-expressing inner endocardial cells of Tg(fli1:GFP) embryos (Fig. 1F,b,c). Thus, VCAP1X2 is a membrane protein that is exclusively expressed in the myocardium.
VCAP1X2 deficient embryonic heart displays a DCM phenotype
To characterize the role of VCAP1X2 in heart development, pseudodynamic 3D cardiac imaging was applied to zebrafish embryos, and cardiac function was evaluated at 72 hpf (Fig. 2A)20. Both end-systolic volume (ESV) and end-diastolic volume (EDV) were substantially increased, while ejection fraction (EF) was decreased in VCAP1X2 mutant compared to Tg (myl7: EGFP; myl7: H2AFZ mcherry) controls (Fig. 2B,a–c). In parallel, we determined the end-diastolic length and width of VCAP1X2 mutant hearts by recording a video of the beating heart. Increased end-diastolic length and width, with reduced fractional shortening (FS) were observed in VCAP1X2 mutants relative to controls (Fig. 2B,d–f), further demonstrating that VCAP1X2 mutant embryos have impaired cardiac function.
Histological analysis revealed a thinner ventricular compact layer (one-cell layer) in VCAP1X2 mutant heart compared to WT (three-cell layer) embryos at 120 hpf, and injection of VCAP1X2 full-length (FL) but not LacZ mRNA restored the thickness (Fig. 2C). Significantly reduced ventricular cardiomyocyte number and a decreased percentage of BrdU+ cardiomyocytes were detected in ventricles of VCAP1X2 mutant compared to control embryos at 72 hpf (Fig. 2D,c,d), while similar low levels of apoptosis were detected in ventricles of VCAP1X2 mutant and control hearts by TUNEL labeling at 72 hpf (Fig. 2E,F). Thus, the thinner ventricular compact layer with reduced cardiomyocyte number in VCAP1X2 mutants may be due to decreased cell proliferation. Together, these results show that VCAP1X2 deficient embryonic heart exhibits DCM phenotypes, including impaired cardiac function, enlarged ventricular chamber and thinner ventricular compact layer with decreased cardiomyocyte number.
Two tyrosine residues within the VCAP1X2 cytoplasmic domain are required for heart contractility and cardiomyocyte proliferation
Because phosphorylation of two cytosolic tyrosine residues in platelet endothelial cell adhesion molecule-1 (PECAM1) is known to activate extracellular signal-regulated kinase (ERK)21, we wondered if the two cytosolic tyrosine residues (Y333 and Y341) of VCAP1X2 may also play important functional roles (Fig. 3A). We generated five VCAP1X2 variants, including SP: only signal peptide, ΔN Y2F: deleted extracellular ICAM and Ig domains with two cytosolic tyrosines mutated to phenylalanines, Y2F: full-length VCAP1X2 with cytosolic tyrosine to phenylalanine mutations, ΔN: deleted extracellular ICAM and Ig domains with two cytosolic tyrosines intact, and FL: full-length VCAP1X2 (Fig. 3B). Initially, we compared heart contractility by measuring FS at 48 hpf in WT or VCAP1X2 mutant embryos, or mutants injected with SP or FL mRNA. Injection of FL, but not SP mRNA, restored ventricular FS to levels comparable with WT (p < 0.001, Fig. 3C). Interestingly, injection of human vascular cell adhesion protein 1 isoform c precursor (hVCAM1) mRNA also rescued decreased FS in injected mutant embryos, despite the fact that VCAP1X2 only shares 27% amino acid sequence identity with human VCAM1 isoform c (Supplementary Fig. S2A). At 72 hpf, FS was restored in VCAP1X2 mutants injected with ΔN, FL or hVCAM1 mRNA, but not in those injected with SP, ΔN Y2F, Y2F or LacZ mRNA (p < 0.001, Fig. 3D). Furthermore, injection of ΔN, FL or hVCAM1, but not SP, ΔN Y2F or Y2F mRNA restored ventricular cardiomyocyte number (p < 0.001, Fig. 3E,F) at 72 and 96 hpf. Injection of FL or hVCAM1 increased the percentage of ventricular PCNA+ cardiomyocytes in VCAP1X2 mutants more robustly (p < 0.001) than ΔN (p < 0.05) at 72 hpf (Fig. 3G). However, at 96 hpf, only injection of FL, but not ΔN or hVCAM1, significantly restored percentage of ventricular PCNA+ cardiomyocytes in VCAP1X2 mutants (p < 0.001, Fig. 3H).
We then designed a splicing morpholino (spMO) that targets the acceptor site of intron 1 and monitored the inhibition of splicing between exon 1 and exon 2 of VCAP1X2 by RT-PCR (Supplementary Fig. S3A,B). VCAP1X2 immunofluorescence was detected in heart ventricles at 96 hpf in Tg(myl7:EGFP; myl7:H2AFZ mCherry) transgenic fish and 5 mm spMO-injected control embryos but not spMO-injected morphants (Supplementary Fig. S3C). A thinner ventricular compact layer was detected in VCAP1X2 morphant heart, which was rescued by co-injection of VCAP1X2 but not LacZ mRNA (Supplementary Fig. S3D). Co-injection of ΔN or FL, but not of SP, ΔN Y2F or Y2F mRNA also completely restored reduced ventricular FS and ventricular cardiomyocyte number (p < 0.001, Supplementary Fig. S3E,F). Together, these results demonstrate that the two cytosolic tyrosine residues of VCAP1X2 are essential for heart contractility, compact layer thickness and cardiomyocyte proliferation in the ventricle of embryonic heart.
VCAP1X2 cytosolic tyrosine residues are essential for maintaining pAKT and pERK levels
Cardiomyocyte proliferation is regulated by several signaling pathways, including neuregulin-1/ErbB222 and insulin like growth factor 2 signaling23. PI3K-AKT and MAPK-ERK1/2 transduce upstream signals to activate genes involved in different biological processes, including cell proliferation. Since ventricular cardiomyocyte proliferation was reduced in VCAP1X2 mutants (Fig. 2), we investigated whether pAKT and pERK levels were also affected in the ventricle of VCAP1X2 mutant embryonic hearts.
Both pAKT and pERK were diminished in VCAP1X2 mutant heart compared to WT at 96 hpf (Figs 4A,B and 5A,B). Injection of ΔN or FL mRNA, but not SP, ΔN Y2F or Y2F mRNA restored ventricular pAKT and pERK levels (Figs 4C–G and 5C–G). Western blots further revealed reductions in pAKT (4.2-fold) and pERK (1.8-fold) levels in VCAP1X2 mutant heart (Figs 4H,I, 5H,I, Supplementary Fig. S4). Injection of FL, but not Y2F mRNA restored pAKT and pERK to levels that were comparable to Tg (myl7: EGFP; myl7: H2AFZ mcherry) embryonic hearts at 96 hpf (p < 0.05, Figs 4I, 5I). Notably, Y2F mRNA injection further reduced ventricular pERK level in VCAP1X2 mutant hearts (p < 0.01, Fig. 5I).
VCAP1X2 variant mRNAs were overexpressed in Tg (myl7: EGFP; myl7: H2AFZ mcherry) transgenic embryos to investigate the effects on cardiomyocyte number, and pAKT and pERK levels. Reductions in pAKT, pERK and cardiomyocyte number were identified in ventricles of transgenic embryos overexpressing ΔN Y2F or Y2F, but not ΔN or FL mRNA at 96 hpf (Supplementary Figs S5C–F, S6C–F, S7). Together, these results indicate that the two cytosolic tyrosine residues of VCAP1X2 are essential for maintaining proper ventricular pAKT and pERK levels as well as cardiomyocyte number in embryonic hearts.
VCAP1X2 deficiency and inhibition of pAKT or pERK induced sarcomere disorganization
Since heart contractility was decreased in VCAP1X2 mutants (Fig. 2), we investigated whether sarcomere assembly was affected in the ventricle. Sarcomere organization was examined by transmission electron microscopy, which revealed intact and well-organized structures with clear Z-disc boundaries in WT heart at 96 hpf. By contrast, sparse and irregular myofibrils bordered by blurred Z-discs were observed in VCAP1X2 mutant heart (Fig. 6A). Blurred Z-discs were also detected by α-actinin immunofluorescence in VCAP1X2 mutant heart compared to WT at 96 hpf (Fig. 6B,a,b). Injection of ΔN or FL mRNA, but not SP, ΔN Y2F or Y2F mRNA rescued the blurred Z-discs (Fig. 6B,c–g). We further measured the resting Z-disc length and sarcomere length in the embryonic hearts. Compared to WT, VCAP1X2 mutant heart ventricles exhibited highly significantly reduced Z-disc length (p < 0.001, Fig. 6C). Again, injection of ΔN, FL or hVCAM1 mRNA, but not SP, ΔN Y2F or Y2F mRNA restored the defect (p < 0.001, Fig. 6C). Similar sarcomere lengths were measured in all groups (Fig. 6D). Therefore, we conclude that two cytosolic tyrosine residues of VCAP1X2 are essential in regulating sarcomere organization.
Since sarcomere disorganization and decreased pAKT and pERK were detected in the ventricles of VCAP1X2 mutant hearts (Figs 4–6), we wondered whether inhibition of pAKT or pERK expression in WT embryos would affect sarcomere organization. We treated WT embryos with an inhibitor of PI3K (LY294002) or pMEK(U0126) at 63 hpf and fixed embryos for α-actinin immunofluorescence at 72 hpf. Blurred shortened Z-discs were detected in heart ventricles of Tg (myl7: EGFP; myl7: H2AFZ mcherry) embryos treated with PI3K or pMEK inhibitor compared to DMSO-treated embryos (Fig. 6E,a–c,F,G), but sarcomere length was similar. We also treated VCAP1X2 mutant embryos with pAKT activator (SC-79) for a similar period prior to α-actinin immunostaining. Complete restoration of Z-disc alignment and length were detected in heart ventricles of SC-79-treated VCAP1X2 mutants compared to DMSO-treated controls (Fig. 6E,d,e,F,G). Together, these results demonstrate that two cytosolic tyrosine residues of VCAP1X2 and proper pAKT and pERK levels are essential for sarcomere organization in the heart ventricle.
Two cytosolic tyrosine residues of VCAP1X2 are essential for epicardium formation and epicardial cell proliferation
A prior study revealed VCAM1-deficient mouse hearts exhibited thinned compact layer, reduction of intraventricular septum and epicardium formation defects16. Because VCAP1X2 mutant embryonic heart showed similar defects of thinner ventricular compact layer and reduced cardiomyocyte number (an effect that could be rescued by hVCAM1 mRNA; Figs 2, 3), we next investigated whether epicardium formation was also affected in VCAP1X2 mutants. Indeed, ventricular epicardium formation was affected in VCAP1X2 mutants, as evidenced by tcf21 RNA staining at 96 hpf (Supplementary Fig. S8A). To further investigate the effect of VCAP1X2 deficiency on epicardium formation, we knocked down VCAP1X2 in Tg(tcf21:nucEGFP) transgenic embryos and labeled proliferative epicardial cells with PCNA antibody at 96 hpf (Supplementary Fig. S8B). A reduced number of ventricular Tcf21+ epicardial cells was observed in VCAP1X2 morphants and co-injection of ΔN or FL, but not SP, ΔN Y2F or Y2F mRNA, completely restored Tcf21+ epicardial cell number compared to un-injected or 5 mm spMO-injected transgenic embryos (p < 0.001, Supplementary Fig. S8C). Similarly, a reduction in the percentage of ventricular Tcf21+/PCNA+ cells was completely rescued by co-injection of ΔN or FL, but not SP, ΔN Y2F or Y2F mRNAs. (p < 0.001, Supplementary Fig. S8D). Thus, the two cytosolic tyrosine residues of VCAP1X2 are also required for proper epicardial cell proliferation and epicardium formation.
VCAP1X2 deficiency leads to reduced gene expression for proteins involved in sarcomere assembly and calcium homoeostasis
To further investigate the mechanism by which VCAP1X2 regulates heart contractility and epicardium formation, we compared gene expression levels of proteins involved in sarcomere assembly, calcium homeostasis, and epicardium formation. Embryonic hearts of VCAP1X2 mutant and WT zebrafish were evaluated by RT-qPCR at 96 hpf (Fig. 7). Since VCAP1X2 mutant embryonic heart exhibited a DCM phenotype, we selected several zebrafish orthologues of known human DCM-associated genes for comparison24 (Fig. 7A). Expression of genes encoding sarcomere proteins, such as myosin heavy chain 6 (myh6), myosin heavy chain 7 (myh7), ventricle myosin heavy chain like (vmhcl), troponin T (tnnt2a), myopalladin (mypn) and four and a half LIM domains 2a (fhl2a), was significantly downregulated in VCAP1X2 mutants compared to WT. Expression levels of two natriuretic peptides (nppa, nppb) were substantially increased in VCAP1X2 mutant heart, similar to other DCM animal models25. We then compared expression levels of the sarcomere modulator, smyd1b6, stress response regulators [integrin subunit beta1 binding protein 2 (itgb1bp2)26 and ankyrin repeat domain 1 (ankrd1)27] and chaperones (hsp90aa1.1, unc45b, αB-crystallin/cryabb)28,29 in VCAP1X2 mutant and WT hearts (Fig. 7B). Significant downregulation of smyd1b, and upregulation of hsp90aa1.1 and cryabb were detected in VCAP1X2 mutants.
Because the contraction and relaxation of cardiac muscle is regulated by cytosolic Ca2+ concentration30, we investigated whether expression of calcium regulators was affected in VCAP1X2 mutant hearts (Fig. 7C). Diminished expression of sarcoplasmic reticulum (SR) ryanodine receptor 2b (ryr2b) and sarcolemma solute carrier family 8, member 1a (slc8a1a) were detected in VCAP1X2 mutants. We also compared gene expression for proteins that are essential for epicardium formation, such as GATA4, Tbx5 and BMP431,32. A trend toward downregulation of gata4 and tbx5a was observed in mutants, with no alteration of bmp4 expression (Fig. 7D). Together, these finding suggest that VCAP1X2 regulates cardiac contractility by modulating expression of smyd1b, mypn or fhl2a for proper sarcomere assembly and Ca2+ regulatory proteins (ryr2b, slc8a1a) for calcium homeostasis.
smyd1b mRNA injection restored cardiac dysfunction, sarcomere organization and cardiomyocyte proliferation in VCAP1X2 mutant embryonic heart
We focused on downregulation of smyd1b as a primary effector in VCAP1X2 mutants, because Smyd1-myosin interaction is essential for thick filament assembly, and zebrafish fla/smyd1 mutants exhibit disrupted sarcomere assembly in both heart and skeletal muscle6. In a rescue experiment, injection of smyd1bwt or HMT-deficient smyd1bY247F mRNA, but not myosin-interaction-site-absent smyd1b278del or LacZ mRNA, restored FS, blurred Z-discs and Z-disc length in heart ventricles of VCAP1X2 mutants at 96 hpf (Fig. 8A–D). Furthermore, the reduced cardiomyocyte number and decreased percentage of PCNA+ cardiomyocytes were also restored at 96 hpf in mutant embryos injected with smyd1bwt but not with smyd1b278del mRNA (p < 0.001, Fig. 8E,F). Together, these finding suggest that VCAP1X2 regulates cardiac contractility and cardiomyocyte proliferation by modulating expression of smyd1b, a myosin-interacting protein that is required for proper sarcomere assembly.
The costamere-Z-disc interaction is important for transmission of mechanical force and molecular signals between the sarcomere, sarcolemma and extracellular matrix. Identifying proteins that connect the sarcolemma to sarcomeres can further our understanding of mechanical and signal transduction in cardiac muscle. Here, we identified a novel cell adhesion molecule, VCAP1X2, which is expressed in the sarcolemma and functions as an important signal transducer from the sarcolemma to the sarcomere and nucleus in cardiac muscle. This protein maintains cardiac contractility as well as proliferation of cardiomyocytes and epicardial cells in zebrafish embryonic heart ventricle.
VCAP1X2 deficient embryonic hearts exhibited DCM phenotypes, including enlarged ventricular chamber, thinner ventricular compact layer with decreased cardiomyocyte number, and impaired cardiac function. Sparse and irregular myofilaments, bordered by blurred Z-discs with decreased length, were observed in VCAP1X2 mutant heart and may be attributed to decreased expression of a sarcomere modulator (smyd1b), proteins associated with Z-discs and I-bands (mypn, fhl2a) and calcium regulators (ryr2b, slc8a1a). Smyd1b is expressed in cardiomyocytes, where it localizes to the M-line of the sarcomere and interacts with myosin. Disrupted myofibrillogenesis was previously reported in cardiomyocytes of zebrafish fla mutants with defective smyd1b. Knockdown of smyd1b induced upregulation of two chaperones (hsp90α1 and unc45b) and resulted in cardiac sarcomere disorganization33. Although unc45b expression was not increased in VCAP1X2 mutant heart, cryabb, which binds to titin in the I-band and protects it from stress28, was significantly induced. These upregulated chaperones may be responding to misfolded thick filament myosin during myofibrillogenesis. Furthermore, injection of smyd1bwt but not myosin-interaction-site-absent smyd1b278del mRNA restored FS, Z-disc alignment and Z-disc length in VCAP1X2 mutants, suggesting that Smyd1b-mediated thick filament assembly is regulated by VCAP1X2.
In addition to Smyd1b, Mypn binds α-actinin at the Z-disc and interacts with ANKRD1 in the I-band to maintain sarcomere integrity34. While ANKRD1 is a transcriptional regulator, it also interacts with titin, where it serves as a scaffold to facilitate ERK1/2 phosphorylation of GATA4. Upon GATA4 phosphorylation, the ANKRD1/ERK/GATA4 complex translocates to the nucleus to regulate hypertrophic gene expression27. GATA4 is known to regulate cardiac muscle-specific gene (α-MHC, β-MHC and MLC) expression35 and promoted cardiomyocyte proliferation by modulating cardiomyocyte cell cycle activity36. In VCAP1X2 mutant embryonic heart, decreased smyd1b expression affected thick filament assembly and sarcomere organization and downregulation of mypn may disrupt the interaction between Mypn and Ankrd1a to affect Ankrd1a/ERK/GATA4 complex formation and decrease ERK phosphorylation. Reduced GATA4 phosphorylation would affect its transcriptional activity, resulting in decreased expression of thick filament genes (myh6, myh7, vmhcl) and reduced expression of cell cycle genes such as cyclin D2, cyclin A2 and CDK437. As a result, reduced cardiomyocyte proliferation was detected in ventricles of VCAP1X2 mutant hearts, which was restored by injection of smyd1bwt but not smyd1b278del mRNA. In addition, GATA4 was reported to be an essential factor for the production of proepicardium in mice, and decreased GATA4 transcriptional activity may also affect the epicardium formation in VCAP1X2 mutant zebrafish31.
Besides the effects on cardiac muscle or cell cycle gene expression by reduced ERK-mediated GATA4 transcriptional activity, PI3K-AKT signaling was also shown to promote expression of genes involved in cardiac structure and Z-disc signaling38. Experiments using caPI3K and dnPI3K in transgenic mice indicated that genes encoding several cardiac muscle-related proteins, including Melusin (Itgb1bp2), were differentially expressed. Melusin was shown to associate with the cytosolic domain of integrin β1 at the costamere and interact with the p85 regulatory subunit of PI3K. Thinner myofibrils and reduced Z-disc alignment were identified in dnPI3K transgenic heart, and PI3K (p110α) inhibitors prevented the formation of mature myofibers in response to IGF1 stimulation. Therefore, upon IGF1 stimulation, PI3K interacts with insulin receptor substrate 1 and Melusin to activate downstream AKT kinase and regulate expression of several cardiac muscle-associated proteins. Consistent with these studies in mice, we also detected blurred Z-discs with reduced length in heart ventricles of zebrafish embryos treated with inhibitor of PI3K or pMEK. Furthermore, such defects in ventricles of VCAP1X2 mutant hearts could be completely restored by treatment with PI3K activator. The PI3K-AKT pathway interacts with Wnt/β-catenin or Hippo-Yap signaling to regulate cardiomyocyte proliferation36. PI3K-AKT promoted β-catenin nuclear translocation by inhibiting GSK3β phosphorylation and induced Yap nuclear translocation by dissociation of the Mst-Sav-LATS1/2 complex and the inactivation of LATS1/2 via recruiting PDK1 to the plasma membrane39. β-catenin and Yap then interact with cofactors to activate expression of cell cycle genes.
Mammalian VCAM1 isoforms contain a highly conserved 19-amino-acid cytoplasmic domain that is not found in VCAP1X2 or zebrafish VCAM1 (Supplementary Fig. S2B). Within the 19-amino-acid sequence, S728 and Y729 are necessary for mouse VCAM1 activation of calcium flux, while S730 and S737 are required for VCAM1 activation of Rac1 during leukocyte transendothelial migration40. Similarly, VCAP1X2 possesses a short 30-amino-acid cytoplasmic domain with two tyrosine residues. We demonstrated that these two cytosolic tyrosine residues are essential for VCAP1X2 function. Although the VCAP1X2 antibody was not suitable for immunoprecipitation experiments to confirm phosphorylation of the cytosolic tyrosines, we did mutate the residues to phenylalanines to block auto- or trans-phosphorylation.
VCAM1-deficient mouse embryonic hearts displayed thinned compact layer, reduction of intraventricular septum and epicardium formation defects, and the interaction between VCAM1 and integrin α4 was shown to be important for heart development16,17. Our finding that VCAP1X2 mutant heart exhibits a thin ventricular compact layer and epicardium formation defect is similar to the phenotype in VCAM1-deficient mice. Our data further suggested that decreased pAKT and pERK levels in VCAP1X2 mutant heart caused sarcomere disorganization and reduction in cardiomyocyte and epicardial cell proliferation, which then resulted in morphologic and functional defects. In contrast, zebrafish VCAM1 was shown to interact with Integrin α9 rather than Integrin α4 and was essential for lymphatic development41.
Tyrosine phosphorylation was shown to be crucial during myofibrillogenesis and embryonic heart development in axolotls42. Our results further provide a potential molecular mechanism to explain how tyrosine phosphorylation may affect sarcomere structure and function. Like PECAM1, VCAP1X2 may possesses a nonconventional ITAM motif with two tyrosine residues in the cytoplasmic domain that may provide a scaffold for adaptor molecule signaling43. SH2-containing phosphatase may bind to phosphorylated tyrosine together with Gab1 to recruit PI3K docking or activate Ras-Raf-MEK-ERK44. Alternatively, SH2-containing PI3K p85 may directly bind to phosphorylated tyrosine45. Activated AKT kinase would then regulate expression of sarcomeric modulators (smyd1b, mypn, fhl2a) for proper sarcomere assembly that in turn enhances GATA4 activity by promoting translocation of Ankrd1a/ERK/GATA4 complex to the nucleus and ERK-GATA4 signal transduction, driving cardiomyocyte and epicardium proliferation31. In addition to regulating cardiomyocyte and epicardium proliferation, activated AKT also regulates Ca2+ cycling by modulating expression of slc8a1a and ryr2b and activity of the Atp2a2a calcium pump for proper cardiac contractility (Supplementary Fig. S9). A previous study showing that VCAM1 activated PI3K and ERK kinase in human airway smooth muscle cells46 provides additional support for our finding that VCAP1X2 enhances pAKT and pERK levels.
In conclusion, using a zebrafish gene trap mutant, we demonstrate that VCAP1X2, a cell adhesion molecule, localizes to the myocardium sarcolemma, where it regulates cardiac contractility, and proliferation of cardiomyocytes and epicardial cells by modulating pAKT and pERK expression via two tyrosine residues within the cytoplasmic domain. These results advance our understanding of signal transmission between the sarcolemma and sarcomeres in cardiac muscle cells.
Zebrafish strains and maintenance
Adult zebrafish, including wild-type ASAB, Tg (myl7: EGFP; myl7: H2AFZ mcherry)cy1, Tg(tcf21:NTR; tcf21:nucEGFP), Tg(tcf21:nucEGFP), Tg(fli1:GFP) and VCAP1X2 mutant, were maintained in 20 L aquaria supplied with filtered fresh water and aeration, or high density, self-circulation systems (Aqua Blue) with a photoperiod of 14 h light/10 h dark cycle. Embryos were raised at 28.5 °C and embryonic stages were determined by morphological criteria defined as described47. All animal procedures were approved by the Academia Sinica Institutional Animal Care & Utilization Committee (AS IACUC) (protocol # 12-12-482). All experimental methods were performed in accordance with the approved guideline.
Generation of VCAP1X2 mutant line
The Tol2 transposon based SAGVG (splice acceptor-Gal4-VP16;UAS:eGFP) plasmid DNA and Tol2 transposase RNA were injected into ASAB 1–2 cell zygotes19,48,49. Injected F0 embryos were raised to adulthood and outcrossed with WT to produce F1. F1 Embryos that exhibited intense heart-specific EGFP expression by fluorescence microscopy were raised to adulthood and the F2 generation was produced. Subsequent ligation mediated-PCR showed that the Gal4-VP16; UAS:EGFP vector had been inserted into intron 1 of the VCAP1X2 gene. To determine whether the splicing of VCAP1X2 exon 1 and exon 2 was disrupted, RT-PCR was conducted using forward (5′- GATCTACAACAGTGACAGGACA-3′) and reverse (5′-TGTAAATAGTGACTGGGAGAG-3′) primers, and total RNA isolated at 24 and 48 hpf. Homozygous VCAP1X2 mutant embryos are able to survive to adulthood. Homozygous populations of adult female and male VCAP1X2 mutants were maintained.
Morpholino and mRNA injection and RNA synthesis
Individual MOs (3.3 ng/embryo) were diluted with DEPC dH2O and injected into randomly selected 1–2 cell zygotes using an IM-300 microinjector (NARISHIGE). For rescue experiments, 33 pg of individual capped VCAP1X2 mRNA variants together with or without VCAP1X2 spMO were injected into randomly selected 1–2 cell zygotes of Tg (myl7: EGFP; myl7: H2AFZ mcherry) transgenic fish or VCAP1X2 mutant. For smyd1b rescue experiments, 50 pg of capped smyd1bwt, smyd1bY247F or smyd1b278del mRNA was injected into randomly selected 1–2 cell zygotes of VCAP1X2 homozygous mutant embryos. For overexpression, 66 pg of individual capped VCAP1X2 mRNA variants were injected into randomly selected 1–2 cell zygotes of Tg (myl7: EGFP; myl7: H2AFZ mcherry) transgenic fish.
To synthesize sense mRNA for rescue experiments, VCAP1X2 variants in the T7TS vector were linearized by SalI digestion and used as template to synthesize capped mRNA by the T7 mMESSAGE mMACHINE Kit (Thermo Fisher Scientific). Myc-DDK tagged Homo sapiens vascular cell adhesion molecule 1 (hVCAM1), transcript variant 3 (NM_001199834) plasmid (OriGene Technologies) was linearized by FseI digestion and used to synthesize hVCAM1 mRNA by a T7 mMESSAGE mMACHINE Kit. To synthesize sense RNA for rescue experiment, smyd1bwt, smyd1bY247F or smyd1b278del mRNA were generated by a SP6 mMESSAGE mMACHINE Kit by digestion with NotI. To synthesize RNA probes for RNA in situ hybridization, plasmids were linearized by restriction enzymes and antisense RNA probes were produced as follows (restriction sites and promoters in parentheses): myl7 (NcoI/sp6), tcf21(NcoI/sp6) and VCAP1X2 (NcoI/sp6).
Analysis of heart function
72 hpf Zebrafish larvae were anesthetized with a mixture of tricaine (100 ppm) and isoflurane (100 ppm). Green fluorescent images of Tg(myl7: EGFP; myl7: H2AFZ mcherry) and VCAP1X2 mutant hearts were captured as described20. Pseudodynamic 3D imaging of the contracting heart was used to measure cardiac function, and the images were constructed and displayed with commercial software (Imaris, Bitplane). End-diastolic volume (EDV) and end-systolic volume (ESV) correspond to the ventricular volume immediately before a contraction and at the end of the contraction, respectively. Stroke volume (SV), the volume of blood ejected in a single heartbeat, was determined from the equation: SV = EDV - ESV, and ejection fraction (EF) was defined as EF = SV/EDV. The FS of zebrafish embryonic hearts was determined as described50. In brief, heart beating was recorded at 37 frames per second for 30 s with a camera (AxioCam HRC) installed on a microscope (Zeiss Imager M1). Videos were converted to frames and the ventricular widths and lengths were measured from single frames showing the maximum ventricular systole (VS) and ventricular diastole (VD) by Image J software. FS (%) of the ventricle was calculated as (Width VD − Width VS )/Width VD × 100%.
Immunofluorescence staining of zebrafish embryos was performed as described with some modifications51,52. For immunostaining of α-actinin, pERK or pAKT, antigen retrieval was conducted by incubating embryos with 150 mM Tris-HCl (pH 9.0) for 15 min at 70 °C. Embryos were then permeabilized by 10 μg/mL proteinase K for 20 min at 28 °C and acetone for 20 min at −20 °C. Embryonic hearts were then dissected using two 22-gauge needles and transferred to slides. Tissues were blocked with 10% serum for 1 h at room temperature (RT), followed by incubation with primary antibody including α-actinin (1:100, Sigma-Aldrich), pAKT (1:100, Cell Signaling) or pERK (1:100, Cell Signaling) at 4 °C overnight. For immunostaining of VCAP1X2 (1:100), embryos were permeabilized in acetone for 7 min at −20 °C and 2% Triton-X100 in PBS for 1.5 h at RT, followed by antigen retrieval in 150 mM Tris-HCl (pH 9.0) for 15 min at 70 °C. For immunostaining of MEF2 or PCNA, embryos were permeabilized by 2% Triton-X100 in PBS for 1.5 h at RT, followed by antigen retrieval in 150 mM Tris-HCl (pH 9.0) for 15 min at 70 °C. Embryonic hearts were then dissected and blocked with 10% serum. Tissues were incubated with primary antibody including MEF2 (1:150, Santa Cruz Biotechnology) or PCNA (1:200, Abcam) at 4 °C overnight. After several PBST washes, embryonic hearts were incubated with goat anti-rabbit Alexa Fluor 647 (1:200, Invitrogen) or mouse Alexa Fluor 568 (1:200, Invitrogen) and later were mounted in ProLong Gold Antifade Mountant with DAPI (Invitrogen). To conduct immunostaining of VCAP1X2 on Tg(fli1:GFP) transgenic embryos, fixed Tg(fli1:GFP) embryos were permeablized with 2% Triton-X100 in PBS for 1.5 h at RT, followed by antigen retrieval in 150 mM Tris-HCl (pH 9.0) for 15 min at 70 °C and then acetone treatment for 8 min at −20 °C. After several PBST washes, embryos were blocked with 10% serum for 1 h at RT and incubated with anti-VCAP1X2 antibody (1:100) at 4 °C overnight. After several PBST washes, embryos were incubated with goat anti-rabbit Alexa Fluor 647 (1:200, Invitrogen).
BrdU incorporation was conducted as described53. 70.5 hpf embryos were incubated with freshly diluted BrdU (final 10 mM, Sigma-Aldrich) in 15% dimethylsulfoxide on ice for 20 min, and then incubated in egg water at 28.5 °C for 1 h after washes. Samples were fixed in 4% PFA at 72 hpf at 4 °C overnight. Fixed embryos were dehydrated and rehydrated through a methanol/PBST series, followed by permeabilization with 10 µg/mL proteinase K for 30 min and exposure of the BrdU epitope with 2 N HCl for 1 h. Embryos were blocked with 10% serum at 4 °C overnight and incubated with BrdU antibody (1:7, BD Biosciences) at 28 °C for 2 h. After incubation with goat anti mouse Alexa Fluor 647 (1:200, Invitrogen) for 1 h at RT, embryos were washed with PBST before imaging.
Embryos were fixed at 72 hpf in 4% PFA at 4 °C overnight. Fixed embryos were dehydrated and rehydrated through a methanol/PBST series, followed by permeabilization with 10 µg/mL proteinase K for 30 min and refixation in 4% PFA for 20 min at RT. Apoptotic cells were labeled by TdT-mediated dUTP nick-end labeling (TUNEL) assay using in situ cell death detection kit (Roche), according to the manufacturer’s instructions. After washing with PBST, embryos were blocked with 5% serum in PBST for 2 h at RT, and incubated with anti-fluorescein-POD antibody (1:1000, Roche) and anti-MEF2 antibody (1:150, Santa Cruz Biotechnology) overnight at 4 °C. After several PBST washes, apoptosis signal was enhanced by TSA Plus Cyanine 5 and Flourescein System (Perkin Elmer), then embryos were incubated with goat anti-rabbit Alexa Fluor 568 (1:200, Invitrogen) for 3 h at RT. Later, embryonic hearts were dissected using two 22-gauge needles and transferred to slides, where the tissues were mounted in ProLong Gold Antifade Mountant with DAPI (Invitrogen). Both DNase I-treated positive control and negative control, without TdT enzyme, were included.
Embryonic heart purification and Western blot analyses
Embryonic heart purification was performed as described54. Hearts of 96 hpf Tg(myl7: EGFP; myl7: H2AFZ mcherry) transgenic embryos were isolated by fluid force driven by syringe with a 19-gauge needle and filtered through 100 μm and 40 μm cell strainer (Corning). Isolated hearts were washed with L-15 medium containing 10% charcoal-stripped FBS and examined under a fluorescent microscope for intactness. For Western blot analyses, isolated embryonic hearts were extracted by homogenization in 1X RIPA lysis buffer (Millipore) containing 1X Protease Inhibitor Cocktail tablet (Roche), 1X Phosphatase Inhibitor Cocktail 2 and 3 (Sigma-Aldrich) and 1 mM PMSF (Sigma-Aldrich). Protein concentration was determined by Bradford assay. Total protein was separated by electrophoresis on a 11% SDS-PAGE and transferred to Immobilon®-P Transfer Polyvinylidene difluoride (PVDF) Membranes were immunoblotted with total AKT (1:1000), phospho-AKT (1:2000), total ERK (1:2000) or phospho-ERK T202/Y204 (1:2000) antibody from Cell Signaling company, or β-actin (1:10000, Millipore) antibody. Peroxidase-conjugated goat anti-rabbit immunoglobulin G (IgG) (1:5000, Jackson) and peroxidase-conjugated goat anti-mouse IgG (1:5000, Jackson) were used as secondary antibodies. Immuno-reactive signals were detected by Luminata Crescendo Western HRP Substrate (Millipore). Protein concentrations from visualized bands were quantified by Image J.
Electron Microscopy and sarcomere length quantification
Transmission electron microscopy analysis was conducted as described55. For sarcomere length quantification, embryos were incubated with relaxation buffer (20 mM imidazole, 5 mM EGTA, 7 mM MgCl2, 5 mM creatine phosphate (Sigma-Aldrich), 10 mM ATP (Sigma-Aldrich), 100 mM KCl) for 1.5 h before fixation56. Z-discs were labelled with anti α-actinin antibody. Sarcomere length and Z-disc length were measured according to immunofluorescent α-actinin signal by Image J.
Treatment of inhibitors or activator of PI3K or pMEK
To inhibit PI3K-AKT or MAPK-ERK1/2 signaling, Tg(myl7: EGFP; myl7: H2AFZ mcherry) embryos were treated with 30 μM PI3K inhibitor (LY294002, Calbiochem) or 40 μM pMEK inhibitor (U0126, Promega) in egg water (final 0.1% DMSO) from 63 to 70.5 hpf, respctively. To activate pAKT, VCAP1X2 mutant embryos were treated with 1 μM SC-79 (Sigma) in egg water (final 0.1% DMSO) from 63 to 70.5 hpf. These embryos were then replaced with relaxation buffer (final 0.1% DMSO) containing the same amount of inhibitors or activators for 1.5 h before fixation at 72 hpf in 4% PFA overnight at 4 °C. Control embryos were incubated with egg water or relaxation buffer with 0.1% DMSO.
Microscopy and statistics
Bright-field images of embryos were taken using a camera (AxioCam HRC) installed on a microscope (Zeiss Imager M1) in DIC mode. Immunofluorescence stained embryos or hearts were observed with a confocal microscope (Leica TCS-SP5-MP). Ultrastructural sections were examined using an electron microscope (FEI Tecnai G2 TF20 Super TWIN, Thermo Fisher Scientific). The total number of embryos analyzed (biological replicates) was reported as the sample number for each experiment. Biological replicates were pooled from at least three separate experiments, conducted independently. For RT-qPCR analysis, each biological replicate was analyzed with three technical replications. Two-tailed Student’s t-test with unequal variance was performed in Microsoft Excel. ANOVA with post hoc multiple comparisons (Bonferroni method) was performed in R57.
Plasmid construction, morpholino sequence, VCAP1X2 antibody generation and purification, histological analyses, RNA in situ hybridization, reverse transcription polymerase chain reaction (RT-PCR) and quantitative real-time reverse-transcription PCR (RT-qPCR) were performed as described in the Supplementary Information.
All data analyzed during this study are included in this published article (and its Supplementary Information files).
Hwang, P. M. & Sykes, B. D. Targeting the sarcomere to correct muscle function. Nat Rev Drug Discov 14, 313–328, https://doi.org/10.1038/nrd4554 (2015).
McNally, E. M., Golbus, J. R. & Puckelwartz, M. J. Genetic mutations and mechanisms in dilated cardiomyopathy. J Clin Invest 123, 19–26, https://doi.org/10.1172/JCI62862 (2013).
Hershberger, R. E., Hedges, D. J. & Morales, A. Dilated cardiomyopathy: the complexity of a diverse genetic architecture. Nat Rev Cardiol 10, 531–547, https://doi.org/10.1038/nrcardio.2013.105 (2013).
Willis, M. S., Schisler, J. C., Portbury, A. L. & Patterson, C. Build it up-Tear it down: protein quality control in the cardiac sarcomere. Cardiovasc Res 81, 439–448, https://doi.org/10.1093/cvr/cvn289 (2009).
Gottlieb, P. D. et al. Bop encodes a muscle-restricted protein containing MYND and SET domains and is essential for cardiac differentiation and morphogenesis. Nat Genet 31, 25–32, https://doi.org/10.1038/ng866 (2002).
Just, S. et al. The myosin-interacting protein SMYD1 is essential for sarcomere organization. J Cell Sci 124, 3127–3136, https://doi.org/10.1242/jcs.084772 (2011).
Frank, D. & Frey, N. Cardiac Z-disc signaling network. J Biol Chem 286, 9897–9904, https://doi.org/10.1074/jbc.R110.174268 (2011).
Knoll, R., Buyandelger, B. & Lab, M. The sarcomeric Z-disc and Z-discopathies. J Biomed Biotechnol 2011, 569628, https://doi.org/10.1155/2011/569628 (2011).
Peter, A. K., Cheng, H., Ross, R. S., Knowlton, K. U. & Chen, J. The costamere bridges sarcomeres to the sarcolemma in striated muscle. Prog Pediatr Cardiol 31, 83–88, https://doi.org/10.1016/j.ppedcard.2011.02.003 (2011).
Lapidos, K. A., Kakkar, R. & McNally, E. M. The dystrophin glycoprotein complex: signaling strength and integrity for the sarcolemma. Circ Res 94, 1023–1031, https://doi.org/10.1161/01.RES.0000126574.61061.25 (2004).
Hannigan, G. E., Coles, J. G. & Dedhar, S. Integrin-linked kinase at the heart of cardiac contractility, repair, and disease. Circ Res 100, 1408–1414, https://doi.org/10.1161/01.RES.0000265233.40455.62 (2007).
Bakkers, J. Zebrafish as a model to study cardiac development and human cardiac disease. Cardiovasc Res 91, 279–288, https://doi.org/10.1093/cvr/cvr098 (2011).
Poon, K. L. & Brand, T. The zebrafish model system in cardiovascular research: A tiny fish with mighty prospects. Glob Cardiol Sci Pract 2013, 9–28, https://doi.org/10.5339/gcsp.2013.4 (2013).
Ley, K., Laudanna, C., Cybulsky, M. I. & Nourshargh, S. Getting to the site of inflammation: the leukocyte adhesion cascade updated. Nat Rev Immunol 7, 678–689, https://doi.org/10.1038/nri2156 (2007).
Muller, W. A. Mechanisms of leukocyte transendothelial migration. Annu Rev Pathol 6, 323–344, https://doi.org/10.1146/annurev-pathol-011110-130224 (2011).
Kwee, L. et al. Defective development of the embryonic and extraembryonic circulatory systems in vascular cell adhesion molecule (VCAM-1) deficient mice. Development 121, 489–503 (1995).
Yang, J. T., Rayburn, H. & Hynes, R. O. Cell adhesion events mediated by alpha 4 integrins are essential in placental and cardiac development. Development 121, 549–560 (1995).
Uosaki, H. et al. Efficient and scalable purification of cardiomyocytes from human embryonic and induced pluripotent stem cells by VCAM1 surface expression. PLoS One 6, e23657, https://doi.org/10.1371/journal.pone.0023657 (2011).
Davison, J. M. et al. Transactivation from Gal4-VP16 transgenic insertions for tissue-specific cell labeling and ablation in zebrafish. Dev Biol 304, 811–824, https://doi.org/10.1016/j.ydbio.2007.01.033 (2007).
Lin, K. Y., Chang, W. T., Lai, Y. C. & Liau, I. Toward functional screening of cardioactive and cardiotoxic drugs with zebrafish in vivo using pseudodynamic three-dimensional imaging. Anal Chem 86, 2213–2220, https://doi.org/10.1021/ac403877h (2014).
Osawa, M., Masuda, M., Kusano, K. & Fujiwara, K. Evidence for a role of platelet endothelial cell adhesion molecule-1 in endothelial cell mechanosignal transduction: is it a mechanoresponsive molecule? J Cell Biol 158, 773–785, https://doi.org/10.1083/jcb.200205049 (2002).
Odiete, O., Hill, M. F. & Sawyer, D. B. Neuregulin in cardiovascular development and disease. Circ Res 111, 1376–1385, https://doi.org/10.1161/CIRCRESAHA.112.267286 (2012).
Huang, Y. et al. Igf Signaling Is Required for Cardiomyocyte Proliferation during Zebrafish Heart Development and Regeneration. Plos One 8, https://doi.org/10.1371/journal.pone.0067266 (2013).
Shih, Y. H. et al. Cardiac transcriptome and dilated cardiomyopathy genes in zebrafish. Circ Cardiovasc Genet 8, 261–269, https://doi.org/10.1161/CIRCGENETICS.114.000702 (2015).
Rangrez, A. Y. et al. MicroRNA miR-301a is a novel cardiac regulator of Cofilin-2. PLoS One 12, e0183901, https://doi.org/10.1371/journal.pone.0183901 (2017).
Brancaccio, M. et al. Melusin, a muscle-specific integrin beta1-interacting protein, is required to prevent cardiac failure in response to chronic pressure overload. Nat Med 9, 68–75, https://doi.org/10.1038/nm805 (2003).
Zhong, L. et al. Targeted inhibition of ANKRD1 disrupts sarcomeric ERK-GATA4 signal transduction and abrogates phenylephrine-induced cardiomyocyte hypertrophy. Cardiovasc Res 106, 261–271, https://doi.org/10.1093/cvr/cvv108 (2015).
Bullard, B. et al. Association of the chaperone alphaB-crystallin with titin in heart muscle. J Biol Chem 279, 7917–7924, https://doi.org/10.1074/jbc.M307473200 (2004).
Myhre, J. L. & Pilgrim, D. B. At the Start of the Sarcomere: A Previously Unrecognized Role for Myosin Chaperones and Associated Proteins during Early Myofibrillogenesis. Biochem Res Int 2012, 712315, https://doi.org/10.1155/2012/712315 (2012).
Gerilechaogetu, F. et al. Current concepts in the role of mechanosensing in the regulation of cardiac contractile function. Austin. J. Clin. Med 1, 11015 (2014).
Watt, A. J., Battle, M. A., Li, J. & Duncan, S. A. GATA4 is essential for formation of the proepicardium and regulates cardiogenesis. Proc Natl Acad Sci USA 101, 12573–12578, https://doi.org/10.1073/pnas.0400752101 (2004).
Liu, J. & Stainier, D. Y. Tbx5 and Bmp signaling are essential for proepicardium specification in zebrafish. Circ Res 106, 1818–1828, https://doi.org/10.1161/CIRCRESAHA.110.217950 (2010).
Li, H. et al. Smyd1b is required for skeletal and cardiac muscle function in zebrafish. Mol Biol Cell 24, 3511–3521, https://doi.org/10.1091/mbc.E13-06-0352 (2013).
Bang, M. L. et al. Myopalladin, a novel 145-kilodalton sarcomeric protein with multiple roles in Z-disc and I-band protein assemblies. J Cell Biol 153, 413–427 (2001).
Molkentin, J. D., Kalvakolanu, D. V. & Markham, B. E. Transcription factor GATA-4 regulates cardiac muscle-specific expression of the alpha-myosin heavy-chain gene. Mol Cell Biol 14, 4947–4957 (1994).
Ponnusamy, M., Li, P. F. & Wang, K. Understanding cardiomyocyte proliferation: an insight into cell cycle activity. Cell Mol Life Sci 74, 1019–1034, https://doi.org/10.1007/s00018-016-2375-y (2017).
Rojas, A. et al. GATA4 is a direct transcriptional activator of cyclin D2 and Cdk4 and is required for cardiomyocyte proliferation in anterior heart field-derived myocardium. Mol Cell Biol 28, 5420–5431, https://doi.org/10.1128/MCB.00717-08 (2008).
Waardenberg, A. J. et al. Phosphoinositide 3-kinase (PI3K(p110alpha)) directly regulates key components of the Z-disc and cardiac structure. J Biol Chem 286, 30837–30846, https://doi.org/10.1074/jbc.M111.271684 (2011).
Fan, R., Kim, N. G. & Gumbiner, B. M. Regulation of Hippo pathway by mitogenic growth factors via phosphoinositide 3-kinase and phosphoinositide-dependent kinase-1. Proc Natl Acad Sci USA 110, 2569–2574, https://doi.org/10.1073/pnas.1216462110 (2013).
Marchese, M. E., Berdnikovs, S. & Cook-Mills, J. M. Distinct sites within the vascular cell adhesion molecule-1 (VCAM-1) cytoplasmic domain regulate VCAM-1 activation of calcium fluxes versus Rac1 during leukocyte transendothelial migration. Biochemistry 51, 8235–8246, https://doi.org/10.1021/bi300925r (2012).
Yang, Y. et al. Cell Adhesion Mediated by VCAM-ITGalpha9 Interactions Enables Lymphatic Development. Arterioscler Thromb Vasc Biol 35, 1179–1189, https://doi.org/10.1161/ATVBAHA.114.304997 (2015).
Meng, F. et al. Relationship between cardiac protein tyrosine phosphorylation and myofibrillogenesis during axolotl heart development. Tissue Cell 35, 133–142 (2003).
Ilan, N. & Madri, J. A. PECAM-1: old friend, new partners. Curr Opin Cell Biol 15, 515–524 (2003).
Kiyatkin, A. et al. Scaffolding protein Grb2-associated binder 1 sustains epidermal growth factor-induced mitogenic and survival signaling by multiple positive feedback loops. J Biol Chem 281, 19925–19938, https://doi.org/10.1074/jbc.M600482200 (2006).
Pawson, T. Protein modules and signalling networks. Nature 373, 573–580, https://doi.org/10.1038/373573a0 (1995).
Lazaar, A. L., Krymskaya, V. P. & Das, S. K. VCAM-1 activates phosphatidylinositol 3-kinase and induces p120Cbl phosphorylation in human airway smooth muscle cells. J Immunol 166, 155–161 (2001).
Kimmel, C. B., Ballard, W. W., Kimmel, S. R., Ullmann, B. & Schilling, T. F. Stages of embryonic development of the zebrafish. Dev Dyn 203, 253–310, https://doi.org/10.1002/aja.1002030302 (1995).
Kawakami, K. et al. A transposon-mediated gene trap approach identifies developmentally regulated genes in zebrafish. Dev Cell 7, 133–144, https://doi.org/10.1016/j.devcel.2004.06.005 (2004).
Urasaki, A., Morvan, G. & Kawakami, K. Functional dissection of the Tol2 transposable element identified the minimal cis-sequence and a highly repetitive sequence in the subterminal region essential for transposition. Genetics 174, 639–649, https://doi.org/10.1534/genetics.106.060244 (2006).
Hoage, T., Ding, Y. & Xu, X. Quantifying cardiac functions in embryonic and adult zebrafish. Methods Mol Biol 843, 11–20, https://doi.org/10.1007/978-1-61779-523-7_2 (2012).
Inoue, D. & Wittbrodt, J. One for all–a highly efficient and versatile method for fluorescent immunostaining in fish embryos. PLoS One 6, e19713, https://doi.org/10.1371/journal.pone.0019713 (2011).
Yang, J. & Xu, X. Immunostaining of dissected zebrafish embryonic heart. J Vis Exp, e3510, https://doi.org/10.3791/3510 (2012).
Verduzco, D. & Amatruda, J. F. Analysis of cell proliferation, senescence, and cell death in zebrafish embryos. Methods Cell Biol 101, 19–38, https://doi.org/10.1016/B978-0-12-387036-0.00002-5 (2011).
Burns, C. G. & MacRae, C. A. Purification of hearts from zebrafish embryos. Biotechniques 40, 274, 276, 278 passim (2006).
Chen, Y. H. et al. Zebrafish cdx1b regulates differentiation of various intestinal cell lineages. Dev Dyn 238, 1021–1032, https://doi.org/10.1002/dvdy.21908 (2009).
Chen, Z. et al. Depletion of zebrafish essential and regulatory myosin light chains reduces cardiac function through distinct mechanisms. Cardiovasc Res 79, 97–108, https://doi.org/10.1093/cvr/cvn073 (2008).
R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/ (2017).
We thank Professor Michael J. Parsons for providing the Tol2 transposon based SAGVG (splice acceptor-Gal4-VP16;UAS:eGFP) plasmid and ligation mediated-PCR protocols, Professor Steffen Just for providing various smyd1b variant constructs, Professor Yu-Ting Yan for his valuable comments, Mr. Zong-Han Song for his assistance in Western blot analyses and Ms. Mei-Jen Chen for her maintenance of fish stocks. The authors thank Professor Kenneth D. Poss and the Taiwan Zebrafish Core Facility at Academia Sinica (TZCAS) for providing Tg(tcf21:NTR; tcf21:nucEGFP) and Tg(tcf21:nucEGFP), TZCAS for providing Tg (myl7: EGFP; myl7: H2AFZ mcherry), and the Taiwan Zebrafish Core Facility at NHRI (TZeNH) for providing Tg(fli1:GFP) transgenic fish. We also thank the Core Facility at ICOB, Academia Sinica for assistance with transmission electron microscopy, RT-qPCR analyses and confocal imaging. This work was supported by the Ministry of Science and Technology, Taiwan [NSC 102-2628-8-001-001-MY3 and MOST 105-2313-B-001-005-MY3 to S-P. L. H.].
The authors declare no competing interests.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.