Wt1+/Itga4+ progenitors drive epicardial-derived cell expansion via non-canonical Wnt signalling

doi:10.21203/rs.3.rs-4842079/v1

Download PDF

Article

Wt1⁺/Itga4⁺ progenitors drive epicardial-derived cell expansion via non-canonical Wnt signalling

https://doi.org/10.21203/rs.3.rs-4842079/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The correct formation of the epicardium – the outermost tissue layer of the heart – is crucial to proper cardiac embryonic morphogenesis. However, our knowledge of the molecular and cellular mechanisms regulating epicardial progenitor cell conversion into primitive epicardial cells is limited. By combining in vivo mouse models for cell lineage tracing, high throughput transcriptomic technologies, and in vitro assays, we have characterised the proepicardium (PE), the transient extracardiac structure that contains epicardial progenitors. Our study reveals that only Wt1⁺/Itga4⁺ PE cells truly represent the progenitors of primitive epicardial cells and epicardium-derived cells (EPDCs). We show that early EPDCs display a high proliferative activity that is regulated by non-canonical Wnt signals (Wnt5a/c-Jun axis). This study redefines the concept of PE/epicardial progenitors, uncovers relevant aspects of the mechanisms controlling early EPDC formation, and lays the foundation for future research on EPDC differentiation in clinically relevant cardiac interstitial cells.

Biological sciences/Developmental biology/Organogenesis

Biological sciences/Developmental biology/Embryology

Proepicardium

epicardium

proliferation

cell lineage tracing

single-cell RNA-seq

Wnt signalling pathway

In the last two decades of the 20th century, various experimental studies demonstrated that the epicardium forms from extracardiac coelomic cells located at the venous pole of the heart^1–3, and does not form from the so-called myoepicardium, a theoretical embryonic progenitor of the myocardium and the epicardium forming the wall of the primitive cardiac tubular primordium^4,5. In mammals, epicardial progenitors are grouped at the proepicardium (PE)⁶, a structure located at the pericardial surface of the septum transversum (ST), which is a folding of the splanchnic lateral mesoderm that contributes to the separation of the thoracic and abdominal coelomic cavities.

Since the primary fate of the PE is to give rise to the embryonic primitive epicardium, i.e., the epithelium that covers the heart anlage⁷, this small cell cluster is widely regarded as the first stage of a developmental continuum⁸. Among the characteristically expressed PE/primitive epicardial genes, Wilms tumour suppressor 1 (Wt1)⁹, T-box Transcription factor 18 (Tbx18)¹⁰ and Transcription Factor 21 (Tcf21)¹¹ are known to be important regulators of (pro)epicardial biology.

Proper embryonic epicardial development is indispensable to cardiac morphogenesis. Indeed, primitive epicardial cells synthesise molecules that accumulate to form the subepicardium, an originally acellular matrix sandwiched between the epicardium and the myocardium that later supports early coronary vascularisation. Primitive epicardial cells are then able to undergo a full phenotypical transformation, giving rise to a novel population of mesenchymal cells known as epicardium-derived cells (EPDCs¹²). EPDCs are proliferative, migratory mesenchymal cells that progressively invade the myocardial layers from the subepicardium. Finally, EPDCs display multilineage differentiation, and materially contribute to the formation of coronary vessels^2,8,13,14, atrioventricular valves^12,15,16, and the cardiac interstitium^17–19. Since embryonic cardiac interstitial EPDCs are known to persist in adulthood and become a major cellular effector of heart responses to pathologic stimuli^17,20, the relevance of EPDCs is not restricted to embryonic life.

Primitive epicardial cell shift towards the EPDC phenotype is mediated by an epithelial-to-mesenchymal transition (EMT), which is a tightly regulated cellular mechanism^21,22. Some controversy remains on the molecules that actually trigger epicardial EMT²³, as it is still unclear why not all primitive epicardial cells are rendered competent to undergo EMT. The non-transforming primitive epicardial cells mature into a squamous epithelium that protects the outer myocardial surface, and once primitive epicardial EMT stops, the primitive epicardium transforms into a secondary, more quiescent epicardium (Wu et al., 2010).

While many studies have addressed the cell heterogeneity and molecular mechanisms that regulate the biology of embryonic EPDCs^25–28, only a few have focused on these mechanisms in the PE^13,29. These first approaches suggested that the PE is an intrinsically heterogeneous structure, and that such heterogeneity has functional implications for its EPDC progeny. Many of these studies have limitations due to the difficulties of unambiguously distinguishing the PE territory from surrounding tissues such as the septum transversum^8,30.

Thus, the cell heterogeneity of the PE remains unclear and, furthermore, the molecular mechanisms that govern the first steps in PE conversion into EPDCs are poorly characterised. Here, we perform a spatial and functional analysis of the posterior region of the cardiac inflow, allowing us to re-define the concept of PE/epicardial progenitor cells, providing novel information on the cellular and molecular mechanisms underlying the biology of the epicardium.

Deciphering cellular heterogeneity in the cardiac inflow

We decided to explore PE cell diversity in an unbiased manner. To do so, we carried out single-cell RNA sequencing (scRNA-seq, 10x Genomics) of E9.5 mouse embryonic posterior inflow tissues (n = 25) (Fig. 1a-a’). These tissues were manually micro-dissected and disaggregated into single cell suspension using an enzymatic procedure. To define the different cell clusters within our dataset, we first performed an unsupervised analysis of the 13,479 cells that passed the quality controls (Suppl. Fig. S1). Then, we used the annotations previously described by other studies dealing with similar embryonic samples^31–33. Our analysis identified a total of 13 different clusters, including cardiomyocytes, liver cells, endothelial cells, and some other cardiac related cells, such as second heart field (SHF) and juxta cardiac field (JCF) progenitors (Suppl. Fig. S2; Suppl. dataset 1). Of these clusters, one (PE cluster) was enriched in the classical gene markers associated with the proepicardium, such as Wt1⁹, Tbx18¹⁰, or Tcf21¹¹. Moreover, two additional clusters were found to be enriched in ST mesenchyme marker genes (STMI, II clusters)³², although only one of them (STMI) also displayed a transcriptional signature similar to that of the PE cluster.

scRNA-seq transcriptomics segregates the PE from the ST

Considering the similarities between the STMI and the PE clusters, we decided to re-cluster them into a single cell pool, named “proepicardium region” (PER), allowing for a deeper characterisation of cell diversity in this domain of the cardiac inflow. This re-analysis of PER cells (4,452 cells) defined 7 clusters (PER1-7) (Fig. 1b; Suppl. dataset 2) that showed a characteristic mixed STMI + PE transcriptional profile, including genes such as Twist1, Lhx2, Wt1, Tcf21, and Tbx18 (Fig. 1c). Among the PER clusters, PER2 showed a unique gene expression signature including genes previously shown to be critical for PE development, such as Itga4^34,35, Scx²⁹, and Tbx18¹⁰ (Suppl. Fig. S3). Many of these genes match GO terms related to extracellular matrix organisation and cell migration (Suppl. Fig. S4). In particular Itga4 gene expression was very strong and restricted to this cluster (Fig. 1c).

Only Wt1 ⁺ / Itga4 ⁺ PE cells are true epicardial progenitors

Based on our previous findings, we reasoned that only PER2 includes true PE cells. To validate this hypothesis, we performed immunofluorescence and multiplexed RNA in situ hybridisation on E9.5 embryos to identify the different marker genes that defined PER clusters. WT1 protein helped us identify the anatomical location of the PE on the ventral part of the ST (Fig. 1d-d’). Analysis of Wt1 mRNA revealed a clear expression gradient for this gene in the PE, with the highest amount of Wt1 transcripts found in the apical (ventral) region of this structure, closer to the myocardium (arrow in Fig. 1E). Wt1⁺ cells were also enriched in the expression of other proepicardial markers identified by our scRNA-seq analysis, such as Tcf21 and SOX9 (arrows in Suppl. Fig. S5 a-b’’)^11,36. Most importantly, apical PE Wt1⁺ cells also co-expressed Itga4, whose expression was restricted to this precise anatomical location (Fig. 1e’-f’’). Remarkably, these are the cells that migrate to the myocardial surface to form the first primitive epicardial cells (arrows in Fig. 1f-f’’). On the contrary, other genes associated with different PER clusters (e.g., PER1, 3, 6, 7), such as Twist1 and Lhx2, are highly enriched in the basal (dorsal) pole of the PE and mostly absent from the apical one (Fig. 1g-h’’). Postn (PER4) and Tbx5 (PER7) display a similar pattern to that described for Twist1/Lhx2, their expression being basal (dorsal) with respect to the Wt1 domain (Suppl. Fig. S5 c-d’’). Tcf21 and SOX9 are also expressed in these cells (arrows in Suppl. Fig. S5 a-b’’).

Next, to confirm that Wt1⁺/Itga4⁺ PE cells are a robust developmental cellular entity, we characterised PE development in cardiac gastruloids (Suppl.Fig. S6 and Table S1). Soon after cardiac mesoderm differentiation was promoted, we found Wt1/Itga4-expressing cells close to the Tnnt2⁺ domains within cardiac gastruloids.

Taken together, our results demonstrate that Wt1⁺/Itga4⁺ cells on the ventral part of the ST constitute a steady cellular entity during epicardial embryonic development, and that these cells should be considered the only real PE cells.

Wt1 ⁺ /Itga4 ⁺ cell identity is conserved in primitive epicardial cell-to-EPDC transition.

Between E9.5 and E10.5, the primitive epicardium spreads and covers the myocardial surface (Fig. 2a-a’). By E12.5, the primitive epicardium has fully covered the myocardial surface and some EPDCs have started to invade the subepicardium tissue (Fig. 2b-b’’). A subset of these cells further migrates into the underlying myocardial. We found that all primitive epicardial cells at E10.5 co-expressed the PE signature Wt1⁺/Itga4⁺ (Fig. 2c-c’), although some of them started to downregulate the expression of Itga4 (arrows in Fig. 2c-c’). However, a gradient in Wt1 expression was observed from the primitive epicardium (highest Wt1 expression) to the subepicardial and myocardial EPDCs (lowest Wt1 expression in both cardiac compartments) at E12.5 (Fig. 2d-d’; Suppl. Fig. S7 a). Surprisingly, only a restricted set of E12.5 epicardial cells co-expressed Wt1⁺/Itga4⁺ (arrowheads in Fig. 2D), while most subepicardial EPDCs strongly expressed both genes (Fig. 2d’-d’’). All intramyocardial E12.5 EPDCs express similar Wt1 levels and, among them, Itga4 expression was somewhat rather variable (arrows in Fig. 2d and Suppl. Fig. S7 a-a’’). Our results confirm that PE cells forming the primitive epicardium retain a Wt1⁺/Itga4⁺ transcriptional profile that is also conserved in epicardial cell transformation into subepicardial EPDCs.

To corroborate this finding, we performed a restricted (pseudo-clonal) PE cell tracking analysis using the inducible Wt1-Cre^ERT2 mouse line³⁷. To specifically label PE cells, we administered a single dose of 4OHT to E8.75 pregnant dams, just before PE formation (Fig. 2e; Suppl. Fig. S8 a-c). On average, 20 YFP⁺ cells were detected in the PE structure in each E9.5 embryo (Fig. 2f; Suppl. Fig. S8 d). Most of these cells (63.3 ± 5.7%) displayed a Wt1⁺/Itga4⁺ expression profile and were located on the PE surface (Fig. 2g-h). At E12.5, YFP⁺ PE-derived cells (200–900 per heart) (Suppl. Fig. S8 d) form discrete primitive epicardial cell patches over the entire heart (Fig. 2i). We inspected these primitive epicardial cells and observed that some YFP⁺ cells had detached from the epicardial layer and occupied the underlying subepicardium/myocardium, confirming that they are bona fide EPDCs (Fig. 2i’-i’’).

Then, we analysed the expression of some gene markers in the EPDC pseudo-clones. According to our spatiotemporal analysis, only a subset of E12.5 Yfp⁺ epicardial cells (37.6 ± 5.0%) expressed Itga4 (Fig. 2j-j’’’, l), while most Yfp⁺ subepicardial EPDCs (85.9 ± 2.2%) expressed Itga4⁺ (Fig. 2k-k’’’, l). Altogether, these results suggest a direct lineage linkage between Wt1⁺/Itga4⁺ PE/primitive epicardial cells and early Wt1⁺/Itga4⁺ EPDCs.

Transcriptional dynamics in epicardial cell lineage

To further confirm the relationship between Wt1⁺/Itga4⁺ primitive epicardial cells and EPDCs, we re-analysed a published E12.5 Wt1-cell lineage scRNA-seq dataset²⁵. This dataset encloses 3,405 cells sequenced from E12.5 Wt1-Cre^ERT2;TdTomato embryos (recombination was induced at E9.5 and E10.5 by 4OHT maternal administration). Our analysis delineated 5 transcriptional epicardial-lineage cell clusters (EP1-5) (Suppl. Fig. S9 a; suppl. dataset 3), all of which were enriched in Wt1 expression (Suppl. Fig. S9 b). Interestingly, EP1, 2, 3 and 5 also showed the expression of epithelial gene markers such as Krt18, Krt19, and Pdpn, whereas EP4 was enriched in Itga4, and the mesenchymal/EMT associated genes Snai2 and Twist1 (Suppl. Fig. S9 b). In particular, EP4 gene profile matches GO terms related to EMT (Suppl. Fig. S10). Therefore, we reasoned that clusters EP1, 2, 3 and 5 correspond to epicardial epithelial cells, whereas cluster EP4 identifies Wt1⁺/Itga4⁺ early mesenchymal EPDCs.

Finally, to test whether the Wt1⁺/Itga4⁺ signature is conserved in the PE to epicardium and EPDC transition, we decided to integrate our E9.5 PE dataset (PER2 from our analysis) and the published E12.5 scRNA-seq dataset ²⁵ (Suppl. Fig. S11 a). This data integration shows a high degree of transcriptional similarity between the PE cluster and EP1, 2, 3 and 5 clusters from Quijada et al.; the cells in these latter clusters most likely represent primitive epicardial cells (Suppl. Fig. S11 b-c). Interestingly, early EPDCs (cluster 5 in the integration dataset) undergo a significant change in their transcriptomic profile relative to their parental cells, despite maintaining expression of Wt1/Itga4 (Suppl. Fig. S11 d).

Wt1⁺/Itga4⁺ epicardial cells and early EPDCs are highly proliferative

Both epicardial cells and EPDCs form from a small, limited pool of PE progenitors. We therefore decided to investigate cell proliferation dynamics along the PE-primitive epicardium-EPDC lineage (E9.5-E18.5). Our results indicate that WT1⁺ PE/primitive epicardial cells are proliferative until E12.5 (Fig. 3a, b; Suppl. Fig. S12). Thereafter, significant proliferation is mostly restricted to subepicardial and intramyocardial WT1⁺ cells until E18.5, when proliferation is significantly reduced in all the studied compartments. Interestingly, most proliferating primitive epicardial and subepicardial/intramyocardial cells co-express Wt1 and Itga4 at E12.5 (Fig. 3a’, c), suggesting that the Wt1⁺/Itga4⁺ EPDC profile is responsible for the growth of this mesenchymal population.

The observed differences in cell proliferation identified between epicardial and subepicardial tissues at E12.5 led us to study the connection between cell proliferation and EPDC formation. To do so, we first cultured an EPDC cell line (EPIC³⁸) and set up a cell cycle arrest (G2/M) assay using nocodazole, a disrupter of microtubule assembly (Suppl. Fig. S13 A-D). Then, we performed ex vivo cultures of E11.5 hearts whose cell cycle was arrested using an optimized nocodazole dose (Fig. 3d-f; Suppl. Fig. S13 e-g). No differences were found in the number of WT1⁺ cells located in either the subepicardium or the ventricular wall between conditions (Fig. 3g; Suppl. Fig. S13 h-j), revealing that WT1 primitive epicardial cells can populate the subepicardium (transition to EPDC) in the absence of cell proliferation.

Finally, to trace the origin of these Wt1⁺/Itga4⁺ early EPDCs, we reproduced the same ex vivo E11.5 heart experiments using samples isolated from Wt1-Cre^ERT2;YFP embryos and treated them for 24 h with nocodazole or vehicle. Our results show no differences in the number of YFP⁺ cells between the experimental and control groups (Fig. 3h-j), confirming that cell proliferation is not necessary for the appearance of EPDCs in the subepicardial space. As expected, some of these EPDCs express Itga4^High (Fig. 3j-k’).

Our results reveal that Wt1⁺/Itga4⁺ PE cells proliferate to form the epithelial layer (primitive epicardium) (Suppl. Figure 13a). Following this first epicardial lineage cell expansion, the primitive epicardium transits to a quiescent state around E12.5. This coincides with the initiation of the Wt1⁺/Itga4⁺ primitive epicardial cell transition towards the subepicardial EPDC phenotype and subsequent massive cell proliferation.

Non-canonical Wnt signals regulate the expansion of primitive epicardial cells and early EPDCs

To investigate the molecular regulation of Wt1⁺/Itga4⁺ cell proliferation, we performed a comparative bulk RNA-seq between the PE and the adjacent ventricle at E9.5, assuming that these mechanisms are active in PE cell progenitors (Fig. S14 a, b; suppl. dataset 5). Our approach highlighted different signalling pathways that were upregulated in the PE as compared to the myocardium (Suppl. Fig. S14 c). Among them, we found that Wnt signalling elements were asymmetrically expressed by both tissues. As the Wnt signalling pathway plays an important role in promoting cell proliferation during cardiovascular development³⁹, we aimed to identify Wnt related genes (ligands, receptors and modulators) in our RNA dataset. We found that many were upregulated in the PE (Suppl. Fig. S14 d). Among these genes, we focused on Jun, a well-known effector of cell proliferation^40,41 (Suppl. Fig. S14 e) whose nuclear transcriptional activity is typically triggered by the WNT5A ligand, a prototypical component of the non-canonical Wnt pathway^42,43. We indeed found that E9.5 PE cells and E10.5 primitive epicardial cells show widespread nuclear accumulation of C-JUN (Fig. 4a, a’; Suppl. Fig. S15 a, b) and expression of both Wnt5a (Fig. 4b, b’; Suppl. Fig. S15 c, c’), and its Ror2 receptor (Fig. 4b’’; Suppl. Fig. S15 d-e). At E12.5, C-JUN expression was mainly found in the subepicardium (Fig. 4c; quantified in d); Wnt5a remained widely expressed in the epicardial and myocardial cell compartments (Fig. 4e), and many subepicardial Itga4⁺ EPDCs were found to be Ror2⁺ (Fig. 4f; quantified in g).

These data suggest that the Wnt5a/Ror2/c-Jun molecular axis could be a major regulator of the expansion of epicardial cell population. To test this hypothesis, we evaluated cell proliferation in both primitive epicardial cells and EPDCs in vivo. Most C-JUN⁺ cells in the epicardium and subepicardium were proliferating (Fig. 4h, i). Remarkably, our analysis pinpoints that most proliferating Itga4⁺ primitive epicardial and subepicardial cells also express Ror2 (Fig. 4j-k). Accordingly, WNT5A promoted cell proliferation in primary epicardial explants in vitro (Fig. 4l) as well as nuclear accumulation of C-JUN in the EPIC cell line (Fig. 4m). Altogether, these results confirm an important role for the Wnt5a/Ror2/c-Jun molecular axis in the regulation of epicardial-lineage cell deployment.

The cardiac inflow tract is a highly complex region of the developing embryo. It encompasses part of the splanchnic lateral mesoderm and contains many cell types, including ST and PE cells that will support the formation of the diaphragm and epicardial progenitors, respectively. The ST and the PE have long been held to be a histological continuum, and difficulties in considering them as separate morphological entities have been reported⁷. From a molecular standpoint, distinguishing PE from ST is by no means easier. Different PE-expressed marker genes, like Wt1, Tbx18, Tcf21, Scx or Sema3d, are also transcribed in ST cells ^8,30. Interestingly, genetic cell lineage tracing based on some of these cardiac inflow-expressed genes has been used to describe the PE as a structure composed of different transcriptional compartments²⁹.

In this study, we aimed to characterise bona fide PE cells. To do so, we first took an unbiased approach using scRNA-seq technologies. Our scRNA-seq results were higher resolution (in terms of the number of cells analysed and depth of sequencing) than other similar studies^31,44,45 and enabled us to unambiguously segregate ST mesenchyme from putative PE cells, which are characterised by the robust expression of genes such as Wt1, Tcf21 and Itga4. Among these genes, Wt1 and especially Itga4 were identified as landmark genes in the PE cell population transcriptional signature. This corroborates in vivo studies that revealed that both genes are indispensable in early epicardial progenitor (PE) cell biology^34,35,46,47.

Our work has demonstrated that robust Wt1/Itga4 expression is restricted to the ventral, outer surface of the coelomic cell outgrowth, classically referred-to as the PE^48,49. In contrast, Wt1⁺/Twist1⁺ and Wt1⁺/Lhx2⁺ cells were found dorsally to the Wt1⁺/Itga4⁺ domain (i.e., at the base of the classical PE), in continuity with the ST mesenchyme. Taken together, all these findings strongly suggest that the PE concept needs to be reconsidered, as the term PE was specifically coined to name the tissue that delivers primitive epicardium progenitors to the surface of the heart alone⁶. We now conclude that only the cortical surface (ventral part) of this classical PE can be considered as containing bona fide PE cells.

Results from our study also show that Wt1⁺/Itga4⁺ expression in the E9.5 PE translocates to the E12.5 nascent subepicardial EPDC population. A direct lineage relationship between these cells is confirmed by genetic (Wt1) cell tracing experiments, which also allow the identification of primitive epicardial cell delamination towards the subepicardial space. All these data indicate that during epicardial EMT, the earliest EPDCs continue expressing Itga4⁺ as much as their parent cells do.

The significance of EPDC proliferation, a cell expansion event preceding differentiation into multiple cardiovascular cell types, has been widely overlooked. We herein confirm that early EPDC proliferation is high, especially when compared to primitive epicardial epithelium proliferation, and that this high EPDC proliferation rate is maintained in the first EPDCs as they infiltrate the myocardium. In addition, our results show that nocodazole-induced epicardial proliferation arrest does not block subepicardial EPDC appearance, suggesting that epicardial delamination remains active in the absence of cell proliferation. Taking into account that we found massive subepicardial proliferation of EPDC, it is tempting to hypothesise that cell proliferation machinery is put on hold during primitive epicardial cell transformation, to be released only when the phenotypical switch is completed. These results contrast the suggested role of cell proliferation in the conversion of epicardial epithelial cells into EPDCs via EMT²⁴.

To progress in our identification of the molecular mechanisms that control EPDC proliferation, we carried out a comparative RNA-seq study using classical PE (coelomic surface and part of the ST) and ventricular samples. This strategy identified the Wnt signalling pathway as displaying the most evident gene expression changes between the PE and ventricular tissues. We thus inspected the in situ expression of three key Wnt signalling machinery genes, the Wnt5a ligand, its Ror2 receptor, and the non-canonical Wnt signalling-mediated cell proliferation effector Jun^42,43,50.

Our results indicate that the Wnt5a/Ror2/Jun axis is key in the regulation of Wt1⁺/Itga4⁺ cell proliferation (E9.5-12.5). The spatial expression pattern of Wnt-related molecules points to the myocardially secreted Wnt5a as the triggering ligand activating epicardial and epicardial-derived cell proliferation. In this model, Ror2⁺ cells would initiate a signalling cascade including the effector C-JUN, eventually resulting in cell mitosis initiation⁴³. This hypothetical scenario is further supported by our in vitro assays, showing that cultured epicardium-derived mesenchymal cells respond to Wnt5a by increasing the expression of C-JUN. In summary, this set of results points to Wnt5a as a key signal in epicardium and EPDC development, a finding supported by previous Wnt5a loss-of-function studies that suggested this molecule was involved in ventricular cardiac wall growth promotion^51,52.

Collectively, our work redefines the PE as a developmental entity formed by Wt1⁺/Itga4⁺ cells. Moreover, we have found that these true PE progenitors first convert into primitive epicardial cells and then transform into mesenchymal EPDCs while retaining the Wt1⁺/Itga4⁺ transcriptional signature. We also show that these cells display a high proliferative potential, first in the epicardium and then in the subepicardium, which is likely to be regulated by autocrine and paracrine mechanisms involving non-canonical Wnt signalling. This finely tuned regulation of epicardial lineage cell proliferation could be crucial for later differentiation into different cardiovascular cell types. Moreover, the tightly controlled proliferation of this progenitor pool during development might be advantageous for maintaining balanced differentiation in the changing embryonic milieu. In this regard, some adult tissues renew through the activation of transit amplification progenitors (TAP) derived from the limited proliferation of resident stem cells⁵³. Perhaps, early EPDCs represent an embryonic-TAP like cell type^53,54. In line with this hypothesis, other relevant questions arise: does an early EPDC-like reservoir exist in the heart postnatally? If this is the case, then these cells could on the one hand play a significant role in heart response to damage^55–57 but on the other hand, also participate in detrimental physiopathological changes in cardiac function following injury¹⁷. Therefore, our work on the developmental mechanisms of EPDC formation pinpoints the subepicardial niche as a promising target to modulate cardiac responses to pathological stimuli and opens exciting avenues for future cardiac therapeutic strategies focused on the modulation of cardiac connective cells in the context of advanced regenerative therapies in heart disease.

Animal models

All experiments were conducted in accordance with the ethical regulations for animal research of the University of Malaga (UMA), the French Ministry of Higher Education and Research and of the EU and were approved by the Ethics Committee of UMA. Most experiments were conducted on mice with a CD1 background, including the single-cell analysis of the PE, spatial transcriptomics, RNA in situ hybridisation experiments and in vitro cell/tissue cultures. The Wt1-Cre^ERT2 mouse strain expresses the Cre^ERT2 fusion protein in the presence of tamoxifen or its derivatives under the control of the Wt1 gene promoter³⁷. The RosaR26R-EYFP line⁵⁸ was used to tag Wt1-lineage cells. All embryos were staged according to vaginal plug observation time point (E0.5 stage). Specific mouse strains were genotyped via PCR as documented in the original reports.

To induce recombination, pregnant Wt1-Cre^ERT2;R26R-EYFP females received a single intraperitoneal administration of 4-hydroxytamoxifen (4OHT) (Sigma, H6278) dissolved in Cremophor® (Sigma, C5135)⁵⁹ at E8.75 stage. Doses of 1 mg of 4OHT per mouse were used (300 µL total administration volume).

Ex vivo E11.5 heart culture

E11.5 embryos were collected from either CD1 or Wt1-Cre^ERT2;R26R-EYFP mouse lines, washed in sterile PBS, and their pericardium removed. Hearts were then excised and placed individually in 96-round bottom well plates (Deltalab). Hearts were incubated at 37ºC and 5% CO₂ in a medium comprising DMEM (Gibco, 11965) supplemented with 10% FBS (Pansera, P30-2602), 1% L-glutamine (Sigma, G7513), 1% penicillin/streptomycin (Gibco, 15140130), and 2 ng/mL bFGF (R&D Systems, 233-FB) as previously described ⁶⁰. To arrest the cell cycle, the experimental growth medium was supplemented with 500 ng/mL nocodazole (Sigma, M1404). Hearts were cultured for a total of 24h.

In vitro culture of epicardial explants and EPICs

To grow primary epicardial cells, hearts from E11.5 embryos were excised and cultured in 4-well plates (Nunc) on glass coverslips coated with 0.1% gelatine (Panreac). Cells were incubated at 37ºC and 5% CO₂ in the standard growth medium described above. Initially, a minimal volume of growth medium was used to prevent the displacement of the beating heart. After one day of culture, some epicardial cells had extended beyond the heart-gelatine interface. Subsequently, the heart was removed, leaving epicardial cells only. Then, the explants were starved overnight in a culture medium containing 1% FBS (starving medium). Finally, epicardial explants were incubated in the starving medium conditioned with 100 ng/mL WNT5A (Sigma, A2547).

EPICs are an in vitro model for epicardial-derived mesenchymal cells. EPICs were originally derived from E11.5 embryonic epicardium³⁸. These cells were cultured under the same conditions previously described, including incubation in WNT5A conditioned medium. Cells were passaged using 0.05% trypsin-EDTA (Gibco, 25300-054) when they reached 70% confluence.

Proepicardial differentiation in cardiac gastruloids

R1 (R1 (R1/E #1036) (ECACC - cat. Number 07072001) mouse embryonic stem cells (mESC) were cultured at 37ºC and 5% CO₂ in GMEM-BHK21 medium (Gibco, 21710-025) supplemented with 8% embryonic stem cell qualified FBS (Pansera, P30-2602), 1% non-essential amino acids (Gibco, 11140-035), 1% sodium pyruvate (Gibco, 11360-039), 1X penicillin/streptomycin (Gibco, 15070-063) and 0.1 mM β-mercaptoethanol (Sigma, M3148). This basal growth medium was completed with 3 µM CHIR99021 (Sigma SML1046), 1 µM PD0325901 (Sigma PZ0162) and 5 ng/mL LIF (Sigma LIF2050). Cells were cultured in 0.1% gelatin-coated labware. For passaging, cells were washed with DPBS (Ca-, Mg-, Gibco, 14190-094) and detached with 0.05% trypsin-EDTA. Cells were passaged no more than 10 times before the differentiation protocols were started.

Gastruloids were generated as previously described^61,62. 400 R1 mESCs were plated in 40 µL N2B27 medium in 96-well Clear Round Bottom Ultra-Low Attachment Microplates (Corning, 0707). N2B27 medium included Neurobasal™ medium (Gibco, 21103049), DMEM/F12 (Gibco, 11320074), 0.5% B27 supplement (Gibco, 17504-044), 0.5% N2 supplement (Gibco, 17502-048), 1X GlutaMax™ (Gibco, 35050-038) and 0.1 mM β-mercaptoethanol (Sigma, M3148). Plates were centrifuged at 100g for 5 min and placed in a 5% CO₂ incubator for 48 hours (day 2). Then, 150 µL of N2B27 containing 3 µM CHIR99021 (Sigma SML1046) was added to each well to induce gastrulation. At day 3 (72 h), the culture medium was replaced by N2B27 without CHIR90921. At day 4 (96 h), the medium was replaced by N2B27 supplemented with 30 ng/µL FGF2 (PMG0034 Gibco), 5 ng/µL VEGF (PHC9394 Gibco), and 0.5 mM ascorbic acid (Sigma, 255564) (Cardiogenic medium, N2B27⁺⁺⁺). Gastruloids were then cultured at 37ºC and 5% CO₂ on an orbital shaker (100 rpm). From day 5 (120 h) onward, the medium was changed daily with N2B27⁺⁺⁺ medium until day 7 (included). After day 8, N2B27 medium was used and changed daily until day 10 of incubation.

Digestion of E9.5 cardiac inflow region and single-cell RNA sequencing

25 E9.5 embryos were harvested from 3 CD1 pregnant females. Once extracted, embryos were washed in cold HBSS (Ca²⁺-/Mg²⁺-, Sigma, 14170112) and membranes were excised using microdissection tools. Subsequently, embryos were transferred to a plate containing cold DMEM (Gibco, 11965) supplemented with 5% FBS. Using tungsten needles and microsurgery forceps, the posterior part of cardiac inflow was excised and placed in 1.5 mL LoBind® Eppendorf tubes (Eppendorf, 0030108051). Tissue explants were pooled and maintained in a small DMEM 5% FBS drop. Then, the tissue samples were washed in HBSS using the minimum centrifuge velocity (400g). Single-cell dissociation was carried out by incubating tissue explants in 60 µL 0.05% trypsin-EDTA at 37ºC for 6 min (gently rocking the tube by hand every 2 min). Tissue dissociation was blocked using HBSS-10% FBS. The cell suspension was transferred to a new LoBind® Eppendorf tube using a 40-µm cell strainer (pluriSelect, 43-10040-40). Cells were washed twice in HBSS-5% FBS and finally diluted in HBSS-5% FBS until 900 single-cells/µL was reached according to the user guide for Chromium Next GEM Single Cell 3’ Reagent Kits v3.1 (10X Genomics, CG000204 Rev D). Viability and cell number were estimated using trypan blue in both the Countess™ 3 Automated Cell Counter (Invitrogen) and the KOVA™ Glasstic™ slide system (Fisher Scientific, 87144). All the experimental procedures before library construction took less than 2 hours. scRNA-seq libraries were constructed from 10,000 individual cells using 10X Genomics Chromium Next GEM Single Cell 3’ kit v3.1 (PN-1000268) and the Chromium controller. Libraries were then sequenced on an Illumina NovaSeq 6000.

scRNA-seq analysis

Raw data were initially analysed using 10X Genomics Cell Ranger 6.0.1. First, output data was converted to fastq and aligned on the Mus musculus reference genome GRCm38/mm10. Finally, Cell Ranger count analysis provided various parameters such as estimated number of cells, mean reads per cell and median number of genes per cell. As a result, 3 files were created: one for each technical replicate and a third by merging the two technical replicates. Then, each dataset was analysed with the R package Seurat v4.3.0, which is the R toolkit for single-cell genomics⁶³.

First, we evaluated the data quality through a linear correlation between the number of genes (features) and reads (counts) detected. After this analysis, we decided to keep cells with a range of 200 to 7,500 genes per cell, and genes detected in at least 10 cells. Then, we set cutoffs on cells with more than 5% mitochondrial gene expression, more than 10% haemoglobin genes and less than 10% of ribosomal genes. Cell doublets were predicted and removed using the R package DoubletFinder v2.0.3⁶⁴.

After completing the quality control steps, expression values across cells were normalised and scaled using the sctransform framework⁶⁵. Then, principal component analysis (PCA) reduced the dimensionality of the data by identifying principal components that captured the most significant variation in gene expression. Subsequently, a graph-based clustering approach was applied to cells and non-linear dimensional reduction UMAP was set up for data visualisation. Differential expression analysis was also conducted applying the findMarkers function providing the gene markers that defined each cluster. The analysis was restricted to genes with an FDR < 0.05 and a minimum fold change of 0.2 in 75% of the clusters to recover genes that were markers shared by closely related cells. Cell identities were manually annotated using previous studies as references^31–33.

The Harmony algorithm⁶⁶ was used to integrate our scRNA-seq dataset of the posterior cardiac inflow with the published scRNA-seq dataset of E12.5 Wt1-lineage cells²⁵. The RunHarmony function was used with a 2.5 value for the theta parameter and 30 dimensions of the Harmony reduction were used for calculating the UMAP.

Total RNA isolation and sequencing

E9.5 stage embryos from more than 3 CD1 mouse females were collected and washed in sterile PBS. Using tungsten needles and microsurgery forceps, two tissue samples were excised: the proepicardium (PE), corresponding to the ventral part of the STM, and part of the heart (HEART). Microdissected samples were pooled (n = 40) and put into TRIzol™ (Invitrogen, 15596026). RNA from three sample replicates were obtained. Total RNA was isolated using the phenol-chloroform method and stored at -80ºC. RNA integrity and concentration of all samples were measured using Agilent 2100 Bioanalyzer (Agilent Technologies). The RNA Integrity Number (RIN) for each sample was > 9, except for the first PE sample. Before RNA sequencing, gene expression of distinct marker genes for each anatomical region (Wt1 and Tbx18 for PE and Tnnt2 for HEART) was measured by qPCR (data not shown). The same RNA samples that were previously analysed by qPCR were submitted to sequencing. First, cDNA was prepared with at least 50 ng of total RNA using the standard Illumina TrueSeq RNASeq library preparation kit. Then, libraries were sequenced by Novogene in a NovaSeq 6000 system using the 150 bp paired-end sequencing strategy.

RNA-seq analysis

For quality control and preprocessing, SeqtrimBBsoftware was used with specific configurations⁶⁷. The plugins used included PluginAdapters for adapter removal, PluginPolyAt for trimming poly-A tails, PluginContaminants with a ribosomal contaminants database and a minimum ratio threshold, PluginQuality for quality filtering, and PluginLowComplexity for removing reads with low complexity regions. The minimum read length was set to 40 bases with a minimum quality PHRED score of 20. GRCm39 was used as the reference genome for alignment with sequences from Mus musculus. Gene expression quantification was performed using the featureCounts software. The expression levels were normalised using the counts per million (CPM) method to account for library size differences between samples. Data analysis was performed with iDEP 1.1 (integrated Differential Expression and Pathway analysis)⁶⁸. Briefly, after uploading the read count data, iDEP filtered out genes expressed at extremely low levels (in three samples, a gene must have more than 0.5 CPM) using the cpm function in edgRand conserved gene IDs according to Ensembl gene ID. Then, iDEP applied the regularised log transformation from the DESeq2package. Genes with an adjusted p-value (false discovery rate, FDR) less than 0.05 and a fold change greater than 2 were considered significantly, differentially expressed.

Immunohistochemistry

Tissues were fixed for 1–4 h in 4% PFA (depending on their size) and then washed three times in PBS. Samples were then incubated for 15–30 min in an ethanol series (50º, 70º, 80º, 90º, 96º and 100º) followed by a double 15-min incubation in butanol before paraffin embedding. Samples for cryo-sectioning were initially cryoprotected in 15% and 30% sucrose-in-PBS solutions. Then, samples were embedded in pure OCT medium (Tissue-Tek®, 4583) and flash-frozen in liquid N₂-cooled 2-methylbutane (Sigma, 277258). Paraffin and cryosections of 8 µm were cut and mounted in poly-L-lysine- or AAS (3-aminopropyltriethoxysilane)-treated slides, respectively.

Paraffin-embedded samples were de-waxed in xylene and re-hydrated in an ethanol series (100º, 96º, 90º, 80º, 70º, 50º), finishing in water and 1X PBS incubation steps. Antigen retrieval was performed using a 0.1 M citrate buffer (0.0825M sodium citrate and 0.0175M citric acid (pH 6) in a pressure cooker for 10 min. Non-specific binding sites were blocked in SBT buffer for 1h at room-temperature (RT) (SBT: 10% goat serum, 1% bovine serum albumin, and 0.5% Triton X-100 in Tris-PBS). For frozen samples, sections were initially rehydrated in PBS, followed by a non-specific binding site blocking step in SBT buffer at RT for 1h in SBT. Fixed cell culture samples were directly immersed in SBT after washing in 1X PBS. Then, samples were incubated overnight (O/N) at 4ºC with primary antibodies diluted in SBT. Primary antibodies used were Wt1 (1:100; Abcam, ab89901), Troponin I (1:100; Santa Cruz, sc133117), GFP (1:500; Abcam, ab13970), PHH3 (1:100; Millipore, 06570), Sox9 (1:100; Millipore, ab5535), and Tbx5 (1:100; Santa Cruz, sc515536). After that, samples were washed in TBST (0.1% Triton X-100 in PBS) and incubated with the corresponding secondary antibodies (anti-chicken AF488, Jackson 703-545-155; anti-rabbit AF647, Jackson 715605-151; anti-mouse AF546, Invitrogen A11030; anti-mouse Cy5, Jackson 715-175-150). Nuclei were counterstained with DAPI (Sigma, D9542) and cell membranes were stained with wheat germ agglutinin (WGA). Samples were mounted with PBS:Glycerol (1:1). All samples were analysed using the Leica SP5 HyD Confocal Microscope and images were processed using ImageJ 1.6 software.

RNA in situ hybridisation (RNAScope®)

Embryos were harvested at different developmental stages, washed in 1X PBS and fixed in 4% PFA O/N at 4ºC. After that, samples were washed in PBS, dehydrated using 30-min ethanol series (50º, 70º, 80º, 90º and 96º), and incubated overnight in 100º ethanol. After incubation in xylene for 15 min samples were embedded in paraffin and stored at 4ºC. Gastruloids were collected, washed in 1X PBS, and fixed in 4% PFA O/N at 4ºC. After that, samples were washed with 1X PBS and embedded in heat-liquified HistoGel™ (Epredia, HG-4000-012). Solidified HistoGel blocks were post-fixed with 4% PFA for 1h at RT and embedded in paraffin as described above.

Tissue sections of 8 µm were cut and mounted in Superfrost® Plus slides (Epredia™J1800AMNZ, 10149870) following the protocol provided by the RNAscope™ Multiplex Fluorescent v2 kit (ACD, 323110). The HybEZ™ Hybridisation System (ACD, 310010) was used for hybridization and incubation steps. Briefly, tissue sections were pre-treated – including deparaffinising and rehydrating – and submitted to treatment with an antigen retrieval reactive and protease solution. Then, RNA-specific probes designed for different fluorescent detection channels were hybridised to RNA targets (ACD™ Bio Wt1, ref. 432711; Itga4, ref. 542901; Twist1, ref. 1266861; Lhx2, ref 485791; Tcf21, ref. 508668; Ror2, ref. 430041; Wnt5a, ref. 316791; Gfp, ref. 400281; Postn, ref. 418581). Different Opal™ fluorescent reagents were used to detect gene probes (Akoya Opal™ 520, ref. FP1487001KT; Opal™ 570, ref. FP1488001KT; Opal™ 570, ref. FP1496001KT). After three specific signal amplifications, single RNA transcripts in distinctly fluorescent channels were imaged using the SP5 Confocal Microscope (Leica).

qPCR

Total RNA was isolated using the phenol-chloroform method and measured in a NanoDrop™ 2000/2000c Spectrophotometer (Thermo Scientific). Samples with a 1.8-2 260/280 ratio were considered for the cDNA synthesis. Prior to cDNA preparation, genomic DNA was removed using DNase I (Invitrogen, 18068015). Then, 500 ng of total RNA was converted into cDNA using the AffinityScript Multiple Temperature cDNA Synthesis Kit (Agilent, 200436). qPCR was run in QuantStudio 5 Real-Time PCR System according to PowerUp SYBR Green instructions (Thermo Scientific, A25742). The forward and reverse primers were used at 0.8 µM and in each qPCR reaction the amount of cDNA equivalent to 2.5-5 ng RNA was used. The amplification protocol consisted of 5 min at 95ºC followed by 45 cycles of 10 sec at 95ºC, 20 sec at 60ºC, and 20 sec at 72ºC. qPCR primers were designed using Primer-BLAST (NCBI). The melting curve analysis and size fractionation by agarose gel electrophoresis were used to confirm amplification of the expected products⁶⁹. Primers are listed in Supplementary Table S1. Relative gene expression (N0) was normalised with Tbp as a reference gene.

Cell proliferation analysis

Both EdU (5′-ethynyl-2′-deoxyuridine, Sigma, 900584) and BrdU (5-Bromo-2′-desoxiuridina, Sigma, B5002) were used to measure cell proliferation as they are nucleoside analogues of the thymidine that incorporate into the cell nuclei during the DNA synthesis phase of the cell division. Adult pregnant females were intraperitoneally injected with 50 mg EdU/kg body weight (diluted in 0.9% NaCl solution) 2h before embryo harvesting. Ex vivo cultured hearts were incubated with 10 µM EdU for 1h. Primary cultures of epicardial cells were incubated with 5 µg/mL BrdU for 1h. EdU detection on tissue sections was based on a Click-iT™ reaction between the EdU molecule and the fluorescent dye azide-fluoride 545 (Sigma, 760757). BrdU detection required a pre-treatment of 4% PFA-fixed cells (0.2N HCl incubation for 20 min followed by sodium tetraborate incubation for 10 min) and immunocytochemistry using a mouse anti-BrdU antibody (DSHB, G3G4) and a secondary anti-mouse antibody.

Cell cycle analysis by flow cytometry

Following ex vivo culture as described above, E11.5 hearts were collected and washed in PBS. Single-cell dissociation was carried out by incubating the hearts in 0.125 µg/mL Liberase TH (Roche, 5401135001) at 37ºC while rocking for a total of 10 min. Samples were filtered through 40-µm cell strainers and washed in cold PBS. Then, cells were fixed in cold 70% ethanol and stored at -20ºC for at least 24 h. Finally, samples were washed twice in cold PBS and incubated with Propidium Iodide (PI) solution (1 mg/mL PI, 0.1% Triton X-100, 200 µg/mL RNase I). The ratio between the number of cells and the volume of PI solution was maintained in all samples. Samples were analysed using the FACSCanto II flow cytometer (BD Biosciences). Downstream analysis of FCS files was performed using Kaluza v2.2 and the Michael H. Fox algorithm (included in the software) was used to analyse the cell cycle.

Protein extraction and quantification

EPIC cytoplasmic and nuclear protein fractions were isolated following the instructions in the Pierce™ NE-PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Scientific, 78833). Protein extracts were stored at -80ºC. The Pierce™ BCA Protein Assay Kit (Thermo Scientific, 23225) was used to determine the protein concentration. BSA and protein sample absorbance were measured at 595 nm in 96-well microplates (Deltalab, 900015) in the Biochrom Asys Expert Plus spectrophotometer (Biochrom Ltd, microplate reader G0 20150), and the protein concentration was determined using a linear regression from multiple BSA dilutions.

Western blot

A total of 30 µg of proteins was prepared in 1X Laemmli buffer (Thermo Scientific, J60015 AD), denatured at 95ºC for 5 min and separated in 12% Bis/Tris polyacrylamide gels. The proteins were transferred onto PVDF membranes (0.45 µm Immobilon-P®, Millipore, IPVH00010) using a Trans-Blot Turbo Transfer System (Bio-Rad). Then, membranes were blocked using 5% (m/v) non-fat dry milk diluted in TBS-T (10 mM Tris-HCl, 150 mM NaCl, 0.005% Tween-20) for 1h at RT, and incubated overnight with agitation at 4ºC (primary antibodies were prepared in blocking buffer). The proteins were incubated with rabbit anti-c-Jun (1;1000, Cell Signalling, 9165), mouse anti-Lamin B1 (2 ug/mL; Invitrogen, 33-2000) or rabbit anti-Gapdh (1:1000; Cell Signalling, 2118S). After three washes with TBS-T, HRP-conjugated anti-rabbit IgG (1:10.000; Sigma A0545) or anti-mouse IgG antibodies (1:10.000; Sigma A5278) were used in the following incubation step. Membranes were revealed using the SuperSignal™ West Pico PLUS (Thermo Scientific, 34580) or the SuperSignal™ West Femto kit (Thermo Scientific, 34095), according to the estimated levels of protein, and revealed in the ChemiDoc™ MP Imaging System (Bio-Rad).

Cell quantification

Itga4 ⁺ cells were quantified on the posterior part of the cardiac inflow from tissue sections from E9.5 Wt1-Cre^ERT2;R26R-EYFP embryos (n = 6) in co-localisation with Wt1 and Yfp gene expression. In sections of E12.5 Wt1-Cre^ERT2;R26R-EYFP embryos (n = 5), Itga4⁺/Yfp⁺ cells were quantified in the epicardium and subepicardium cardiac compartments using WGA staining to delimit cell membranes and basal lamina. In the in vivo proliferation analysis, Wt1⁺/Itga4⁺ (n = 9) and Itga4⁺/Ror2⁺ cells (n = 8) were quantified in the above-mentioned cell compartments in co-localisation with EdU staining. A minimum of 100 epicardial and subepicardial cells were analysed in each heart section. WT1⁺ and YFP⁺ cells were quantified in sections from wild-type and Wt1-Cre^ERT2;R26R-EYFP E11.5 hearts cultured ex vivo for 24h in vehicle and nocodazole conditions (n = 5) using troponin I staining to delimit the epicardium-myocardium compartments and WGA to mark cell membranes. C-JUN⁺ cells were quantified in sections from E12.5 embryos (n = 6) in co-localisation with EdU and using troponin I staining to delimit the epicardium-myocardium boundaries. Between 30–100 epicardial and subepicardial cells were analysed in each heart. BrdU⁺ cells were quantified in primary culture of epicardial cells (n = 8). In all cases, Dapi was used to stain cell nuclei and quantification was performed using ImageJ software.

Statistical analysis

Data were expressed as mean ± SEM for bar graph data presented and statistically significant differences among samples were measured by unpaired T-test assuming both populations have the same SD. P values < 0.05 were considered statistically significant. Bar graph data analysis was performed using GraphPad Prism (version 8).

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

CODE AVAILABILITY

All transcriptomic analyses were performed using standard protocols with previously described R packages in the methods. R scripts mentioned in this manuscript are available upon request.

DATA AVAILABILITY

Bulk RNA-seq data have been deposited in the GEO database under accession code “GSE269674”. Single-cell RNA-seq data generated in this study have been deposited in the GEO database under accession code “GSE269675”. Original western blot images have been deposited at Mendeley (DOI: 10.17632/swk4b7cnks.1). All other relevant data supporting the key findings of this study are available from the corresponding author upon request.

COMPETING INTERESTS

The authors declare no competing interests.

AUTHOR CONTRIBUTIONS

E.M.S. designed the experiments, analysed data and contributed to the writing of the manuscript. J.C.F. performed most bioinformatic analyses. A.S.M. and E.C. contributed to sample collecting, processing and immunofluorescence. L.D.M. analysed bulk RNA-seq data. F.L. contributed to the scRNA-seq and gastruloid experiment. A.R.V. supervised the research and contributed to the writing/supervision of the manuscript. J.M.P.P. supervised the research and wrote the manuscript with input from all authors.

ACKNOWLEDGEMENTS

We are grateful to the Genomics and Bioinformatics Platform (GBiM) from the U1251/Marseille Medical Genetics lab, in particular Catherine Robert and Camille Humbert, for performing the single-cell RNA-seq experiments and primary analysis respectively; Henk Buermans for his assistance with single-cell preparation (10X Genomics); María Isabel Castillo for her support in mouse breeding at the IBIMA-Platform Bionand Animal Housing Facility; Cristina Pogontke for her technical support (DCA laboratory, University of Malaga); Laurent Argiro for his support with the culture of gastruloids. This research was supported by the Spanish Ministry of Science and Innovation-National Research Agency grants PID2021-122626-OB-100 and RED2022-134485-T (to J.M.P.P), and PID2020-119430RJ-100 (to A.R.V.). E.M.S. was supported by a personal predoctoral fellowship from the Spanish Ministry of Education (FPU18/05219).

Männer J (1993) Experimental study on the formation of the epicardium in chick embryos. Anat Embryol (Berl). 187
Mikawa T, Fischman DA (1992) Retroviral analysis of cardiac morphogenesis: discontinuous formation of coronary vessels. Proc Natl Acad Sci 89:9504–9508
Virágh S, Challice CE (1981) The origin of the epicardium and the embryonic myocardial circulation in the mouse. Anat Rec 201:157–168
Kölliker A. Entwicklungsgeschichte des Menschen und der Thiere. In: 2., Engelmann (1879) 171–173
De Haan RL, Ursprung H (1965) Morphogenesis of the vertebrate heart. Organogenesis. New York, Holt, Reinhardt & Winston, pp 377–419
Virágh S, Gittenberger-de Groot AC, Poelmann RE, Kálmán F (1993) Early development of quail heart epicardium and associated vascular and glandular structures. Anat Embryol (Berl). 188
Männer J, Pérez-Pomares JM, Macías D, Muñoz-Chápuli R (2001) The origin, formation and developmental significance of the epicardium: a review. Cells Tissues Organs 169:89–103
Cano E et al (2016) Extracardiac septum transversum/proepicardial endothelial cells pattern embryonic coronary arterio-venous connections. Proc Natl Acad Sci U S A 113:656–661
Moore AW et al (1998) YAC transgenic analysis reveals Wilms’ tumour 1 gene activity in the proliferating coelomic epithelium, developing diaphragm and limb. Mech Dev 79:169–184
Kraus F, Haenig B, Kispert A (2001) cloning and expression analysis of the mouse T-box gene Tbx18. Mech Dev 100:83–86
Quaggin SE, Vanden Heuvel GB, Igarashi P (1998) Pod-1, a mesoderm-specific basic-helix-loop-helix protein expressed in mesenchymal and glomerular epithelial cells in the developing kidney. Mech Dev 71:37–48
Gittenberger-de-Groot AC, Vrancken-Peeters MPFM, Mentink MMT, Gourdie RG, Poelmann RE (1998) Epicardium-derived cells contribute a novel population to the myocardial wall and the atrioventricular cushions. Circ Res 82:1043–1052
Mikawa T, Gourdie RG (1996) Pericardial mesoderm generates a population of coronary smooth muscle cells migrating into the heart along with ingrowth of the epicardial organ. Dev Biol 174:221–232
Pérez-Pomares JM, Macías D, García-Garrido L, Muñoz-Chápuli R (1998) The Origin of the subepicardial mesenchyme in the avian embryo: an immunohistochemical and quail–chick chimera study. Dev Biol 200:57–68
Männer J (1999) Does the subepicardial mesenchyme contribute myocardioblasts to the myocardium of the chick embryo heart? A quail-chick chimera study tracing the fate of the epicardial primordium. Anat Rec 255:212–226
Wessels A et al (2012) Epicardially derived fibroblasts preferentially contribute to the parietal leaflets of the atrioventricular valves in the murine heart. Dev Biol 366:111–124
Ruiz-Villalba A et al (2015) Interacting resident epicardium-derived fibroblasts and recruited bone marrow cells form myocardial infarction scar. J Am Coll Cardiol 65:2057–2066
Pogontke C, Guadix JA, Ruiz-Villalba A, Pérez-Pomares JM (2019) Development of the myocardial interstitium. Anat Rec 302:58–68
Pogontke C et al (2022) Dynamic epicardial contribution to cardiac interstitial c-Kit and Sca1 cellular fractions. Front Cell Dev Biol. 10
Forte E et al (2020) Dynamic interstitial cell response during myocardial infarction predicts resilience to rupture in genetically diverse mice. Cell Rep 30:3149–3163
Chen T, You Y, Jiang H, Wang ZZ (2017) Epithelial–mesenchymal transition (EMT): A biological process in the development, stem cell differentiation, and tumorigenesis. J Cell Physiol 232:3261–3272
Thiery JP, Acloque H, Huang RYJ, Nieto MA (2009) Epithelial-mesenchymal transitions in development and disease. Cell 139:871–890
Perez-Pomares JM, De La Pompa JL (2011) Signaling during epicardium and coronary vessel development. Circ Res 109:1429–1442
Wu M et al (2010) Epicardial spindle orientation controls cell entry into the myocardium. Dev Cell 19:114–125
Quijada P et al (2021) Coordination of endothelial cell positioning and fate specification by the epicardium. Nat Commun 12:1–18
Mantri M et al (2021) Spatiotemporal single-cell RNA sequencing of developing chicken hearts identifies interplay between cellular differentiation and morphogenesis. Nat Commun 12:1771
Deng Y, He Y, Xu J, He H, Li G (2023) Heterogeneity and functional analysis of cardiac fibroblasts in heart development. bioRxiv
Streef TJ et al (2023) Single-cell analysis of human fetal epicardium reveals its cellular composition and identifies CRIP1 as a modulator of EMT. Stem Cell Rep 18:1421–1435
Katz TC et al (2012) Distinct compartments of the proepicardial organ give rise to coronary vascular endothelial cells. Dev Cell 22:639–650
Lupu IE, Redpath AN, Smart N (2020) Spatiotemporal analysis reveals overlap of key proepicardial markers in the developing murine heart. Stem Cell Rep 14:770–787
de Soysa TY et al (2019) Single-cell analysis of cardiogenesis reveals basis for organ-level developmental defects. Nat 572:120–124
Lotto J et al (2020) Single-Cell transcriptomics reveals early emergence of liver parenchymal and non-parenchymal cell lineages. Cell 183:702–716
Tyser RC et al (2021) Characterization of a common progenitor pool of the epicardium and myocardium. Sci 371:6533
Palmquist-Gomes P et al (2023) Origin of congenital coronary arterio-ventricular fistulae from anomalous epicardial and myocardial development. Exp Mol Med 55:228–239
Yang JT, Rayburn H, Hynes RO (1995) Cell adhesion events mediated by alpha 4 integrins are essential in placental and cardiac development. Dev 121:549–560
Zhang H et al (2017) Fibroblasts in an endocardial fibroelastosis disease model mainly originate from mesenchymal derivatives of epicardium. Cell Res 27:1157–1177
Zhou B et al (2008) Epicardial progenitors contribute to the cardiomyocyte lineage in the developing heart. Nat 454:109–113
Ruiz-Villalba A, Ziogas A, Ehrbar M, Pérez-Pomares JM (2013) Characterization of epicardial-derived cardiac interstitial cells: differentiation and mobilization of heart fibroblast progenitors. PLoS ONE 8:17–19
Günthel M, Barnett P, Christoffels VM (2018) Development, proliferation, and growth of the mammalian heart. Mol Ther 26:1599–1609
Wilkinson DG, Bhatt S, Ryseck RP, Bravo R (1989) Tissue-specific expression of c-jun and junB during organogenesis in the mouse. Dev 106:465–471
Wisdom R (1999) c-Jun regulates cell cycle progression and apoptosis by distinct mechanisms. EMBO J 18:188–197
Ho HYH et al (2012) Wnt5a–Ror–Dishevelled signaling constitutes a core developmental pathway that controls tissue morphogenesis. Proc Natl Acad Sci 109:4044–4051
Yamanaka H et al (2002) JNK functions in the non-canonical Wnt pathway to regulate convergent extension movements in vertebrates. EMBO Rep 3:69–75
Asp M et al (2019) A Spatiotemporal organ-wide gene expression and cell atlas of the developing human heart. Cell 179:1647–1660
Feng W et al (2022) Single-cell transcriptomic analysis identifies murine heart molecular features at embryonic and neonatal stages. Nat Commun 13:7960
Guadix JA et al (2011) Wt1 controls retinoic acid signalling in embryonic epicardium through transcriptional activation of Raldh2s. Dev 138:1093–1097
Moore AW, McInnes L, Kreidberg J, Hastie ND, Schedl A (1999) YAC complementation shows a requirement for Wt1 in the development of epicardium, adrenal gland and throughout nephrogenesis. Dev 126:1845–1857
Kalman F, Viragh S, Módis L (1995) Cell surface glycoconjugates and the extracellular matrix of the developing mouse embryo epicardium. Anat Embryol (Berl) 191:451–464
Männer J (1992) The development of pericardial villi in the chick embryo. Anat Embryol (Berl) 186:379–385
Yang Y, Mlodzik M (2015) Wnt-Frizzled/Planar cell polarity signaling: cellular orientation by facing the wind (Wnt). Annu Rev Cell Dev Biol 31:623–646
Schleiffarth JR et al (2007) Wnt5a is required for cardiac outflow tract septation in mice. Pediatr Res 61:386–391
von Gise A et al (2011) WT1 regulates epicardial epithelial to mesenchymal transition through β-catenin and retinoic acid signaling pathways. Dev Biol 356:421–431
Li H et al (2019) Rate of progression through a continuum of transit-amplifying progenitor cell states rgulates blood cell production. Dev Cell 49:118–129
Rangel-Huerta E, Maldonado E (2017) Transit-amplifying cells in the fast lane from stem cells towards differentiation. Stem Cells Int. 1–10 (2017)
Dubé KN et al (2017) Recapitulation of developmental mechanisms to revascularize the ischemic heart. JCI Insight 2:e96800
Russell JL et al (2011) A dynamic notch injury response activates epicardium and contributes to fibrosis repair. Circ Res 108:51–59
Sayed A et al (2022) Hypoxia promotes a perinatal-like progenitor state in the adult murine epicardium. Sci Rep 12:9250
Srinivas S et al (2001) Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev Biol 1:4
Chevalier C, Nicolas JF, Petit AC (2014) Preparation and delivery of 4-hydroxy-tamoxifen for clonal and polyclonal labeling of cells of the surface ectoderm, skin, and hair follicle. Methods Mol Biol 1195:239–245
Mellgren AM et al (2008) Platelet-derived growth factor receptor β signaling is required for efficient epicardial cell migration and development of two distinct coronary vascular smooth muscle cell populations. Circ Res 103:1393–1401
Rossi G et al (2021) Capturing cardiogenesis in gastruloids. Cell Stem Cell 28:230–240
Argiro L et al (2023) Cardiopharyngeal mesoderm specification into cardiac and skeletal muscle lineages in gastruloids. bioRxiv
Hao Y et al (2021) Integrated analysis of multimodal single-cell data. Cell 184:3573–3587
McGinnis CS, Murrow LM, Gartner ZJ (2019) DoubletFinder: doublet detection in single-cell RNA sequencing data using artificial nearest neighbors. Cell Syst 8:329–337
Choudhary S, Satija R (2022) Comparison and evaluation of statistical error models for scRNA-seq. Genome Biol 23:27
Korsunsky I et al (2019) Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods 16:1289–1296
Bushnell B (2018) BBTools: a suite of fast, multithreaded bioinformatics tools designed for analysis of DNA and RNA sequence datae. Joint Genome Inst Published https://sourceforge.net/projects/bbmap/
Ge SX, Son EW, Yao R (2018) iDEP: an integrated web application for differential expression and pathway analysis of RNA-Seq data. BMC Bioinformatics 19:534
Ruiz-Villalba A, van Pelt-Verkuil E, Gunst QD, Ruijter JM, van den Hoff (2017) M.J. Amplification of nonspecific products in quantitative polymerase chain reactions (qPCR). Biomol Detect Quantif. 14, 7–18

There is NO Competing Interest.

2024.08.01MarinSedenoetalsourcedataandexceldatasets.zip
Source data and DEGs files
2024.08.01MarinSedenoSupplementalFigures.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Wt1+/Itga4+ progenitors drive epicardial-derived cell expansion via non-canonical Wnt signalling

Status:

Version 1

Abstract

Figures

INTRODUCTION

RESULTS

Deciphering cellular heterogeneity in the cardiac inflow

scRNA-seq transcriptomics segregates the PE from the ST

Transcriptional dynamics in epicardial cell lineage

Wt1+/Itga4+ epicardial cells and early EPDCs are highly proliferative

Non-canonical Wnt signals regulate the expansion of primitive epicardial cells and early EPDCs

DISCUSSION

METHODS

Animal models

Ex vivo E11.5 heart culture

In vitro culture of epicardial explants and EPICs

Proepicardial differentiation in cardiac gastruloids

Digestion of E9.5 cardiac inflow region and single-cell RNA sequencing

scRNA-seq analysis

Total RNA isolation and sequencing

RNA-seq analysis

Immunohistochemistry

RNA in situ hybridisation (RNAScope®)

qPCR

Cell proliferation analysis

Cell cycle analysis by flow cytometry

Protein extraction and quantification

Western blot

Cell quantification

Statistical analysis

Reporting summary

Declarations

CODE AVAILABILITY

DATA AVAILABILITY

COMPETING INTERESTS

AUTHOR CONTRIBUTIONS

ACKNOWLEDGEMENTS

References

Additional Declarations

Supplementary Files

Status:

Version 1

Wt1⁺/Itga4⁺ progenitors drive epicardial-derived cell expansion via non-canonical Wnt signalling

Wt1⁺/Itga4⁺ epicardial cells and early EPDCs are highly proliferative