Abstract
Unlike mammals, zebrafish can regenerate injured hearts even in the adult stage. Cardiac regeneration requires the coordination of cardiomyocyte (CM) proliferation and migration. The TGF-β/Smad3 signaling pathway has been implicated in cardiac regeneration, but the molecular mechanisms by which this pathway regulates CM proliferation and migration have not been fully illustrated. Here, we investigated the function of TGF-β/Smad3 signaling in a zebrafish model of ventricular ablation. Multiple components of this pathway were upregulated/activated after injury. Utilizing a specific inhibitor of Smad3, we detected an increased ratio of unrecovered hearts. Transcriptomic analysis suggested that the TGF-β/Smad3 signaling pathway could affect CM proliferation and migration. Further analysis demonstrated that the CM cell cycle was disrupted and the epithelial–mesenchymal transition (EMT)-like response was impaired, which limited cardiac regeneration. Altogether, our study reveals an important function of TGF-β/Smad3 signaling in CM cell cycle progression and EMT process during zebrafish ventricle regeneration.
Introduction
Myocardial infarction (MI) is a major cause of mortality worldwide (). Adult mammalian hearts cannot replenish the large number of cardiomyocytes (CMs) that are lost due to MI, and the loss of CMs eventually leads to heart failure (). In contrast, zebrafish can effectively regenerate injured hearts through the dedifferentiation, proliferation, and migration of pre-existing CMs (; ; ).
Cardiomyocyte proliferation after injury is usually accompanied by the disassembly of sarcomere (; ) and the re-expression of cardiogenic genes, such as nkx2.5, gata4, and hand2, which suggests the occurrence of CM dedifferentiation (). Mammalian CMs retain the ability to proliferate only for a short period after birth (; ; ; ), and the exact mechanism by which this ability is lost is still unclear (; ). Cell migration is also necessary for cardiac regeneration (), and ventricular and atrial CMs can migrate into and replenish injured areas (; ). Epicardial cells can also enter the myocardial layer through the process of epithelial–mesenchymal transition (EMT), during which the expression of EMT marker genes, such as snail, twist, and vimentin, are upregulated (; ). However, the regulation of cell migration and how it coordinates with other cellular processes, such as cell proliferation, during cardiac regeneration remain largely unknown.
The TGF-β signaling pathway regulates the differentiation, proliferation, and migration of a variety of cell types during cardiovascular development () and is also vital for cardiac regeneration (; ). The expression of tgfb1, tgfb2, and tgfb3 is upregulated in injured hearts (). Activated by integrins, proteases, or matrix proteins (), latent Tgfβ proteins in the extracellular matrix are released from complexes, bind to the Tgfbr2 and Tgfbr1 receptors on cell membranes, and then activate downstream Smad-dependent or Smad-independent signaling pathways (; ; ). The TGF-β/Smad-dependent signaling pathway is activated in a model of MI (), and Smad3 and Smad2 exert opposite effects in multiple cellular processes (; ). The mechanism by which the TGF-β/Smad3 signaling specifically affects CM proliferation and migration warrants detailed investigation.
In this study, we used a zebrafish model of ventricular ablation to determine the function of the TGF-β/Smad3 signaling pathway in cardiac regeneration. Multiple components of this pathway were upregulated/activated after injury. Treatment with SIS3, a specific inhibitor of Smad3, increased the ratio of unrecovered hearts. RNA-seq data suggested that the TGF-β/Smad3 signaling pathway was involved in multiple processes during regeneration. Further analysis revealed that the cell cycle progression and EMT-like response of CMs were impaired upon inhibition of Smad3, which led to the failure of cardiac regeneration. In conclusion, our study provides novel insights into the role of the TGF-β/Smad3 signaling pathway in regulating cardiac regeneration.
Results
TGF-β/Smad3 Signaling Is Activated During Zebrafish Ventricular Regeneration
We first examined the involvement of the TGF-β/Smad3 signaling pathway in a model of larval ventricular regeneration using the transgenic line Tg(vmhc:mCherry-NTR), which exhibits injury and death of ventricular CMs after ablation with metronidazole (MTZ) treatment at 3 days post-fertilization (dpf) (). Whole-mount in situ hybridization (WISH) showed that multiple TGF-β ligands, including tgfb1a, tgfb1b, tgfb2, and tgfb3, were strongly expressed in the ablated hearts at 5 dpf/2 days post-treatment (dpt), whereas their expressions in the control hearts were very weak (Figures 1A–D′), whose signals could not be seen under the same staining condition as ablated group but showed up after extended staining. Similarly, the gene expressions of TGF-β receptors alk5a, alk5b, and cofactor smad3a were also dramatically upregulated in the ablated hearts (Figures 1E–G′). Immunofluorescence staining showed that the number of cells expressing phospho-Smad3, the active form of R-SMAD, was dramatically increased in the ablated hearts compared with the control hearts at 5 dpf/2 dpt (Figures 1H,I, 78.0 ± 8.6 vs. 9.1 ± 1.4 per heart, N = 11 and 8, respectively). The pSmad3-positive cells were not restricted in the myocardium only but also located in the epicardium and the outflow tract (OFT). Overall, these results indicated that the TGF-β/Smad3 signaling pathway was activated during zebrafish ventricular regeneration.
FIGURE 1
Inhibition of the TGF-β/Smad3 Signaling Pathway Impedes Ventricular Regeneration
After ventricle ablation, most hearts could fully regenerate and contract normally like the control hearts at 7 dpf/4 dpt (Figures 2A,B, type 1). The unrecovered hearts could be classified into two categories based on their cardiac morphology: one category included hearts with tiny shrunken or elongated ventricles (Figure 2C, type 2), and one category included hearts with partially regenerated ventricles (Figure 2D, type 3), presumably due to failure in CM proliferation and/or migration. The percentages of the different types of regenerated hearts at 7 dpf/4 dpt were 73, 17, and 9%, and 1% of the larvae died (Figure 2G, N = 221).
FIGURE 2
To investigate the role of the TGF-β signaling pathway in cardiac regeneration, we acquired a tgfb1a–/– mutant line () and generated a tgfb1b–/– mutant line via CRISPR/Cas9 technique. The tgfb1b–/– mutant possessed a 5-bp deletion in the first exon of genomic DNA, which presumably resulted in a truncated peptide product of 62 aa instead of the full-length protein of 379 aa (Supplementary Figures 1A,B). The cardiac development and morphology appeared normal in both tgfb1a–/– and tgfb1b–/– single mutants, and mutation of either tgfb1 homolog seemed to have no inhibitory effect on the recovery of the injured ventricles (Supplementary Figures 1C–F). Homozygous tgfb1a–/–; tgfb1b–/– mutants were indistinguishable from their siblings in embryonic stages, but we failed to acquire adult homozygous double mutants. Careful examination revealed a subset of juvenile fish with shorter body length and lower body weight in a population of mixed genotypes (Supplementary Figures 2A–F). These fish were prone to death and confirmed as homozygous tgfb1a–/–; tgfb1b–/– mutants via genotyping using tailfin genomic DNA (Supplementary Figure 2G). Therefore, we used several pharmacological inhibitors targeting different components of TGF-β signaling, including SIS3, SB431542, and LY364947 (), to examine the function of TGF-β signaling in ventricular regeneration. All three inhibitors dramatically reduced the percentages of recovered hearts at 7 dpf/4 dpt (Supplementary Figure 3A).
Next, we focused on the effect of SIS3, a selective Smad3 inhibitor that attenuates Tgfβ1-dependent Smad3 phosphorylation and DNA binding (; ). A 24-h treatment scheme revealed that SIS3 treatment at 0–1 and 1–2 dpt significantly reduced the ventricle recovery ratio from 76 to 17% and 45% at 7 dpf/4 dpt, respectively, while SIS3 treatment at 2–3 and 3–4 dpt had no obvious effect (Figure 2E). The most dramatic impact was observed after the 48-h SIS3 treatment at 0–2 dpt, in which the recovery ratio decreased to 5% (Figure 2F). Forty-four percent of the larvae showed tiny ventricles, 27% had partially regenerated ventricles, and the other 24% died (Figure 2H and Supplementary Figure 4). Immunofluorescence staining of phospho-Smad3 and phospho-Smad1/5/9 was performed to confirm the specificity of the SIS3 treatment. Although the numbers of both pSmad3-positive cells and pSmad1/5/9-positive cells increased in the myocardium, epicardium, and OFT of regenerating hearts after ventricle ablation, SIS3 specifically inhibited the phosphorylation of Smad3 but not the phosphorylation of Smad1/5/9 (Figures 2I–K and Supplementary Figures 5, 6).
To explore the function of Smad3 in ventricular regeneration, we carried out transcriptomic analysis using four pools of approximately 600, 800, 1,200, and 6,500 manually dissected hearts from the control, SIS3-treated control, ablated, and SIS3-treated ablated groups, considering that fewer CMs remained in the SIS3-treated ablated hearts. Differentially expressed genes were enriched with gene ontology (GO) terms associated with cell differentiation [gata4 ()], cell proliferation [plk1 (), tnni3k (), and erh ()], and cell migration in wound healing [tmsb4x () and fn (; )] (Figures 2L–N). Taken together, our results suggest that the TGF-β/Smad3 signaling pathway is required for ventricular regeneration and may be involved in multiple processes during regeneration.
Smad3 Inhibition Has Limited Effects on Sarcomere Disassembly and Cardiogenic Factor Reactivation During Regeneration
Previous reports showed sarcomere disassembly occurred in the injured hearts, which allowed CM to proliferate (; ). To examine the effect of Smad3 inhibition on sarcomere structure, we used Tg(myl7:actinin-EGFP) transgenic fish to observe the morphological changes in sarcomeres during ventricle regeneration (Figures 3A–D′). While the control and SIS3-treated control hearts displayed intact striated sarcomeres, SIS3-treated ablated hearts showed a pattern of sarcomere disarray similar to that in the ablated hearts at 4 dpf/1 dpt. These results suggested that Smad3 inhibition did not affect cardiac regeneration by blocking sarcomere disassembly.
FIGURE 3
Cardiac regeneration was also accompanied by reactivation of cardiogenic genes. WISH showed that the expressions of cardiogenic transcription factors, such as nkx2.5, gata4, and tbx5a, were significantly increased in the ablated hearts. Smad3 inhibition via SIS3 treatment abolished the reactivation of nkx2.5 at 5 dpf/2 dpt; however, it had no obvious effect on the reactivation of gata4 and tbx5a (Figures 3E–P). These results suggested that Smad3 inhibition had a limited effect on cardiogenic factor reactivation during ventricle regeneration.
Smad3 Inhibition Reduces Cardiomyocyte Proliferation Through Cell Cycle Arrest During Regeneration
To observe CM proliferation in real time during ventricle regeneration, we acquired the transgenic line Tg(myl7:mAG-zGeminin), which uses fluorescent ubiquitylation-based cell cycle indicator (FUCCI) technology () that can label CMs in the S/G2/M phases with green fluorescence (Figures 4A–E and Supplementary Figure 7). zGem+ CMs could be observed in the control hearts at 4 and 5 dpf, mainly in the ventricle due to trabecular initiation and growth at this time point. After ventricle ablation, the number of zGem+ CMs in the ablated hearts, in both the ventricle and atrium, was dramatically increased compared with that in the control hearts (35.1 ± 3.9 vs. 20.7 ± 2.8 at 4 dpf/1 dpt, N = 14 and 12, respectively; 32.2 ± 5.8 vs. 8.9 ± 2.2 at 5 dpf/2 dpt, N = 13 and 13, respectively). Surprisingly, Smad3 inhibition did not significantly reduce the number of zGem+ CMs in the SIS3-treated ablated hearts (28.6 ± 4.9 at 4 dpf/1 dpt, N = 11; 33.0 ± 6.9 at 5 dpf/2 dpt, N = 12), which is contradictory to a previous report that Smad3 inhibition impaired CM proliferation (). Our results also showed that SB431542 treatment reduced EdU incorporation in CMs at 5 dpf/2 dpt (Supplementary Figures 3B–D). Thus, we examined the change of CM number during regeneration with or without SIS3 treatment using Tg(myl7:H2B-EGFP) transgenic fish, which specifically labeled CM nucleus (Figures 4F–N and Supplementary Movies 1–4). The CM numbers were similar in ablated and SIS3-treated ablated groups at the beginning (118.3 ± 9.9 vs. 125.9 ± 12.6 at 4 dpf/1 dpt, 133.1 ± 11.6 vs. 120.8 ± 12.8 at 5 dpf/2 dpt, N = 8 for each group). However, the CM number increased in ablated group along with progression of regeneration, but it remained unchanged or even declined in SIS3-treated ablated group with statistically significant differences (165.1 ± 11.1 vs. 111.4 ± 15.5 at 6 dpf/3 dpt, 227.3 ± 25.9 vs. 104.4 ± 13.8 at 7 dpf/4 dpt, N = 8 for each group).
FIGURE 4
To resolve this discrepancy, we examined CM proliferation after Smad3 inhibition during ventricular regeneration using traditional phospho-histone H3 (pH3) immunostaining as well as pulsed EdU incubation and staining (Figures 4O–R). The results showed that the number of pH3+ CMs (Figure 4O′, arrowheads) was increased in the ablated hearts (6.9 ± 1.5, N = 10) compared with the control hearts (4.2 ± 1.4, N = 6). Upon SIS3 treatment, although there were still strong pH3 signals in the non-CMs (Figure 4P), the number of pH3+ CMs decreased in the SIS3-treated control hearts (1.0 ± 0.2, N = 11) and SIS3-treated ablated hearts (2.8 ± 0.4, N = 10) (Figure 4Q). EdU staining exhibited a similar pattern (Figure 4O″, arrow). SIS3 treatment reduced the number of EdU+ CMs in both the control hearts (28.2 ± 2.6 vs. 14.9 ± 1.4, N = 6 and 10, respectively) and the ablated hearts (48.1 ± 4.5 vs. 17.5 ± 3.9, N = 9 and 11, respectively) (Figure 4R), which was consistent with a previous report ().
We noticed that the number of zGem+ CMs was slightly lower, though not statistically significantly lower, in the control hearts treated with SIS3 for 24 h than in those not treated with SIS3 at 4 dpf (Supplementary Figure 7E, 13.7 ± 1.7 vs. 20.7 ± 2.8, N = 9 and 12, respectively). However, when the SIS3 treatment was extended to 48 h, the number of zGem+ CMs in the control hearts dramatically increased at 5 dpf (Figure 4E, 27.3 ± 3.0 vs. 8.9 ± 2.2, N = 18 and 13, respectively). zGem+ cells are in the middle-to-late S phase, G2 phase, and early-to-middle M phase of the cell cycle, while transient EdU labeling and pH3 staining indicate cells in the S and M phases, respectively (). Thus, we speculated that SIS3 treatment may cause cell cycle arrest at the G2 phase. We co-stained Tg(myl7:mAG-zGeminin) for EdU and pH3 simultaneously (Supplementary Figure 8) and subtracted the numbers of zGem+/EdU+ CMs and zGem+/pH3+ CMs from the total number of zGem+ CMs in the same heart. The results revealed that the number of zGem+/pH3–/EdU– CMs was significantly increased in the control hearts after SIS3 treatment at 5 dpf (51.5 ± 5.7 vs. 23.7 ± 3.3, N = 13 and 11, respectively), suggesting that Smad3 inhibition via SIS3 treatment may block CM proliferation through cell cycle arrest.
Smad3 Inhibition Impairs Cardiomyocyte Migration by Weakening the Epithelial–Mesenchymal Transition-Like Response During Regeneration
It has been reported that CM migration occurs during cardiac regeneration (; ) and that CMs undergo EMT-like response in a mouse cardiac infarction model (). Thus, it is of great interest to explore if CMs acquire EMT-like response during ventricle regeneration as well as the roles of the TGF-β/Smad3 signaling pathway in this process. We first examined the expression of the EMT marker Twist1b, a transcription factor that plays an essential role in metastasis (). The average fluorescence intensity of the Twist1b protein in the ablated hearts was much higher than that in the control hearts (Supplementary Figure 9). WISH showed that twist1b gene expression was upregulated in the ablated hearts compared with the control hearts and that SIS3 treatment abolished this upregulation (Figures 5A–D).
FIGURE 5
Next, we examined the expression of Snail, another transcription factor and EMT marker (). Immunostaining showed that Snail+ CMs were uniformly distributed in the ventricles of the control hearts (Figure 5E and Supplementary Movie 5) and SIS3-treated control hearts (Figure 5F and Supplementary Movie 6). In the ablated hearts, Snail+ CMs accumulated around the lost area in the ventricle and scattered throughout the atrium (Figure 5G and Supplementary Movie 7). Further analysis demonstrated that Snail+ CMs were distributed in both the ventricular compact layer and trabecular layer in the control hearts (Figures 4E′,E″), which was consistent with trabecular formation at this stage. In the ablated hearts, the number of Snail+ CMs in the trabecular layer decreased, while the number in the compact layer and atrium slightly increased (Supplementary Figure 10), and Snail+ CMs frequently extruded from the myocardium into the outer layer (Figures 4G′,G″, arrowheads). Smad3 inhibition via SIS3 treatment reduced the number of Snail+ CMs in the whole ablated hearts at 5 dpf (Figure 5I, 56.0 ± 4.0 vs. 72.8 ± 4.0, N = 8 and 6, respectively), and Snail+ CMs often located away from the lost area (Figure 5H and Supplementary Movie 8). Snail+ CM number reduction was also observed after SB431542 treatment (Supplementary Figures 3E–G).
We also examined the changes in the expression of other EMT markers during ventricle regeneration after Smad3 inhibition. WISH of vimentin, a gene encoding an intermediate filament protein that is important for cell migration, revealed that its upregulation in the ablated hearts could also be abolished by SIS3 treatment (Figures 6A–D). Immunostaining of N-cadherin (Figures 6E–N), a cellular adhesion molecule that is vital for cell contact and migration, demonstrated regular punctate N-cadherin distribution along the lateral side of CMs in the trabecular and compact layers in the control hearts (Figures 6E–E″,I–I″′, arrowheads). In the ablated hearts, the N-cadherin signal was absent in CMs adjacent to the ablated area (Figures 6G–G″, open arrowheads) or displayed an irregularly dispersed distribution in other areas (Figures 6K–K″′, arrows). Interestingly, we observed a much thicker outer layer with N-cadherin signal but no myocardial markers in the ablated hearts than in the control hearts, and the identity and physiological function of this layer remain to be explored. SIS3 treatment did not globally affect the distribution of N-cadherin (Figures 6H–H″,L–L″′), although the average fluorescence intensity of N-cadherin in the SIS3-treated ablated hearts increased (Figure 6M). The thickness of the outer layer was also reduced in the ablated hearts after SIS3 treatment (Figure 6N). Taken together, our results indicated that CMs possessed EMT-like response during cardiac regeneration and that Smad3 inhibition via SIS3 treatment weakened the EMT-like response, which may lead to CM migration defect.
FIGURE 6
Discussion
The TGF-β/Smad3 signaling pathway has been reported to be vital for scar resolution in an adult zebrafish model of heart cryoinjury (), and our results indicated that this pathway was activated after ventricle ablation and played important roles in ventricle regeneration. The exacerbation of different types of unrecovered hearts after SIS3 treatment and the transcriptomic analysis suggested that TGF-β/Smad3 signaling may affect CM proliferation and migration. Further examination revealed that Smad3 inhibition caused cell cycle arrest and weakened the EMT-like response during regeneration. However, since the control hearts still develop and undergo trabeculation at this stage, we have interpreted the data with extra caution by comparing different compartments of the heart including the ventricular trabecula, compact layer, and atrium.
The Reactivation of nkx2.5 Was Repressed by Smad3 Inhibition
During cardiac regeneration, cardiogenic transcription factors, such as nkx2.5, gata4, and tbx5a, are reactivated (); and inhibition of the Notch signaling pathway abolishes the reactivation of these factors (). However, our study revealed that Smad3 inhibition via SIS3 treatment only repressed the expression of nkx2.5 but not the expression of other cardiogenic transcription factors, suggesting a more specific function of the TGF-β/Smad3 signaling pathway in ventricle regeneration. It has been reported that inhibition of the TGF-β signaling pathway by A83-01 promotes the proliferation and differentiation of nkx2.5+ CMs and improves cardiac function after MI (; ). A83-01 inhibits the ALK5, ALK4, and ALK7 receptors, blocking the phosphorylation and activation of Smad2. In contrast, SIS3 attenuates TGFβ1-dependent Smad3 phosphorylation but has no effect on Smad2 (; ). Considering the opposing roles of Smad2 and Smad3 in CM proliferation during development (), we speculate that Smad2 and Smad3 also have opposing functions in nkx2.5 reactivation and CM proliferation during ventricle regeneration, which warrants further investigation if specific Smad2 antibodies become available.
Cardiomyocyte Cell Cycle Arrest by Smad3 Inhibition
The FUCCI system can reveal cell cycle progression in real time (); thus, we used the transgenic line Tg(myl7:mAG-zGeminin) to label CMs in the S/G2/M phases. CM proliferation in the control hearts was mainly in the ventricle due to trabecular initiation and growth. However, zGem+ CMs in the ablated hearts, mainly in the atrium, remained fluorescent after Smad3 inhibition. The number of zGem+ CMs was not changed after a 24-h SIS3 treatment but was significantly increased in the 48-h SIS3-treated control hearts at 5 dpf. Through co-staining with pH3 and transient EdU labeling in the same heart, we revealed that the number of zGem+/pH3–/EdU– CMs increased, suggesting cell cycle arrest at G2 phase. It has been reported that the TGF-β signaling inhibitor SB431542 causes CM cell cycle arrest at the G2/M phase through p27 inhibition (; ). We found no difference in the expression of p27 in the SIS3-treated hearts compared with the control hearts (data not shown). This result may occur due to the low basal expression level of p27, making it difficult to detect a reduction after SIS3 treatment.
Cardiomyocyte Epithelial–Mesenchymal Transition-Like Response Is Reduced by Smad3 Inhibition
Cardiac regeneration requires coordination between CM proliferation and migration, and cardiac regeneration in both mammalian and zebrafish is accompanied by CM migration (; ; ; ). How the EMT-like process plays a role in CM migration during regeneration remains to be explored. We discovered that the expression of the EMT marker twist1b was increased in the ablated hearts compared with the control hearts. CMs that express Snail, which is another EMT marker, were mainly distributed in the trabecular layer due to trabeculation in the control hearts at this stage (), while in the ablated hearts, Snail+ CMs were enriched near the wounded area and scattered in the atrium. Interestingly, a few Snail+ CMs extruded from the myocardium into the outer layers, which were multiple cells thick in the ablated hearts, as revealed by N-cadherin and DAPI staining, instead of the single epicardial layer observed in the control hearts. The localization of N-cadherin also changed from a regular punctate distribution on the lateral side of the compact layer in the control hearts to an absent or irregular distribution in the ablated hearts, suggesting that CMs were ready to migrate. Smad3 inhibition abolishes twist1b upregulation, reduces Snail+ CM numbers, and increases N-cadherin expression, thus weakening EMT-like response during regeneration. It has been reported that after ventricle resection in newt, a subset of CMs with proliferative abilities migrates toward the epicardium and apex through transformation and then embeds into the regeneration-specific matrix to repair the damaged area (). We hypothesize that during ventricle regeneration in zebrafish, a subset of CMs also undergoes an EMT process and migrates into the extracellular matrix located between the myocardium and epicardium to enter the ablated area. Conventional confocal technique is not able to document this dynamic process in vivo; advanced microscopy, such as light-sheet microscopy with low phototoxicity and photobleaching effect for long-term imaging, combined with novel reporter lines is required for further investigation of migration during zebrafish heart regeneration.
In conclusion, we show that the TGF-β/Smad3 signaling pathway participates in CM cell cycle progression and EMT process during ventricle regeneration, and we lay a foundation for the development of novel therapeutic strategies for MI in the future.
Materials and Methods
Zebrafish Husbandry
All zebrafish were raised under standard conditions in accordance with institutional and national animal welfare guidelines. The transgenic lines used in this study were as follows: Tg(vmhc:mCherry-NTR), Tg(flk:GFP), Tg(myl7:actinin-EGFP), Tg(myl7:H2B-EGFP), Tg(myl7:lifeact-EGFP), and Tg(myl7:mAG-zGeminin). 1-Phenyl-2-thiourea (PTU; Sigma, P7629) was added at 24 hpf to prevent pigmentation.
Generation of tgfb1b–/– Mutant Zebrafish
tgfb1b–/– mutants were generated using CRISPR/Cas9 technique (). sgRNA target sites were identified by online tool CRISPRscan. The sgRNAs were in vitro transcribed as previously reported () and co-injected with Cas9 protein (New England Biolabs) into embryos at the one-cell stage. Positive founders were mated with wild-type or Tg(vmhc:mCherry-NTR) fish to obtain F1 generation. The F1 heterozygous zebrafish with identical frameshift mutations were intercrossed to generate F2 homozygous mutants.
Chemical Treatment
Tg(vmhc:mCherry-NTR) larvae were treated with 5 mM of MTZ (Sigma) in E3 water at 72 hpf for 4 h as previously described (). After being washed, the larvae were then incubated in 2 μM of SIS3 (Selleck), 25 μM of SB431542 (Tocris), 20 μM of LY364947 (Selleck), or 0.2% dimethyl sulfoxide (DMSO) (Thermo Fisher Scientific) as control from 76 to 124 hpf.
In situ Hybridization
Whole-mount in situ hybridization was performed using a modified protocol as previously described (), including nkx2.5, gata4, vimentin, tgfb1a, and smad3a. The probes for tgfb1b, alk5a, alk5b, tgfb2, tgfb3, twist1b, and tbx5a were amplified by PCR with the following primers: tgfb1b F-5′ CC AAGGAACCAGAAGTAGAA, tgfb1b R-5′ CCAGACATA GGAGCAAGAG; alk 5a F-5′ TCGTGTGCCAAGTGAAGAAG, alk5a R-5′ CATGATCTTGGCCATCACAC; alk5b F-5′ GG TGTGTGTGCTGTGTTTCC, alk5b R-5′ ACTGACGGGTCTG ACTGGAC; tgfb2 F-5′ ACGCCAAAGAAGTGCACAAG, tgfb2 R-5′ CGCTCCACAGATACGGACAG; tgfb3 F-5′ AACCTGA GCACCTCCAGGAC, tgfb3 R-5′ GCTGCACTTGCAGGAT TTG; twist1b F-5′ GAAAACACGAGGACCAATG, twist1b R-5′ GAATTGTACTAAAGCTTTGTA. tbx5a F-5′ GCGTT TCCAGCACATCTCAG, tbx5a R-5′ TGTGTCCAGTGCTCCTT TACC.
Immunofluorescence
Immunostaining on whole mount larvae or dissected larval hearts was performed as previously described (). The primary antibodies used in this study included the following: anti-phospho-Smad3 (rabbit; Abcam, 52903), anti-phospho-Smad1/5/9 (rabbit; CST, 13820), anti-phospho-histone H3 (rabbit; Merck Millipore, 06570), anti-MHC (mouse; DSHB, MF20), anti-Twist (mouse; Santa Cruz, 81417), and anti-Snail (rabbit; Genetex, 125918). The secondary antibodies included the following: Alexa Fluor 488 goat anti-rabbit IgG, Alexa Fluor 488 goat anti-mouse IgG, Alexa Fluor 555 goat anti-mouse IgG, and Alexa Fluor 647 goat anti-rabbit IgG from Thermo Fisher Scientific.
EdU Treatment
The larvae at 123 hpf were incubated with 500 μM of EdU for 1 h in E3 water containing PTU and 2% DMSO to facilitate EdU solubilization. After treatment, larvae were rinsed with E3 water, anesthetized with 0.2% tricaine, and fixed overnight in 4% paraformaldehyde (PFA). The CLICK-IT reaction for EdU labeling was performed according to the manufacturer’s instruction (Thermo Fisher Scientific).
Imaging
The larvae and dissected hearts were imaged by LSM710/LSM880 (Zeiss), SP8 (Leica), or A1 (Nikon) confocal microscopes. The acquired confocal z-stack images were processed, and cell counting was performed using ZEN (Zeiss), LAS X (Leica), and Fiji software. The injured area was determined by visually examination of each single optical section and then manually labeled in the maximal projection image.
RNA Extraction
Zebrafish hearts were manually dissected by tweezers from control and ablated larvae after euthanasia with an overdose of anesthetic (0.2% tricaine). The dissected hearts were washed in 10% fetal bovine serum (FBS) medium and stored in Eppendorf tubes with TRIzol solution (Thermo Fisher Scientific). Different batches were pooled together to reach approximately 600, 800, 1,200, and 6,500 hearts for the control, SIS3-treated control, ablated, and SIS3-treated ablated groups. RNA was extracted using TRIzol solution following manufacturer’s instruction; 1.5 μg of total RNA for each group was sent out for RNA-seq.
Bioinformatic Analysis
The sequencing data were obtained by RNA-seq conducted by Genewiz company and had been uploaded to Gene Expression Omnibus (GEO) database (accession number GSE162820). The data were monitored by FASTQ quality trimmer by which low-quality reads were removed. High-quality reads were then mapped with the reference genome using HISAT (). The number of reads was counted using HTseq (). DE-seq () was used to analyze the differentially expressed genes. Supervised two-way clustering was conducted using pheatmap package of R software ().
Statistical Analysis
The regeneration ratio was calculated as the number of recovered larvae over the number of total injured larvae. GraphPad software was used for statistical analysis. Values were presented as mean ± s.e.m. Statistical significance was defined as ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, and ****P < 0.0001, determined by Student’s t-test or chi-square test in the quantification of the percentage of recovered hearts.
Statements
Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.
Ethics statement
The animal study was reviewed and approved by Fudan University Institutional Animal Care and Use Committee.
Author contributions
YP and RZ conceived and designed the project and wrote and revised the manuscript. YP, WW, YF, HH, Y-FH, XL, and HL conducted the experiments and analyzed the data. NC, MP, and J-WX provided critical reagents. All authors reviewed the manuscript.
Funding
This study was supported by National Key R&D Program of China grants 2020YFA0803900 and 2018YFA0801000 to RZ.
Acknowledgments
We thank Haitao Zhou and Guoqin Xu for fish care. We thank Drs Guobao Li and Peilu She for providing reagents; Dr. Shunji Jia for providing tgfb1a–/– mutant fish; Chunxiao Yu and Qi Li for technical support; Ji Tang for help with Fiji software; and Drs Ting Ni, Su Guo, and Jianjian Sun for helpful discussions.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcell.2021.632372/full#supplementary-material
Supplementary Movie 13D reconstruction of live imaging of ablated Tg(vmhc:mCherry-NTR; myl7:H2B-EGFP) heart at 5 dpf/2 dpt.
Supplementary Movie 23D reconstruction of live imaging of SIS3-treated ablated Tg(vmhc:mCherry-NTR; myl7:H2B-EGFP) heart at 5 dpf/2 dpt.
Supplementary Movie 33D reconstruction of live imaging of ablated Tg(vmhc:mCherry-NTR; myl7:H2B-EGFP) heart at 7 dpf/4 dpt.
Supplementary Movie 43D reconstruction of live imaging of SIS3-treated ablated Tg(vmhc:mCherry-NTR; myl7:H2B-EGFP) heart at 7 dpf/4 dpt.
Supplementary Movie 53D reconstruction of fluorescence images of fixed Tg(vmhc:mCherry-NTR; flk:GFP) fish with immunostaining of Snail (white) and MF20 (red) in the control heart at 5 dpf/2 dpt.
Supplementary Movie 63D reconstruction of fluorescence images of fixed Tg(vmhc:mCherry-NTR; flk:GFP) fish with immunostaining of Snail (white) and MF20 (red) in the SIS3-treated control heart at 5 dpf/2 dpt.
Supplementary Movie 73D reconstruction of fluorescence images of fixed Tg(vmhc:mCherry-NTR; flk:GFP) fish with immunostaining of Snail (white) and MF20 (red) in the ablated hearts at 5 dpf/2 dpt.
Supplementary Movie 83D reconstruction of fluorescence images of fixed Tg(vmhc:mCherry-NTR; flk:GFP) fish with immunostaining of Snail (white) and MF20 (red) in the SIS3-treated ablated hearts at 5 dpf/2 dpt.
References
1
AharonovA.ShakkedA.UmanskyK. B.SavidorA.GenzelinakhA.KainD.et al (2020). ERBB2 drives YAP activation and EMT-like processes during cardiac regeneration.Nat. Cell Biol.221346–1356. 10.1038/s41556-020-00588-4
2
AndersS.HuberW. (2010). Differential expression analysis for sequence count data.Genome Biol.11:R106.
3
AndersS.PylP. T.HuberW. (2015). HTSeq–a Python framework to work with high-throughput sequencing data.Bioinformatics31166–169. 10.1093/bioinformatics/btu638
4
AzharM.Schultz JelJ.GruppI.DornG. W.IIMenetonP.MolinD. G.et al (2003). Transforming growth factor beta in cardiovascular development and function.Cytokine Growth Factor Rev.14391–407. 10.1016/s1359-6101(03)00044-3
5
Bock-MarquetteI.SaxenaA.WhiteM. D.DimaioJ. M.SrivastavaD. (2004). Thymosin beta4 activates integrin-linked kinase and promotes cardiac cell migration, survival and cardiac repair.Nature432466–472. 10.1038/nature03000
6
BujakM.RenG.KweonH. J.DobaczewskiM.ReddyA.TaffetG.et al (2007). Essential role of Smad3 in infarct healing and in the pathogenesis of cardiac remodeling.Circulation1162127–2138. 10.1161/circulationaha.107.704197
7
ChablaisF.JazwinskaA. (2012). The regenerative capacity of the zebrafish heart is dependent on TGFbeta signaling.Development1391921–1930. 10.1242/dev.078543
8
ChangN.SunC.GaoL.ZhuD.XuX.ZhuX.et al (2013). Genome editing with RNA-guided Cas9 nuclease in zebrafish embryos.Cell Res.23465–472. 10.1038/cr.2013.45
9
ChenW. P.LiuY. H.HoY. J.WuS. M. (2015). Pharmacological inhibition of TGFbeta receptor improves Nkx2.5 cardiomyoblast-mediated regeneration.Cardiovasc. Res.10544–54. 10.1093/cvr/cvu229
10
ChenW. P.WuS. M. (2012). Small molecule regulators of postnatal Nkx2.5 cardiomyoblast proliferation and differentiation.J. Cell. Mol. Med.16961–965. 10.1111/j.1582-4934.2011.01513.x
11
CherianA. V.FukudaR.AugustineS. M.MaischeinH. M.StainierD. Y. (2016). N-cadherin relocalization during cardiac trabeculation.Proc. Natl. Acad. Sci. U.S.A.1137569–7574. 10.1073/pnas.1606385113
12
ChoiW. Y.GemberlingM.WangJ.HoldwayJ. E.ShenM. C.KarlstromR. O.et al (2013). In vivo monitoring of cardiomyocyte proliferation to identify chemical modifiers of heart regeneration.Development140660–666. 10.1242/dev.088526
13
DenisJ. F.SaderF.GatienS.VilliardE.PhilipA.RoyS. (2016). Activation of Smad2 but not Smad3 is required to mediate TGF-beta signaling during axolotl limb regeneration.Development1433481–3490. 10.1242/dev.131466
14
DibounI.WernischL.OrengoC. A.KoltzenburgM. (2006). Microarray analysis after RNA amplification can detect pronounced differences in gene expression using limma.BMC Genomics7:252. 10.1186/1471-2164-7-252
15
DograD.AhujaS.KimH. T.RasouliS. J.StainierD. Y. R.ReischauerS. (2017). Opposite effects of Activin type 2 receptor ligands on cardiomyocyte proliferation during development and repair.Nat. Commun.8:1902.
16
EngelM. E.McdonnellM. A.LawB. K.MosesH. L. (1999). Interdependent SMAD and JNK signaling in transforming growth factor-beta-mediated transcription.J. Biol. Chem.27437413–37420. 10.1074/jbc.274.52.37413
17
HeJ.MoD.ChenJ.LuoL. (2020). Combined whole-mount fluorescence in situ hybridization and antibody staining in zebrafish embryos and larvae.Nat. Protoc.153361–3379. 10.1038/s41596-020-0376-7
18
ItouJ.OishiI.KawakamiH.GlassT. J.RichterJ.JohnsonA.et al (2012). Migration of cardiomyocytes is essential for heart regeneration in zebrafish.Development1394133–4142. 10.1242/dev.079756
19
JinninM.IhnH.TamakiK. (2006). Characterization of SIS3, a novel specific inhibitor of Smad3, and its effect on transforming growth factor-beta1-induced extracellular matrix expression.Mol. Pharmacol.69597–607. 10.1124/mol.105.017483
20
JoplingC.SleepE.RayaM.MartiM.RayaA.Izpisua BelmonteJ. C. (2010). Zebrafish heart regeneration occurs by cardiomyocyte dedifferentiation and proliferation.Nature464606–609. 10.1038/nature08899
21
KikuchiK.HoldwayJ. E.WerdichA. A.AndersonR. M.FangY.EgnaczykG. F.et al (2010). Primary contribution to zebrafish heart regeneration by gata4(+) cardiomyocytes.Nature464601–605. 10.1038/nature08804
22
KimD.LangmeadB.SalzbergS. L. (2015). HISAT: a fast spliced aligner with low memory requirements.Nat. Methods12357–360. 10.1038/nmeth.3317
23
KimJ.WuQ.ZhangY.WiensK. M.HuangY.RubinN.et al (2010). PDGF signaling is required for epicardial function and blood vessel formation in regenerating zebrafish hearts.Proc. Natl. Acad. Sci. U.S.A.10717206–17210. 10.1073/pnas.0915016107
24
LepilinaA.CoonA. N.KikuchiK.HoldwayJ. E.RobertsR. W.BurnsC. G.et al (2006). A dynamic epicardial injury response supports progenitor cell activity during zebrafish heart regeneration.Cell127607–619. 10.1016/j.cell.2006.08.052
25
LiX.LuQ.PengY.GengF.ShaoX.ZhouH.et al (2020). Primary cilia mediate Klf2-dependant Notch activation in regenerating heart.Protein Cell11433–445. 10.1007/s13238-020-00695-w
26
MahmoudA. I.PorrelloE. R. (2018). Upsizing neonatal heart regeneration.Circulation1382817–2819. 10.1161/circulationaha.118.037333
27
MercerS. E.OdelbergS. J.SimonH. G. (2013). A dynamic spatiotemporal extracellular matrix facilitates epicardial-mediated vertebrate heart regeneration.Dev. Biol.382457–469. 10.1016/j.ydbio.2013.08.002
28
MohamedT. M. A.AngY. S.RadzinskyE.ZhouP.HuangY.ElfenbeinA.et al (2018). Regulation of cell cycle to stimulate adult cardiomyocyte proliferation and cardiac regeneration.Cell173104.e12–116.e12.
29
NaqviN.LiM.CalvertJ. W.TejadaT.LambertJ. P.WuJ.et al (2014). A proliferative burst during preadolescence establishes the final cardiomyocyte number.Cell157795–807. 10.1016/j.cell.2014.03.035
30
PattersonM.BarskeL.Van HandelB.RauC. D.GanP.SharmaA.et al (2017). Frequency of mononuclear diploid cardiomyocytes underlies natural variation in heart regeneration.Nat. Genet.491346–1353. 10.1038/ng.3929
31
PorrelloE. R.MahmoudA. I.SimpsonE.HillJ. A.RichardsonJ. A.OlsonE. N.et al (2011). Transient regenerative potential of the neonatal mouse heart.Science3311078–1080. 10.1126/science.1200708
32
PossK. D.WilsonL. G.KeatingM. T. (2002). Heart regeneration in zebrafish.Science2982188–2190. 10.1126/science.1077857
33
Sakaue-SawanoA.KurokawaH.MorimuraT.HanyuA.HamaH.OsawaH.et al (2008). Visualizing spatiotemporal dynamics of multicellular cell-cycle progression.Cell132487–498. 10.1016/j.cell.2007.12.033
34
SatoF.OtsukaT.KohsakaA.LeH. T.BhawalU. K.MuragakiY. (2019). Smad3 suppresses epithelial cell migration and proliferation via the clock gene Dec1, which negatively regulates the expression of clock genes Dec2 and Per1.Am. J. Pathol.189773–783. 10.1016/j.ajpath.2019.01.006
35
SethiA.JainA.ZodeG. S.WordingerR. J.ClarkA. F. (2011). Role of TGFbeta/Smad signaling in gremlin induction of human trabecular meshwork extracellular matrix proteins.Invest. Ophthalmol. Vis. Sci.525251–5259. 10.1167/iovs.11-7587
36
SoonpaaM. H.KimK. K.PajakL.FranklinM.FieldL. J. (1996). Cardiomyocyte DNA synthesis and binucleation during murine development.Am. J. Physiol. Heart Circ. Physiol.271H2183–H2189.
37
TaharaN.BrushM.KawakamiY. (2016). Cell migration during heart regeneration in zebrafish.Dev. Dyn.245774–787. 10.1002/dvdy.24411
38
TrinhL. A.StainierD. Y. R. (2004). Fibronectin regulates epithelial organization during myocardial migration in zebrafish.Dev. Cell6371–382. 10.1016/s1534-5807(04)00063-2
39
UribeV.RamadassR.DograD.RasouliS. J.GunawanF.NakajimaH.et al (2018). In vivo analysis of cardiomyocyte proliferation during trabeculation.Development145:dev164194. 10.1242/dev.164194
40
Valiente-AlandiI.PotterS. J.SalvadorA. M.SchaferA. E.SchipsT.Carrillo-SalinasF.et al (2018). Inhibiting fibronectin attenuates fibrosis and improves cardiac function in a model of heart failure.Circulation1381236–1252. 10.1161/circulationaha.118.034609
41
ViraniS. S.AlonsoA.BenjaminE. J.BittencourtM. S.CallawayC. W.CarsonA. P.et al (2020). Heart disease and stroke statistics-2020 update: a report from the american heart association.Circulation141e139–e596.
42
WengM. T.LuoJ. (2013). The enigmatic ERH protein: its role in cell cycle, RNA splicing and cancer.Protein Cell4807–812. 10.1007/s13238-013-3056-3
43
WuY.ZhouB. P. (2010). Snail: more than EMT.Cell Adh. Migr.4199–203. 10.4161/cam.4.2.10943
44
XingC. C.GongB.XueY.HanY. C.WangY. X.MengA. M.et al (2015). TGF beta 1a regulates zebrafish posterior lateral line formation via Smad5 mediated pathway.J. Mol. Cell Biol.748–61. 10.1093/jmcb/mjv004
45
YangJ.ManiS. A.DonaherJ. L.RamaswamyS.ItzyksonR. A.ComeC.et al (2004). Twist, a master regulator of morphogenesis, plays an essential role in tumor metastasis.Cell117927–939. 10.1016/j.cell.2004.06.006
46
YangJ.XuX. (2012). Immunostaining of dissected zebrafish embryonic heart.J. Vis. Exp.59:e3510.
47
YangL.BesschetnovaT. Y.BrooksC. R.ShahJ. V.BonventreJ. V. (2010). Epithelial cell cycle arrest in G2/M mediates kidney fibrosis after injury.Nat. Med.16535–543. 10.1038/nm.2144
48
YeL.D’agostinoG.LooS. J.WangC. X.SuL. P.TanS. H.et al (2018). Early regenerative capacity in the porcine heart.Circulation1382798–2808. 10.1161/circulationaha.117.031542
49
YuL.HebertM. C.ZhangY. E. (2002). TGF-beta receptor-activated p38 MAP kinase mediates Smad-independent TGF-beta responses.EMBO J.213749–3759. 10.1093/emboj/cdf366
50
ZhangR.HanP.YangH.OuyangK.LeeD.LinY. F.et al (2013). In vivo cardiac reprogramming contributes to zebrafish heart regeneration.Nature498497–501. 10.1038/nature12322
51
ZhuW.ZhangE.ZhaoM.ChongZ.FanC.TangY.et al (2018). Regenerative potential of neonatal porcine hearts.Circulation1382809–2816. 10.1161/circulationaha.118.034886
Summary
Keywords
ventricle regeneration, TGF-β, Smad3, cell cycle, EMT-like response
Citation
Peng Y, Wang W, Fang Y, Hu H, Chang N, Pang M, Hu Y-F, Li X, Long H, Xiong J-W and Zhang R (2021) Inhibition of TGF-β/Smad3 Signaling Disrupts Cardiomyocyte Cell Cycle Progression and Epithelial–Mesenchymal Transition-Like Response During Ventricle Regeneration. Front. Cell Dev. Biol. 9:632372. doi: 10.3389/fcell.2021.632372
Received
23 November 2020
Accepted
03 February 2021
Published
16 March 2021
Volume
9 - 2021
Edited by
Liqing Zang, Mie University, Japan
Reviewed by
Angel Raya, Institut d’Investigacio Biomedica de Bellvitge (IDIBELL), Spain; Michael Tsang, University of Pittsburgh, United States
Updates
Copyright
© 2021 Peng, Wang, Fang, Hu, Chang, Pang, Hu, Li, Long, Xiong and Zhang.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Ruilin Zhang, zhangruilin@whu.edu.cn
This article was submitted to Molecular Medicine, a section of the journal Frontiers in Cell and Developmental Biology
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.