The characterization of novel mechanisms safeguarding cell fate identity in differentiated cells is crucial to improve our understanding of 1) how differentiation is maintained in healthy tissues or altered in a disease state and 2) the basic mechanisms governing cell fate reprogramming and our ability to use this technology for regenerative purposes. Here, we report on the first identification of a generic and transcription factor-mediated mechanism controlling cell fate stability in differentiated cells via the concomitant regulation of chromatin accessibility and transcription of genes required for large-scale phenotypic changes. Importantly, resulting knowledge could be translated in vivo, as the targeted inhibition of these new fate-stabilizers could significantly improve direct cardiac reprogramming-mediated heart repair post-myocardial infarction in adult mice.

Teleost fish maintain the capacity to regenerate lost cardiac tissue throughout adulthood. However, recent reports suggest that both inter- and intra-species variations exist. Based on these studies, we hypothesised that there is strain-dependent variability in zebrafish heart regeneration. To test this, we characterised the regenerative response of seven wild-type zebrafish strains (AB, NA, SAT, TL, TU, WIK, KCL). Analysing morphology, wound size and composition, apoptosis, proliferation and gene expression, we identified significantly different characteristics between the NA, TL, TU and WIK strains during regeneration. The WIK strain showed increased proliferation in cardiomyocytes bordering the wound at 7dpci . Accordingly, RNA-seq showed upregulation of cell cycle and DNA replication-related genes at 7dpci in the WIK compared to the other strains, resulting in the strongest wound size reduction between 7- and 21dpci. However, no further reduction was observed between 21- and 90dpci, with scar remaining in all WIK hearts at 90dpci, likely caused by elevated ECM gene expression and increased collagen presence in the 7dpci WIK wound. Correlation analysis indeed identified collagen deposition at 7dpci as a predictor of the 90dpci regenerative outcome. At 90dpci, the NA line showed the highest percentage of scar-free hearts whereas 80% of the TU hearts failed to regenerate the compact wall. Comparing gene expression between the TU and NA showed a reduced activation of the metabolic genes required for regeneration in the TU after injury. In conclusion, comparing seven wild-type zebrafish strains has allowed us to identify correlations between different cellular processes occurring during regeneration.

Unlike humans, certain species possess a robust ability to regenerate their hearts. We have identified the Astyanax mexicanus fish as a unique model for heart regeneration research. While Astyanax surface fish regenerate their heart after injury, their cave-dwelling counterparts cannot and, like humans, form a permanent scar. This model provides an opportunity for direct comparison of natural regeneration versus scarring within a single species. 

We have previously shown that there are similar levels of cardiomyocyte proliferation at the wound border zone in both surface fish and cavefish. However, BrdU pulse-chase revealed a higher number of BrdU-positive cardiomyocytes in surface fish compared to cavefish which may be a result of defective cytokinesis resulting in a failure to regenerate. Indeed, examining cavefish cardiomyocytes showed higher DNA content in comparison to the surface fish heart. 

Using RNA-seq to investigate mechanisms underlying cardiomyocyte proliferation in Astyanax hearts after injury, we identified metabolism-related genes among those most differentially expressed. Specifically, glycolytic genes were highly upregulated in surface fish border zone cardiomyocytes. Moreover, glucose levels were reduced in the cavefish heart after injury. Seahorse metabolic assays revealed a reduction in anaerobic glycolysis in the cavefish heart after injury but no difference in oxygen consumption rates or fatty acid utilisation between the fish. This was accompanied by reduced ATP in cavefish hearts compared to surface fish after injury. Inhibiting glucose metabolism in surface fish using 2-deoxy-d-glucose increased cardiomyocyte DNA content showing a similar failure in cell cycle progression to the cavefish heart.

Overall, this suggests an important role for glucose metabolism in cardiomyocyte cell cycle progression, and that inability to maintain sufficient ATP in the cavefish heart might be linked to their inability to support a regenerative response.

Direct cardiac reprogramming (CR) of fibroblasts into induced cardiomyocytes (iCMs) represents a promising therapeutic avenue to promote heart repair post injury. Our lab previously identified four transcription factors playing a conserved role as barriers to CR. Remarkably, knock-down of these barriers increases CR up to six-fold as compared to MGT (MEF2C; GATA4; TBX5) overexpression alone. Next, mechanistic and functional, exploration of downstream transcriptional targets led to the identification of SULF2 and CHST2 as novel regulators of CR, indicating a potential novel role for proteoglycan sulfation-modifying enzymes as regulators of cell fate reprogramming. In this context, functional screening of all genes involved in the regulation of PG modification and sulfation in CR assays, identified CHST7 as most potent and evolutionarily conserved barrier to CR. Collectively, these data confirm that PG sulfation plays a previously unrecognized role in the regulation of CR and cell fate stability in differentiated cells.

Heart failure (HF) is a prevalent disease, projected to affect over 8 million Americans by 2030. Disease burden continues to be high, despite therapy. Thus, new strategies that target HF progression should be developed. Our lab employs the neonatal mouse model of heart regeneration to identify reparative pathways to apply them to HF. We previously showed that deletion of Interleukin 13 (IL13) or its receptor, IL4Rα, impair heart regeneration. However, the cell types mediating this response are not known. IL13 shares the receptor IL4Rα with IL4 and both cytokines are known to polarize macrophages into a pro-reparative phenotype. Here, we hypothesize that IL4/13 signaling to macrophages promotes neonatal heart regeneration after myocardial infarction (MI) and we explore the cell types producing IL13/IL4 during cardiac regeneration. We generated a genetic model whereby IL4Rα is depleted in macrophages by crossing IL4Rα floxed (IL4Rα^fl/fl) mice with transgenic mice whereby Cre recombinase is driven by the CX3CR1 promoter (CX3CR1^Cre). We performed MI on P1 mice and assessed cardiac function by ultrasonography 21 days post-injury. We found that Cre-positive mice had a lower ejection fraction compared to Cre-negative controls. Additionally, we found smaller capillaries in the myocardium of Cre-positive mice. To identify the cellular source of IL4 and IL13 in the heart we used fluorescent reporter mouse lines; IL4-green fluorescent protein and IL13-yellow fluorescent protein. In homeostasis, we found IL4 expression in innate lymphoid cells (ILCs) and T cells. Following P1 MI, IL13 was expressed primarily in ILCs and IL4 in ILCs and myeloid cells. This suggests a novel role for ILCs during cardiac regeneration. In addition, we found that lack of IL4/13 signaling in macrophages impairs cardiac function and capillary morphology after MI in neonatal mice. Future studies will be aimed to identify the downstream signaling in macrophages that mediates this reparative response.

The heart relies on opposing signals from sympathetic and parasympathetic nerves to guide developmental, homeostatic, and repair functions. However, the cardiac innervation patterning that regulates these essential functions has been relatively unexplored. This is partially a result of technical limitations, as histological sectioning prevented nerve morphology to be accurately examined. In this work, we are analyzing the spatial patterning of cardiac innervation by employing organ clearing, whole-mount staining, and three-dimensional (3D) imaging techniques. 

We are particularly interested in mapping the parasympathetic nerves, which are thought to only innervate the atria and nodes; however, their contribution to ventricular innervation has not been analyzed in whole-mount and 3D. Excitingly, our results demonstrate for the first time that dense parasympathetic axon bundles innervate the cardiac ventricles. Moreover, preliminary data demonstrates that parasympathetic and sympathetic nerve fibers are intertwined in the ventricles through a developmental patterning mechanism that we are currently exploring. 

The intricate neuronal networks are susceptible to injury and fatal misfiring following an adult myocardial infarction (MI). Remarkably, an MI in the neonatal mouse results in robust regeneration and restoration of autonomic functions. Our lab discovered that neonatal heart regeneration is regulated by nerve signaling. Here, we demonstrate reinnervation of the regenerating myocardium following injury in contrast to non-regenerating hearts. Moreover, these newly formed nerves exhibit significant crossover with smooth muscle positive arteries, suggesting potential physiological restoration of neurovascular coupling. The results of our research will expand our knowledge of the development and plasticity of the cardiac nervous system during homeostasis, disease, and regeneration. 

Heart failure is a major burden to our societies. Besides patients with left ventricular dysfunction following ischemic insults or hypertension, progress in pediatric surgery to repair cardiac malformations has led to a growing population of now-adult congenital heart diseases (CHD) patients. These patients with right ventricular (RV) failure are left without any efficient pharmacological therapeutic approach. Cell therapy has been an option to regenerate the myocardium although the mechanism of action of such an approach remains questionable.

Here, we used human embryonic stem cell-derived cardiac Nkx2.5+ progenitor cells seeded in a collagen-based patch covering the whole RV to regenerate failing RV of a pig model of repaired tetralogy of Fallot. We report that these cells migrate within the myocardium while reversing the interstitial fibrosis. They then engraft and fully differentiate into small clusters of fetal-like human myocytes within the myocardium. Degradation of the fibrotic extracellular matrix triggers an inflammatory reaction involving resident macrophages releasing cytokines in the neighborhood of myocytes. This leads to the activation of pig myocytes' inflammasome and their Nfkb-dependent reprogramming into Oct4+ cells. The reprogrammed myocytes redifferentiate and proliferate around human myocytes. Altogether, the graft of human CPC triggers a reprogramming of pig myocytes and in turn endogenous regeneration and improved contractility of the RV. 

Post-ischemic tissue remodeling involves the formation of a permanent scar that impairs cardiac function.  Unlike mammals, zebrafish can regenerate its heart by forming new functional tissues without a persistent scar.  The severity of tissue damage and the regenerative capacity depend on the quality, extent and temporal dynamics of the immune response.  However, how the immune system regulates regeneration, particularly how the antigen presentation-adaptive immunity axis plays a role in this process, remains to be elucidated.

In the present study, scRNA-Seq revealed a strong antigen presentation signature in the injured zebrafish heart, with distinct antigen-presenting cell clusters at different stages of the regenerative process.  Therefore, we hypothesized a role for the adaptive immune response during cardiac regeneration.  Indeed, we observed the infiltration of T-cells, especially Cd4+ T-cells, within the injured tissue at intermediate stages of regeneration.  Genetic targeting of Cd74, a key component of MHC class II antigen presentation, dampened regeneration as evidenced by impaired cardiomyocyte cell cycle re-entry and defective scar formation/resolution.  Pharmacological inhibition of T-cell activation further confirmed these data.  Importantly, these defects were accompanied by changes in T-cell populations and by an overall faulty immune response.  Further transcriptomic and immunohistochemical analyses revealed the involvement of the MAPK signaling pathway in the regulation of the regenerative response by antigen presentation.

Our work highlights the importance of antigen presentation in modulating the spectrum of regenerative/reparative outcomes post-cardiac injury.  Ultimately, these data will help identify mechanisms with the potential to ameliorate adverse remodeling and improve key aspects of mammalian heart regeneration.

Limited understanding of the molecular mechanisms of induced cardiomyocyte (iCM) reprogramming is a key obstacle preventing its effective clinical applications. We report here the identification of a phosphorylation code in 14-3-3 binding motifs (PC14-3-3) that greatly stimulates iCM formation. PC14-3-3 is identified in pivotal functional proteins for iCM reprogramming, including transcription factors and epigenetic factors. Akt1 kinase and PP2A phosphatase are a key writer and eraser of the PC14-3-3 code, respectively and PC14-3-3 activation induces iCM formation with only Tbx5 but without Mef2c and Gata4. In contrast, PC14-3-3 inhibition by mutagenesis or inhibitor-mediated code removal abolishes reprogramming. We further discover that key PC14-3-3 embedded factors, such as Mef2c, Nrip1, and Foxo1, form inhibitory nuclear condensates with Hdac4 under hypo-phosphorylation state and PC14-3-3 activation disrupts these condensates to promote cardiac gene expression. This study provides a framework in decoding post-translational modifications for cell reprogramming and organ regeneration.

Objectives: Unlike mammals, zebrafish are able to regenerate their heart after an injury at all stages of life. It has been identified that this process involves cardiomyocyte dedifferentiation/proliferation, but the implication of interstitial cell types during heart regeneration remains poorly understood. Our study aims to describe the non-myocyte cell types present in the heart during regeneration in adult zebrafish as well as focus on the functional role of endothelial Tal1 expression in this process.

Methods: We performed single cell RNA sequencing (scRNAseq) on unamputated, 3-, 7- and 14-days post amputation (dpa) zebrafish hearts to provide a transcriptomic analysis of the zebrafish regenerating heart. We identified genes of interest from the dataset obtained. We then used a zebrafish conditional line expressing a dominant negative isoform of Tal1 (DNTal1) specifically in endothelial cells. After validating our DNTal1 line with a developmental characterization, adult fish were amputated and their hearts were extracted at different timepoints to be fixed and stained in order to assess their regenerative abilities.  

Results: Using scRNAseq, we described the different interstitial cell types present in the zebrafish heart and the changes they undergo during heart regeneration. We observed an increase in the number of Tal1 expressing endothelial cells during heart regeneration compared to unamputated controls. After validating our zebrafish DNTal1 line showing it exerts cardiac developmental defects already described in the literature, we saw DNTal1 expression inhibited heart regeneration primarily by disrupting the formation of the regenerating vascular plexus.

Conclusion: We showed Tal1 expression is essential for the endothelial regenerative response and its inactivation leads to a failure of regeneration.

Upon ischemic cardiac injury, the epicardium, the outer layer of the heart which is essential for cardiac development, becomes re-activated and displays reparative potential. In this process, epicardial epithelial-to-mesenchymal transition (epiMT) is an essential step. To understand and increase epicardial activation, we aim to identify novel epiMT-inducing pathways by performing a small molecule screen.

Primary human epicardial cells were derived from human heart auricles. These epicardial derived cells (EPDCs) were cultured as epithelial-like cells maintaining a cobblestone morphology, and could be induced to undergo EMT by adding e.g. TGFbeta. Using this cell culture model, a phenotypic screen was performed on epithelial-like EPDCs using the LOPAC1280 small molecule library to identify epiMT-inducing compounds. The screen was performed 3 times to exclude patient variability. EpiMT was confirmed using αSMA-positive immunostaining as a hallmark for a phenotypic switch to a mesenchymal cell.
After validation of the positive hits, five compounds were selected that reproducibly induced epiMT, as shown by: 1) a phenotypic switch from epithelial (cobble) to mesenchymal (spindle-shaped) cells, 2) a decrease in CHD1 expression, and 3) an increase in EMT-related transcription factors (Snail, Slug) and mesenchymal markers (a-SMA, PSTN). To further establish a potential mechanism, cells were treated for 3 hours with the two most promising compounds and were subjected to RNA sequencing. These results suggest novel EMT regulators in human epicardium which are currently being validated.

In conclusion, high-throughput experiments using human primary epicardial derived cells to identify novel epiMT-inducing compounds is feasible. Using this model, we have identified several novel inducers of epiMT.
This work is funded by the DHF (2017T059, to AMS)

The stimulation of terminally differentiated cardiomyocyte proliferation represents one of the main therapeutic approaches for heart regeneration and repair. We uncovered a role for the Fibroblast Growth Factor 10 (FGF10) signalling in regulating both fetal and adult cardiomyocyte proliferation. Using Fgf10 gain and loss of function mouse models together with experimental mouse model of myocardial infarction (MI) we demonstrated FGF10 regenerative potential.

Our study aims to determine the clinical relevance of FGF10 therapy for heart regeneration. Thus, using AAV strategy, we aim to evaluate Fgf10 gene transfer therapeutic impact after myocardial infarction (MI).

We first aimed to determine the most efficient route of delivery for AAV9 administration by considering short- and long-term heart expression but also side organ targeting. We thus analyzed the spatiotemporal pattern of Fgf10 expression following AAV9-Fgf10 injection (retro-orbital or tail vein) with 10¹¹ viral particles of either AAV9-GFP or AAV9-Fgf10. Our results reveal that tail-vein administration route is more appropriated for a rapid, long term and efficient cardiac FGF10 treatment, with a pic of Fgf10 expression at 3 weeks post injection and a maintained expression 10 weeks post-injection. In order to investigate the impact of AAV9-Fgf10 administration following MI on cardiac function and remodeling, mice subjected to MI were tail vein-injected with 10¹¹ particles of either AAV9-GFP or AAV9-Fgf10 at the time of surgery. Our preliminary experiments indicate that AAV9-Fgf10 treatment promotes cardiomyocyte proliferation and prevents cardiac fibrosis infiltration. Moreover, echocardiographic measurements show that AAV9-Fgf10 treatment preserves cardiac remodeling and function post-MI.

Altogether, our results are thus of particular interest regarding the use of AAV9-Fgf10 strategy as a clinical perspective to promote heart repair following myocardial injury.

A promising strategy to repair the injured mammalian heart is by the induction of cardiomyocyte dedifferentiation followed by their proliferation. Thus, there is now a focus on identifying factors and pathways that can drive the production of new cardiomyocytes through proliferation. However, a largely underappreciated concept is the redifferentiation and maturation of cardiomyocytes following proliferation.  This becomes particularly prominent as current methods to induce cardiomyocyte proliferation often result in cardiomegaly due to uncontrolled cell numbers. Unlike mammals, the zebrafish adult heart can robustly regenerate its heart following injury.  Here, we have studied zebrafish heart regeneration to discover an essential mechanism that restricts proliferation and stimulates the redifferentiation and maturation of cardiomyocytes after injury. Using an ex vivo imaging technique that we developed, we observed differences in intracellular calcium dynamics in cardiomyocytes located at the wound border zone, where cardiomyocyte proliferation is induced. These changes in calcium dynamics provided evidence that after a period of strong proliferation, cardiomyocytes redifferentiate and mature. An adaptor protein lrrc10 was identified to be able to specifically halt proliferation and promote cardiomyocyte redifferentiation. Finally, these maturation mechanisms were similarly observed in mouse and human iPSC-derived cardiomyocytes. Hence demonstrating that rather than being a passive process, redifferentiation is actively regulated and highlights the importance on also determining its underlying mechanisms in order to generate fully functional cardiomyocytes.

The adult zebrafish is able to fully regenerate its heart after injury, making it an attractive model to identify and study mechanisms that regulate heart regeneration. In addition, a wide range of transgenic tools are available for zebrafish which allows the live imaging of cellular events. Thus far, live imaging in adult zebrafish have been hampered by the opacity and movement of the heart. In addition, ex vivo cultures don’t allow for deep tissue imaging and show rapid cell death. To overcome these issues, we present the use of cardiac slices of the zebrafish heart to study heart regeneration ex vivo. Cardiac slices can be cultured for several days while retaining characteristics of the in vivo heart. We have used these slices to perform live imaging on proliferating adult cardiomyocytes within their native tissue context for the first time. Currently, we are expanding the use of cardiac slice cultures to study the dynamics of calcium cycling and energy metabolism to get a better understanding of the role of these processes during heart regeneration. In conclusion, culturing cardiac slices of the adult zebrafish heart allows us to gain temporal insights of cardiomyocyte biology during physiological and pathological conditions. 

Unlike the adult mammalian heart, which has limited regenerative capacity, the zebrafish heart can fully regenerate following injury. Reactivation of cardiac developmental programmes is considered key to successfully regenerating the heart, yet the regulatory elements underlying the response triggered upon injury and during development remain elusive. Organ-wide activation of the epicardium is essential for zebrafish heart regeneration and is considered a potential regenerative source to target in the mammalian heart. Here we compared the transcriptome and epigenome of the developing and regenerating zebrafish epicardium by integrating gene expression profiles with open chromatin ATAC-seq data. By generating gene regulatory networks associated with epicardial development and regeneration, we inferred genetic programmes driving each of these processes, which were largely distinct. We identified wt1a, wt1b, and the AP-1 subunits junbb, fosab and fosb as central regulators of the developing network, whereas hif1ab, zbtb7a, tbx2b and nrf1 featured as putative central regulators of the regenerating epicardial network. By interrogating developmental gene regulatory networks that drive cell-specific transcriptional heterogeneity, we tested novel subpopulation-related epicardial enhancers in vivo. Taken together, our work revealed striking differences between the regulatory blueprint deployed during epicardial development and regeneration. These findings challenge the dogma that heart regeneration is essentially a reactivation of developmental programmes, and provide important insights into epicardial regulation that can assist in developing therapeutic approaches to enable tissue regeneration in the adult mammalian heart.

Factors responsible for cardiomyocyte (CM) proliferation may serve as a potential therapeutic to stimulate endogenous myocardial regeneration. It is established that RUNX1 induction in CMs increases after injury. Here, we examine the effect of RUNX1 on CM cell cycle and establishment of the presumed proliferative-competent Mononuclear Diploid CM population (MNDCM) during postnatal development and cardiac regeneration using both CM-specific gain-and loss of function mouse models. We hypothesize that RUNX1 overexpression (OE) increases CM cell cycle activity with expansion of the MNDCMs, thereby extending the neonatal regenerative window and positively impacting adult cardiac remodeling post injury. 

During postnatal development, RUNX1 KO decreased postnatal CM cell cycle activity, while RUNX1 OE extended the period of cell cycle activation. This extension observed in RUNX1 OE mice is complete with cytokinesis resulting in an expansion of the MNDCMs and total CM endowment. To determine whether RUNX1 could similarly regulate CM cell cycle in a regenerative and non-regenerative model, we induce P6 and 8-week MIs to measure cell cycle activity and cardiac function post injury. RUNX1 OE neonatal mice with a P6 MI displayed no difference in CM cell cycle activity 7 days post injury compared to control littermates. However, RUNX1 OE in adult mice with an 8-week MI showed increased CM cell cycle activity with completion of cytokinesis 2 weeks post injury and limited improvement in cardiac function 28 days post injury. RUNX1 influences CM cell cycle activation in the context of normal postnatal development and adult MI. Conversely, this phenomenon does not appear to translate to the neonatal injury context. We are examining this possible discrepancy with further experiments to fully understand the role of RUNX1 in CM cell cycle activity and its impact on cardiac regeneration.

The last two decades of research have significantly increased our understanding of mechanisms regulating zebrafish heart regeneration.  However, an aspect that remains largely underexplored is how newly proliferated cardiomyocytes invade and repopulate fibrotic tissue following cryoinjury.  We have previously shown that blocking the function of AP-1 transcription factors leads to defects in regeneration, partly due to a defect in cardiomyocyte protrusion and invasion into injured tissue.

Here, we show that cardiomyocyte protrusion peaks at 10 days post cryoinjury, shortly following the peak in cardiomyocyte proliferation.  Furthermore, using live imaging of myocardial slices, we observe that border zone (BZ) cardiomyocytes are dynamic, extending multiple protrusions into the collagen-containing injured area.  This dynamic behavior is accompanied by collagen degradation by BZ cardiomyocytes, which may be regulated by Mmp14b, a membrane-tethered matrix metalloprotease with collagen-degrading ability.  In addition, we find that macrophages are very closely associated with protruding cardiomyocytes at the BZ.  In irf8 mutants, which lack macrophages, we observe a decrease in the length of cardiomyocyte protrusions while the number of extended protrusions remains unchanged.  Single-cell RNA-seq of BZ cells suggests that both macrophages and cardiomyocytes contribute to regulation of the BZ microenvironment to promote invasion of cardiomyocytes to replenish lost tissue following injury.

Altogether, our data reveal a crosstalk between BZ cardiomyocytes and macrophages to facilitate repopulation of injured tissue by proliferating cardiomyocytes.  Detailed understanding of these mechanisms may have therapeutic implications, e.g., to promote the engraftment of exogenous cardiomyocytes following myocardial infarction in the mammalian heart.

Heart development and rhythm control are highly Tbx5 dosage-sensitive. TBX5 haploinsufficiency causes congenital heart defects and conduction disorders, whereas slightly increased levels of TBX5 in human heart samples have been associated with atrial fibrillation. To gain insight into the impact of slight dosage changes of Tbx5 in vivo, we deleted the mouse orthologue of a conserved atrial fibrillation-associated regulatory region in the TBX5 locus (RE(int)^-/- mice). Postnatal RE(int)^-/- mouse atria showed slightly increased Tbx5 expression levels (30%) and increased susceptibility to atrial arrhythmia. Strikingly, we observed increased Tbx5 (30%) in the ventricles before birth, accompanied by increased heart size. We observed an increase in fetal cardiomyocyte proliferation rates specific to the left ventricle. Transcriptional profiling of ventricles of fetal control and RE(int)^-/- mice revealed induction of several cell cycle related genes and Prrx1, a transcription factor associated with atrial fibrillation. When expression of Prrx1 was reduced by introducing a Prrx1 enhancer deletion in RE(int)^-/- mice, we observed a normalization of heart size. These data indicate prenatal cardiomyocyte proliferation rates are Tbx5 dosage sensitive and act in part through moderating Prrx1 levels. To examine the potential of Tbx5 to induce postnatal cardiomyocyte renewal, AAV9-TNNT2-TBX5 was retro-orbitally administered to juvenile mice. Preliminary results show cardiomyocyte-specific expression of TBX5, and induction of target genes (Gja5, Nppa) and genes involved in proliferation. Future experiments will focus on the capacity of Tbx5 to induce cardiomyocyte proliferation and the underlying mechanisms.

Cardiovascular diseases account for nearly 18 million deaths per year, and majority of these deaths occur due to myocardial infarction (MI).  To augment cardiac repair and regeneration following MI, mechanistic insights derived from the regenerative zebrafish heart are considered to be of high translational value.  Unlike the mammalian heart, the zebrafish heart is able to restore lost cardiomyocytes (CMs) through mechanisms that allow spared cardiomyocytes to dedifferentiate, proliferate, and repopulate the injured tissue.  This regenerative ability is attributed to a stringent spatiotemporal regulation and crosstalk between the various cardiac cell types.  Therefore, for an effective bench-to-bedside application of cardio-regenerative strategies, it is imperative that the distinct roles and spatiotemporal dynamics of the injury-responsive cardiac cell populations in zebrafish are well characterized.  

The injury-responsive endocardial cells remain poorly characterized.  Since they provide direct structural support, and may activate several signaling pathways in the CMs, we aim to further investigate their role in facilitating cardiac regeneration in adult zebrafish.  Using single-cell and bulk transcriptomic approaches on the ET(krt4:EGFP^sqet33-1A) line, in which endocardial cells express GFP, we have identified the injury-specific endocardial sub-populations enriched for EndoMT (serpine1, vim), and immune-cell (lgals2a) genes. Expression of these genes was mapped to the injury area and wound-border zone interface using the available Tomo-seq data for injured hearts.  In addition, several markers of mature endothelial/endocardial cells were identified, and observed to be repressed in these clusters.  These data corroborate recent findings in mouse that show EndoMT in endothelial cells of the infarcted hearts. Thus, we will perform functional studies on a few identified candidates, aiming to identify factors that promote CM dedifferentiation and/or proliferation.

Afterload, also known as resistance to contraction, and preload, also known as stretch, are key mechanical stimulation parameters that influence ventricular maturation but are difficult to test in vivo. Current in vitro mechanical stimulation bioreactors can mimic preload via cyclically stretching cardiac tissue but fail to simultaneously mimic afterload due to the usage of high afterload anchorage points. Therefore, we developed a mechanical stimulation bioreactor system that better recapitulates the developmental cardiac cycle. This system can apply duty cycles that vary the afterload and preload during the stretch regimen, which can be used to assess how these mechanical parameters affect maturation of biomimetic cardiac tissues. To determine the optimal stage to construct biomimetic cardiac tissues, we conducted single cell RNA sequencing on embryonic chicken ventricular cells spanning four key ventricular development stages. This data revealed embryonic day 7 as an important ventricular maturation transitional stage where epicardial-derived cells are actively contributing to the compacting myocardium. Embryonic day 7 ventricular biomimetic tissues were created and cultured statically for seven days followed by three different stimulation regimens for seven additional days: static, cyclic stretch without a duty cycle, or cyclic stretch with a duty cycle. Cyclically stretched biomimetic tissues maintained their overall spontaneous contractile function and led to significant increases in contractile frequency and mechanical stiffness compared to static controls. Strikingly, biomimetic tissues conditioned with a duty cycle led to the greatest improvement in cardiomyocyte elongation, alignment, and sarcomere organization. This data demonstrates duty cycles are a driver of ventricular developmental maturation.

Therapeutic induction of cardiomyocyte proliferation in vivo by modulating Wnt, Hippo, neuregulin, and other signaling pathways has shown significant promise to treat congenital and acquired heart diseases. However, when these signaling molecules are applied in vitro to expand cardiomyocytes (CMs) from pluripotent stem cells, the extent of proliferation has generally been modest (<5 fold), precluding the use of these cells for industrial-scale applications. Here, we demonstrate the massive expansion of beating human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) in vitro by glycogen synthase kinase-3 beta (GSK-3β) inhibition using CHIR99021 (CHIR), a small molecule GSK-3β antagonist, together with the removal of contact inhibition by low-density serial passaging, resulting in a total of 100-250-fold increase in CM number. Prevention of cell-cell contact stimulated hiPSC-CM cell cycle activity, providing 10 to 25-times greater expansion beyond GSK-3β inhibition alone. To disambiguate the role of Wnt/b-catenin signaling on hiPSC-CM proliferation upon CHIR-treatment and low-density passaging, we found that the maintenance of a more immature sarcomeric structure and gene expression was dependent on LEF/TCF transcriptional activity while cell cycle activation was regulated by cell contact-mediated AKT phosphorylation. To demonstrate the functionality of expanded hiPSC-CM, we engineered heart tissues and showed that the contractile performance is equivalent between expanded and unexpanded hiPSC-CMs. In summary, we uncovered a molecular interplay between GSK-3β inhibition and cell-cell contact that enables the massive expansion of functional hiPSC-CMs thereby facilitating large-scale drug screening efforts and engineering of therapeutically relevant-sized cardiac tissues.

During vertebrate development, cardiomyocytes undergo cellular rearrangements to form complex myocardial structures.  Malformations of these structures lead to diseases that include cardiomyopathies and cardiac arrythmias.  Previous studies have mostly focused on the formation of the trabecular network in the ventricle; however, morphogenetic processes that drive atrial myocardial complexity, which is crucial to propagate the action potential for cardiac contraction, have largely been overlooked.  Our study uses zebrafish larvae to elucidate cardiomyocyte behaviors during atrial development, as they allow for high-resolution live imaging and are easily amenable to genetic manipulations.

Using live 3D confocal imaging of zebrafish hearts, combined with mosaic labelling and temporal tracking of individual atrial cardiomyocytes, we found that atrial myocardial morphogenesis is driven by complex cell behaviors.  Specifically, we observed that atrial cardiomyocytes in zebrafish larvae form membrane protrusions and adopt an elongated shape in a non-stochastic orientation that establishes atrial tissue-level polarity.  These shape changes lead to restricted multilayering between neighboring cardiomyocytes, and the formation of new cell contacts, resulting in populations of elongated cardiomyocytes that span the atrium in an orientation parallel to the direction of blood flow.  These cell behaviors lead to the appearance of muscle ridges on the inner surface of the atrium.  Notably, these atrial cardiomyocyte behaviors appear to be independent from factors important in ventricular morphogenesis such as Erbb2 and Notch signaling.  Altogether, these data suggest that atrial morphogenesis is driven by oriented cell elongation as well as by distinct molecular and environmental/physical factors, all of which need further investigation.

Congenital heart disease (CHD) is the leading cause of death from congenital anomalies, with 7.4% of the cases caused by abnormal development of the atrioventricular (AV) canal. The AV canal gives rise to the atrioventricular valves and node as well as the septae in mammalian hearts, all of which are essential for heart function.  A previous transcriptomic analysis performed in our laboratory identified the transcription factor gene egr3 as a novel endocardial AV valve marker.  Here, we investigate egr3 function during AV valve development and valve leaflet formation using zebrafish as a model system.  We generated a promoter-trap line and several mutant alleles, including a full locus deletion, by Crispr/Cas9-mediated genome engineering.  We find that egr3 expression is specific to the presumptive endocardial AV canal, and that loss of Egr3 function leads to the absence of AV valve leaflets, resulting in severe retrograde blood flow.  In addition, we observed that egr3 expression in the AV endocardial cells was lost in non-beating hearts, and that it was expanded in perturbed cardiac flow conditions, suggesting that it is responsive to mechanical forces.  Altogether, this work identifies Egr3 as a critical regulator of AV valve formation, downstream of blood flow, thereby uncovering a new candidate gene for valve-related CHDs.

Introduction: Defective outflow tract (OFT) elongation causes congenital heart disease (CHD) resulting in newborn lethality. OFT formation involves the addition of second heart field (SHF) cells to the nascent heart tube, and regulation of SHF cell proliferation, migration, and differentiation in the OFT. Fibronectin (Fn1) is an extracellular matrix glycoprotein essential for the development of cardiac structures, including the OFT. Although, the requirement for Fn1 in the OFT elongation is clear, the cellular and molecular mechanisms underlying the role of Fn1 in OFT development remain unknown. Materials and Methods: We ablated Fn1 from the anterior mesoderm, including the SHF, using the Mesp1^Cre/+ knock-in strain of mice. To study the role of mesodermal Fn1, we performed immunostainings and western blots using Fn1^f/+;Mesp1^Cre/+ (control) and Fn1^f/-;Mesp1^Cre/+ (mutant) embryos isolated at different stages of development. Results: The ablation of mesodermal Fn1 results in a shortened OFT. This defect was due to an altered SHF cell shape and polarity, defects in cell-cell adhesion, proliferation, and premature cell differentiation in the SHF. Interestingly, even when mesodermal Fn1 was depleted, Fn1 expression between the SHF and endoderm was similar in control and mutants. However, the presence of endodermal Fn1 was not sufficient to rescue SHF cell defects in mutant embryos, suggesting that mesodermal Fn1 has a cell-autonomous role in OFT elongation. Discussion: In these studies, we have gained new insights into mechanisms regulating the addition of SHF-derived cells to the primitive heart tube during the formation of the OFT. Our work will lead to a deeper understanding of mechanisms regulating normal cardiac development and alterations that cause CHD.  Future work will be concentrated to understand if SHF cells migration is affected in vivo and why mesodermal Fn1 is required in a cell-autonomous manner.

Transgenic mouse models are an essential tool to understand mammalian genetics and cell biology. We used a piggyBac-on-BAC strategy to generate transgenic mice carrying a Bacterial Artificial Chromosome (BAC) containing the human ADAMTS19 enhancer and gene, with a tamoxifen-inducible Cre (CreERT2) replacing the first exon. hADAMTS19-CreERT2 mice (F0-F3) were crossed with Rosa26-tdTomato reporter mice and timed-pregnant females dosed with tamoxifen at E13.5. Surprisingly, E16.5 whole embryonic hearts showed three separate, but complementary and founder lineage-specific tdTomato expression patterns that together represent all mouse Adamts19-expressing cell types: atrial and ventricular (AV) cardiomyocytes, outflow tract (OFT) cells, and valve interstitial cells (VICs). Nanopore gDNA sequencing of a selection of founder lineages revealed that the hADAMTS19-CreERT2 BAC had uniquely integrated in a 5’ to 3’ orientation in either N-Cadherin (Cdh2) intron 14 (AV line) or Plexin A2 (Plxna2) intron 9 (OFT line). Immunofluorescence and mouse heart single cell data analyses confirmed that tdTomato expression patterns were consistent with Cdh2 or Plxna2 expression patterns, suggesting that the genomic region surrounding the BAC dominated CreERT2 expression. In contrast, analysis of the VIC subline revealed BAC integration in a reverse (3’ to 5’) orientation in intron 9 of RNA Polymerase II Associated Protein 2 (Rpap2), which is ubiquitously expressed. This reverse orientation allowed CreERT2 expression to be regulated by the hADAMTS19 BAC enhancer. We conclude that the genomic location and insertion orientqtion of BAC transgenes affects their expression pattern.

The heart is the first functional organ to form during vertebrate development. First, a linear heart tube is formed, followed by cardiac looping, chamber formation, and maturation. As for many organs, the heart arises from a simple epithelium with planar polarity properties. It has been established that cardiac chamber remodeling is coordinated through tissue-scale polarization of actomyosin. However, the mechanism governing the looping process is not fully understood. Here, using Zebrafish as a model, we describe the role of actomyosin to generate and distribute the tension forces necessary across the cardiac epithelium during cardiac looping and chamber formation. We found a supracellular actomyosin network required for this process to occur, which is regulated mainly by two kinases (Mylk3 and Rock2a) to achieve the proper distribution of phosphorylated myosin across the ventricular myocardium. Moreover, we also identified that the planar cell polarity pathway (non-canonical Wnt signaling) modulates the appropriate phosphorylation pattern in the cardiac epithelium by controlling the kinase activity in a spatio-temporal manner. Our findings describe a mechano-molecular mechanism necessary for proper looping and chamber formation during cardiogenesis.

Congenital heart disease (CHD) affects 0.8-1.1% of live births each year, making it the most common birth defect globally. Although the significant role of genes in the pathogenesis of CHD has been previously established, the etiology of 60% of CHD remains unexplained. To uncover additional genetic causes of CHD, we analyzed 3D microCT images of lethal and subviable mouse mutant strains generated by the International Mouse Phenotyping Consortium  and looked for evidence of CHD at embryonic days 15.5 and 18.5. We evaluated 430 mutant lines and 2,479 idividual mutants, and scored them for the following CHDs: atrial septal defects (ASD), ventricular septal defects (VSD), patent ductus arteriosis (PDA), truncus arteriosis (TA), double outlet right ventricle (DORV), atrioventricular septal defect (AVSD), transposition of the great arteries (TGA), pulmonary atresia, dextrocardia, situs inversus totalis (SIT), and cleft palate. From our data we identified 28.5% of null mutant mouse strains that exhibited at least a 25% penetrance of CHD. 65.1% of the mutant lines meeting our penetrance threshold were novel, meaning that they have not been associated with CHD previously. Thus, we have identified 79 new genes that give rise to CHD when mutated. We therefore hold that this type of analysis is valuable to identify novel genes implicated in CHD that have not been revealed by other conventional methods. We will extend our findings by performing pathway analyses to uncover novel pathways associated with CHD. We will also use sequence data from CHD patients to uncover novel mutations associated with CHD in patients. Results will eventually be used to help to guide clinical prognoses and treatments.  

Defective sarcomeric assembly is implicated in the pathophysiology of several inherited cardiomyopathies. While the mature sarcomere structure has been extensively studied, the molecular mechanisms that drive de novo sarcomere assembly are largely unknown. One of the earliest components in de novo sarcomere assembly is alpha-actinin 2 (ACTN2). ACTN2 localizes to the z body which fuses to form the z disc that serves as a linker for adjacent sarcomeric units.  Here we used human pluripotent stem cell-derived cardiomyocytes (hPSC-CMs), which closely follow this temporal progression of sarcomere formation, to interrogate the early ACTN2 interactome.  We performed RNAseq and mass spectrometry on hPSC-CMs and ACTN2 co-IP components at different times during cardiac differentiation. We find that ACTN2 sequentially associates with structural components such as intermediate filament and junctional proteins, and functional regulators such as calcium channel and metabolic proteins reflecting a coordinated program over time. The early cardiomyocyte coordinates translation and processing of sarcomeric proteins with their spatial organization and incorporation into developing myofibrils. ACTN2 in the early z body associates with chaperones and actin binding proteins, suggesting that the z body acts as a nidus for de novo actin filament assembly. In support of this role WNT and ROCK inhibition result in an increase of ordered sarcomeric filaments. We further identified ribosomal subunits and RNA binding proteins in the ACTN2 interactome, suggesting a role of the z body as a dynamic condensate. We confirmed this dynamic state of the z body using ACTN2 FRAP. We further show that epiblast-specific loss of Ddx3x, an early ACTN2 interactor and RNA binding protein, leads to embryonic lethality and cardiac defects in the mouse. Together these findings highlight the importance of the z body in the assembly process and provide a window to study the mechanisms of sarcomere organization.

Polyploidization is a normal cellular adaptation for a variety of cell types in mammals, including cardiomyocytes (CMs); however, the development of polyploidization understudied. Recent work suggests that genetics contribute to diverse displays of ploidy in the adult murine heart. Thus, we hypothesized that the developmental progression to differing end-states may similarly be affected. Here, we assessed CM endowment, cell cycle, and ploidy temporally across two diverse inbred mouse strains, A/J and C57Bl/6J. Consistent with previous work, C57Bl/6J hearts displayed rapid cell cycle activation in the first postnatal week. Total CM numbers are relatively unchanged after P7, while final ploidy states are generally constant from P14 on. In contrast, A/J mice displayed depressed cell cycle activation in the first week of life.  Polyploidy in A/Js reaches its peak at P21, whereby only ~3.5% of CMs remained mononuclear and diploid. Interestingly, from 3-6 weeks of age, we observed an expansion of total CM numbers and of the 1x2N subpopulation to ~8%, which could not be explained by proliferation of the residual diploid population. Instead, the expanded diploid population appears to come from a polyploid CM. This finding was confirmed by analysis identifying a population of CMs which had completed cytokinesis by 6 weeks which was not present at 3 weeks. We believe this is the first report of ploidy reversal by a mammalian CM, a phenomenon first observed by the hepatocyte field. Ongoing single nuclear RNA seq is examining the transcriptomic differences between A/J and C57Bl/6J CMs at P21 to identify this unique population competent to undergo ploidy reversal. Preliminary results suggest A/J CM nuclei, regardless of their ploidy state, are more likely to be in the cell cycle in comparison to C57Bl/6J. Understanding the developmental paths to end state CM endowment and ploidy states helps us understand the biology of organ size and can be applied to stimulating regeneration.

Macrophages are well-characterized as sentinel immune cells that coordinate cellular responses to injury and infection. However, recent developments in cardiac macrophage biology have broadened our understanding of this critical cell type in heart development and function. Still, it is not known how loss of embryonic-derived macrophages affects heart health into adulthood. Here, we are the first group to show in larval and adult zebrafish as well as in adolescent mice that macrophages reside at the sinoatrial node, where they couple to nodal cardiomyocytes via connexin-45. Using a combination of histology, echocardiograms, electrocardiograms (ECGs), and physiological stress tests, we demonstrate that loss of embryonic macrophages significantly disrupts adult heart health and function, leading to arrhythmia and fibrosis. Adult irf8^st96/st96 embryonic-macrophage knockout zebrafish have significant disruption in epicardial integrity, detachment, and expansion, known indicators of heart stress or damage. Mutant zebrafish also exhibit increased collagen deposition within the myocardium and along the atrial wall. Longitudinal analysis of echocardiographic measurements reveals significant age- and sex-specific changes in irf8 mutant heart function, including arrhythmia with notable variability in diastasis time as well as consistently prolonged isovolumic relaxation time. ECG traces of macrophage mutants following a 1-hr swim tunnel-based endurance test reveal significantly prolonged P-wave durations and PR intervals compared to wildtype. A subset of mutants also exhibited intermittent P-wave absence and potential flutter, indicating sinus or atrial dysfunction. Overall, these data reveal significant cardiac abnormalities following the specific loss of embryonic-derived macrophages and expand our knowledge of critical macrophage functions governing long-term heart health. 

Congenital Heart Defects (CHDs) are structural malformations that occur during development and affect 1% of live births worldwide. Correct heart development requires tight spatiotemporal control of both the Rac1 and Notch signalling pathways, and mutations in components or regulators of both pathways can lead to CHDs. Adams-Oliver Syndrome (AOS) is a multisystemic disorder in which 20% of patients also suffer from CHDs. All known causative mutations in AOS have been identified in Rac1 or Notch signalling pathways, including homozygous recessive mutations in DOCK6 and EOGT which regulate Rac1 and Notch signalling respectively. However, it is unknown how mutations in these genes lead to onset of this disease. We aim to understand how mutations in DOCK6 and EOGT lead to dysregulation of Rac1 and Notch signalling during heart development and cause CHDs in patients with AOS, using zebrafish as a model. 

We find that mRNA of both dock6 and eogt is expressed transiently during early zebrafish cardiovascular development. eogt mRNA expression is restricted to a subset of developing vascular structures, while dock6 exhibits slightly broader expression. Interestingly, both genes also have non-coding transcripts which are expressed in the myocardium during early heart morphogenesis. Using CRISPR mediated genome editing we have created zebrafish disease models of AOS with mutations in both dock6 and eogt, in which we are assessing cardiac morphology throughout development. In addition, we are investigating whether non-coding transcripts of dock6 and eogt play a role in heart development, or the onset of CHDs in AOS patients.

Establishment of the left-right body axis is critical for the placement, morphogenesis and function of organs, with disruption to laterality often associated with congenital heart malformations – known as heterotaxy. Left-right specification has long been proposed to be dependent on asymmetric fluid flow in the embryonic node, driven by motile (nodal) cilia. Nodal flow ultimately results in asymmetric expression of Nodal in the left lateral plate mesoderm, which directs organ morphogenesis.

Positioning of nodal cilia is governed by the highly conserved Wnt planar cell polarity (PCP) pathway, yet the exact effectors of PCP signalling remain elusive. We have examined the role of the highly conserved JNK (c-Jun N-terminal Kinase) gene family – a proposed downstream component of PCP signalling – in the development and function of the zebrafish node.

We show that together jnk1 and jnk2 function to specify length of nodal cilia, generate asymmetric flow in the node and restrict Nodal to the left lateral plate mesoderm. Crucially, loss of asymmetric Nodal expression does not result in disturbances to heart looping morphogenesis, supporting an emerging model that nodal flow may be dispensable for organ laterality.

This work uncovers multiple roles for the JNK gene family acting at different points during embryonic development. Furthermore, it reinforces observations that other node-independent pathways function in parallel to establish cardiac laterality and that genetic causes underlying heterotaxy are likely to be multi-faceted.

Congenital heart disease (CHD) is a major factor in mortality and morbidity in children and adults. To advance our abilities to identify and manage CHD, we need to further understand the key genetic factors involved in causing these maladies. Drosophila melanogaster shares similar cardiac developmental mechanisms with humans and has been an essential model for human heart development. One similarity occurs between Tinman (Tin), a transcription factor in Drosophila vital for cardiac cell differentiation, and its mammalian ortholog NK2 Homeobox 5 (Nkx2.5). These proteins share two conserved regions: the homeobox and the tin (TN) domain. Although the TN domain is completely conserved between these two proteins, there is little known about its significance. By utilizing CRISPR/Cas9 gene editing, I established a line of Drosophila containing an in-frame deletion of the TN domain. Staining Drosophila embryos for a marker of cardial cells revealed mutant embryos generate a significantly more cardial cells than wild type (p<0.01). I also stained embryos for Tin and Svp, a Drosophila protein important for heart development, to determine if the TN domain deletion affected cardial cell specification. Mutant embryos contained more cells expressing tin and more cells expressing svp than wild type embryos. These results support the importance of the TN domain in heart development and indicate the increase in cardial cells occurs before cardial cells differentiate into cells exclusively expressing tin or svp. Characterizing the role of the conserved TN domain in Tin and Nkx2.5 can expand our understanding of cardiac formation and maturation

The heart starts beating early in development, and maintains its function throughout dramatic developmental changes in electrophysiology, tissue morphology, and cell fate specification. However, little is known about the heart’s initial transition from silent to active - what are the intermediate dynamics between a quiescent tissue and a periodic and spatially coordinated heartbeat? Capturing the first heartbeats is a severe technical challenge because these sub-second events only occur once in each animal’s life, and could emerge at any point in a window of hours.  To study these initial dynamics, we developed an imaging system to observe calcium dynamics in up to 72 developing zebrafish embryos simultaneously. The first beats, occurring 20-21 hours post fertilization, were initially infrequent and irregular but from onset were coherent propagating waves that swept the heart field. We then developed genetic tools for simultaneous targeted optogenetic perturbation and voltage or calcium imaging, which revealed that the heart cone supports electrical wave propagation and becomes increasingly excitable before spontaneous activity emerges. Using spatially resolved calcium imaging, we mapped the spatial origin of the first heartbeats. This origin is highly variable between individuals, can move suddenly within a short time interval, and is not correlated with genetic markers of future pacemaker cells. Together our data suggest that broad excitability and electrical connectivity impose biophysical constraints that enforce initial spatial organization and timing of the first heartbeats, despite little specialization of cardiac cell function.

During development, the heart tube comprises two cellular layers, an outer myocardium and an inner endocardium, between which lies a layer of extracellular matrix (ECM). This cardiac ECM creates specialised extracellular environments important in mediating the chemical and biomechanical signals that shape the heart tube into a three-dimensional (3D) organ. Visualising and linking ECM and tissue morphology is challenging since sample processing may impact matrix hydration and tissue structure. Using morphoHeart, a novel and in-house-developed quantitative image analysis tool that allows the 3D segmentation of the heart layers  -including the ECM- from live embryos, we have identified an early left-right axis of ECM-regionalisation linked to heart looping and orientation of chamber ballooning.

To understand if defects in heart tube lateralisation result in abnormal ECM-regionalisation and hence disrupted morphogenesis, we investigated zebrafish hearts in which spaw (zebrafish homolog of Nodal responsible for left-right laterality) is disrupted. Analysis of cardiac ECM distribution in spaw mutants identified that ECM asymmetry in the early tube is maintained but mispositioned. Irrespective of subsequent looping direction, this early ECM asymmetry translates into a highly regionalised ECM in the inner and outer curvatures of the atrium at looping and ballooning stages. Further linking laterality, ECM and heart morphogenesis, quantification of heart size in spaw mutants demonstrated a role for Nodal in timely chamber growth and chamber-specific ECM ‎remodelling, even in dextrally-looped mutant hearts. Together we propose a model whereby laterality cues orient an early asymmetric cardiac ECM that intrinsically sets an axis for chamber growth and looping morphogenesis to occur.

Dilated cardiomyopathies (DCM) including ischemic- and genetic-related DCM are characterized by cardiomyocyte necrosis and fibrosis associated with impaired cardiac function leading to severe heart failure. LMNA is mutated in 10% of DCM cases in humans and is responsible for rapidly progressing DCM with severe cardiac defects. While the role of LMNA has been extensively studied in adult heart, its role during heart development remained to be determined. We recently identified the fibroblast growth factor 10 (FGF10) as a potential target to promote cardiac regeneration in ischemic DCM.

This study thus aims to uncover the role and the cell-type requirement of Lmna during heart development, maturation and homeostasis and to evaluate FGF10 as a therapeutic target in Lmna-related DCM.

Transgenic mouse lines will be used to perform temporal (development, maturation and adult) and cell-type specific (cardiomyocytes and cardiac fibroblasts) deletion of the Lmna and conditional Fgf10 overexpression.

Expression profiling revealed increased LMNA myocardial expression from embryonic to postnatal stages. Interestingly, our results revealed that LMNA preferentially accumulates in cardiac fibroblasts in fetal and postnatal heart. Cardiomyocyte specific Lmna deletion in fetal, postnatal and adult heart displays rapid cardiac function alteration with dramatic left ventricular dilatation and massive fibrosis, unveiling a critical requirement in developing cardiomyocytes. Interestingly, specific Lmna-deletion in embryonic cardiac fibroblasts leads to embryonic lethality, revealing an unsuspected role for LMNA in cardiac fibroblasts. Finally, we evaluated the impact of FGF10 treatment on Lmna-related DCM and our results suggest that FGF10 preserves cardiac fibrosis infiltration and cardiac remodelling in Lmna-related DCM.

Altogether, this study unveils the key developmental role of LMNA and identifies FGF10 as a therapeutic target for cardiac regeneration in Lmna-related DCM.

Bicuspid aortic valve (BAV) is the most common congenital heart defect that affects around 1.5% of population. The majority of patients develop aortic valve stenosis, aortic root dilatation or infective endocarditis. As a consequence, BAV accounts for nearly 50% of all aortic valve replacements. The genetical heterogeneity of BAV makes identification of its genetic causes complicated, thus, to date they remain largely elusive.

Based on unpublished GWAS study of 2236 BAV patients and 11604 controls, we identified a new risk locus for BAV – MUC4. The goal of this study was to functionally characterize muc4 in zebrafish and analyze its involvement in cardiac valve morphogenesis.

We used two alternative approaches for functional analyses of muc4 in zebrafish atrioventricular (AV) valvulogenesis: transient F₀ CRISPR/Cas9-mediated gene knockout and antisense oligonucleotide Morpholino- (MO)-mediated gene knockdown. The 1-cell stage embryos were injected either with a mix of Cas9 protein and muc4-gRNAs or with muc4-MOs. At 56-58 hours post fertilization (hpf) the crispant and morphant embryos were fixed, genotyped and imaged using confocal microscopy against Alcam and kdrl:GFP.

Using both approaches, we observed a delay in AV valvulogenesis. In the wild-type at 56 hpf, a few AV endocardial cells had ingressed into the extracellular matrix (ECM) to form a multilayered valve leaflet. However, in muc4 crispants or morphants, the valve remained monolayered or only one cell had ingressed into the ECM. This phenotype had a penetrance of approx. 50%.

Here, we have functionally characterized a new BAV candidate gene, MUC4. Knockout and knockdown of this gene in zebrafish revealed that muc4 may impact cardiac AV valvulogenesis by causing a developmental delay during the multilayering of valve leaflets. Our research constitutes the basis for future studies on the relevance of MUC4 in BAV as one of the candidate genes, contributing to this polygenic pathology.

The Wilms tumor 1 homolog (Wt1) encodes a zinc finger protein essential for heart development. Over the last ten years, several Wt1 transgenic mouse models have been generated and used by several laboratories to study the behaviour of WT1 expressing cells during heart development and repair. In this study, we generated a new mouse model of the Wt1 gene, the Wt1^LoxP/GFP;Wt1^Cre mice. In these mice, one copy of the Wt1 allele is flanked by loxP sites, whereas the other copy has a GFP knock-in in exon 1, which disrupts WT1 expression. In addition, the Wt1^Cre line used to generate this model is a BAC transgenic line that has been extensively used in the cardiovascular field. The genotyping of postnatal mice of our model revealed a sub-Mendelian distribution of the Wt1^LoxP/GFP;Wt1^Cre mice. The observed lethality of Wt1^LoxP/GFP;Wt1^Cre mice probably corresponds to the late embryonic stages since the proportion of this genotype was not reduced when embryos were examined on days E14.5. Interestingly, all mutant mice lacked mature gonads and displayed genital tracts containing both male and female genital structures and ambiguous genitalia. In addition, the histological examination of embryonic Wt1^LoxP/GFP;Wt1^Cre mice demonstrated severe heart defects, which could be the main cause of lethality of some mutant mice. Our results call for cautious interpretation of data obtained by conditional gene deletion experiments using this Wt1^Cre line, as mouse models in which Wt1 has been deleted in the embryonic heart are embryonic lethal.

The prevalence of CHD in 22q11.2 deletion syndrome (22q11.2DS) patients is 65% as compared to the general population at ~0.7%, of which most have conotruncal heart defects (CTD).  To understand the basis of variable phenotypic expression, we analyzed whole-genome sequence data of 456 CTD cases and 537 controls with 22q11.2DS retaining the most damaging coding/splicing rare variants (MDRV), as identified by two independent variant calling pipelines with multiple annotation algorithms and examined 19 gene-sets for the association.  We found a significantly higher burden of MDRVs in three interdependent gene-sets comprising evolutionarily constrained genes (OR=3.31; P-value, 4.00E-07), chromatin regulator genes (OR=4.16; P-value, 7.00E-05) and known congenital heart disease genes (OR=6.63; P-value, 5.9E-04) based upon a weighted gene-set based test.  When removing chromatin regulators and considering constrained genes, most are neuronal synapse genes, while the CHD genes are associated with cardiac development.  Further, by focusing on recurrently affected genes, we found the chromatin regulatory genes were overrepresented in cases, but no enrichment was found in controls.  Overall, using both methods, there are a total of 46 affected chromatin genes in 42 CTD cases accounting for 9.21% of the total, with six recurrently affected chromatin genes including EP400, KAT6A, KMT2C, KMT2D, NDS1, and PHF21A, involved in histone peptidyl-lysine modification.  Although ubiquitously expressed, the chromatin genes have enriched expression in cardiac progenitor cells in the pharyngeal apparatus, where TBX1, the transcription factor gene for 22q11.2DS is expressed.  Modifier genes identified might act in the genetic and epigenetic pathways of TBX1 function.

Clinical manifestation of atherogenic cardiovascular disease differs between men and women. Men often present with classical symptoms i.e., chest pain radiating to left arm and jaw. Women, on the other hand, seem to show a wider range of symptoms that are considered less typical, including but not limited to back pain and nausea. The anatomical pathways feeding cardiac pain perception have not been properly characterized with respect to sex-related differences.

In our study, we explored the UK Biobank cohort to perform sex-stratified genome-wide association studies for myocardial infarction, coronary artery disease and a variety of pain types to identify instrumental variables for Mendelian randomization. This approach enables determination of sex-differences of the causality between heart disease and its clinical manifestation, thus allowing for large-sample size evidence of the differences in atherogenic cardiovascular disease manifestation between sexes.

We observed that a genetically-influenced chest pain was associated with a higher risk of atherogenic cardiovascular disease in men (OR: 1.49 95% CI: 1.01 – 2.19). Surprisingly, for women, this chest pain was not linked with atherogenic cardiovascular disease (OR: 1.06, 95% CI: 0.96 – 1.17), suggesting chest pain by itself is a less reliable predictor of heart disease. Furthermore, we characterized sex-stratified causal relationships of other pain types with myocardial infarction and coronary artery diseases.

Our study identified differences in clinical manifestation of myocardial infarction and coronary artery disease between men and women, that form the basis for our research on developmental and anatomical differences in pain perception between the sexes. 

Critical developmental roles for matrix molecules in proliferation and differentiation and in generating tissue structural integrity are evident by the range of clinical disorders induced by mutations in extracellular matrix (ECM) genes. In the cardiovascular system, recent data underscore the dynamic nature of ECM composition during embryogenesis. This emerging focus on the function of ECM in cardiac development and disease has advanced our understanding of the morphogenetic basis of congenital heart defects (CHD). Yet, we have limited appreciation of the cellular and biophysical mechanisms regulated by individual ECM proteins during formation of the cardiac outflow tract (OFT). Our data reveal that Fibulin molecules, a family of glycoproteins that regulate cell behaviors and structural contributions to the ECM, have previously unrecognized requirements in establishing the proper dimensions of the vertebrate OFT. Zebrafish fibulin mutants also exhibit decreased TGF-β signaling in second heart field (SHF) progenitors of the linear heart tube. Consequently, fibulin-deficient SHF progenitors supply fewer endothelial and smooth muscle cells, resulting in a narrower arterial pole. In addition to effects on SHF progenitor differentiation, we find that deposition of Elastin, a crucial ECM component of the OFT, is impaired as evidenced by decreased expression, fiber number and alignment in fibulin mutant embryos. Importantly, tissue stiffness of the OFT auxiliary chamber is impaired by disrupted ECM integrity and these factors govern biomechanical function. Our insights will yield valuable strategies to improve engineered biomaterials for surgical intervention in congenital OFT anomalies and will fuel further identification of novel disease genes in CHD.

Diabetic pregnancy has increased risk of asymmetric septal hypertrophy in the fetus, and we aim to elucidate how the nutritional and metabolic environment of the embryo regulates cardiac morphogenesis. In a previous study, we revealed that cardiomyocyte maturation is impaired in a high glucose environment in vitro, however it remains unclear how ventricular chamber undergoes asymmetric hypertrophy at organ scale.

To understand morphogenetic mechanisms, it is essential to clarify the hierarchical relationship among organ, tissue and cellular activities. Histological analysis revealed disarrayed muscle fibers in the hypertrophic tissue at cellular scale, however little is known about the tissue scale deformation process. Based on histological observation, we hypothesized that the tissue scale morphogenetic process of hypertrophy is isotropic growth caused by disruption of proper orientation in tissue elongation. To test this hypothesis, we developed a new method to estimate the size and direction of local tissue growth during heart morphogenesis. Specifically, 3D whole organ imaging is performed to detect the cells of the same origin in a fixed heart using MADM-based tracking. Then, based on the detected spatial distribution of these cells, thickening of myocardial wall and the direction of tissue stretch at each local tissue are calculated. In this presentation, we provide a proof of concept for this method through the analysis of artificial data. We discuss how to apply this method to reveal the anisotropy of tissue growth and how to dissect the cellular mechanisms that cause tissue scale dynamics.

This new methodology for estimating the tissue scale deformation process during organogenesis will help us understand the multiscale morphogenetic mechanisms of organ development.

Nr2f transcription factors play an integral role in vertebrate atrial development and have been linked to congenital heart defects (CHDs) in humans. Adult patients with structural CHDs affecting the atrium often present with secondary complications, such as pulmonary hypertension and vascular remodeling. Here, we report a zebrafish nr2f1a mutant allele with embryonic atrial defects that can survive to adulthood. Remarkably, despite essentially lacking an atrium, the adjacent sinus venosus (SV) in these mutants undergoes adaptive remodeling, which entails substantial thickening due to increased cell proliferation. Transcriptomic analysis of the bulbus arteriosus (BA) and SV, transgenic reporters, and genetic lineage tracing showed previously unknown molecular and cellular similarity between the wild-type BA and SV, including the presence of vascular smooth muscle (VSM) and neural crest cells. Furthermore, single cell RNA-seq revealed novel heterogeneity of VSM cell populations in the BA and SV and that VSM and endothelial cells within the adapting SV of the nr2f1a mutants take on transcriptional signatures of the BA. Echocardiography showed retrograde blood flow in adult nr2f1a mutant hearts, suggesting that aberrant blood flow contributes to the adaptive remodeling of the SV. Consistent with this hypothesis, prolonged treatment with a vasodilator mitigated thickening, while increased exercise exacerbated the adaptive thickening. Altogether, our data demonstrate an unexpected latent similarity between the large chamber-like vessels at the arterial and venous poles of the zebrafish heart and adaptive remodeling of the SV in nr2f1a mutants due to hemodynamic forces, which may provide insights into complications associated with CHDs in humans.

The ascending aorta harbors a heterogenic smooth muscle cell (SMC) population. Embryonic lineage tracing studies have shown that extracardiac cells contributing to the vascular outflow tract (OFT), originate in the cardiac neural crest and second heart field (SHF). The cardiac neural crest is important for proper outflow tract septation and formation of the inner SMCs layer. The SHF mainly contributes to the SMC population of the pulmonary trunk and a small part of the ascending aorta. Currently, the origin of the outer medial SMC and adventitial fibroblasts is unknown. Recently we described a distinct Wilms Tumor 1 (WT1) positive cell population in the area of the OFT. We hypothesize that these cells contribute to the aortic SMC and fibroblast population. Using the Cre-LoxP recombination system driven by an epicardial specific promotor (WT1-Cre^ER2) and the ROSA^mT/mG reporter line we studied the fate of this cell populationfrom stage E9.5 till E17.5 in mouse embryos.

Immunohistochemistry revealed that this  population of cells shares other SHF markers.. GFP-labeled epicardial cells undergo -mesenchymal transformation and migrate into the arterial vessel wall of the ascending aorta. During this process,these cells lose their epicardial phenotype and contribute to the mural cells of the outer layer of the aorta and pulmonary trunk. Our results demonstrate a new embryonic origin of SMC which specifically contributes to media and adventitia of the ascending aorta but not to the aortic arch or descending aorta. Future studies will focus on their specific role in normal development and during pathological circumstances.

Abnormal cardiac innervation plays a role in arrhythmogenicity after myocardial infarction(MI). Nerve quantification is challenging as visualization of  nerves requires high magnifications. We developed a method to analyze nerve density(ND) in large transmural biopsies.

Myocardial biopsies from 4 swine, 3 months after MI,  were stained with Sirius Red(fibrosis) and β3Tubulin(autonomic nerves). Fibrosis cut-offs classified biopsies in: MI core(>59.8%), borderzone(BZ)(14.3-59.8%) or remote zone(RZ)(<14.3%). Using a custom software pipeline, biopsies were graphically divided into 1x1mm grids. Nerve and myocardial tissues were thresholded and after  quality check and artefact removal, divided into 0.1x0.1mm squares. ND/square was calculated and classified into denervation, hypoinnervation, normal innervation and hyperinnervation according to control derived cut-offs(5th, 95th %, n biopsies=38). Squares were located back to their original biopsy position for visualization and quantification of innervation types.

All 83 MI biopsies showed variable innervation patterns. Denervation, hypo- and hyperinnervation were largest in MI core and BZ. Innervation was normal in the RZ: 94.9%(93.8-96.0%), decreased in the BZ: 87.9%(85.1-92.4%, p<0.001), and even lower in the MI core: 72.8%(66.3-75.8%, p<0.001). Denervation and hypoinnervation were highest in core biopsies, 16.7%(13.5-23.5%) and 3.8%(2.9-5.0%) resp., lower in the BZ, 2.6%(1.6-6.5%, p<0.008) and 1.2%(0.8-1.7%, p=0.015) and even lower in the RZ, 1.2%(0.7-1.8%, p<0.001), 0.6%(0.3-1.3, p<0.001). Hyperinnervation was observed mainly in the BZ, 5.3%(3.7-9.4%) and in the core, 5.3%(3.7-9.4%, p>0.999), whilst being low in the RZ 2.6%(1.7-4.3%, p=0.002 and p=0.038 respectively).

This method allows detailed visualization and quantification of ND after cardiac damage and can be broadly applied to high resolution nerve imaging. Variable innervation types within the same MI-biopsies indicate a heterogeneous arrhythmia substrate.

In the past decades, attention on sex differences in the prevalence and outcomes of a wide range of cardiac diseases has increased. Next to overt sex differences in disease presentation and outcome, also differences in autonomic function between males and females have been exposed. After myocardial infarction, male patients have an increased risk for ventricular arrhythmias and sudden cardiac death. Part of these arrhythmias have been attributed to an increase in cardiac sympathetic nerve fibers occurring after cardiac damage. Although mechanical studies of post-myocardial infarction cardiac sympathetic hyperinnervation have raised growing awareness on the role of the autonomic nervous system in arrhythmogenesis, data on the role of sex herein are still scarce and not conclusive. We found that co-cultures of male superior cervical ganglia with male myocardium and male mesenchymal epicardium-derived cells (EPDCs) results in significant higher neurite directional outgrowth towards myocardium compared to entirely female co-cultures. Moreover, male EPDCs in a female setting enhanced the female neurite outgrowth comparable to an entire male environment. We are currently exploring by RNA sequencing of male and female EPDCs, whether  male and female EPDCs have a differential expression of axon repellent guidance cues. Our results confirm the stimulating effect of EPDCs on neurite outgrowth and also demonstrate that sympathetic nerve outgrowth and density differs between male and female in vitro. Our data underlines the potential relevance of sex differences in post-myocardial infarction cardiac hyperinnervation.

Congenital heart defects (CHD) affect 1% of all live births in the US annually. The etiology of CHD is multifactorial, that is, both pathogenic genomic variation and environmental risk factors act as CHD contributors. Among the environmental teratogens, maternal pre-gestational diabetes mellitus (matDM) is associated with up to ~5-fold increase in the risk of having an infant with CHD.  Generation of excess reactive oxygen species (ROS) due to maternal hyperglycemia is routinely observed in embryonic hearts exposed to matDM but the teratogenic effect of ROS on cardiac developmental pathways is not fully understood. The Notch and Nitric oxide (NO) signaling pathways, which are highly expressed in the endocardium, are critical for normal heart development.  Previously, we identified a reduction in NO bioavailability and Notch signaling in mouse embryonic hearts exposed to matDM and reported a gene-environment interaction between Notch1 haploinsufficiency and matDM in the development of CHD. Here, we investigated the effect of modulating intracellular levels of oxidative stress in endocardial and endocardial-derived cell lineages to determine the role of ROS in matDM-associated CHD. Endocardial-derived cells exposed to oxidative stress elicit similar responses as hyperglycemia in reduction of NO bioavailability and Notch signaling. To study the role of oxidative stress in vivo, we generated WT and Notch1^+/- embryos overexpressing the antioxidant gene, SOD1, to determine whether mitigating oxidative stress can rescue matDM associated CHD. We have confirmed a reduction of oxidative stress in SOD1-tg embryonic hearts and preliminary data shows lower incidence of CHD in SOD1-tg compared to WT embryos exposed to matDM. The role of oxidative stress in matDM-associated CHD requires further investigation, as mitigation of endocardial oxidative stress may be a promising approach to lower the incidence of CHD in high-risk populations.

Mammalian heart development depends on the contributions of two major sources of cardiac progenitor cells, the first and the second heart field (FHF/SHF). While FHF cells establish the early heart tube and are major contributors to the left ventricle (LV), SHF progenitors enter the heart tube at the poles and are essential for outflow tract (OFT) and right ventricle (RV) formation. The cardiac transcription factors (TFs) Gata4 and Hand2 are linked in a key gene regulatory module orchestrating SHF progenitor specification, differentiation, proliferation and migration. Mouse embryonic hearts deficient for these TFs exhibit OFT and RV defects, pointing to essential roles of the cis-regulatory landscapes of these genes in controlling SHF progenitors and mirroring congenital heart defects. Using transgenic reporter assays, previous studies have identified sets of enhancers with overlapping activities in SHF progenitors in each locus. However, the related subregional specificities and functions of these enhancers have not yet been characterized in vivo, leaving their functional necessity and involvement in heart disease unexplored. Here, we have generated mouse lines harboring individual deletions of Hand2 and Gata4 cardiac enhancers. Our results based on qualitative and quantitative transcriptional profiling and cardiac marker analysis have revealed considerable cis-regulatory robustness in both, Gata4 and Hand2 topologically associated domains. Nevertheless, we have identified a genomic cardiac enhancer module essentially contributing to Gata4 gene dosage in a heart compartment-specific manner. Our results highlight the complexity of cardiac TF enhancer landscapes and provide a framework for assessment of heart disease-related loss-of-function effects of genomic heart enhancers in vivo.

The cardiac neural crest is a transient embryonic cell population that contributes to portions of the heart, including the septum of the cardiac outflow tract. Here, we elucidate regulatory changes that may confer ectomesenchymal ability onto the cardiac neural crest, enabling them to contribute to elements of the heart. To this end, we analyzed transcriptional and epigenetic changes in the cardiac neural crest as a function of time from induction to migratory stages. Analysis of single-cell RNA-seq data reveals groups of genes with differential expression profiles, with those associated with ectomesenchymal potential expressed at premigratory stages and then up-regulated during migration. In contrast, genes associated with specification show a reciprocal profile, peaking at the premigratory stages. We next looked for transcription factors co-expressed in the premigratory cardiac but absent from the cranial neural crest at a similar developmental timepoint, identifying FoxC2, FoxP1, and Twist1 as putative regulators of ectomesenchymal fate. Targeted knockouts in the cardiac neural crest using CRISPR-Cas9 revealed that individual loss of each transcription factor results in severe persistent truncus arteriosus. Finally, to build regulatory linkages, we analyze the chromatin landscape of cardiac neural crest cells using Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq). The results reveal numerous enhancers activated in the cardiac crest population at premigratory stages that are further upregulated during migration, characterized by GO terms associated with heart morphogenesis. Taken together, our results suggest that cardiac crest-specific transcription factors expressed during the specification stage may be critical for mediating their ectomesenchymal differentiation at later stages of heart development.

Congenital heart defects (CHD) are a common cause of fetal death and termination of pregnancy. The aim of this study was to phenotype all fetal cardiac specimens of our  collection, and to create a user-friendly database.

Each specimen was thoroughly examined according to segmental analysis. The main CHD was determined according to the 11 categories and 23 subcategories of the clinical and anatomical classification of CHD. Associated lesions were coded using IPCCC ICD-11 Nomenclature. Codes and photographs for each of the 1236 specimens were linked into the database. Among them, 120 were normal hearts (10%). The three main groups of CHD were ventricular outflow tracts anomalies (32%), functionally univentricular hearts (23%), and anomalies of atrioventricular junctions and valves (14%). The most frequent lesions were valvar anomalies (55%), ventricular septal defects (VSD, 40%), ventricular hypoplasia (35%), interatrial communications (35%). A cladistic grouping can superimpose on the anatomical classification to identify similar characteristics and raise embryological hypotheses. For example, the term “conotruncal” anomaly comprises defects supposed to have the same embryological origin/”common ancestor”. A phenetic approach explores how a segmental characteristics associates with other defects. For example, common arterial trunk is almost always associated with outlet VSD, but also with intact interventricular septum, hypoplastic ventricle, or right aortic arch. Clustering segmental cardiac characteristics without a priori developmental hypotheses may help identifying new mechanisms for CHD. Detailed anatomical phenotyping of fetal hearts should allow us to identify rare associations of malformations, and can help elaborating different approaches of describing and grouping CHD.

Cardiovascular calcification (CVC) is one of the most frequent forms of cardiovascular disease, with increasing prevalence showing high morbidity and mortality rates worldwide.  CVC is characterized by the progressive dysfunction of the cardiac muscle, arteries, and cardiac valves due to ectopic calcification of their extracellular matrix (ECM).  Previously considered a passive accumulation of calcified tissue, it is now recognised that these diseases result from a cell-mediated active process.  In fact, it has been suggested that before calcification cardiovascular cells differentiate to an osteoblastic (bone-cell) fate.  Previous work has shown that overexpression of Runt-related Transcription Factor 2 (Runx2) in mouse is sufficient to induce calcification in the heart and aorta.

Here we use the zebrafish to study the initiation of CVC in vivo using genetic tools and live-imaging at single-cell resolution.  We overexpressed runx2a in cardiovascular cells using cell-specific promoters and confirmed the differentiation of these cells to a pro-osteoblastic lineage. Importantly, these cells were in direct contact with ectopic calcification sites and triggered the recruitment of immune cells.  To determine the role of the immune system in the progression of the CVC phenotype we are now using cell type-specific genetic ablation tools.

Overall, we introduce the zebrafish as a unique model to study the events leading to the initiation and progression of CVC, aiming to identify new therapeutic targets.

Patients presenting with congenital heart diseases (CHDs), Ebstein’s anomaly (EA) and left ventricular noncompaction (LVNC), suffer higher morbidity than either CHD alone. The genetic etiology and pathogenesis of combined EA/LVNC remain largely unknown. In a familial case of EA/LVNC associated with the KLHL26 (p.R237C) variant, we discovered an altered electrostatic surface profile of the variant protein, which likely decouples the CUL3-interactome and alters protein turnover. 

To investigate the underlying mechanisms, we differentiated a family trio (EA/LVNC-unaffected daughter and EA/LVNC-affected mother and daughter) of induced pluripotent stem cells into cardiomyocytes (iPSC-CMs), assessed for significant differences in CM structure and function (P<0.05), and synthesized cellular and pathway insights. We characterized morphology via immunofluorescence and electron microscopy and contractile function via biomechanical and automated video methods. We analyzed differential mRNA expression via RNA Sequencing of iPSC-CMs and cardiac tissues, and iPSC-CM protein abundances via label-free quantification.

In comparison to the unaffected, EA/LVNC-affected iPSC-CMs exhibited aberrant morphology and contractile ability, mainly a distended endo/sarcoplasmic reticulum (ER/SR), decreased beat rate, and altered calcium transients. The affected daughter iPSC-CMs further showed decreased apoptosis and increased proliferation. From pathway enrichment analyses, we saw a suppression of myocardial contraction and an activation of ER lumen and lipid metabolism. Via RT-PCR, relative to GAPDH and unaffected iPSC-CMs, we found significantly decreased expression of Sarcoplipin (SLN) and increased Junctophilin-2 (JPH2).

Together, these results from cell and developmental studies suggest that this familial EA/LVNC associated with the KLHL26 variant develops dysregulated ER/SR signaling and altered calcium transients with subsequent lesions in CM contractility and cell cycle progression.

Abnormalities in early development of Cardiac Progenitor Pool (CPC) located in the Second Heart Field can progress to a myriad of congenital heart defeats that can contribute to neonatal mortality. Thus, understanding the regulatory mechanisms that contribute to cell lineage differentiation and fine tuning in early human development will help to prevent or cure heart defects. Specifically, we focus on Retinoic Acid since it is a key regulator in cardiogenesis where it promotes the posterior specifications of CPCs and their progression toward atrial Cardiomyocyte (CM) lineages. Despite its importance, the specific mechanisms of RA are not well understood. In Embryonic stem cells YAP functions differentially and is controlled by multiple different signaling pathways. We have identified YAP1;TEAD 4 factors as important cell fate determinants of atrial CM in cooperation with RA signaling. It has been shown that RA induces NR2F2 (Nuclear Receptor Subfamily 2 group F member 2) expression to confer atrial identity during CPC differentiation. NR2F2 is an orphan nuclear receptor that binds DNA and it forms heterodimers with the retinoid X receptor to regulate transcription of RA target genes. Our data suggest that RA recruits YAP to the chromatin through NR2F2 and that there is a collaborative effect in CM differentiation with YAP and NR2F2. By combining a range of molecular approaches we aim to establish a Link between NR2F2 and YAP;TEAD in vitro and in vivo. Overall, our ongoing studies suggest that RA signaling induces the activation of non-canonical YAP:TEAD enhancers which are integrated with the cardiac transcription factor network of CPCs essential for the specification of atrial lineages.

We implicated a novel CHD candidate, SHROOM3, in cardiac defects within mice and humans. Mechanistic studies have focused on SHROOM3 interaction with ROCK1/2 to impact the cytoskeleton. However, a recent study indicates SHROOM3 interaction with cell adhesion protein N-CADHERIN is important during neural tube development. Given the range and severity of defects attributed to SHROOM3, we hypothesize mechanism beyond ROCK1/2 are important during cardiac development. To test this hypothesis we performed complementary analysis of the transcriptome, global proteome, and phospho-proteome, in Shroom3^+/+ vs Shroom3^gt/gt embryo heart lysates using RNA-Seq and tandem mass tag labeling, mass spectroscopy (MS). Pathway analysis of both the transcriptome and proteome revealed altered G-Protein coupled receptor signaling. However, GO analyses of the phospho-proteome, indicate the top pathway disrupted in Shroom3^gt/gt embryos is cardiac cell-cell adhesions. An in vitro cell dissociation assay confirmed SHROOM3-loss-of-function downregulates cell-cell adhesions.  Next, we generated a SHROOM3 protein interactome using co-immunoprecipitation/MS (co-IP/MS), with both full length GFP-SHROOM3 and deletion constructs. The SHROOM3 protein interactome confirms N-CADHERIN is a top binding component and the deletion constructs co-IP/MS reveal a likely interaction site, with the deletion of a.a. 286-776, disrupting SHROOM3-N-CADHERIN interaction. Finally, we interrogated exome sequencing of 283 patients with left-right patterning and cardiac defects, and identified 12 rare, potentially damaging variants in SHROOM3. Whereas no variants fell within the ROCK1/2 binding region, one third cluster to the proposed N-CADHERIN binding region, indicating these variants may interrupt N-CADHERIN interaction and cell adhesions during cardiac development, resulting in CHD. Taken together these data indicate novel role for SHROOM3 interacting with N-CADHERIN driving cell adhesions during cardiac development.

The presence of cartilage tissue in the embryonic and adult hearts of different vertebrate species is a well-recorded fact. However, while the embryonic neural crest has been historically considered as the main source of cardiac cartilage, recently reported results on the wide connective potential of epicardial lineage cells suggest they could also differentiate into chondrocytes. In this work, we describe the formation of cardiac cartilage clusters from proepicardial cells, both in vivo and in vitro. Our findings report, for the first time, cartilage formation from epicardial progenitor cells, and strongly support the concept of proepicardial cells as multipotent connective progenitors. These results are relevant to our understanding of cardiac cell complexity and the responses of cardiac connective tissues to pathologic stimuli.

The transcription factors GATA4, GATA5 and GATA6 are key regulators of vertebrate heart muscle differentiation (cardiomyogenesis), but specific target genes regulated by these individual cardiogenic GATA factors remain unknown. We have identified genes that are specifically regulated by each of them, as well as those regulated by either of them using genome-wide transcriptomics analysis in Xenopus laevis. The genes regulated by gata4 are particularly interesting because GATA4 is able to induce differentiation of beating cardiomyocytes in Xenopus and in mammalian systems. Among the specifically gata4-regulated transcripts we identified two SoxF family members, sox7 and sox18. Experimental reinstatement of gata4 restores sox7 and sox18 expression, and loss of cardiomyocyte differentiation due to gata4 knockdown is partially restored by reinstating sox7 or sox18 expression, while (as previously reported) knockdown of sox7 or sox18 interferers with heart muscle formation. In order to test for conservation in mammalian cardiomyogenesis, we confirmed in mouse embryonic stem cells (ESCs) undergoing cardiomyogenesis that knockdown of Gata4 leads to reduced Sox7 (and Sox 18) expression and that Gata4 is also uniquely capable of promptly inducing Sox7 expression.  Our genome-wide transcriptomics analysis therefore identifies an important and conserved gene regulatory axis from gata4 to the SoxF paralogs sox7 and sox18 and further to heart muscle cell differentiation. Our identification of genes that are differentially regulated by each of cardiogenic gata factors also provides a platform for future investigations on the gene regulatory network underpinning embryonic cardiomyogenesis.

Wnt5a is a known regulator of planar cell polarity (PCP) signals in second heart field (SHF) progenitors. However, the exact identity of the cell source of Wnt5a has not been fully elucidated. We studied murine heart development following neural crest cell (NCC)- and SHF-specific conditional knock out of Wnt5a. Our results demonstrate that SHF-derived Wnt5a regulates PCP within the early SHF cells required for initial OFT elongation. Mef2c-cre mediated loss of SHF-derived Wnt5a leads to reduced SHF incorporation into the early OFT and a spectrum of OFT phenotypes, ranging from common arterial trunk arising from RV (25%) to double-outlet right ventricle (DORV, 25%) to normal OFT development (50%). As NCC migrate into the developing OFT, they also secrete Wnt5a. Wnt1-cre mediated loss of NCC-derived Wnt5a causes downregulation of PCP signaling and migratory arrest of more cranial, late wave SHF progenitors. The resulting foreshortened OFT is unable to align over the ventricles such that the appropriate rotation and development of the pulmonary trunk is perturbed. All NCC-mutants demonstrate DORV (100%), with additional pulmonary outflow defects observed in 62%. Neither conditional mutant demonstrates disrupted NCC migration into OFT or venous pole defects. Overall, these results demonstrate that different subsets of SHF progenitors respond to PCP signals from SHF and NCC depending on their location and timing during embryonic development. Apart from providing new insights into the source of Wnt5a during heart development, our data are also the first to demonstrate a novel role for NCC in regulating PCP in SHF cells.

Down syndrome (DS) is caused by trisomy of human chromosome 21 (Hsa21) and is the most common cause of congenital heart defects (CHDs), with approximately half of all DS births being born with a form of CHD. Most of these arise from aberrant heart septation during development, leading to a combination of shunting between the left and right heart chambers at either the ventricular or atrial levels and incorrect valve formation. However, the developmental origins of the CHDs in DS remain poorly understood. To address this, we are using the Dp1Tyb mouse model of DS, which contains an extra copy of mouse chromosomal regions orthologous to Hsa21, and recapitulates human DS CHD phenotypes, with ~60% of Dp1Tyb embryos at embryonic day 14.5 (E14.5) showing defects in cardiac septation. Analysis of morphogenetic changes during ventricular septation in Dp1Tyb E10.5-E13.5 embryos found no difference in growth of the muscular ventricular septum but showed abnormalities in the outflow tract cushions (OFTCs), the transient developmental structures that complete the final stages of ventricular septation and separation of the pulmonary artery from the aorta. Results indicate that Dp1Tyb OFTCs have altered morphology and reduced cell density. Since most of the OFTCs are derived from cardiac neural crest cells, these results imply that the CHDs found in DS may be caused by changes in neural crest differentiation, potentially linking these defects to other DS-associated abnormalities such as craniofacial dysmorphology.

Heart valve defects are the most common birth defects in new-borns worldwide. Endocardial (EC) cells are a subpopulation of endothelial (ET) cells that have a unique ability to undergo endothelial-to- mesenchymal (EndoMT) transition, a process critical for heart valve formation. To date, only one molecular maker, NFATc1, is known that uniquely labels EC cells during valve development. However, it is not the master regulator of EC cell specification since mice deleted for Nfatc1 forms underdeveloped heart valves. The ability of EC cells to undergo EndoMT makes them distinct from ET cells, despite both sharing a common precursor origin during cardio-genesis. We hypothesize that EC cells become distinct from ET cells via the expression of a sub-set of genes that are regulated differently at genetic and epigenetic levels.

The integrated analysis of bulk RNA-sequencing and whole-genome bisulphite sequencing data alongside single-cell RNA-sequencing datasets from public domain has identified candidate genes involved in EC-fate determination. These candidates have been validated at the gene expression level. Additionally, inducible shRNA knockdowns of candidate genes have been studied in an Nfatc1-mCherry mouse ES cell line using an in vitro hanging drop culture differentiation system. This has provided a functional assay for the temporal knockdown effects of these genes on the EC differentiation and specification. One of the top candidates is Zfpm1 and the effect of its knockdown on the functional ability of EC cells to undergo EndoMT is assessed using trans-well invasion and migration assays. Altogether, this will not only facilitate our understanding of the role of Zfpm1 in heart development but will also provide a basis for stem cell-based therapies for valve-related defects. 

Kdm8 demethylates the di-methylated form of lysine 36 of histone H3 (H3K36me2) and prevents spurious gene expression. Kdm8 is enriched in the developing mouse heart, and its constitutive inactivation is embryonically lethal. Moreover, H3K36me distributes abnormally genome wide in the failing heart, in which metabolism shifts towards glycolysis. However, the function of Kdm8 in the heart or the contribution of metabolic imbalance to initiate the events leading to heart failure are unknown. We show that Kdm8 maintains the expression of genes controlling mitochondrial metabolism by repressing Tbx15 to prevent dilated cardiomyopathy leading to lethal heart failure. Kdm8 inactivation in cardiomyocytes by cre-mediated homologous recombination increased global levels of H3K36me2, caused adverse myocardial remodeling beginning at 4 months, and dead due to heart failure by 9 months of age. Genome wide mRNA profiling revealed that Tbx15 target genes controlling NAD⁺ metabolism were downregulated in Kdm8 mutant hearts before the initiation of adverse myocardial remodelling. Moreover, metabolomics showed decreased NAD⁺ pathway intermediates, and NAD⁺ treatment blunted the initiation of cardiac deterioration towards heart failure. Furthermore, Tbx15 overexpression in cardiomyocytes blunted the respiratory increase induced by NAD⁺. This suggests that dysregulation of a KDM8 – TBX15 axis sits high in the hierarchy of events initiating cardiac deterioration towards heart failure. Indeed, KDM8 was downregulated in human hearts affected by heart failure. Furthermore, higher expression of TBX15 tracked with stronger downregulation of genes encoding mitochondrial proteins. Our findings suggest that epigenetically controlled metabolic gene networks are dysregulated to initiate adverse myocardial remodelling towards heart failure.

A dynamic Atlas with cellular resolution is an essential resource for understanding at a multiscale level the key mechanics of cardiac morphogenesis and assessing variability between embryos.
Reconstructing a 4D atlas of the mouse heart is a challenge; indeed, although new live-cell imaging techniques allow a crucial understanding of cellular and tissue kinetics, they do not guarantee a whole-organ acquisition at a spatiotemporal resolution necessary for a detailed multidimensional reconstruction.
To overcome this technological limitation, we designed a dedicated computational framework that, starting from a high-resolution stereotyped atlas, incorporates cellular kinetics and dynamic data of the myocardium provided by time-lapse images from a collection of specimens.
The incorporation of live-embryo data into the Atlas is performed in two fundamental steps: a temporal staging and a spatial mapping of cardiac shapes. Our framework implements a staging system that, through a morphometric feature, assigns to each frame of the time-lapse a developmental stage of the Atlas. Given the temporal matching, each cardiac shape is then mapped, by combined applications of image processing and registration techniques, in the respective Atlas shape.
This last step allows to map the individual cells tracked in the time-lapse images and to project dynamic features such as growth rate, anisotropy and strain to which the myocardium is subjected during its morphogenesis.
This computational method is intended as a novel tool to compare different live-embryos in a single time-space convention, which is necessary to investigate key processes in heart development.

During development, cells adopt diverse strategies to sculpt epithelial tissues into characteristic 3D shapes with molecularly and mechanically distinct regions. The vertebrate heart tube is a good example, as it extends by progressive addition of second heart field (SHF) progenitor cells within an epithelial sheet in the dorsal pericardial wall. Initially contributing to the entire cardiac primordium across the dorsal mesocardium, SHF cell addition is restricted to the poles of the heart tube after dorsal meoscardial breakdown. Perturbation of SHF deployment results in a spectrum of congenital heart defects (CHD). T-box transcription factors implicated in CHD regulate the emergence of a boundary segregating progenitor cells to alternate cardiac poles. In addition, the epithelial properties of SHF cells, including cell elongation, clonal anisotropy and epithelial tension, have been identified as regulatory targets during heart tube elongation. However, the relationship between epithelial properties, mechanical tissue stress and progenitor cell patterning in the SHF remains unknown. Indeed, understanding how cellular and tissue forces contribute to growth and form during organogenesis is a major challenge in developmental biology. Here we present quantitative phenomenological analysis of epithelial cellular properties of cells throughout the dorsal pericardial wall, including cell elongation, polarity, anisotropy and cell contacts.  Patterns of epithelial stress and tension in the SHF are integrated in our dataset using force inference image analysis. We document the dynamic relationship between epithelial features and progenitor cell patterning at embryonic days 8.5 and 9.5, before and after dorsal mesocardium breaks down, and in T-box mutant embryos. By integrating findings from cell scale feature patterning, mechanical tissue stress and cell biology our project will provide new mechanistic insights into cardiac morphogenesis and the origins of CHD.

Second heart field (SHF) progenitors allow the rapid elongation of the embryonic heart tube. The precise control of SHF deployment is a prerequisite for correct heart tube elongation and any alteration results in congenital heart defects (CHD). We previously identified the transcriptional repressor Hes1 as a critical regulator of SHF deployment and preliminary results suggest that regionalized SHF expression of the cAMP-dependent protein kinase (PKA) regulatory subunit RIα (Prkar1a) may control Hes1 expression in the SHF. In the adult heart, PKA is known to regulate heart contraction nevertheless its role during heart development still remains unknown. In human, mutations in PRKAR1A gene cause haploinsufficiency which results in a rare Carney Complex syndrome, associated with benign tumors including cardiac myxomas described to be associated with congenital heart defects.

To evaluate the role of R1α during early heart morphogenesis, we developed transgenic mouse lines allowing conditional deletion of Prkar1a in diverse cardiac progenitor cell lineages.

Our results demonstrate a key role for RIα in controlling SHF deployment. Indeed, RIα deletion in SHF progenitors results in impaired SHF proliferation, associated with early embryonic lethality. Upregulated Hes1 expression was detected in RIα mutants and increased phenotype severity observed in RIα/Hes1 compound mutants strongly support RIα and Hes1 genetic interaction. The extracardiac cell population, cardiac neural crest (CNC), plays a critical role in heart morphogenesis and our results suggest key role for RIα in CNC cell deployment. Additionally, heterozygous RIα deletion in CNC leads to a 5 week-old lethality associated with cardiac hyperplasia and CNC lineage tracing reveals the existence of subendocardial CNC progenitors in the adult heart, suggesting a potential role for CNC derivatives in the development of cardiac myxomas.

Altogether our study reveals a critical role for RIα in early cardiac development.

The arterial pole of the heart is a hotspot for life-threatening forms of congenital heart defects (CHDs). This cardiac region results from the elongation of the embryonic outflow tract (OFT) that requires the addition of Second Heart Field (SHF) progenitor cells to provide the myocardium at the base of the ascending aorta and pulmonary trunk. Investigating the transcriptional program of future sub-aortic and sub-pulmonary myocardium we found that the gene encoding the lipid sensor Pparγ is preferentially expressed in the inferior OFT wall. Here we show that TBX1 is required for the regional expression of PPARγ and its downstream target genes in the inferior OFT. Using mouse genetics and ex vivo embryo culture in the presence of PPARγ agonists or antagonists, we demonstrate that Pparg controls cardiac progenitor cell proliferation and addition of the future subpulmonary myocardium to the OFT, both critical determinants for normal arterial pole development. Moreover, we show that the non-canonical DLK1/NOTCH/HES1 pathway negatively regulates Pparγ in future subaortic myocardium. In conclusion, we present evidence that Pparg is a regulatory component of the cross-circuitry controlling regional identity at the arterial pole of the heart, thus providing new insights into gene, and potential gene-environment, interactions involved in CHD.

The heart is built by distinct progenitor populations that contribute to specific cardiac chambers, valves and vessels. These progenitors arise from different germ layers in the gastrulating embryo. The mesoderm-derived cardiac progenitors include a population called the cardiopharyngeal mesoderm (CPM), that contributes to second heart field-derived cardiac regions, but also to head skeletal muscles by populating the pharyngeal arches. While the development and differentiation of cardiac cell types has been examined closely, the developmental mechanisms regulating fate commitment of the head muscle progenitors in the CPM are unclear. I have addressed the early development of the CPM, and its differentiation into skeletal muscle. Using mouse embryos and an embryonic stem cell model, our work shows that inhibition of Wnt/beta-catenin and Nodal pathways underlies the fate specification of these progenitors. Using our findings, we directed embryonic stem cells to differentiate into the CPM and subsequently, into cardiomyocytes and skeletal muscles. Our in vitro model also suggests that there may be a distinct myogenic cue for skeletal muscles derived from the CPM, as compared to skeletal muscles derived from somitic mesoderm. By generating derivatives in vitro, specifically via the CPM, this approach can provide a tractable model of syndromes including both congenital heart diseases and muscular dystrophies.

The existence of an early cardiac progenitor (CP) population has long been supported by fate mapping experiments which identify presumptive CPs at specific embryonic positions as gastrulation begins. These presumptive CPs then migrate to the anterior lateral plate mesoderm and engage a conserved transcriptional network to direct specification, migration, and differentiation of the cardiac lineage. While later stages of cardiac development are well characterized, initial specification remains poorly understood, due to a lack of specific molecular markers for early CPs. To address the outstanding question of CP specification our group is studying the Apelin receptor (Aplnr), a highly conserved GPCR, first shown to regulate cardiac development in zebrafish where aplnra/b mutants do not express nkx2.5, the earliest specific CP marker, following gastrulation and do not develop cardiomyocytes. Fate mapping analysis has shown that migration of presumptive CPs during gastrulation requires Aplnr. Following loss of Aplnr function Nodal signalling fails to activate correctly and this failure disrupts mesendoderm specification. scRNA-seq has identified that with loss of Aplnr function a cardiac-like mesoderm population fails to develop during gastrulation. Our results indicate that Aplnr is required for the initial commitment of mesodermal cells to the cardiac lineage. Current work is focused on characterizing this specification failure to understand both the function of Aplnr and the mechanisms regulating early CP development.

The embryonic heart elongates by progressive addition of second heart field (SHF) cardiac progenitor cells to the arterial and venous poles of the early heart tube, derived from the first heart field (FHF). The SHF gives rise to the right ventricle, outflow tract and part of the atrial myocardium. Defects in SHF deployment contribute to a spectrum of congenital heart defects (CHD) affecting the cardiac poles. Septation defects account for over 50% of CHD. The primary atrial septum and dorsal mesenchymal protrusion originate in Tbx1 expressing cells in the posterior dorsal pericardial wall where they activate the FHF regulator Tbx5 in response to retinoic acid (RA) signaling. Transient coexpression of Tbx1 and Tbx5 is followed by downregulation of the SHF program, resulting in emergence of a sharp boundary between Tbx1-positive arterial pole progenitors and Tbx5-positive venous pole progenitors. A similar history of expression of both FHF and SHF regulators, followed by downregulation of the SHF program, demarcates boundary formation at the site of ventricular septation. Here we show that transient overlap between the FHF and SHF programs is observed along the dorsal mesocardium during early heart development, followed by downregulation of the FHF program in the midline of the dorsal pericardial wall. Single cell RNA-seq analysis of the SHF lineage identifies venous pole progenitor cells that switch to a FHF program and mouse genetics reveals that SHF genes are downregulated through T-box gene dependent and independent mechanisms. Finally, using a conditional dominant negative RA receptor we show that RA signaling is required at the heart field interface for morphogenesis of the muscular ventricular septum. Our results point to the importance of a transient regulatory module at the heart field interface in boundary formation during early heart morphogenesis and patterning of future septal structures, providing insights into the etiology of CHD and septum evolution.

Heart regeneration after infarction has been an important goal in the last decades. The low regenerative potential of the mammalian heart has set an essential barrier to this achievement, and innovative strategies need to be developed to recover heart function. Cell competition (CC) has been described in multiple tissues, including the mammalian developmental and adult heart. In this context, fitter cardiomyocytes outcompete their neighbors, thus improving the proportion of more functional cells with no deleterious consequences. However, this process has not been thoroughly characterized yet in the mammalian heart, even though understanding it could help unlock cardiomyocytes' regenerative capacity and heal the injured heart.

We have developed a model to closely study cardiomyocyte CC based on Myc mosaic overexpressing hearts (iMOS Myc). By dissociating P1 iMOS Myc;MHC-Cre mouse hearts, we obtain highly reproducible cultures in which the winner population can be tracked over time. Using this approach we have been able to detect the expansion of Myc-overexpressing cardiomyocytes at the expense of the wildtype ones. This indicates that CC is potentially happening in our experimental setting. We are currently characterizing key features of CC, such as winner cell proliferation and loser cell death. Preliminary data suggests increased proliferation in iMOS-Myc cultures compared to controls, although further characterization is warranted.

Additionally, we have set an in vitro system to overexpress factors of our interest in our cardiomyocyte cultures, taking advantage of the specific tropism of Serotype 9 AAVs. So far, we have expressed YAP5SA, a constitutively active form of YAP that elicits LATS1/2 repressive phosphorylation. We found 2,24% proliferating cardiomyocytes at 3 dpi (compared to a 0,16% rate in controls) with clear signs of sarcomere disassembly. We plan to use this model to test for new factors capable of triggering CC.

Congenital heart disease affects nearly 1% of births and includes valvular abnormalities, such as pulmonary valve stenosis, which commonly occurs in the RASopathies. During valvulogenesis, a subpopulation of endocardial cells overlaying the cardiac jelly undergoes endothelial-to-mesenchymal transition (EndMT), producing valvular interstitial cells (VICs). These resident mesenchymal cells of the cardiac valve are responsible for extracellular matrix production and homeostasis. Furthermore, a subpopulation of endocardial cells escapes EndMT and becomes valvular endothelial cells (VECs), which line the outside of cardiac valve leaflets. Critically, a differentiation strategy to generate both cell populations is needed so that human valvular disease can be modeled in vitro. Recently, a feeder-dependent embryoid body differentiation strategy utilizing BMP10 has been shown to generate an endocardial population that is distinct from vascular endothelial cells. Here, we show that iPSCs maintained without feeder cells can be differentiated as a monolayer towards mesoderm with the GSK3 inhibitor CHIR-99021 and subsequently to an endocardial lineage with BMP10 and bFGF. Importantly, these endocardial cells express high levels of endocardial markers, such as NKX2.5, GATA4, GATA5, and NRG1. EndMT can then be induced on this endocardial population. Following six days of EndMT, some of these endocardial cells become VICs and upregulate key mesenchymal genes, such as SOX9, CDH11, and COL1A1. A subpopulation of endocardial cells escapes EndMT to become VECs and maintain high expression of endothelial genes, such as PECAM1, CDH5, and EMCN. These cell populations can be utilized for in vitro modeling of human cardiac valves.

Chamber-like cardiac organoids or « cardioids » constitute a promising in vitro model to study human physio-pathological cardiac development. A WNT-BMP modulation-based protocol for the generation of self-organizing cardioids recently published (Hofbauer & al., Cell 2021) was optimized using several pluripotent stem cell lines. Cardioids featured expected Ca2+ spiking and contractions from differentiation day 14 when chambers were formed. The physiological significance of cardioids was further challenged.

Both the rate and the amplitude of beating cardioids were challenged using several pharmacological agents including the pacemaker channel inhibitor ivabradine, or the positive inotropic agents phenylephrine and epinephrine, a1-adrenergic (Phenylephrine) or b-adrenergic agonist, respectively. Contractility rate of cardioids was not impaired by a pacemaker current inhibitor (Ivabradine) pointing to a myocyte cell-autonomous generation of beating rate. This observation was confirmed by the presence of functional inositol 1,4,5-trisphosphate receptors (IP₃R) within cardiomyocytes, revealed by the inhibition of spontaneous beating rate by Xestospongin.

We identified the early cell aggregation step as a key process which determines both the fate and position of cells in the self-generating cardioids. To improve cell aggregation  and reproducibility  of cardioids over time, micro-patterning used within a Sound Induced Morphogenesis (SIM) device (CymatiX™ MimiX Biotherapeutics), should turn out to be an extremely powerful technology.

Regardless of their still immature states, we report the functional and physiological properties of chamber-like cardioids and thus their relevance as a potent platform to study cardiac development and physiopathology.

Introduction – Transcription factors (TFs) are key regulators that control the temporal extensive variations of gene expression, the hallmarks of sequential stages of human cardiac development. Genetic alterations of these TFs cause human cardiac defects. TFs act within regulatory networks which have been only partially described. Using cardiac differentiation of human induced pluripotent stem cells (hiPSCs), this study aims at unravelling the global TF regulatory network that govern human cardiac development.

Methods & Results – Transcriptomic data were generated daily, from day 0 to day 30, in triplicates, during directed cardiac differentiations of three hiPSC lines reprogrammed from healthy donor somatic cells. The top 3,000 differentially expressed genes (DEGs) along cardiac differentiation were pooled into 12 gene clusters, illustrating successive temporal alterations of gene expression. Functional annotation of these gene clusters and comparison of their expression profiles to publicly available transcriptomic data of murine cardiac development, demonstrated that hiPSC cardiac differentiation accurately recapitulates transcriptomic processes governing cardiac development. Next, a regulatory network that includes all 216 TFs from the 3,000 DEGs was investigated using an expression-based correlation score involving time delay. Overall, 6,817 activation-type and 6,497 inhibition-type relationships were inferred between these 216 TFs. Compared to the literature-based STRING database, our regulatory network was 3.3 fold denser and could also inform about directions, types and time delays of each interaction. Identified sub-networks included well-known interactions, such as between FOXC1, ISL1, MEF2C and HAND2, but also new candidate sub-networks.

Conclusion – This work shows that hiPSC cardiac differentiation is relevant for developmental studies, and provides a valuable database to explore new TF regulation networks during human cardiac development.

Long-standing studies in embryos have shown that polarized extraembryonic signals generate a NODAL:SMAD2/3 gradient in the epiblast that regulates gastrulation patterning and the establishment of the anterior-posterior (A-P) axis. However, epiblast-like cells (hESCs and mESC) organize in 2D and 3D gastruloid structures in vitro in the absence of extraembryonic tissues. These observations highlight the importance of intrinsic self-organizing mechanisms in hESCs, independent of polarized extraembryonic signals. Indeed, human 3D-gastruloids display features of Carnegie-stage-9 embryos, with a high degree of organization in gene expression along the A-P axis, including a posterior-to-anterior signature of somitogenesis. The A-P pattern correlates with the organization of signaling components along the length of the gastruloid. In elongated gastruloids, BMP signals are predominantly anterior, while the expression of WNT3 and NODAL genes are restricted to the posterior end, consistent with a role of the latter in the mammalian tailbud at the onset of gastrulation. However, how the regionalization of the signaling components is achieved in the 3D-gastruloids is unknown. Our ongoing analysis suggest that the Hippo-effector YAP1 regulates the antero-posterior organization of the 3D-gastruloids by restricting the posterior expression of Nodal signaling components. Our data show that YAP1 represses Nodal activity through the recruitment of the Polycomb-repressor complex to the chromatin of Nodal genes in hESCs. Hence, in the 3D-gastruloids, YAP1 deletion lead to the anterior expansion of Nodal:Smad2/3 activity. As a consequence, the YAP1 KO-derived 3D-structures are longer than controls, suggestive of an increase in Nodal-induced mesoderm derivatives. Our findings highlight a crucial interaction between YAP1 and NODAL signaling essential for the germ-layer formation and morphogenetic events occurring during the establishment of the body plan.  

Human pluripotent stem cell-derived cardiomyocytes (hPSC-CMs) offer an attractive experimental platform to model cardiovascular diseases and advance potential regenerative therapies. However, hPSC-CM cultures exhibit heterogeneity and immature characteristics.

In our laboratory, we have developed a “Left Ventricle” (LV) differentiation protocol which efficiently generates LV-like cardiomyocytes (hPSC-LV-CMs), characterised by higher maturity and improved functionality than those generated using the “Standard” WNT-On/WNT-Off differentiation protocol. The CytoCypher MultiCell High Throughput System was used to prove improved contractility, calcium dynamics and increased dependence of β2-adrenergic receptor signalling as well as a positive force-frequency relationship as indicators of superior maturity at 30 days of differentiation. To address if hPSC-LV-like CMs could be further functionally matured, we grew cells from day 20-30 in the presence of fatty acid containing maturation media. Results show this media was unable to improve the cells’ contractility and overall cells loss the ability to display a positive force-frequency response. However, a trend for a higher contraction amplitude was noted. This media also led to improved calcium dynamics, but the calcium amplitude of these cells was lower, suggesting that fatty acid supplementation per se is insufficient to improve functional maturity of LV-like CMs.

Maturity can be measured in multiple ways and not all media supplements improve cellular maturity in the same way. Some might lead to metabolic maturity (fatty acids) and others may help cytoarchitecture maturity (mechanical loading and pacing) or functional maturity, thus raising the need to develop a media to more holistically improve hPSC-CM maturity.

Decreased left ventricle (LV) function caused by genetic mutations or injury often leads to debilitating and fatal cardiovascular disease. LV-cardiomyocytes (LV-CMs) are, therefore, a desirable therapeutical target. Here we made use of developmental-cues to devise a protocol for generating LV-cardiomyocytes from human pluripotent stem cells (hPSC). Voltage clamping confirmed hPSC-LV-CMs display a ventricular action potential shape with hallmarks of maturity, including increased expression of IK1. Moreover, hPSC-LV-CMs have an elongated shape, exhibit well-defined sarcomere structures with a length typical of adult cardiomyocytes, express mature cytoarchitecture markers such as TCAP and have a better respiratory capacity. Functionally, LV-cardiomyocytes display more mature calcium transients, supported by higher RYR expression. Engineered heart tissues (EHTs) generated with hPSC-LV-CMs consist of longitudinally organised cardiomyocyte bundles whereas those generated using the standard WNT-On/WNT-Off protocol (hPSC-Std-CMs) are mostly populated by clumps of cardiomyocytes. LV-EHTs produce more force, have faster contraction dynamics and a slow beating rate but can be paced, confirming LV-CMs are losing the pacemaker potentials typical of immature cardiomyocytes. Collectively, we show that: 1) it is possible to rapidly obtain LV-CMs with mature properties, even before they are exposed to reported maturation regimes; and 2) the generation of EHTs per se is unable to rescue the lag in maturation observed between hPSC-Std-CMs and hPSC-LV-CMs. This work demonstrates that hPSC-LV-CMs are a suitable model to study LV development and disease and could enable more faithful LV-specific cardiotoxicity screens. Moreover, it opens the possibility of like-for-like cell replacement therapy becoming an accessible therapy to treat heart failure patients.

Haploinsufficiency for GATA6 is associated with various forms of congenital heart disease (CHD) including septal and conotruncal defects. Genetic loss-of-function studies in model organisms confirmed that GATA6 regulates critical aspects of heart morphogenesis but have not been useful to model human haploinsufficiency. The phenotypic diversity of CHD in patients containing GATA6 mutations is likely due to an unknown combination of interacting gene variants that converge to influence the cardiac phenotype. A greater understanding of the GATA6 genetic regulatory network that controls human heart development is thus essential for better understanding the pathophysiology and ultimately developing therapies. In the present study, we used cardiac directed differentiation with human embryonic stem cells (hESCs) and induced pluripotent stem cells (iPSCs) as a platform to study GATA6 function. GATA6^-/- hESCs failed to generate cardiomyocytes (CMs) or cardiac progenitor cells (CPCs) during cardiac-directed differentiation. The expression of markers for cardiac mesoderm were markedly reduced in GATA6^-/- cells compared to controls, indicating a defect in mesoderm patterning. Profiling by RNA-seq and CUT&RUN at the mesoderm patterning stage identified genes of the BMP and WNT pathways as being regulated by GATA6, including LGR5, a GPCR associated with WNT signaling and a putative key target. In contrast to the homozygous mutants, GATA6^+/- hESCs generated CPCs and CMs but did so less efficiently than controls. Analysis of an iPSC line containing a heterozygous mutation in GATA6 (c.1071delG) derived from a patient with CHD had similar defects in CM differentiation efficiency, suggesting that it can be used to study in vitro the human disease phenotype. Together, this study provides evidence for a regulatory function for GATA6 during human pre-cardiac mesoderm patterning and describes a system for examining GATA6 haploinsufficiency in vitro. 

The number one cause of fetal death are defects in heart development. Determining the underlying causes faces many challenges, including the complexity and inaccessibility of the embryonic heart, the unclear impact of drugs and environmental factors during pregnancy, and the lack of in vitro models representing all the compartments of the human heart. Here, we established a cardioid organoid platform recapitulating the development of the major compartments of the human embryonic heart, including the right and left ventricles, the atria, the outflow tract, and the atrioventricular canal. These cardioids have the compartment-specific in vivo-like gene expression profile, morphology, and functionality. We use this platform to unravel the developmental electrochemical signal propagation between interacting heart chambers and dissect how genetic and environmental factors cause specific defects in different regions of the developing human heart.

Microinjection of lipophilic dyes, iontophoresis and infection by modified viruses allow tracing the embryonic cell-lineages prospectively.  One the one hand, dyes are easy to use but fail to label single cells. On the other hand, iontophoresis and viral vectors require specialized equipment, unavailable in most laboratories. Here, we developed a new method for cell lineage tracing based on the microinjection of the TAT-Cre recombinase protein in embryos carrying floxed reporters. This approach allows the prospective monitoring of cell progenitors and their descendants from primitive streak stages in mouse (E6.5) to organogenesis (E8.5). Using a low titer of TAT-Cre allows single-cell labelling, which makes this method suitable for prospective clonal analysis. Moreover, TAT-Cre microinjection in a floxed targeted gene embryo yields local genetic ablation at a defined region and developmental stage. In summary, this new method is accessible and efficient for cell lineage tracing system, enabling fate mapping, clonal analysis and targeted genetic perturbation in vivo.

Congenital heart defects (CHDs) are the most common form of birth defects, occurring in nearly 1% of newborns. Retinoic acid (RA) is a signaling molecule synthesized from dietary vitamin A. An excess of RA has dramatic effects on human embryonic development. This can occur in either the offspring of women undergoing therapeutic treatment with the synthetic retinoid isotretinoin (13-cis-retinoic acid), or in the offspring of women with excess dietary Vitamin A supplementation. 13-cis-RA, known as isotretinoin (INN) and used for the treatment of acne, is correlated with a high risk of birth defects. Administration of INN during pregnancy causes multiple types of malformations in infants, including cardiovascular defects. There is also evidence that reduced maternal levels of Vitamin A can also increase the risk of CHD in her offspring. Despite the tremendous progress in deciphering the implication of the retinoic acid signaling in heart development, many gaps regarding the underlying mechanisms of RA-mediated gene regulation remains to be unraveled. Here, we are aiming at generating atlases of retinoic acid cis-regulatory elements in the developing heart of a Vitamin A deficient mouse model as in human cardioids through the use of multiomic approaches. Our pilot results indicate that retinoic acid signaling regulates a highly conserved, disease-associated candidate enhancer regulating sino-atrial development.

Supplementation of Folic Acid (FA) during pregnancy prevents neural tube defects (NTD). Recent studies indicate FA supplementation may similarly prevent heart defects, though the relationship remains unclear. Proposed etiologies for FA prevention of birth defects include a role in methylation, cell proliferation/differentiation and G-Protein signaling, with disruption affecting susceptible cell populations, including neural crest cells. However, the mechanism remains undetermined. To determine the mechanistic role of FA supplementation during cardiac development, we utilized an established model of high and low FA supplementation (2ppm versus 10ppm) in pregnant wild type mice, and in the resulting embryo hearts at day E12.5, we performed complementary analysis of the cardiac transcriptome, global proteome, and phospho-proteome utilizing RNA-Seq and tandem mass tag mass spectroscopy (TMT/MS). Ingenuity pathway analysis of differentially expressed genes and proteins indicates altered protein translation, with eiF2 signaling topping the list in the transcriptome and global proteome. Within the phospho-proteome, pathway analysis revealed alteration in DNA methylation, cell cycle control, tight junction signaling and RHOA signaling. The protein SHROOM3 is similarly associated with DNA methylation, cell proliferation, tight junctions and RHOA signaling and is expressed int the neural crest. SHROOM3 loss-of-function leads to NTDs that can be reversed with high FA supplementation. We recently demonstrate SHROOM3-loss-of-function leads to cardiac defects. We utilized short term FA supplementation in SHROOM3-loss-of-function mice, and preliminary analysis of 43 embryos reveals reduced penetrance of membranous VSDs from 36% to 11% with high folic acid supplementation. Overall, these findings suggest a mechanistic role for FA during cardiac development that includes DNA methylation, cell proliferation, tight junction and RHOA signaling within susceptible cell populations.

Mammalian heart regeneration has been the focus of study in the past decades, aiming to promote recovery after ischemic injury. However, mammalian cardiomyocytes have a notoriously slow turnover, incapable of compensating for the loss of contractile muscle.

In regenerative models (zebrafish, neonatal mammalian heart), new cardiomyocytes arise form preexisting ones; suggesting the presence of a cardiomyocyte pool able to reenter cell cycle. One hallmark of naïve, proliferative cardiomyocytes is their ploidy levels. Therefore, it has been proposed that mononucleated cardiomyocytes in mice and diploid cardiomyocytes in zebrafish and human can pose a potential population with the ability to replace lost myocardium.

However, defining a trancriptomic signature of these myocytes in order to understand and promote a proliferative phenotype has remained elusive due, in part, to the nature and sensitivity of the cardiac muscle cell.

Here we describe a method that has enabled us to perform single-cell RNA Sequencing in binucleated and mononucleated cardiomyocytes from murine ventricles. This method allows for a very high-quality RNA purification and depth of sequencing, enabling us to define several cardiomyocyte clusters with distinct transcriptomic profiles, both among binucleated and mononucleated populations, suggesting a higher heterogeneity among cardiomyocytes than what was previously described.

Valvular heart disease affects 1-2% of the population, often requiring surgery. However, current replacement valves do not respond normally to biological signals and cannot undergo growth or proper remodeling. An alternative is the use of valves generated via differentiation of human pluripotent stem cells into valve cells, recapitulating in-vivo development. Valve leaflets are composed of 3 layers of extracellular matrix containing valve interstitial cells (VICs), ensheathed by valve endothelial cells (VECs). Valve development involves the formation of endocardial cushions populated by cells undergoing endothelial-to-mesenchymal transition (EndoMT), followed by elongation and layer stratification. The transcriptional programs driving these changes remain insufficiently defined, and little is known regarding VIC subtypes. Moreover, current differentiation protocols generate heterogeneous rather than specific valve cell populations. Our study aims to identify novel factors driving EndoMT and mechanisms driving specification of VIC subpopulations.

We performed scRNA-seq on E8.0-P1 whole mouse hearts and computationally generated cell type-specific clusters. Endocardial, VEC and VIC markers were identified. Transcriptomic changes corresponding to VICs were noted by E9.0, and two distinct VIC clusters were found at E10.25. Increased EndoMT marker expression was noted at E11.5. Early endothelial and VIC cells were analyzed using URD (Farrell et al 2018), which reconstructs differentiation trajectories in the form of a branching tree. Cells at branch points were subject to gene regulatory network analysis using SCENIC (Aibar et al 2017), and top hits included key transcription factors for valve development. CellPhoneDB (Efremova et al 2020) was then used to identify important cell-cell signalling interactions. The knowledge gained is expected to assist in optimization of differentiation protocols for VICs, and ultimately generation of “biological” stem cell-derived valves.

The heart is covered by the epicardium, consisting of epithelial cells and a mesenchymal layer. The epicardium has been shown to be essential during cardiac development by contributing cells through epithelial-to-mesenchymal transition (EMT) and the secretion of paracrine factors. In the adult, the epicardium conveys a cardioprotective response after myocardial infarction, albeit suboptimal compared to the epicardial contribution to cardiac development. Direct analysis of the human fetal epicardium is vital as it provides new insights into the cellular and biochemical interactions within the developing heart, which can potentially contribute to enhancing the post-injury response. Epicardial layers were isolated from human fetal hearts (14-15 weeks gestation) and digested into single cells. Single-cell RNA sequencing was used to determine the cellular composition of human fetal epicardium, and data were further explored to identify regulators of epicardial EMT. Analysis of 2073 cells reveals an enrichment of epicardial derived populations and displays a clear clustering of the epicardial epithelial and mesenchymal populations. Importantly, we found that in humans ‘classical’ markers, such as Wilms’ Tumor 1, Transcription factor 21 and T-box transcription factor 18, are not specific enough to reliably classify the epicardium. Our analysis has provided markers that do allow for robust identification of the epicardium validated by immunohistochemistry. To establish the regulation of epicardial activation we focus on the process of EMT within our dataset. Here we present a novel pathway involved in epicardial EMT and several promising candidates that influence key developmental processes.

This work was funded by a Senior Researcher Dekker grant by the Dutch Heart Foundation (2017T059)

Approximately 30% of individuals with Cornelia de Lange Syndrome (CdLS) exhibit congenital heart defects (CHDs). The most common form of CdLS is caused by haploinsufficiency for NIPBL, which encodes a cohesin-associated protein. In Nipbl+/- mice, NIPBL-haploinsufficiency causes hundreds of small gene expression changes in every cell and CHDs similar to those in CdLS. Using the Nipbl+/- mouse as a model system for discovering new causal factors for CHDs, we documented abnormal heart development by cardiac crescent (CC) stage, suggesting that CHDs originate as early as gastrulation. To investigate this, we performed single-cell RNA sequencing on both CC- and gastrulation-stage wildtype and Nipbl+/- mouse embryos. Strikingly, we found that Nipbl+/- embryos dramatically overexpressed Nanog. Nanog encodes a transcriptional repressor required for pluripotency in pre-implantation embryos and is normally transiently re-expressed during gastrulation. In Nipbl^+/- mice, however, Nanog fails to shut off appropriately post-gastrulation, so that by CC-stage its levels are many-fold above normal. Interestingly, numerous gene expression changes detected in Nipbl^+/-mice involve post-implantation-stage targets of Nanog. These include genes associated with pluripotency (Pou5f1/Oct4), left-right patterning (Tdgf1, Lefty2, Nodal), anterior-posterior patterning (Hox genes), and primitive erythropoiesis (Tal1, Lmo2, Hbb-bh1). Accompanying these changes are changes in the allocation of cells to clusters representing Mesp1-expressing cardiac progenitors, the first and second heart fields, and neural crest. These results suggest that a failure to downregulate Nanog expression after gastrulation, and the transcriptional dysregulation that ensues, lead to misallocation and/or dysfunction of cardiogenic cell populations. Funded by NIH-NHBLI.

Congenital heart defects are frequent and collectively represent a significant burden for human health. Understanding the developmental origins of such anomalies is key to improving diagnoses, prognoses and therapies. The Human Developmental Cell Atlas (HuDeCA) consortium in France gathers as much data and metadata as possible about normal human embryonic and early fetal tissues donated to research, complete with strict quality control procedures and use of standardized annotations, as a complement to the international Human Cell Atlas and a baseline for the scientific community worldwide. Here, we contribute new data about the identities and trajectories of cells in human male and female hearts from 8 to 12 post-conceptional weeks through judicious comparisons of individual hearts assessed through single-nucleus and spatial transcriptomics and advanced imaging approaches. We integrate and validate this information both across our datasets and with existing atlases, using multiple technical platforms to provide four-dimensional analyses at varying levels of resolution. Delineating the gene expression networks deployed within individual cells at specific positions has revealed an unsuspected diversity of cellular states. This data enables new hypotheses about the paracrine influences exerted by and on the many lineages necessary for the complex functions of the first and longest-lived vital organ.

* These authors contributed equally to this work and are joint first authors

** These authors share joint last authorship; correspondence may be addressed to both (stephane.zaffran@inserm.fr, heather.etchevers@inserm.fr).

Mammalian cardiomyocyte maturation entails phenotypic and functional optimization during the

late fetal and postnatal phases of heart development, processes that are driven and coordinated by

complex gene regulatory networks. Cardiomyocytes derived from human induced pluripotent stem

cells (iPSCs) are heterogeneous and immature, barely resembling their adult in vivo counterparts.

To characterize relevant developmental programs and maturation states during human iPSC-cardiomyocyte

differentiation, we performed single-cell transcriptomic sequencing. This revealed

six cardiomyocyte subpopulations, with heterogeneity defined by cell cycle and maturation states.

Among those subpopulations, two of them demonstrated a non-proliferative and mature

transcriptional profile. To further investigate the proliferation-maturation transition in

cardiomyocytes, we induced LMNB2 loss-of-function, which represses cell cycle progression in in

vivo cardiomyocytes. This resulted in increased maturation, characterized by transcriptional

profiles related to myofibril structure and energy metabolism, in LMNB2-inactivated

cardiomyocytes. Unique maturation signatures and two maturational trajectories were identified

for control and LMNB2-inactivated cardiomyocytes. By comparing these datasets with single-cell

transcriptomes of human fetal hearts, we were able to define spatiotemporal maturation states in

human iPSC-cardiomyocytes. Our results provide an integrated approach for comparing in vitro differentiated

cardiomyocytes with their in vivo counterparts and suggest a strategy to promote

cardiomyocyte maturation.

Hypoplastic left heart syndrome (HLHS) is a severe form of congenital heart disease characterized by underdevelopment of the left ventricle. There are conflicting theories regarding the pathogenesis of this disease. We sought to further define the genetic landscape of HLHS through exome sequencing and mechanistic interrogation using a novel, functional genomics approach (fly, mouse, human). We previously reported whole exome sequencing of a cohort of patients with HLHS identifying the mutational landscape. Now, we extend our analysis, using a functional genomic Drosophila screen to analyze our identified variants more completely for developmentally important genes. We identified new loci that were previously filtered out by in silico techniques, and disregarded. In Drosophila, slu7 knockdown with human SLU7 rescue experiments demonstrate a critical role for SLU7 in heart tube formation. In mice, SLU7 IHC demonstrated cardiac-enriched expression in the developing heart. We created Slu7 gene-targeted mice, that were embryonic lethal, demonstrating an essential role for Slu7 in mammalian development. SLU7 knockdown resulted in defects in transcriptional and alternative splicing of developmental programs, many of which are essential for cardiac embryogenesis. One of the aberrantly spliced products is the chromatin-modifying gene KMT2D, one of the most common genes mutated in HLHS. Furthermore, our studies demonstrate SLU7 as a critical factor in RNA processing and cardiac development in multiple species. These pathways include novel, as well as known (KMT2D), genes, building a knowledge base to enable the development of rational therapeutic and counseling strategies for this complex set of patients.

A single layer of cells could be perceived as insignificant in comparison to a whole organ. However, extensive evidence has shown that the epicardium, the outermost layer of the heart, is essential for cardiac development and regeneration. Epicardium-derived cells (EPDCs) and epicardial signalling are crucial for coronary vasculature development and myocardial growth.

MEIS transcription factors (TFs) are important regulators of cardiac development. Single Meis1 or Meis2-null mice die at midgestation with severe cardiac alterations (Stankunas et al. 2008; Machon et al., 2015). Here, we show that MEIS TFs are expressed and play a function in the epicardium during cardiac development. Meis1 and Meis2 epicardial-specific conditional deletion show prenatal agenesis of the lymphatic vasculature. This reveals a previously unknown non-autonomous function of the epicardium in promoting cardiac lymphangiogenesis. We characterize a population of subepicardial, fibroblast-like lymphatic-associated EPDCs (LEPDCs) that completely enseaths lymphatic vessels as they grow towards the apex of the ventricles. LEPDC and Lymphatic Endothelial Cell (LECs) association is disrupted in Meis mutants, which suggests that LEPDC-LEC crosstalk is important for cardiac lymphangiogenesis. Transcriptomic analysis of Meis-mutant epicardium/subepicardium showed a downregulation of Vegfc and Vegfd lymphoangiocrine signals. Epicardial-specific Vegfc deletion provokes less developed and immature coronary lymphatic vessels, whereas analysis of the lymphatic vasculature of Vegfd knockout hearts shows that VEGFD is important for the development of ventral coronary lymphatics. These results show Meis-regulated direct cellular interactions and paracrine signalling from the epicardium/EPDCs is essential for cardiac lymphatic development.

Myxomatous valve degeneration (MVD) is a common valve disease affecting approximately 2.5% of individuals in the western world. While a lot is known about the clinical manifestations of MVD, the cellular mechanisms that cause MVD largely remain to be elucidated. Using the Wilms Tumor 1 epicardial cre mouse, it has been demonstrated that during development, cells from the epicardium undergo an epicardial-to-mesenchymal transformation and invade the atrioventricular (AV) junction. These epicardial-derived cells (EPDCs) preferentially populate the parietal valve leaflets which are derived from the lateral AV cushions during embryonic valvulogenesis. The mechanisms governing the directional migration of AV-EPDCs into these parietal leaflets are slowly emerging, yet their functional role in regulation of valve development and maturation remains unknown. In order to gain insight into the role of EPDCs in valve development, we conditionally knocked out SOX9, a transcription factor implicated in EMT, valve development, and cell migration, utilizing the Wt1 epicardial cre recombinase. This leads to a significant reduction in the number of EPDCs that populate the parietal leaflets.  Postnatal Wt1cre;SOX9fl/fl specimens have a valvular phenotype reminiscent of myxomatous valve degeneration (MVD) with abnormalities in extracellular matrix (ECM) organization. Transcriptomic analysis of these leaflets display aberrant regulation of genes associated with cell adhesion, regulation of ECM disassembly, and extracellular structure organization. Together, these data indicate an important role for EPDCs in valve development and homeostatic remodeling of the maturing valve and points to the possible involvement of the epicardium in MVD pathogenesis.

Aortic valve stenosis is characterized by an active calcification process driven by high mechanical stress-induced inflammation in the aortic valve leaflets. Interestingly, Singleton-Merten-Syndrome (SMS), a rare monogenetic disease, which can be caused by gain-of-function mutations in the retinoic acid-inducible gene I (RIG-I), is also characterized by diverse calcification abnormalities including calcification of aortic valve and proximal aorta. Therefore, we aim to investigate the role of the RNA receptor RIG-I in the development and progression of aortic valve stenosis. By using the mouse model of wire-injury, aortic valve stenosis was induced in wildtype (WT), RIG-I and MAVS (mitochondrial antiviral signaling protein) knockout (KO) mice. The development of the disease was monitored by ultrasound, while tissue samples were collected 6 weeks after the induction of aortic valve stenosis and analyzed using immunohistochemistry and fluorescence microscopy. The results showed a significant reduction of aortic peak blood velocity, an indicator of aortic stenosis, in mice deficient for RIG-I as well as for MAVS, the central adaptor molecule of RIG-I. Immunofluorescence staining revealed less CD45 positive immune cells infiltrating the aortic valve in MAVS KO mice compared to WT mice. In addition, MAVS deficiency led to a reduced activation of valvular cells in response to injury, indicated by the decreased expression of the mesenchymal cell marker aSMA in MAVS KO mice. The reduced severity of the aortic valve stenosis in RIG-I and MAVS KO mice suggests a possible role of RIG-I like receptor pathway in the pathophysiology of this disease.

Valvular endothelial cells (VECs) covering the aortic valve leaflets are exposed to different stresses, in particular wall shear stress (WSS). Perturbation in the mechanical environment of the valve can lead to valve leaflet remodeling which ultimately can result in calcification or degeneration of the valve. EGR1 and KLF2 are known as flow responsive transcription factors. However, the pathway activating these factors in response to mechanical signals is not well understood. Our study aims to decipher how EGR1 and KLF2 are activated in response to the WSS. We used a unique fluid activation device that applies physiologically relevant pulsatile WSS to identify the signal required to activate EGR1 and KLF2. We performed gene expression and/or western blot analyses to assess the downstream effect of EGR1 and KLF2. VECs exposed to the WSS were treated with U0126 (ERK1/2 inhibitor). Interestingly, VECs treated with U0126 show a downregulation of EGR1 and KLF2 transcription factors suggesting that the MAPK pathway and more precisely ERK1/2 is required for their activation.  We also used adenovirus or siRNA to overexpress EGR1 and KLF2 in valvular cells. We found that addition of EGR1 adenovirus results in the transcriptional activation of TGFB1 and eNOS while KLF2 is downregulated under the same condition. However, valvular cells treated with EGR1 siRNA over-expressed KLF2, suggesting that both genes are reversibly regulated. Interestingly, we found that EGR1 is a direct activator of the eNOS transcription as confirmed by the increase of luciferase activity when EGR1 is added to the experiment. Our ex vivo model using explanted valve confirm these data. In conclusion our results support that EGR1 and KLF2 are upregulated in VECs exposed to WSS and that this activation is dependent on the MAP kinase pathway such as ERK1/2. We showed that EGR1 is enabled to upregulate the transcription of eNOS an essential regulator of valve homeostasis.

INTRODUCTION: Understanding Mitral Valve Dystrophy (MVD) pathophysiological mechanisms is a current challenge. We have generated a unique knock-in (KI) animal model bearing the FlnA-P637Q mutation associated with MVD in humans and confirmed the presence of MVD in 3-week-old KI rats. The main signaling pathway identified by RNA-seq referred to immune cell recruitment, along with extracellular matrix remodeling and response to molecular stress. Those findings were corroborated with cytometry showing the increased proportion of myeloid cells in 3-week-old KI MV.

OBJECTIVE: The aim of the present study was to analyze earlier developmental time points and the specific role of endothelial, interstitial and immune cells in the onset and the progression of MVD.

METHODS: 7-day post-natal (D7) KI and WT rats were studied. Classical histology, FACS and qPCR experiments were performed to determine the presence and the nature of the valve remodeling, the proportion and the molecular signature for each cellular subpopulation.

RESULTS: At D7, histological analyses showed a myxomatous remodelling in KI MV with thickened and proteoglycan-rich MV as previously found at 3-week-old. On the other hand, contrary to 3-week-old, no difference was observed in myeloid cell proportion at this age (6% vs 6% for 12 WT and 10 KI animals, respectively; p=0.63). CD45+ leukocytes and CD206+ developmental macrophages were located more diffusely in KI MV. The ongoing molecular signature characterization for the different cellular subpopulation uses primers targeting Esm1, Cspg4, Wnt2 and Grem1.

CONCLUSIONS: Our results revealed a specific role of myeloid cells in the progression of MVD rather than the onset of the disease. The typical inflammatory, ECM, and cellular activation markers currently under investigation at D7 need to be tested at 3-week-old. Cellular cross-talk of endothelial, interstitial and immune cells needs to be further studied to decipher the molecular mechanisms leading to MVD.

Congenital aortic valve disease (AVD) is a most common type of congenital heart defect, but the underlying mechanisms remain largely unknown. NOTCH1 mutations are associated with AVD, and we previously reported that Notch1;Nos3 mutant mice display AVD. However, this model has limitations since ~65% of mutant mice suffer neonatal lethality. GATA5 is also implicated in human AVD, and Gata5^-/- mice exhibit bicuspid aortic valve but with only 25% incidence, showing an associated reduction in Notch signaling. We hypothesized that the interaction of Notch1 and Gata5 was critical for aortic valve morphogenesis, and generated novel murine models by intercrossing Notch1 and Gata5 heterozygote mice. Mice heterozygous for Notch1 and Gata5 (Notch1^+/-;Gata5^+/-), and heterozygous for Notch1 and null for Gata5 (Notch1^+/-;Gata5^-/-) demonstrated no neonatal lethality. Echocardiographic analyses demonstrated aortic valve stenosis (AVS) in ~50% of Notch1^+/-;Gata5^+/- mice and ~90% of Notch1^+/-;Gata5^-/- mice by 16 weeks of age. Notably, progression of AVS, as measured by an increase in aortic velocity from 6 to 16 weeks, was found in Notch1;Gata5 compound mutant mice. Furthermore, thickened and malformed aortic valves were found by histological examination in Notch1;Gata5 compound mutant mice at 16 weeks and 1 year as well as embryonic day 18.5 and postnatal day 10, consistent with a congenital phenotype. Interestingly, the expression of NICD and Nos3 was decreased in aortic valves of Notch1;Gata5 compound mutant mice as compared to wildtype littermates. In conclusion, our findings demonstrate a novel genetic interaction between Notch1 and Gata5. These new compound mutant mouse models display highly penetrant congenital and progressive AVS without lethality, and allow for unbiased genomic approaches to identify novel therapeutic targets for disease initiation and progression.

Neutrophils typically roll along the endothelium, ready to extravasate at a moment’s notice to the sites of tissue damage and/or pathogen invasion. The ability of extravasated neutrophils to re-enter the vasculature was first described in the zebrafish and later confirmed in the mouse model, establishing its relevance to mammalian systems. This endothelial re-entry has implications for  a rapid termination of inflammatory cytokine release by neutrophils. We examined neutrophil migration using zebrafish embryos at 2 days of development. After tail fin amputation, embryos were treated with selective chemical compounds targeting a number of receptor tyrosine kinases and intracellular signaling molecules (VEGFR, PDGFR, Eph, Abl, PI3K, mTOR and gamma-secretase) at concentrations that did not induce visually observable harm to tissues and the functioning vasculature. At least 50 embryos were scored for neutrophil migration at the end of 2h. Robust immobilization of neutrophils was observed upon treatment with selective inhibitors for the PI3K-delta and PI3K-gamma isoform (AS605240 and AS-252424), but not with pan-PI3K or PI3K-alpha-selective inhibitors. Unexpectedly, treatment with DAPT, an inhibitor of gamma-secretase, accelerated neutrophil migration. This observation and was confirmed using another, structurally distinct compound, PF03084014. So far, neutrophil-active compounds identified through library screening have reduced neutrophil migration rates. One compound derived from a medicinal extract, tanshinone IIA (protein target unknown), accelerated reverse migration. From our work, gamma secretase activity appears to negatively regulate directed, injury-induced neutrophil migration rates. Additional assays are underway to examine how these chemicals may alter netosis and/or vascular re-entry. Results from this study could provide novel targets for the development of neutrophil modifying drugs for research and for therapy.

Although most cardiovascular defects result from genetic mutations, many of the symptoms may be detectable only in adulthood.  Zebrafish is a well-established genetic model to investigate cardiovascular development, disease and regeneration.  However, the tools to analyze cardiovascular adult phenotypes in vivo remain limited.

Here, we combine multiple imaging techniques including echocardiography, magnetic resonance imaging, micro-computed tomography, electrocardiography, and light-sheet microscopy to study cardiovascular phenotypes in adult zebrafish.

To determine the sensitivity of our approach we analyzed genetic models with variable and mild cardiovascular phenotypes only manifested in adulthood.  Using these mutant and cell-ablation models, we identified functional impairment resulting from morphological alterations.  Specifically, we performed a correlation analysis of the different parameters and found that cardiac valve injury and reduced outflow tract wall thickness cause blood regurgitation, confirming the association between haemodynamic defects and structural alterations of the heart.

Overall, we propose a new approach to identify phenotypes in adult zebrafish that are too variable or subtle to identify with conventional histological analyses.  Importantly, this study helps set the zebrafish as an important model to analyze adult cardiovascular phenotypes potentially translatable to clinical settings.

Endothelial and erythropoietic lineages arise from a common developmental progenitor.  Etv2 is a master transcriptional regulator required for the development of both lineages.  However, the mechanisms through which Etv2 initiates the gene regulatory networks (GRNs) for endothelial and erythropoietic specification and how the two GRNs diverge downstream of Etv2 remain incompletely understood.  Here, by analyzing a hypomorphic Etv2 mutant, we demonstrate different threshold requirements for initiation of the downstream GRNs for endothelial and erythropoietic development.  We show that Etv2 functions directly in a coherent feedforward transcriptional network for vascular endothelial development, and a low level of Etv2 expression is sufficient to induce and sustain the endothelial GRN.  In contrast, Etv2 induces the erythropoietic GRN indirectly via activation of Tal1, which requires a significantly higher threshold of Etv2 to initiate and sustain erythropoietic development.  These results provide important mechanistic insight into the divergence of the endothelial and erythropoietic lineages.

Inflammation is triggered immediately after cardiac injury, and it has been hypothesized to promote regeneration in regenerative models such as zebrafish.  Myeloid differentiation factor 88 (MyD88) is a central adaptor molecule for the Toll-like receptor and interleukin-1 receptor signaling pathways, both of which are essential regulators of inflammation.  However, how specific molecules, such as MyD88, regulate inflammation after cardiac injury remains unclear.  In this project, we are testing the hypothesis that the MyD88 signaling axis plays a central role in the initiation of the inflammatory response after injury and that this early response is required for cardiac regeneration in zebrafish.  Our data indicate that zebrafish lacking functional MyD88 display features linked to impaired cardiac regeneration potential, including reduced immune cell recruitment to the injured tissue and reduced coronary endothelial cell proliferation at the border zone.  Additionally, transcriptomic analysis reveals that myd88 mutants exhibit a downregulation of immune-related processes after cardiac cryoinjury.  To understand the mechanisms activated downstream of MyD88, we identified effectors of the MyD88 signaling pathway and are studying their role with gain- and loss-of-function tools.  To determine the cell types in which MyD88 signaling plays a role during cardiac regeneration, we are using a transgenic approach to block MyD88 function in a cell type-specific manner.  This project will provide insights into the cellular and molecular mechanisms involved in the inflammatory response during zebrafish cardiac regeneration.  By modulating specific immune components and their response to injury, we could potentially improve the regeneration outcome in mammalian models including in humans.

The embryonic epicardium undergoes epithelial-to-mesenchymal transformation, contributing cells for formation of coronary vasculature. It is also a source of paracrine cues that are essential for fetal cardiac growth, coronary vessel patterning, and regenerative heart repair.

We have modified an established model of early embryonic transmural myocardial cryoinjury (resulting in incomplete repair by smooth muscle) by performing it at the stage, where the heart is already covered by the epicardium. We hypothesized that in the presence of the epicardium, the mechanisms of reparation will be different, and a complete regeneration will occur similar to adult fish or mammalian neonatal models.

Myocardial injury was induced by cryoprobe on ventricle of avian embryos (chicken and quail) at Hamburger-Hamilton (HH) stage 21, which were then re-incubated until HH31. Myocardial regeneration was analyzed electrophysiologically by epicardial optical mapping, and morphologically by immunohistochemical markers for endothelium (QH1), myocytes (MF20), and smooth muscle actin (SMA).

Functional analysis showed changes in epicardial signal spreading in the cryoinjured area. There were profound changes in the epicardium, which was thickened, detached and wrinkled in the affected, but also in the remote area. Numerous newly formed microvessels (QH1, SMA) and myocytes (MF20-positive) were present in the subepicardial region, confirming the regenerative capability.

In conclusion, embryonic myocardial cryoinjury at trabeculated stage induces epicardial activation with enhanced vascularization and myocardial regeneration. This could be potentially used for devising novel regenerative strategies in humans that could use epicardial activation to repair myocardial infraction.

Shroom3 is a member of a small family of proteins that regulate cell shape by coordinating cell signaling and cytoskeletal architecture. Shroom3 is comprised of a PDZ domain that facilitates localization, and two ASD domains that mediate actin and myosin dynamics. Through mouse and human studies, Shroom3 mutations have been implicated in congenital heart defects, suggesting that Shroom3 may be important in cardiac development. In the heart, shroom3 is primarily expressed in cardiomyocytes (CMs). Using CRISPR/Cas9 technology, we generated a global homozygous null mutation (shroom3^-/-) in zebrafish that survive to adulthood in contrast to the embryonic lethal gene-trap mutagenesis observed in mice. Differential gene expression analysis, ascertained by from wildtype versus shroom3^-/- zebrafish whole heart RNAseq, revealed significant changes in several signaling pathways known to regulate cardiac regenerative capacity including Hippo, PI3K-Akt, HIF-1, and ErbB in addition to metabolic alterations indicative of heart failure development. Although external heart morphometric analysis revealed no obvious differences, shroom3^-/- maturation is delayed overall compared to wildtypes. Functional cardiovascular assessment through an intermittent-flow variable swim tube fitness test illustrated that mutants are unable to metabolize oxygen efficiently and fatigue faster than wildtype counterparts. These findings were further supported via high frequency echocardiography which showed that shroom3^-/- had a statistically lower stroke volume, ejection fraction, and fractional shortening. Finally, an impaired regenerative response was observed in mutants at both 7 and 30 days post ventricular cryoinjury. These preliminary findings suggest that Shroom3 likely contributes to cardiac function and response to injury in zebrafish.

The characterization of novel mechanisms safeguarding cell fate identity in differentiated cells is essential for 1) our understanding of how differentiation is maintained in healthy tissues, and 2) the development of new strategies to improve therapeutic reprogramming. In this context, we recently identified a set of four transcription factors (TFs) that collectively promote cell fate stability and restrict direct lineage reprogramming in differentiated cells. Among those factors, JUNB was found to oppose reprogramming by limiting reprogramming TF (i.e MEF2C) ability to access its target DNA and remodel chromatin. AP-1 TFs, such as JUNB, are known to mediate their function via homo and hetero-dimerization with other proteins, thus we hypothesized that yet to be identified JUNB-interacting proteins might cooperate with JUNB to regulate chromatin accessibility and promote cell fate stability. Here we describe the identification and characterization of H1F0 as a novel JUNB interactant mediating cell fate stability and opposing cardiac reprogramming.

Arrhythmias are a hallmark of myocardial infarction (MI) and contribute to poor prognosis. Increasing evidence implicates the ventricular conduction system (VCS) in generating arrhythmias. How cardiac injury affects the global architecture and cellular composition of the VCS, and whether the VCS can endogenously regenerate or repair itself, have not been systematically described. The neonatal mouse heart can regenerate following left anterior descending coronary artery ligation at postnatal day 1 (P1). The P1 heart replaces lost cardiomyocytes, removes temporary scarring, and recovers full function. This transient regeneration capacity is lost during the first week, such that P7 hearts fail to regenerate myocardium and instead maintain a permanent fibrotic scar.

We used the knock-in reporter line Cx40-eGFP to investigate VCS changes after injury in P1 versus P7 mice. By developing advanced tissue-clearing and wholemount imaging, we determined the 3D anatomy of the postnatal VCS. We observed loss of VCS cells near the injury site in infarcted P1 and P7 hearts, resulting in discontinuity within the Purkinje network. Longitudinal analysis showed an increased density of connexin-40-positive cells surrounding the injury zone at P1. We are currently using single-cell transcriptomic analysis to identify molecular pathways associated with VCS response to injury, (re-) growth and function. Collectively, our work identifies specific morphological changes in the Purkinje network following MI across the regenerative window, with evidence of recovery at early neonatal stages. These findings provide significant insights into VCS pathophysiology during heart injury and may identify molecular targets to maintain synchronous contraction in infarcted adult hearts.

Bmp and Fgf signaling are widely involved in multiple aspects of embryonic development. Non-coding RNAs, such as microRNAs, have also been recently reported to play essential roles during embryonic development. We previously demonstrated that microRNAs are capable of modulating cell fate decision during proepicardial/ septum transversum (PE/ST) development in chicken, since over-expression of miR-23 blocked while miR-125, miR-146, miR-223 and miR-195 enhanced PE/ST-derived cardiomyogenesis, respectively. Importantly, regulation of these microRNAs is distinct modulated by Bmp and Fgf administration. In this study, we aim to dissect the functional role of Bmp and Fgf signaling during mouse ST/PE development, their implication regulating these microRNAs and their impact on lineage determination. Mouse PE/ST explants and epicardial/endocardial cell cultures were distinctly administrated Bmp and Fgf family members. qPCR analyses of microRNAs, cardiomyogenic, fibrogenic differentiation markers as well as key elements directly epithelial to mesenchymal transition were evaluated. Our data demonstrate that neither Bmp2/Bmp4 nor Fgf2/Fgf8 signaling is capable of inducing cardiomyogenesis, fibrogenesis or inducing EMT in mouse ST/PE explants, yet upregulation of miR-146 and miR-195 is observed. RNAseq analyses in mouse PE/ST and embryonic epicardium identified novel Bmp and Fgf family members that might be involved in such cell fate differences, such as Bmp6, Bmp7, Bmp19, Fgf5, Fgf7 and Fgf10. Administration of these Bmp and Fgf family members in mouse PE/ST lead to limited modulation of EMT induction and cardiomyogenic and/or fibrogenic differentiation. Thus, our data support the notion of species-specific differences regulating PE/ST cardiomyogenic lineage commitment.

The Wilms tumor suppressor gene (Wt1) encodes a C2H2-type zinc-finger transcription factor that participates in transcriptional regulation, RNA metabolism and protein-protein interactions. WT1 is critically involved in the development of several organs, including kidneys and gonads, spleen, adrenals, liver, and diaphragm (Hastie, 2017). WT1 is highly expressed in the embryonic epicardium where it regulates a process of epicardial-mesenchymal transformation and the development of the epicardial-derived cells.

We have recently shown evidence of a transient Wt1 expression in about 25% of cardiomyocytes of mouse embryos. Conditional deletion of this expression in the cardiac troponin T lineage caused abnormal sinus venosus and atrium development, thin ventricular myocardium and, in some cases, interventricular septum and cardiac wall defects (Díaz del Moral et al., Front Cell Dev Biol. 2021;9:683861).

We aimed to know if Wt1 is also expressed in adult cardiomyocytes and what could be the consequences of its conditional deletion for cardiac homeostasis and/or in the response to damage induced by isoproterenol and doxorubicin treatments.  For conditional deletion of Wt1 in cardiomyocytes, we generated tamoxifen inducible Wt1 mutants by crossing aMHC^MerCreMer mice with homozygous Wt1 conditional mice, where the first exon of Wt1 is flanked by loxP sites.

We have found experimental evidence of a low expression of Wt1 in postnatal murine cardiomyocytes, using reporter and lineage tracing models as well as qPCR. Our preliminary data suggest that conditional deletion of Wt1 in cardiomyocytes induces interstitial fibrosis, increased oxidative stress markers, altered metabolism and mitochondrial dysfunction in Wt1-deficient cardiomyocytes. In addition, conditional deletion of Wt1 in adult cardiomyocytes increases the damage induced by doxorubicin and isoproterenol treatments. These findings suggest a novel role of Wt1 in myocardial physiology and protection against damage.

Cardiomyocyte renewal by dedifferentiation and proliferation has fuelled the field of regenerative cardiology in recent years, while the reverse process of redifferentiation remains largely unexplored. Redifferentiation is characterised by the restoration of function that is lost during dedifferentiation and is key to the healing process following injury. Previously, we showed that ERBB2-mediated heart regeneration has these two distinct phases: dedifferentiation, followed by redifferentiation. Here, using temporal RNAseq and proteomics, we survey the landscape of the dedifferentiation-redifferentiation process in the adult mouse heart. We find well characterised dedifferentiation pathways, such as reduced oxphos, increased proliferation and increased EMT-like features, largely return to normal, though surprisignly, elements of residual dedifferentiation remain, even after contractile function is restored. These hearts appeared rejuvenated and showed robust resistance to ischaemic injury. We find that redifferentiation is driven by negative feedback signalling, notably through LATS1/2 Hippo pathway activity. Disabling LATS1/2 in dedifferentiated cardiomyocytes augments dedifferentiation in vitro and prevents redifferentiation in vivo. Taken together, our data reveal the non-trivial nature of redifferentiation, whereby elements of dedifferentiation linger in a surprisingly beneficial manner. This cycle of dedifferentiation-redifferentiation protects against future insult, in what could become a novel prophylactic treatment against ischemic heart disease for at-risk patients.

Objectives

There is increasing evidence of the importance of interstitial cells in cardiovascular development and disease. In contrast to the adult mammalian heart, the adult zebrafish heart has the ability to regenerate. However, little is known about how interstitial cell populations contributing to this. We aim to better characterise interstitial cells with an emphasis on cardiac fibroblasts, during the regeneration process.

Methods

Using immunohistological analysis of the zebrafish heart from multiple transgenic reporter lines, we were able to quantify the various cell types that comprise the adult zebrafish heart. In addition, we performed single-cell RNA sequencing (scRNA-seq) on uninjured and amputated hearts at various time points in order to uncover the transcriptional properties of cardiac fibroblasts, as well as other interstitial cell populations that support adult zebrafish heart regeneration. The data was cross-referenced to published scRNA-seq data from murine studies to identify features of zebrafish cardiac fibroblasts.

Results

We were able to establish that, as in the adult mouse heart, endothelial cells and cardiomyocytes where the main cell types. However, fibroblasts where far less abundant in the adult zebrafish heart. Indeed, they represented 1-2% versus 11% of total cells in adult zebrafish versus mouse heart, respectively. ScRNA-seq analysis revealed features of interstitial cells that may support regeneration. Notably, in zebrafish, we found very limited fibroblast to αSMA⁺-myofibroblast transition as compared with injured mouse heart. By pharmacologically blocking regeneration, we found that αSMA⁺ expression was upregulated, but not in fibroblasts expressing high levels of collagen.

Conclusion

Our study suggests that features of myofibroblast transition, such as increased aSMA expression, are less pronounced in zebrafish cardiac fibroblasts as compared to fibroblasts in regenerating neonatal or infarcted adult mouse heart.

Myocardial infarction (MI) induces death of cardiac muscle and replacement by a non-contractile fibrotic scar, which ultimately leads to heart failure. Current treatments restore blood flow and assist with cardiac workload but none are regenerative, and clinical trials using stem-cell- based approaches have been disappointing. We have previously shown that the cardiac lymphatic system is activated following MI and that further stimulation with the lymphatic endothelial specific isoform of VEGF-C, VEGF-CC156S, resolves the immune response and improves cardiac function. Whilst this study demonstrates the potential of this approach, VEGF-CC156S is sub-optimal for clinical use. 

Our aim is to target the lymphatic vasculature to enhance the intrinsic regenerative potential of the infarcted adult heart using small molecules. To achieve this, we have established a sprouting assay using human lymphatic endothelial cells (hLEC), to mimic lymphangiogenesis in vitro and to provide the basis of a phenotypic screen. We have miniaturized this cell-based assay into a 384 well-plate format and have begun to screen focused libraries of compounds (epigenetic regulators, kinase inhibitors and stem cell modulators). Using automated imaging and post-hoc analyses we have selected, clustered and quantified hit molecules, which are currently being validated and further characterised through combined medicinal chemistry and preclinical models of adult cardiac injury.

Cardiac valve disease is a prevalent cause of morbidity and mortality worldwide, in which the cardiac valves that are normally ensuring unidirectional blood flow during the cardiac cycle, become defective. In severe cases, the diseased valve is surgically removed and replaced by decellularized implants, which generally rely on the repopulation by endogenous cells from the host. However, this recellularization process which ensures the stability and growth of the implant is greatly challenged in mammals due to the limited regenerative potential of their cardiac tissues.

We recently showed that adult zebrafish are capable of regenerating their cardiac valves. With the use of the nitroreductase-metronidazole genetic ablation system, we showed that valve-cell ablation in the adult zebrafish triggers the recruitment of endothelial cells and kidney marrow-derived cells, leading to the formation of a new functional valve. However, studying this process in histological sections of fixed tissues limits the assessment of the cellular contributions and rearrangements that take place during cardiac valve regeneration in vivo.

To overcome the limitations associated with adult specimens, we are now using zebrafish larvae to study the cellular dynamics and rearrangements involved in valve cell recruitment through live imaging. We have observed that also in larval stages, genetic ablation triggers a regenerative program leading to the recruitment of new valve cells, clearly visible at five days post-ablation. We also observed that this program is accompanied by an increase in the number of immune cells in the vicinity of the valve. We are now characterizing the different cellular sources and molecular pathways contributing to the formation of the new valve through live-imaging microscopy.

Overall, with this approach we expect to clarify the in vivo mechanisms of new valve cell recruitment after valve-cell ablation and determine the key factors promoting cardiac valve regeneration.

The mouse heart can regenerate during a short time window after birth, which coincides with maturation of the ventricular myocardium and compaction of trabeculae. These transient invaginations at the inner surface of the developing heart give rise to both contractile cardiomyocytes (CM) and Purkinje fibers of the ventricular conductive system (VCS). Purkinje fibers are recruited throughout embryonic and fetal development in successive waves, including a major wave of around the time of birth. This depends on maximal Nkx2-5 levels and Nkx2-5 heterozygous mice present a hypoplastic VCS. Here we investigate the contribution of trabecular myocardium to the regeneration of contractile and conductive cardiac tissues after neonatal myocardial infarction (MI). 

Neonatal MI was performed by permanent ligation of the left coronary artery on one-day-old mice. Trabecular derivatives were genetically traced using Connexin40 (Gja5) driven inducible Cre recombinase, revealing an active participation and increased proliferation of trabecular-derived cells during cardiac repair. Moreover, subendocardial myocardium re-expresses Cx40 under ischemia, consistent with a switch towards an immature phenotype. Surprisingly, we observed that regenerated hearts possess a global and permanent hyperplasia of the distal VCS despite spatially restricted damage following MI. This MI induced hyperplasia is not found in Nkx2-5+/- mice, suggesting that the underlying mechanism is Nkx2.5 dependent. We hypothesize that VCS hyperplasia under conditions of regeneration is an extension of the normal perinatal recruitment wave.

Together our data uncover cellular mechanisms involved in postnatal regeneration that resemble normal maturation of the ventricular myocardium. Trabecular-derived CM dedifferentiate toward an immature bipotent phenotype. The proliferation of these cells helps in repopulating the myocardial ischemic zone but produces an excess of PF which leads to VCS hyperplasia.

Myocardial infarction causes irreversible loss of cardiomyocytes and formation of a fibrotic scar, impairing cardiac function in patients. While the mammalian heart is incapable of regenerating the lost cardiomyocytes following injury, zebrafish can completely regenerate their heart through cell cycle re-entry of pre-existing cardiomyocytes. The molecular mechanisms underlying differences in regenerative capacity between species have yet to be unraveled. We employed spatial transcriptomics on mouse and zebrafish hearts post injury and compared the transcriptomes of their respective border zones to identify overlapping and diverging expression patterns. This dataset will be a helpful tool to identify drivers of cardiac regeneration. Here we will present the dataset and its validation. Furthermore, the dataset was used to select candidate genes for functional analysis in zebrafish using a knock-out approach of which the results will be presented here.

Dysregulated sympathetic innervation pattern and activity are observed in a number of cardiovascular diseases including post myocardial infarction.  Upon cardiac injury, sympathetic nerves (SNs) in the neonatal mouse heart migrate into the injury site and ablation of SNs post-injury impairs neonatal mouse heart regeneration.  In contrast, SNs fail to migrate into the injury site and accumulate at the border zone in the post-infarction adult mouse heart and it fails to regenerate.  While these data suggest that SNs play an important role in heart regeneration, when and how the SNs migrate into the injury site and their role(s) in heart regeneration remain unknown.

Here using the zebrafish heart, which regenerates post-injury, we have analyzed when and how SNs migrate into the injured tissue.  Using various reporter lines and antibodies, we find that the zebrafish SNs migrate rapidly into the injured tissue.  A subset of nerves persists within the injured tissue for several days and the innervation pattern is re-established within 30 days post-injury.  This rapid response of SNs is accompanied by the disintegration of the pre-existing neural network.  Altogether our studies indicate that after cardiac injury, cardiac SNs are very dynamic.  Hence, analyzing the function of SNs after cardiac injury will significantly advance our understanding of their role(s) during organ regeneration and cardiac disease.

Clinical and preclinical studies reported insufficient cell engraftment and delayed cardiac functional recovery problems in stem cell-based therapy in heart failure. Therefore, pretreatment of stem cells with an exogenous factor may improve regenerative efficiency. Here we demonstrated the effect of estrogen administration on the regenerative capacity of CPCs with in vitro and in vivo experiments. Explant-derived CPCs were isolated from adult mouse hearts and molecular, functional, and electrophysiological changes in CPCs upon estrogen treatment were investigated. Estrogen treatment increases the expression of proliferation, pluripotency and stem cell markers like Abcg2 and Islet-1. The migration and vasculogenic potential of CPCs was also induced by estrogen. Considering the paracrine effect of transplanted progenitor cells, we have examined the changes in the exosomal miRNA profile of CPCs. We found that estrogen regulates exosomal miRNAs known as modulators in neo-angiogenesis and cell migration. These results provided molecular evidence that estrogen pretreatment improves the regenerative capacity of CPCs inducing their self‐renewal, potency and repairment mechanisms. We have also monitored ohmic behaviour for K-channels like inward rectifier K+-current and significant hyperpolarization-shift in inward currents upon estrogen treatment suggesting the establishment of the resting membrane potential and differentiation induced by estrogen treatment. We have observed the inductive role of estrogen on cardiomyocyte differentiation of CPCs. To support in vitro data, we have used mouse heart failure model established by isoproterenol injection. Molecular and electrophysiological changes after the intracardiac allogenic transplantation of estrogen pretreated CPCs were also demonstrated. All these results indicate that ex vivo pretreatment of CPCs with estrogen may propagate the transplantation efficiency.

Cardiomyocyte proliferation is a critical source of new myocardium during heart development and regeneration. Novel therapies designed to replace missing or lost cardiomyocytes in the context of congenital heart disease (CHD) or myocardial infarction (MI) represents the holy grail of cardiac regenerative medicine. To boost cardiomyocyte proliferation in either setting, critical regulators must be identified. We reported that Reptin (also known as Ruvbl2), a highly conserved and widely expressed AAA+ ATPase, is a potent suppressor of cardiomyocyte proliferation during heart development and regeneration. Specifically, reptin loss-of-function mutants display ventricular hyperplasia at 3 days post-fertilization (dpf), while cardiomyocyte-specific reptin overexpression causes ventricular hypoplasia. Moreover, heat-inducible reptin overexpression during adulthood also lowers the myocardial proliferation index and results in scarring. Building on these published data, I am currently investigating whether the suppressive activity of Reptin on cardiomyocyte proliferation is conserved in mice and dissecting the downstream molecular mechanisms mediating this function. To address the former, I created cardiomyocyte-specific Reptin knock-out mice by crossing cTnt:Cre;Reptin^fl/+ X Reptin^fl/fl. Of 38 liveborn pups, zero mutant animals were recovered. I found that the conditional mutants die by E12.5, demonstrating a strict requirement for murine heart development. I am currently investigating cardiomyocyte proliferation in E10.5 and E11.5 conditional knock-out embryos and analyzing differentially expressed genes identified by RNA-seq from wild-type and reptin mutant cardiomyocytes in zebrafish. 

Cardiovascular diseases remain the leading cause of death worldwide, and few effective treatment options are available. Myocardial injury, such as myocardial infarction, causes irreversible damage of the heart muscle and its replacement by scar, leading to a chronic decrease in heart function. In contrast to humans, the injured zebrafish heart muscle regenerates efficiently through robust proliferation of myocardial cells. Thus, the zebrafish presents a beneficial vertebrate model for studying genetic programs behind cardiac regeneration, which may be present, albeit dormant, in the adult human heart.

To this end, we established a novel platform for studying heart regeneration after cardiomyocyte ablation in zebrafish larvae. The specific ablation of cardiomyocytes is achieved through a transgenic construct inducing the expression of nitroreductase, a bacterial enzyme, in a pool of ventricular cardiomyocytes. Subsequent treatment with antibiotics induces cell death specifically in nitroreductase-expressing cells. In combination with automated 3D heart imaging, this platform can be used for medium-throughput screening of genes and compounds with presumed effects on regeneration. Our results confirm that we induce a robust loss of the targeted cardiomyocytes, which are replaced through the proliferation of remaining cardiomyocytes within 4 days post injury. Our results further show that treatment with known anti-regenerative molecules causes a significant delay in regeneration kinetics, providing a proof of principle for this platform in identifying anti-regenerative effects of genes and drugs. Using this platform, we aim to discover therapeutic targets and drugs that will allow us to activate the dormant regenerative potential of the human heart.

Cardiac progenitor cells (CPC), whose regenerative capacity is reduced due to cardiac damage, play a role in heart failure that develops after heart damage. Regenerative capacity of multipotent CPCs is being evaluated by clinical phase-1 studies. Due to the low rates of inoculation and differentiation, treatment modalities based on CPC transplantation need to be improved. Our aim in this study is to show how estrogen replacement affects cardiac regeneration and differentiation capacity of CPCs in female ischemic animals. Recent studies  have shown that estrogen has an active role in cardiac regeneration both in functional and molecular levels. In addition, in another study by our group, we showed that the effects of estrogen on CPC differentiation have a positive effect on the regenerative capacity of CPCs. Evaluation of estrogen-related changes in the regenerative capacity of female ischemic CPCs and healthy CPCs is important in order to better understand the role of estrogen in cardiac regeneration. To investigate this, we examined changes in endothelial, fibroblastic, and cardiac differentiation capacities of CPCs derived from healthy and ischemic heart explants upon estrogen replacement. Our results showed that estrogen has a positive effect on these parameters. We also examined estrogen-dependent changes in EMT markers (Vimentin, Snail, Slug, CDH1, and CDH2) affecting the engraftment rate. Our in vitro results showed a shift towards mesenchymal characteristics upon estrogen treatment. Our results with this study are a step towards a better understanding of the effects of estrogen on the regenerative capacity of the hearts of healthy or ischemic females.

Cardiac progenitor cells (CPCs) are a diverse group of cells found in the adult heart that can develop into distinct cardiac lineages. CPC transplantation has poor stem cell engraftment and delayed cardiac functional recovery difficulties in stem cell-based heart failure treatment. Therefore, exogenous factor therapy of CPCs may boost regeneration efficiency. Estrogen is recognized to play a role in heart regeneration on both a functional and molecular level. In vitro experiments, we wanted to see how estrogen treatment affected Sca1+ regenerative ability. The Sca1+ CPC subpopulation was isolated exclusively from mouse cardiac explants. When compared to mRNA levels in heart tissue, the expression level of pluripotency markers in sorted Sca1+ cells is significantly higher. Estrogen also promotes colony formation and proliferation. In wound healing and angiogenesis experiments, estrogen therapy promoted migration and micro vascularization in Sca1+ cells. The expression of mesenchymal markers has also increased, implying that the outcomes are functional. We also looked for ohmic activity in K-channels, which was identical to the inward rectifier K+-current, as well as a considerable hyperpolarization shift in inward currents after estrogen therapy. These data show that estrogen therapy promotes the development of the resting membrane potential and differentiation. All of these findings suggest that preconditioning CPCs with estrogen ex vivo before transplantation may improve transplantation efficiency and could be employed as a therapy option in clinical trials.

 Surjya Narayan Dash,1* Suneeta Narumanchi,2* Jere Paavola,2 Sanni Perttunen,2 Hong Wang,1 Päivi Lakkisto,2,3 Ilkka Tikkanen,2,4* and  Sanna Lehtonen1*

¹Department of Pathology, University of Helsinki, Helsinki, Finland; ²Unit of Cardiovascular Research, Minerva Institute for Medical Research, Biomedicum Helsinki, Helsinki, Finland; ³Clinical Chemistry and Hematology, University of Helsinki and Helsinki University Hospital, Helsinki, Finland; and ⁴Abdominal Center, Nephrology, Helsinki University Hospital, Helsinki, Finland

 * equal contribution

Actin is the most abundant and conserved protein in eucaryotic cells and involved in various cellular functions from cell division and migration to muscle contraction. Actin and myosin generate contractile force in cardiac muscle, and mutations in these proteins may lead to heart failure. Septins are small GTPases that associate with actin filaments and play an important role in cytoskeleton organization, however, their role in cardiomyocyte organization and function is not completely understood. We found that septin 7 is expressed in both embryonic and adult zebrafish heart. Knockdown of sept7b, the zebrafish ortholog of human septin 7, reduced F-actin and α-cardiac actin expression in heart and caused disorganization of actin filaments. Electron microscopy revealed disorganization of heart myofibrils and partial detachment from Z-discs in sept7b knockdown larvae. Functional studies showed that the knockdown larvae had reduced ventricular dimensions, contractility and cardiac output. Additionally, depletion of sept7b reduced the expression of retinaldehyde dehydrogenase 2 (raldh2), which catalyzes synthesis of retinoic acid (RA), necessary for heart morphogenesis. Furthermore, we found that sept7b and RA signaling pathways converge to regulate cardiac function. Together, these results identify an essential role for sept7b in zebrafish heart development and function.

During embryonic heart development, epicardial cells residing within the outer layer of the heart undergo epithelial-mesenchymal transition (EMT) and migrate into the myocardium to support and stimulate organ growth and morphogenesis. Disruption of epicardial EMT results in aberrant heart formation and embryonic lethality. Despite being an essential process during development, the regulation of epicardial EMT is poorly understood.  Here we report EMT relative to the topography of the epicardial surface of the embryonic heart at high resolution using scanning electron microscopy (SEM). We identified high- and low-EMT regions within the mesothelial layer of the epicardium and an association of these regions with key components of the extracellular matrix (ECM).  The ECM basement membrane-associated proteoglycan agrin was found to localize in the epicardium in regions actively undergoing EMT. Deletion of agrin resulted in impaired EMT and compromised development of the epicardium, accompanied by down-regulation of the epicardial EMT regulator WT1. Agrin enhanced EMT in human embryonic stem cell-derived epicardial-like cells by decreasing b-catenin and promoting pFAK localization at focal adhesions. In addition, we observed that agrin promoted the aggregation of its receptor dystroglycan within the Golgi apparatus in murine epicardial cells. Loss of agrin resulted in dispersal of dystroglycan throughout the epicardial cells in embryos, disrupting basement membrane integrity and impairing EMT. Our results provide new insights into the role of the ECM in heart development and implicate agrin as a critical regulator of epicardial EMT.  

Proper functional remodeling of atrio-ventricular junction (AVJ) leaving a sole myocardial connection in the annulus fibrosus (His bundle) is crucial for the post-septated heart development. Presence of myocardial accessory pathways (APs) bypassing the annulus fibrosus can lead to the ventricular pre-excitation and could lead to severe tachyarrhythmias. The aim of our study was to functionally characterize remodeling of AVJ and diminution of myocardial atrio-ventricular connection throughout chick cardiogenesis. At the embryonic day (ED) 14, 16 and 18, we used optical mapping of retrogradely perfused heart to determine atrio-ventricular delay and localize the APs. Measurements were performed during normal conditions and after atrioventricular node conduction slowing by adenosine injection. Hematoxylin-Eosin /Alcian Blue and Picrosirius Red staining were used for morphological validation of the APs. We observed a decreased proportion of hearts with electrically active APs from 60 % at ED 14 to 50 % at ED 16 and 20 % ED 18, respectively. At ED 14, the majority of the APs were detected during normal conditions and were distributed all around the AVJ. However, the most of the APs were detected after adenosine perfusion only at 16 and 18 ED, and were mostly right-sided. This was accompanied by a trend to decrease in atrio-ventricular delay during the studied period.  Progression of AVJ fibrotization with APs disappearance was confirmed by histology. We provide a view on functional AVJ remodeling throughout chick development. This could help to elucidate spatiotemporal characteristic of proper AVJ isolation and understanding of arrhythmias associated with presence of APs.

Crucial for heart function is its partition into a left and right pump. Left-right asymmetry of cardiac morphogenesis begins with the rightward looping of the embryonic heart tube, which positions cardiac chambers relative to each other to establish the double blood flow. In humans, disturbances in this process have been associated with severe heart defects in the heterotaxy syndrome. We have previously demonstrated that the left signal Nodal is required in cardiac precursors for orienting and shaping the embryonic heart loop (Desgrange et al., 2020). Nodal mutants later develop an heterotaxy syndrome with a spectrum of heart defects. However, the heterogeneity of the heterotaxy phenotype has remained poorly understood. Since Nodal mutants at E9.5 display four categories of abnormal heart loops, we reasoned that heterotaxy may be stratified. For example, in the clinical nomenclature, the final position of the ventricles is considered to reflect the direction of the embryonic heart loop. However, this correlation has never been tested experimentally. To tackle the challenge of monitoring a specific phenotype at two stages of development in a single embryo, we developed a multimodality imaging pipeline, combined with advanced 3D image analyses. With this unique approach, we can identify heart loop features that are predictive of specific cardiac malformations. This study provides novel insights into the emergence of complex congenital defects in the heterotaxy syndrome.

Pediatric and adult patients with Chronic Atrial and Intestinal Dysrhythmia (CAID) have a recessive point mutation (A > G) in the first coding exon of the cohesin interacting protein Shugoshin 1 (SGO1). This leads to a lysine-to-glutamic acid change at the highly conserved amino acid 23 in SGO1 (p.Lys23Glut: SGO1-K23E). To investigate the role of Sgo1 in vivo, we generated Sgo1-K23E and Sgo1-flox mouse models. Affected Sgo1-K23E/K23E homozygous mouse mutants die embryonically and show global developmental delay, consistent with an early embryonic requirement for Sgo1. Non-affected Sgo1-K23E/K23E homozygous mouse mutants survive postnatally at reduced Mendelian ratios and in a genetic background-dependent manner, but develop normally with no overt signs of functional or structural defects. Likewise, early postnatal conditional ‘knock-in’ of Sgo1-K23E similarly does not affect cardiac or intestinal function, but, surprisingly, leads to lack of neonatal hair growth. Mesodermal, second heart field, and neural crest lineage-specific Sgo1 knock-out mutants show complete embryonic lethality and lineage-specific defects shortly upon activation of Cre expression. Together, our results indicate that mouse Sgo1-K23E induced defects are genetic background dependent and that Sgo1 is likely globally required throughout development.

The genetic control of heart development has been extensively studied, partly due to the prevalence of congenital heart disease (CHD). CHDs primarily arise from mutations affecting gene dosage or altering spatial/temporal expression patterns, making non-coding elements that control these aspects of gene expression particularly important. However, the annotation and functional dissection of non-coding elements is lacking when compared to the coding genome.

Recently, our lab identified 8866 human regulatory elements deeply conserved between zebrafish and humans. Dubbed accessible conserved non-coding elements (aCNEs), these open chromatin regions are enriched for enhancers and are conserved at the DNA sequence level between zebrafish and humans. Due to the deep conservation of these regions, we hypothesize that a subset of aCNEs are conserved critical regulators of cardiac development and implicated in cardiac disease.

To identify regions with a likely role in cardiac development, I used publicly available ChromHMM data from human fetal hearts to identify aCNEs with chromatin signatures representing enhancer states. I then used published ChIPseq datasets to identify regions bound by known cardiac transcription factors (TF), including NKX2.5, GATA4/5/6, and TBX5. These criteria identified 329 regions, capturing previously identified cardiac enhancers, and predicting 258 novel cardiac enhancers.

Functional examination of 17 of these novel regions showed that 88% were able to drive reporter gene expression in the developing zebrafish heart. Strikingly, two of these regions contain ultra-rare SNVs in patients with familial Left-sided cardiac legions. Currently, we are working on linking upstream TFs and downstream target genes to aCNEs to gain molecular insight into how these regions function in cardiac development and how alterations may contribute to CHDs. Altogether, this work identified putative cardiac enhancers crucial in cardiac development and disease.

Congenital heart defects (CHDs) are the most common human birth defect, occurring in 1% of newborns. Cardiac neural crest cells (cNCCs) are a migratory and multipotent cell population known to aid in the development of the cardiac outflow tract (OFT), valves, and interventricular septum. We aim to establish the contribution of Yap and Taz effectors of the fundamental Hippo signaling pathway in neural crest (NC)-derived cardiac development. Conditional knockout (CKO) mice were created using a cre-lox system, by Wnt1^cre and Wnt1^cre2SOR, NC-specific drivers.Yap^+/-;Taz^-/- CKO mutants produced various OFT and remodeling defects including ventricular septal defect (VSD), tetralogy of Fallot, double outlet right ventricle, and cardiac valve aberrations, varyingly between E14.5, E16.5, and E18.5. Interestingly, Yap^+/-;Taz^+/- double-heterozygous hearts exhibited external morphology similar to that of controls, but sectioning revealed mild VSD, along with cardiac valve leaflet irregularities, at E16.5 and E18.5. Cell apoptosis and proliferation were unchanged in Yap^+/-;Taz^-/- CKO mutants. Compelling preliminary data utilizing various methodologies including RNA-sequencing, Reverse Phase Protein Array (RPPA), Assay for Transposase-Accessible Chromatin-sequencing (ATAC-seq), and transwell migration assay, indicate a potential role for Yap/Taz in regulating NC migration. Preliminary migratory studies are followed by both OFT culture and live-imaging. Together, our data indicate that Yap/Taz play a critical role in proper cNCC-derived cardiac formation.

Cornelia de Lange syndrome (CdLS) is a rare genetic and developmental disorder affecting about 1:10 000/1:30:000 children. Cardiac defects are observed in more than 50% of patients. CdLS children feature septal defects, and outflow tract defects including the hypoplastic aorta, stenosis, or coarctation of great arteries as well as Tetralogy of Fallot. We found that Nipbl+/- mice on a C57Bl/6 genetic background featuring a significant decrease in Nipbl mRNAs as well as a decrease in the protein in the heart also featured a severe delay in embryonic and postnatal growth. High-Resolution Episcopic Microscopy analysis of the heart at birth revealed ventricular hypertrophy, ventricular septation defects associated with a persistent truncus arteriosus. The adult hearts then feature a severe aortic phenotype with an enlargement of the intima including senescent cells and stenosis resulting in an increase in aortic flux velocity and persistent left ventricular hypertrophy. The aortic phenotype was recapitulated when Nipbl was deleted in smooth muscle cells using Sma^CreERT2 x Nipbl^fl/fl mice. Using proteomics and RNA-sequencing of the outflow tract, we identified a dysregulated TGFb and retinoic acid pathways in embryonic Nipbl+/- hearts as well as the presence of P21+  and pERK+ senescent cells as early as in E13.5 Nipbl+/- embryonic hearts.

Human CdLS patient iPS cells differentiated in smooth muscle cells also featured a severe senescence phenotype. The phenotype was reversed by an ALK5 inhibitor

Treatment of pregnant mice with an ALK5 inhibitor from E9.5 to E13.5 also rescued the cardiac phenotype as well as the body size of mice at birth. 

Down Syndrome (DS), caused by trisomy of human chromosome 21 (Hsa21), results in a range of phenotypes, including congenital heart defects (CHD). These phenotypes arise from increased dosage of one or more of the ~230 genes on Hsa21, however the gene(s) needed in three copies to cause CHD and the underlying mechanisms remain unknown.

The lab has generated mouse model for DS (Dp1Tyb) carrying a duplication of 23Mb on Mmu16, orthologous to Hsa21. These mice show the types of CHD seen in babies, such as ventricular septal defects and atrioventricular septal defects. The overall aim is to identify causative genes and understand the pathological mechanisms underlying CHD in DS.

Using our mapping-panel, we found that a critical gene for the CHD phenotype is Dyrk1a. Our data show that DS fetal hearts have decreased oxygen consumption and proliferation, suggesting that this is an important cause of the developmental defects. We show that three copies of Dyrk1a are required for CHD in Dp1Tyb. Three copies of Dyrk1a are also required for the mitochondrial and proliferation defects in embryonic cardiomyocytes.

Our preliminary analysis of the atrioventricular cushion (AVC) shows a reduction of cellular density as well as a reduction of NFATC1 signaling pathway, involved in valve morphogenesis. Thus, decreased NFAT signaling in the AVC might lead to decreased mesenchymal proliferation and cause the reduced AVC density in Dp1Tyb mice.

Taken together, we propose that CHD in DS arise in part from increased DYRK1A activity in cardiomyocytes leading to mitochondrial and proliferation dysfunction.

The epicardium is a layer of epithelial cells covering the surface of the heart that provides important cellular contributions for embryonic heart formation. In addition, seminal works during the last decade have pointed out that the adult epicardium is re-activated after heart damage contributing to cardiac remodeling. Several fate mapping and cell lineage studies have demonstrated that coronary vascular smooth muscle cells (cVSMC) and cardiac fibroblasts develop from the EPDCs in a multi-step process involving cell proliferation, epithelial-to-mesenchymal transition (EMT) and cell migration. However, the molecular signalling cascades regulating EPDC specification and migration are poorly understand. Here we show that miR-200b is expressed from E12.5 to E15.5 during heart development. FISH in situ hybridization analysis in Wt1Cre‐YFP, embryos, as well as, qPCR of sorted cells, showed that miR-200b is present in a cell subpopulation of EPDCs at these stages of cardiogenesis. RNA-pull-down together with RNAseq analyses reveal that all miR-200b-targets in epicardial cells are related to cell migration. In vitro experiments of gain and loss of functions, by using EPDCs and whole-organ cultures from mouse embryos, evidenced that miR-200b modifies cell motility. Additionally, analysis in ventricles from infarcted adult mice, strongly demonstrate that miR-200b is upregulated from post-myocardial infarction day 3, peaking at day 7. Collectively, our data suggest that this miRNA might be a key molecule regulating epicardial cell lineage migration during cardiac development, as well as, in EPDC activation after myocardial infarction.

A role for cardiac sympathetic hyperinnervation in arrhythmogenesis after myocardial infarction (MI) has increasingly been recognized. In humans and mice, the heart receives cervical as well as thoracic sympathetic contributions. In mice, superior cervical ganglia (SCG) have been shown to contribute significantly to myocardial sympathetic innervation of the left ventricular anterior wall. Of interest, the SCG is situated adjacent to the carotid body (CB), a small organ involved in oxygen and metabolic sensing. We investigated the remodeling of murine SCG, as well as the proximate CB, over time after MI. Murine SCGs were isolated from control mice, as well as 24 hours, 3 days, 7 days and 6 weeks after MI. SCGs were stained for autonomic nervous system markers (Tubb3, TH and ChAT), as well as for neurotrophic factors (BDNF and NGF) and their receptors. Quantification of the staining-intensity as well as the neuronal size was performed in the entire SCG and in the proximate CB. Our results show that ChAT and TH are co-expressed in SCG neuronal cells in control ganglia. After MI, neuronal remodelling occurs, with a significant increase in size of ganglionic cells and a decreased intensity of ChAT expression. This SCG remodeling was observed as early as 24 hours after infarction, with a peak at day 7, regressing within 6 weeks post-MI to basal levels. Of note, the most robust neuronal remodeling was observed at the region adjacent to the CB. An increase of neurotrophic factors (BDNF and NGF) was observed in the CB and neuronal cells, whereas the high affinity receptors for BDNF and NGF increased in the SCG after MI. These findings were concomitant with an increase in GAP43 expression indicating axonal outgrowth in the SCG. In conclusion, overt neuronal remodeling occurs after MI in the SCG as well as in the CB, suggesting an interaction of these 2 structures that might contribute to pathological cardiac hyperinnervation

During cardiac development the ventricle undergoes trabeculation, where a subset of cardiomyocytes delaminate from the myocardial wall and project into the heart lumen. At the onset of trabeculation the ventricular wall exhibits apicobasal polarity, where the exterior surface is apical and trabecular cardiomyocytes emerge basally. Despite recent advances, the mechanisms by which ventricular wall integrity is maintained as cardiomyocytes delaminate are still poorly understood.

Laminins are extracellular matrix proteins that are typically basally deposited, forming the basement membrane. We identified a dynamic basal-to-apical shift in laminin deposition in the developing ventricle: prior to trabeculation laminin localises to the luminal (basal) ventricular surface, but concomitant with initiation of trabeculation basal laminin is removed and instead deposited at the apical exterior of the ventricle. Analysis of the ventricular wall in laminin beta1a zebrafish mutants reveals an increased number of cardiomyocytes undergoing aberrant apical extrusion, suggesting apical laminin helps maintain myocardial wall integrity.

Llgl1 (Lethal(2) giant larvae protein homolog 1) is a component of the Scribble complex, which is important for maintaining apicobasal polarity including regulating Crumbs, an apical transmembrane protein required for ventricular wall organisation and trabeculation. We therefore hypothesised Llgl1 plays a role in ventricular wall polarisation, organisation, or trabeculation. Analysis of llgl1 zebrafish mutants revealed a delayed basal-apical laminin shift, alongside defects in Crumbs relocalisation. These defects are accompanied by an increase in ventricular cardiomyocytes undergoing apical extrusion. Together our data indicate that llgl1 maintains ventricular wall integrity during trabeculation through promoting timely establishment of an apical laminin sheath around the ventricle.

The mammalian adult heart is a post-mitotic organ characterized mainly by polyploid cardiomyocytes (CMs). The post-mitotic status of adult CMs poses a clear limit to heart regeneration. Recent studies have revealed that the PIDDosome, a multi-protein complex consisting of Pidd1, Raidd, and caspase-2, controls p53 activation in response to supernumerary centrosome and limits scheduled as well as accidental polyploidization events. Since PIDDosome function in heart remains elusive, the overall goal of this project is to unveil the role of the PIDDosome in CM polyploidization, as barrier to tissue regeneration.

Flow cytometry-based analyses of adult CM nuclei showed that the absence of either Pidd1, Raidd, or caspase-2 results in an increase of tetraploid CM nuclei. Moreover, cardiac-specific deletion of caspase-2 resembles the ploidy phenotype observed in Casp2 knockout mice, suggesting that the increased CM ploidy is a cell autonomous process. Comparative analyses between different postnatal stages highlighted a gradual increase in tetraploid CM nuclei in hearts from Pidd1, Raidd or Casp2 knockout mice compared to wild-type mice at postnatal day 7 narrowing down the time of PIDDosome action. Previous studies have demonstrated that Pidd1 must be anchored to mother centriole via Ankrd26 in order to activate the PIDDosome pathway. Preliminary data from Ankrd26 knockout mice revealed that similar to Pidd1, Raidd and Casp2  knockout mice, the absence of Ankrd26 leads to an increase in tetraploid CM nuclei. On the other hand, in p53 knockout mice the ploidy is not statistically increased compared to wild-type animals, suggesting that the PIDDosome regulates CM ploidy in a p53-indipendent way.

The atria serve as a contractile blood reservoir of the cardiac pump. During embryogenesis, the atria and ventricles are soon distinguished from the persisting tubular segments (atrioventricular canal, outflow tract) by disappearance of the cardiac jelly and formation of the pectinate muscles and trabeculae, respectively. Unlike the well-characterized and frequently investigated ventricular myoarchitecture, the arrangement of muscular bundles in the atria is insufficiently studied. Here we present the effects of experimental hemodynamic alterations on embryonic and adult structural and electrical properties of the atria in a variety of animal models. Chick embryonic left atrial ligation (LAL) led to inhibition of pectinate muscle development in the excluded portion of the left auricle and appearance of ectopic pacemaking activity in that region. In rat volume overload model induced by aorto-caval fistula (ACF), the atrial weights were significantly increased (similar to ventricles), with more pronounced response in the right atrium; this increase was significantly attenuated by renal denervation. In another experimental rat model of pulmonary hypertension (SUGEN-hypoxia), an extreme rise in right atrial and ventricular pressures was observed, with accompanying profound structural remodeling of both ventricles as well as the atria. Analysis of these models will provide us with clues regarding the mechanisms governing the normal development of atrial form and function and their perturbations in overload-induced remodeling and arrhythmogenesis.

Muscular dystrophies (MD) are a group of diseases that cause progressive weakness and loss of muscle mass. Mutations in components of the dystrophin-glycoprotein complex (DGC) cause several types of MDs, such as Duchenne. There are no effective cures available for MD and current treatments only modestly alter disease progression. This is mostly attributed to the lack of understanding of the precise causal mechanisms. While both cardiac and skeletal muscles share many functional and molecular similarities, DGC function in both tissues is not identical. Thus, in both animal models and human patients, skeletal muscle dysfunction develops much faster than abnormalities in the heart. To better understand cardiac involvement in MD, we used a mouse model of cardiomyocyte-specific knockout (KO) of Dystroglycan- a main component of the DGC. We evaluated cardiac function and molecular changes in mice during homeostasis at different ages, and during acute injury settings. We show that KO of Dystroglycan in cardiomyocytes leads to a broader loss of other DGC proteins. However, normal cardiac function in mutant mice is not affected until the age of six months, after which it is drastically impaired. Surprisingly, a panel of acute heart injuries in young mutant mice do not reveal exacerbated damage as compared to WT. Yet, when stressing young mutant mice by β-adrenergic receptor stimulation we observe a marked deterioration in heart function. Late induction of Dystroglycan KO in both young and adult mice causes a temporary impairment in cardiac function, which returns to normal within a few weeks. Taken together we show that, as in the clinic, mice lacking the DGC can function quite well (if not stressed) as young adults suggesting the existence of compensatory mechanisms to cope with the loss of DGC in cardiomyocytes. Understanding these compensatory mechanisms might reveal novel targets for protection in both cardiac and skeletal muscle tissues affected in MDs.

Branchiomeric head and neck muscles have a shared origin with second heart field cardiac progenitor cells in cardiopharyngeal mesoderm (CPM). Clonal analysis has identified a series of common skeletal and cardiac muscle lineages along the anterior-posterior axis of the developing pharynx. Progenitor cells giving rise to jaw opening muscles and the right ventricle, or facial expression muscles and the cardiac outflow tract, are found in anterior CPM. Posterior CPM gives rise to neck muscles, including the trapezius, and both arterial and venous pole myocardium. The retinoic acid (RA) signalling pathway is required for normal deployment of cardiac progenitor cells in posterior CPM and blocking RA signaling during a defined early time window leads to conotruncal and atrial septation defects. Here we show using pharmacological approaches that blocking RA signaling also results in a strikingly selective loss of the trapezius muscle, without affecting other branchiomeric or somite-derived trunk and limb muscles. This reveals that, although regulators such as the transcription factor and 22q11.2 deletion syndrome gene TBX1 are broadly required for branchiomeric myogenesis, regulatory subprograms differ in posterior and anterior CPM. RA is required to specify the trapezius anlagen upstream of expression of myogenic determination factors of the MyoD family. Lineage specific activation of a dominant negative RA receptor suggests that, unexpectedly, this effect is not mediated by direct RA signaling to trapezius progenitor cells in CPM, but indirectly through the somitic lineage. These findings suggest a model in which trapezius development is dependent on a signaling cascade between cell types across the head/trunk interface. Our results provide insights into the mechanisms driving myogenic development within CPM and contribute to a better understanding of muscle pathology and evolution.

During development, the heart growths through addition of progenitor cells to the poles of the primordial heart tube.  In the zebrafish, wilms tumor 1 transcription factor a (wt1a) and b (wt1b) are expressed in the pericardium, at the venous pole of the heart. From this pericardial layer, the proepicardium emerges. Proepicardial cells are subsequently transferred to the myocardial surface and form the epicardium, covering the myocardium. We found that while wt1a/b expression is maintained in proepicaridal cells, it is downregulated in those pericardial cells contributing to cardiomyocytes from the developing heart. Sustained wt1 expression impaired cardiomyocyte maturation by reducing chromatin accessibility of specific genomic loci. Strikingly, a subset of wt1a/b-expressing cardiomyocytes changed their cell adhesion properties, delaminated from the myocardium and upregulated epicardial gene expression. Thus, wt1 acts as a break for cardiomyocyte differentiation and ectopic wt1 expression in cardiomyocytes can lead to their transdifferentiation into epicardium.

Haploinsufficiency of TAB2 causes Congenital Heart Disease (CHD). TAB2 mediates non-canonical TGF-β signalling via interactions with TAK1 and TRAF6. Mutations of TAK1 cause Frontometaphyseal Dysplasia (FMD) and cardiospondylcarpofacial syndrome (CSCFS), rare multisystem syndromes, where CHD may appear in the clinical spectrum.

Whole exome sequencing data from a cohort of 1,471 syndromic CHD patients (sCHD), 2,405 patients with isolated CHD (iCHD) and 45,082 controls show increased burden of rare TAB2 and TAK1 variants in sCHD, but not in iCHD. We hypothesized that TAB2 and TAK1 are functionally important in heart development and addressed this experimentally in vitro and in vivo. Using in vitro models and embryonic tissues, we show that TAB2 and TAK1 are expressed in embryonic hearts and localise to primary cilia. Furthermore, mutation of Tak1 and Tab2 inhibits cardiomyogenesis of P19CL6 stem cells. In vivo zebrafish models show that mutation of tab2 and tak1 cause cardiac developmental defects. In addition, deep phenotyping of tak1 mutant zebrafish showed that several phenotypic features of FMD and CSCFS appear in the zebrafish model during early stages of development. This suggests a potential novel in vivo model for investigating FMD and CSCFS. In summary, our data suggests an important role for TAB2 and TAK1 in cardiac development and CHD, coordinated via the primary cilium.

　　　　　A subset of endocardial cells is hemogenic during early embryogenesis. Hemogenic endocardial cells are enriched in the cushion region and undergo endocardial-hematopoietic transition via Nkx2-5-dependent manner, suggesting that Drosophila tinman-dependent cardio-hematopoietic program is conserved in mammals. These hemogenic endocardial cells give rise to a subset of cardiac tissue macrophages that are essential for the cushion remodeling.

　　　　　To examine the regulatory network of Nkx2-5-dependent endocardial hematopoiesis, we analyzed scRNA-seq data from wildtype and Nkx2-5-null embryonic hearts. As expected, Nkx2-5-null hearts were devoid of clusters for hemogenic endocardium (e.g. Runx1⁺, Cd41⁺) and cushion endocardium (e.g. Twist1⁺, Msx1⁺). Interestingly, scRNA-seq analysis further revealed that genes related to Notch signaling pathway are significantly downregulated in Nkx2-5-null endocardium. To examine whether Notch signaling induces Nkx2-5-dependent endocardial hematopoiesis, we performed genetics experiments. Nkx2-5-null endocardium showed significant downregulation of Notch1⁺ endocardial cells, and NICD overexpression drastically activated hematopoiesis in the endocardium. Interestingly, impaired hematopoiesis and cushion defect in the Nkx2-5-null heart were both rescued by overexpression of NICD. Further gene regulatory network analysis identified that Dhrs3, which is known to catalyze the reduction of all-trans-retinaldehyde to all-trans-retinol and attenuates retinoic acid signaling, was a signature gene of the hemogenic endocardial cells downstream of Nkx2-5. Forced activation of NICD increased the number of not only Dhrs3⁺ hemogenic endocardial cells but also Dhrs3⁺ macrophages, suggesting a possible involvement of Dhrs3⁺ macrophages in cushion remodeling.

　　　　　This study demonstrated that the Nkx2-5/Notch signaling axis plays a pivotal role in endocardial-hematopoietic transition during early embryogenesis, thereby facilitating local tissue remodeling.

Most cases of congenital heart disease (CHD) occur sporadically and are, in part, attributable to rare de novo variants in chromatin regulatory genes, shown to account for ~3% of sporadic CHD by the Pediatric Cardiac Genomics Consortium (PCGC). This increases to 20% in cases of CHD that co-occur with neurodevelopmental disabilities (NDDs) and/or extra-cardiac anomalies. We analyzed the exome from whole genome sequence in 456 cases with conotruncal heart defects (CTDs) versus 537 controls with no heart anomalies, all with 22q11.2 deletion syndrome (22q11.2DS), to identify genetic modifiers. 22q11.2DS is ideal for this study as it has variable phenotypic expressivity of CHD and NDDs, along with other congenital abnormalities. Among the most interesting genes found was a subset of chromatin regulators that account for 9.21% of the CTD patients. We cross-referenced our list of 40 chromatin regulators with 90 identified in sporadic CHD by the PCGC, finding significant overlap (n=12; p<3.16e^-4) between the two groups. This is particularly compelling when considering that the identification of these variants was performed using different patient cohorts and statistical methods. Further, the chromatin regulatory genes in our cohort are overwhelmingly associated with genetic syndromes with increased incidence of CHD, NDDs, and extra-cardiac anomalies, making it of interest to determine whether they may also serve as modifiers for other syndromic anomalies. Thus, the data from our 22q11.2DS cohort reinforces the importance of chromatin regulatory genes as risk factors for sporadic CHD with syndromic features and suggests that modifiers of 22q11.2DS have inter-syndromic significance.

The cardiovascular system, comprised of the heart and the blood vessels, distributes oxygen and nutrients to the body. Accordingly, defective cardiovascular development can cause congenital heart disease (CHD). We have previously elucidated the mechanisms by which the transmembrane protein NRP1 promotes the remodelling of the foetal cardiac outflow tract (OFT) into the base of the pulmonary artery and aorta. NRP1 also enables the remodelling of the primitive pharyngeal arch arteries (PAA) into the great vessels that distribute blood from the heart into the lungs and body. We are seeking to (a) define the temporal window in which NRP1 enables PAA morphogenesis, (b) elucidate which specific cell types require NRP1 for PAA morphogenesis and (c) identify the NRP1 ligand(s) that mediate PAA morphogenesis. Here we show that NRP1 acts in both endothelial cells from the second heart field to promote PAA formation and in the cardiac neural crest lineage to promote vascular smooth muscle cell differentiation on the PAA, thereby promoting two consecutive stages of PAA morphogenesis. We have also begun to compare the requirement of different NRP1 ligands for these consecutive stages of great vessel formation. Defining the molecular and cellular mechanisms by which NRP1 enables PAA morphogenesis will increase our understanding of the normal and abnormal development of the cardiovasculature. In the long run, this knowledge may help improve the early diagnosis and treatment of CHD that involve the aortic arch arteries.

Wt1 encodes a zinc finger protein whose best-known function is its role as a transcription factor. Although for many years WT1 has been considered as one of the main hallmarks of the embryonic epicardium, new pieces of evidence have demonstrated the expression of WT1 in coronary endothelial cells during the vascularization of the embryonic heart and following myocardial infarction. Here we have used Wt1 reporter and lineage tracing mouse models to fully characterize the expression of WT1 in the endothelium through all developmental stages of coronary formation. To investigate its role during coronary vessel formation we have generated an inducible endothelial-specific Wt1 KO mouse model (Wt1-KO^ΔEC). The analysis of these mutant mice has demonstrated that WT1 expression in coronary endothelial cells during coronary plexus formation is required for the correct development of the coronaries and the compact myocardium. The transcriptomic analysis of coronary endothelial cells from Wt1-KO^ΔEC mice demonstrates that WT1 works as a master regulator of crucial processes of endothelial cells. We next used the postnatal retina, to investigate whether Wt1 functions in endothelial cells are conserved among other vascular beds. We observed profound alterations in the retinal endothelial network of postnatal Wt1-KO^ΔEC mice, revealing a reduced endothelial migratory phenotype and severely reduced vascular complexity. Our results shed new light on the functions of WT1 in the endothelium and suggest interesting roles for WT1 during physiological blood vessel formation. We speculate that pharmacologic interventions of WT1 or WT1 downstream targets may be useful for stimulating coronary vascular growth.

Heart development begins with the formation of a tube, as contralateral cardiac progenitors migrate and merge medially. Defective cell movement results in improper heart tube formation and congenital heart defects. However, the mechanisms of cell migration in early cardiogenesis remain unclear. In Drosophila, the embryonic heart is a linear structure composed of 52 pairs of bilateral cardiac precursors (cardioblasts, CB), that migrate dorsally and medially to form the heart tube. Using quantitative time-lapse microscopy, we found that CBs took periodic forward and backward steps as they migrated. The forward steps were greater in both amplitude and duration, resulting in net forward movement. Force generation by the molecular motor non-muscle myosin II is critical for cell movements. We found that myosin displayed an alternating pattern of localization between the leading and trailing ends of migrating CBs, forming oscillatory waves that traversed the cells. Live imaging revealed that the alternating myosin polarity was associated with alternating contractions of the leading and trailing edges. Mathematical modeling predicted that forward migration requires the presence of a boundary at the trailing edge of CBs that restricts backward movement. Consistent with this, we found a supracellular actin cable at the trailing edge of the CBs. Releasing the tension sustained by the supracellular actin cable increased the amplitude of the backward steps of CBs, thus reducing the speed of migration. Our results indicate that periodic cell shape changes coupled with a supracellular actin cable result in asymmetrical forces that guide cardioblast migration.

Congenital heart disease (CHD) remains the largest cause of birth defects worldwide, affecting about 1% of live births. Human genetics studies have identified genes potentially implicated in CHD, but the function of many of these genes in heart development and the specific cell types affected by their perturbation remain unknown. Furthermore, recent bulk and single-cell omics approaches have extensively characterized the molecular landscape of heart development, also identifying candidates expressed at various developmental stages. However, the functional relevance of many of these candidates remains largely unknown. The enormous search space for functionally important players in heart development and disease collectively created by these studies is challenging to approach with individual functional studies such as mouse knockout models. Here, we take advantage of the well-established system of hPSC differentiation to cardiac cell types, coupled with the recently established CRISPR-based perturbation method, CRISPR activation/inhibition, to perform a screen of 500 candidates potentially involved in the specification and differentiation of cardiac cell types. We will first establish our screen in atrial and ventricular cardiomyocytes, with the goal of expanding our approach to other cardiac cell types in the future. Taking advantage of the faithful recapitulation of key developmental stages in the hPSC system, we expect to uncover genes essential for the formation of differentiated cell types as well as their intermediary progenitor stages. Thus, by moving from expression to function at large scale, we hope to contribute to an enhanced understanding of cell-type specific molecular mechanisms of heart development and disease.

The Second Heart Field (SHF) is a population of cardiac progenitor cells that contributes to the developing heart after the initial formation of the primary heart tube. At the arterial pole, the SHF contributes to the outflow tract (OFT), the right ventricle (RV), and the ventricular septum (VS). At the venous pole, the SHF contributes to some atrial myocardial structures and to the dorsal mesenchymal protrusion (DMP). The importance of the SHF for normal heart development is demonstrated in studies in which perturbation of gene expression in the SHF has shown to result congenital heart malformations, including atrioventricular septal defects (AVSDs).

The AV mesenchymal complex (AVMC) consists of four separate entities; the two major endocardially-derived AV cushions, the mesenchymal cap situated on the leading edge of the developing primary atrial septum, and the DMP. Proper development of the AVMC is essential for the development of the septal structures at the AV junction

It is well-known that the transcription factor Sox9 plays an important role in the development of the AV cushions and valve formation. Preliminary studies and analysis of data in published papers by others, led us to believe that Sox9 could play a role in SHF-dependent development of the AV valvuloseptal complex. To test this hypothesis, we generated a SHF-specific Sox9 knockout mouse and studied how depleting Sox9 from this cell population affected the formation of the components of the AVMC and the impact of this experimental approach on AV septation.  The results of the study show that SHF-derived Sox9 plays a critical role in atrial as well as ventricular septation.