Endocardial-mesenchymal transition underlies fusion of the conotruncal ridges during normal and bicuspid aortic valve development

doi:10.21203/rs.3.rs-4999094/v1

Download PDF

Article

Endocardial-mesenchymal transition underlies fusion of the conotruncal ridges during normal and bicuspid aortic valve development

https://doi.org/10.21203/rs.3.rs-4999094/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Right-left bicuspid aortic valve (R-L BAV) is the most frequent phenotype of the most common congenital heart disease. Its etiology is based on two associated morphogenetic defects during cardiac outflow tract (OFT) septation: abnormal migration of cardiac neural crest (CNC) cells, and excessive fusion of the conotruncal ridges (CRs). The aim of this study is to elucidate the mechanism involved in the fusion of the CRs responsible for normal and abnormal OFT septation and BAV formation. Two mechanisms have been proposed: endocardial apoptosis and endocardial-mesenchymal transition (EMT). The involvement of these mechanisms in the fusion event was tested in embryos of the hamster model with BAV. Apoptotic cells were absent in the fusion area of the CRs. However, we detected endocardial cells (CD34⁺;VE-Cadherin⁺) showing positive signals for migration markers (α-actin⁺) in the fusion area of the CRs of embryos developing both normal aortic valve and BAV. These cells showed an intermediate morphological phenotype between endocardial and mesenchymal cells. The findings clearly indicate that EMT, and not apoptosis, is the cellular mechanism underlying the normal and excessive fusion of CRs that give rise to tricuspid aortic valve and BAV, respectively. Furthermore, our results show that the fusion of CRs in embryos developing BAV continues after the OFT septation, suggesting over-induction of EMT by abnormally distributed CNC cells.

Biological sciences/Biological techniques/Biological models/Cardiovascular models

Biological sciences/Biological techniques/Biological models/Animal disease models

Biological sciences/Biological techniques/Microscopy/Confocal microscopy

Biological sciences/Biological techniques/Experimental organisms/Model vertebrates/Hamster

Health sciences/Cardiology/Cardiovascular biology/Cardiovascular diseases/Congenital heart defects

Health sciences/Cardiology/Cardiovascular biology/Cardiovascular diseases/Valvular disease

Health sciences/Cardiology/Cardiovascular biology/Heart development

Biological sciences/Developmental biology/Embryogenesis/Epithelial mesenchymal transition

Biological sciences/Developmental biology/Embryology

Biological sciences/Biological techniques/Immunological techniques/Immunohistochemistry

Bicuspid aortic valve

cardiac outflow tract septation

conotruncal ridge

fusion

endocardial-mesenchymal transition

hamster

The cardiac outflow tract (OFT) or conotruncus is the most cranial segment of the embryonic heart. It consists of a single tubular chamber that connects the right ventricle with the aortic sac and the aortic arch arteries (Fig. 1a). Conotruncal septation is the process by which the OFT is divided into two, right and left independent outflows. The conotruncus contains two long, helical, and opposing mesenchymal cushions covered by endocardium. These cushions, called conotruncal ridges (CRs), are involved in the septation process and form the primordia of the semilunar valves (Fig. 1). Septation takes place through the formation of the conotruncal septum (CS), after the fusion of the two facing CRs [1–6]. This fusion event occurs when the endocardium lining the two opposing and contacting CRs disappears, allowing cellular continuity between the CRs, resulting in a unique mesenchymal mass (Fig. 1b). Then, cardiac neural crest (CNC)-derived mesenchymal cells aggregate into this mesenchymal mass to form the CS (Fig. 1b) [1, 2, 5, 7]. Besides the well-known role of CNC cells in conotruncal septation, and the numerous genes that have been shown to be involved in the process (reviewed in [8]), the cellular mechanism underlying the fusion of CRs remains unknown.

The margins of the CRs constitute the primordia of the right and left semilunar valve leaflets and sinuses, the so-called right and left aortic and pulmonary valve cushions (Fig. 1b). Two additional small mesenchymal cushions, named intercalated cushions (ICs), located anteriorly and posteriorly in the OFT (Fig. 1b), form the primordia of the posterior (non-coronary) and anterior leaflets and sinuses of the aortic and pulmonary valves, respectively [6, 9]. As a result of septation, six mesenchymal cushions arise in the OFT: the right and left aortic and pulmonary valve cushions, derived from the CRs, and the anterior pulmonary and posterior or non-coronary aortic valve cushions, derived from the ICs (Fig. 1b). The aortic valve cushions acquire the adult semilunar morphology through a still poorly understood process called excavation [10–12].

Bicuspid aortic valve (BAV) is the most common congenital heart disease in humans, with an incidence of 1–2% in the general population [13, 14]. The anatomical elements that constitute a normal or tricuspid aortic valve (TAV) are reduced in BAVs, and two anatomical morphotypes have been described. The antero-posterior BAV, also called right-left (R-L) or BAV type A, consists of two sinuses and leaflets arranged in antero-posterior orientation. In contrast, the latero-lateral BAVs (BAV type B or right-non-coronary -R-NC-, and BAV type C or left-noncoronary -L-NC-) also show two sinuses and leaflets but located at the right and left side of the aortic root [15, 16]. R-L BAV is the most frequent BAV morphotype (70–80%), followed by R-NC BAV (20–30%) and L-NC BAV (3–6%) [16–18].

The R-L BAV results from a defect during the OFT septation process: the excessive fusion of the CRs accompanied by an altered organization of the CS [8, 19]. As a result of the septation defect, the right and left aortic valve cushions develop as fused structures. CNC cells are involved in both endocardial cushion formation and conotruncal septation [1, 2, 7, 20]. These cells, which come from the neural tube, migrate along the OFT to participate in the correct formation and positioning of CRs and ICs. Subsequently, they induce and regulate the conotruncal septation process, forming the CS. Although the participation of CNC cells in the OFT septation is well established, the cause of the extra-fusion of CRs is not understood in detail due to the lack of knowledge of the cellular mechanism involved in the normal process of fusion of the CRs.

Currently, only one spontaneous animal model of non-syndromic BAV has been described. It consists of an inbred strain (T strain) of Syrian hamster (Mesocricetus auratus) with a relatively high (⁓40%) incidence of R-L BAV [8, 12, 19, 21–24]. This animal model has substantially contributed to the current knowledge on BAV etiology [8, 12, 13, 16, 19, 21, 25]. The anatomy and inheritance of BAV in the hamster model are similar to those in patients [22, 23].

The aim of this study was to elucidate the cellular mechanism involved in the fusion of the CRs of hamster embryos from the T strain and a control strain with null incidence of BAV. Three fusion mechanisms have been proposed to operate during embryonic development: epithelial adhesion, epithelial apoptosis and epithelial-mesenchymal transition (EMT) [26]. Apoptosis and EMT were selected to be tested as possible candidate mechanisms. For this purpose, we examined hamster embryos at the stage of conotruncal septation (embryonic day (ED) 11 and ED12) from the T and the control strain, and assessed several experiments, looking for evidence of apoptosis and EMT in the fusing CRs. Apoptosis was assessed by the TUNEL assay and immunofluorescence for active caspase 3. EMT was assessed by cellular morphology in semithin sections and double immunofluorescence for markers for endocardium (CD34 and VE-Cadherin) and cell migration (α-actin).

Normal conotruncal septation process in the Syrian hamster

In the Syrian hamster, the conotruncal septation begins at ED11. At this stage, the embryonic cardiac OFT is a myocardial tube containing four endocardial cushions (Fig. 2a,g). The CRs are two long, helical, and opposite endocardial cushions extending through the entire OFT, named septal and parietal CRs. The other two endocardial cushions, the anterior and posterior ICs are smaller and restricted to the distal region of the OFT, at the level where the semilunar valves are formed. The CRs and the ICs are composed of mesenchymal cells lined by a layer of endocardial cells. The conotruncal septation process begins when the central portion of the septal and parietal CRs get fused (Fig. 2b,d-f,h,i). The fusion of the CRs starts with the contact of the endocardial linings covering the opposite CRs (Fig. 2a,d), followed by the disappearance of the endocardial cells in the center of the ridges (Fig. 2b,e,f,h,i) and the formation of a mesenchymal condensation called CS (Fig. 2c). The region of the CRs where the contacting endocardial cells disappear delimits the area of fusion, which extends proximally as septation progresses. At ED12, the OFT is divided by the CS into two independent OFTs, each containing three mesenchymal cushions, the valve cushions, which constitute the aortic and pulmonary valves primordia (Fig. 2c). The right and left aortic and pulmonary valve cushions develop from the unfused margins of the CRs, whereas the posterior (non-coronary) aortic and anterior (pulmonary) valve cushions develop from the ICs. The valve cushions differentiate in the adult leaflets and sinuses of the semilunar valves.

Mechanism of fusion of the conotruncal ridges

Inspection of serial sections of the OFT of ED11 and ED12 embryos from both the control and the T strain revealed TUNEL⁺ and caspase 3⁺ cells in the thoracic wall (Fig. 3a,b,d,e) and in the CS (Fig. 3a,c,d,f) of ED12 embryos. However, no positive signals were found in the fusion area of the CRs in ED11 embryos (Fig. 3g-i). The amount and location of positive signals were similar in embryos from the T and control strains.

In order to assess EMT process during the fusion of the CRs (Fig. 4), we first performed a fine morphological study using frontal semithin sections of the fusion area of ED11 embryos. Each CR was composed of mesenchymal cells covered by a monolayer of endocardial cells (Fig. 4a). Mesenchymal cells in the core of the CRs were featured by a small cell body, few long and thin membrane projections (i. e., filopodia and lamellipodia), and a large, round nucleus with disseminated chromatin (Fig. 4b). Most endocardial cells showed a flat morphology with a prominent, flattened, elliptical nucleus that occupied most of the cell body (Fig. 4c). Three consecutive disto-proximal areas could be distinguished in the endocardium covering the CRs (Fig. 4d). In the proximal region (Fig. 4e), septal and parietal CRs were separated, and endocardial cells presented the typical endothelial morphology described above. In the middle region (Fig. 4f), the two opposite endocardial linings were in contact, so that endocardial cells did not face the conotruncal lumen, but exhibited a similar morphology to that described earlier. In the most distal region (Fig. 4g), corresponding to the fusion area, endocardial cells from one CR were also in contact with endocardial cells from the opposite ridge, but showed a distinct morphology. Three types of cells could be described in this location: 1) cells with a normal endothelial morphology (black arrows in Fig. 4c,e,f); 2) cells with mesenchymal morphology (black arrowheads in Fig. 4b,f-h); and 3) cells with both endothelial and mesenchymal features (white arrowheads in Fig. 4g,h). This intermediate phenotype was characterized by an endothelial cell appearance with moderately flattened morphology and hypertrophied nucleus, together with extensions of the plasmatic membrane, compatible with filopodia and lamellipodia, similar to those found in mesenchymal cells (Fig. 4g,h). Cells with mixed phenotype were predominantly present in the central portion of the fusion area (Fig. 1b, Supplemental Fig. 1).

We also conducted double immunofluorescence with endocardial (CD34 and VE-Cadherin) and migration (α-actin) markers in frontal (Fig. 5a-c) and transversal (Fig. 5d-l) histological sections of the fusion area. Lining endocardial cells showed CD34⁺ signals in the outer cell membrane (Fig. 5b,i) and VE-Cadherin⁺ punctate intercellular signals (Fig. 5c,l). a-actin⁺ mesenchymal cells were particularly frequent in the most distal region of the cardiac OFT, just above the area of fusion of the CRs (Fig. 5b,c), along the line of fusion (Fig. 5g,j) and in the central portion of coalesced CRs (Fig. 5h,k). In the fusion area we constantly found some endocardial cells that showed CD34, α-actin and VE-Cadherin positive signals (Fig. 5b,c,h,k). These cells were in close contact with the before mentioned α-actin⁺ mesenchymal cells, and were located where cells with an intermediate morphological phenotype were previously detected (Fig. 5a-c).

Fusion of the conotruncal ridges during bicuspid aortic valve formation

A developing BAV was detected in 15 out of 40 (37.5%) ED11 and ED12 embryos of the T strain. In these embryos, an excessive fusion of the CRs (ED11, Fig. 6a) or fused right and left aortic valve cushions (ED12, Fig. 6j) were observed.

In ED11 embryos developing BAV, the fusion of the CRs was not restricted to their central portions but extended to the posterior margins of the CRs (Fig. 6a). Double immunofluorescence with CD34 or VE-Cadherin and α-actin in transversal (Fig. 6b,c) and frontal (Fig. 6e,f) sections revealed the presence of CD34⁺; α-actin⁺ and VE-Cadherin⁺; α-actin⁺ cells in the region of CRs affected by extra-fusion. These cells were adjacent to the endocardium in the contact area (Fig. 6b,c). The analysis of frontal sections of the fusion area (Fig. 6e,f) revealed α-actin⁺ mesenchymal cells located in the most distal portion, CD34⁺; α-actin⁻ and VE-Cadherin⁺; α-actin⁻ endocardial cells in the most proximal region, and double-labelled cells in the intermediate zone. This cellular distribution pattern was very similar to that described in normal embryos (compare Fig. 5a-c with Fig. 6d-f).

After fusion of the CRs (ED12), the CS develops in the center of the fused area, and the aortic and pulmonary valve cushions become delineated in the OFT (Fig. 6g). In embryos developing TAV, the right and left aortic valve cushions were delimited by the OFT lumen, the CS and the conotruncal myocardium, while the posterior or non-coronary valve cushion was delimited by the OFT lumen and the conotruncal myocardium (Fig. 6g). Double immunofluorescence for CD34 and α-actin showed that α-actin⁺ signals were mainly found in the CS and the conotruncal myocardium, whereas endocardial cells were positive for CD34 endothelial marker (Fig. 6h). The endocardium that delimited the left and right aortic valve cushions reached the CS, demarcating two separated and well-defined cushions (Fig. 6i). No CD34⁺; α-actin⁺ cells were detected near the endocardium (Fig. 6i). By contrast, in embryos developing BAV, the extended fusion of the posterior margins of the CRs had led to fusion of the right and left aortic valve cushions (Fig. 6j). CD34⁺ endocardial cells lined the left and right aortic valve cushions, but the endocardium did not reach the CS, showing two fused valve cushions (Fig. 6j,k). However, some cells of the fused region of the cushions located near the endocardium exhibited positive double signals for CD34 and α-actin (Fig. 6l).

The division of the embryonic common OFT into right and left outflows, usually referred to as OFT septation, is a key developmental process for the conversion of a single to a double circulatory system in mammals [1–3, 6]. Defects in OFT septation lead to multiple cardiac abnormalities, the sum of which constitute a wide majority of the human congenital cardiac malformations [3, 4, 27].

Cardiac OFT septation is an evolutionary conserved embryonic mechanism in birds and mammals [5, 27]. It consists in the fusion of the OFT endocardial cushions that cover the inner surface of the conotruncus, allowing the formation of a mesenchymal CS, and the formation of the aortic and pulmonary valve primordia [1–3, 6]. Therefore, OFT division and semilunar valve formation are embryonically linked processes.

Conotruncal septation in the Syrian hamster occurs between stages ED11 (fusion of the CRs) and ED12 (formation of the CS). The fusion of the CRs entails disappearance of the endocardium covering the two opposite CRs after contact. Three fusion mechanisms have been proposed to operate during embryonic development in different morphogenetic processes, including the fusion of the CRs: epithelial adhesion, epithelial apoptosis and epithelial-mesenchymal transition (EMT) [26]. These mechanisms are typically involved in fusion events such as neural tube formation or palate development. The first mechanism requires the expression of adhesion molecules and the maintenance of the identity of cells in contact, which occurs during early stages of neural tube formation [28]. In the other two mechanisms, epithelial cells covering the fusing structures disappear by apoptosis or by transforming into mesenchymal cells (EMT), which are the cellular mechanisms underlying the palatal shelf fusion [29]. Since the endothelium that constitutes the endocardium is a specialized squamous epithelial tissue [30], we considered that these three mechanisms of fusion can be extrapolated to the endocardium. Taking into account that endocardial cells disappear during the CRs fusion event, we selected apoptosis and EMT as possible candidate mechanisms to be tested.

Experiments carried out to analyze apoptosis, i.e., TUNEL assay and immunofluorescence for active caspase 3, did not show apoptotic cells in the fusion area of CRs (Fig. 3). However, we identified apoptotic cells in other embryonic structures, such as the CS and the thoracic wall, in which the presence of apoptosis had been previously described [31, 32], acting as internal positive controls. Regarding EMT, fine morphological analysis and colocalization experiments using double immunofluorescence determined the presence of three different cell types in the fusion region (Fig. 4,5): 1) CD34⁺;VE-Cadherin⁺ cells in the proximal region of the fusion area, showing a typical endothelial morphology (prominent, elliptical, flattened nucleus occupying most of the cell body; [33, 34]); 2) α-actin⁺ cells in the distal area to the fusion zone, which presented a typical mesenchymal morphology (small cell body, with membrane projections and a large, round nucleus with scattered chromatin particles; [34, 35]); and 3) CD34⁺;VE-Cadherin⁺;α-actin⁺ cells with a mixed phenotype between endothelial and mesenchymal cells, located just in the region where endocardial cells disappeared, in an intermediate position between the two previously described cell types, which had a moderately flattened endothelial morphology, a hypertrophied nucleus and/or plasma membrane extensions compatible with filopodia and lamellipodia. This intermediate phenotype includes morphological features previously ascribed to EMT [35–37] (Fig. 7).

These findings clearly indicate that EMT, and not apoptosis, is the cellular mechanism underlying CR fusion. The endocardial cells in the central region of the opposing CRs disappear by transforming into mesenchymal cells, morphologically indistinguishable from the rest of mesenchyme in the cushions core, and leading to morphological continuity between the CRs (Fig. 7). The distribution of each cell type in the fusion area allows establishing a sequential pattern of the EMT process (Fig. 2,4). It can be inferred that the most proximal CD34⁺; VE-Cadherin⁺; α-actin⁻ endothelial cells have not yet initiated EMT; cells with intermediate phenotypes and positive signals for the three markers (CD34⁺; VE-Cadherin⁺; α-actin⁺) are endothelial cells transforming into mesenchymal cells; and in the most distal α-actin⁺ mesenchymal cells, the transition event is ending (Fig. 5). Thus, this pattern reflects a disto-proximal sequence of cellular events consistent with the EMT mechanism (Fig. 7). Nevertheless, the whole morphogenetic process of conotruncal septation follows the same disto-proximal sequence.

Two cellular mechanisms have been ascribed to the EMT process in the literature [38]. The first is known as partial EMT, according to which transformation may affect only a few cells within the epithelium or may occur through stages, in such a way that the epithelium remains intact during the entire process, being the transformed epithelial cells replaced by neighbors, and eventually yielding two distinct cell types (epithelial and mesenchymal cells) after the EMT process. The second mechanism entails a complete EMT, that takes place when all cells in the epithelium undergo transformation simultaneously, with a complete disintegration of the epithelial cover that transforms into mesenchyme. Both EMT mechanisms have been previously described in processes involved in embryonic development, such as delamination of cranial neural crest cells and sclerotome and dermatome development, respectively [39,40]. The results obtained in this study allow us to suggest that the fusion of CRs is based on a complete EMT mechanism that sequentially affects the underlying endocardial cells located in the center of the CRs. This would explain the presence of cells with different phenotypes in the fusion area (i. e., mesenchymal cells already transformed, transforming cells with intermediate phenotype, endothelial cells waiting for transformation) (Fig. 7), whereas endothelial cells lateral to the fusion area remain unaffected, constituting the endothelial cover of the valve cushions. The specific location of the EMT process in the center of the CRs might be the consequence of the specific and localized induction by CNC cells.

R-L BAV in hamsters results from the excessive fusion of the CRs, leading to the formation of two instead of three aortic valve cushions [8, 19]. We reasoned that the excessive fusion of the CR might be caused by an extension of the EMT process, which is normally restricted to the center of the CRs, to the posterior margins of the ridges, affecting the adjacent areas of the presumptive right and left aortic valve cushions. To test this hypothesis, we examined ED11 and ED12 hamster embryos from the T strain, with relatively high incidence of R-L BAV (⁓40%). Fifteen out of 40 (37.5%) embryos exhibited a developing BAV. At ED11, BAV embryos presented an excessive fusion of CRs that extended towards the posterior margins, as described by Fernández et al. (2009) [19] (Fig. 6a). The immunostainings revealed the existence of CD34⁺; VE-Cadherin⁺; α-actin⁻ signals in the lining endocardium, CD34⁻; VE-Cadherin⁻; α-actin⁺ cells in the mesenchyme core of the CRs, and CD34⁺; VE-Cadherin⁺; α-actin⁺ cells specifically in the region of CRs affected by extra-fusion (Fig. 6b,c). Furthermore, these cells showed a disto-proximal distribution pattern very similar to that observed in the fusion area of normal embryos: CD34⁺; VE-Cadherin⁺; α-actin⁺ cells located in the medial portion of the fusion area, bounded distally by α-actin⁺ mesenchymal cells and proximally by CD34⁺; VE-Cadherin⁺; α-actin⁻ endocardial cells (Fig. 6e,f). These findings suggest that EMT is also involved in the extra-fusion of CRs in BAV embryos. In addition, the results reinforce the hypothesis proposed above of a fusion event based on a sequential and complete EMT mechanism.

Regarding embryos at ED12 developmental stage, control specimens presented a CD34⁺ endocardium that reached the CS and delimited two separated and well-defined left and right aortic valve cushions. However, in BAV embryos from the T strain, CD34⁺ lining endocardium did not reach the CS, allowing the continuity between the two mesenchymal masses, and evidencing the fusion of the left and right aortic valve cushions (Fig. 6i,l). Some CD34⁺; α-actin⁺ signals were constantly detected in cells adjacent to the endocardium of the contacting valve cushions. We interpret that these double-labelled cells are endocardial cells transforming into mesenchyme. These findings show for the first time that the fusion of CRs can continue after the formation of the CS.

We propose that during the formation of R-L BAVs, the number of endocardial cells involved in the EMT process is altered, resulting in a spatial (i. e., spreading of the fusion towards the posterior limits of the CRs) and temporal (i. e., after de CS formation) extension of the fusion of the CRs. The alteration of EMT underlying BAV formation is supported by several genetically modified mouse strains (reviewed in [8]), in which defective expression of genes known to regulate EMT during OFT development, such as eNOS, Brg1, Ift88 and Npr2, leads to BAV as a consequence of dysregulation of EMT that affects cushion formation. In addition, a recent study on BAV-associated aortopathy demonstrated overexpression of EMT genes related to endocardial cushion formation in the aorta of affected patients [41], suggesting a common embryologic defect that disturbs the development of the aortic wall and the aortic valve.

The EMT dysregulation hypothesis seems to match with the fact that R-L BAV originates from an abnormal distribution of CNC cells [19]. It has been shown that the CNC is essential for proper OFT septation and formation of the arterial trunks and semilunar valves. CNC cells form the so-called aortic-pulmonary septation complex, which shows an inverted U-shape and consists of a mesenchymal septum and two cell condensations that spread proximally along the CRs (Fig. 1a). Formation of this complex takes place disto-proximally through the CRs and is responsible for the fusion event [1,2,7,42,43]. We propose that during the sequential formation of the aortic-pulmonary septation complex, CNC cells induce EMT of contacting endocardial cells located close to the forming CS (complete EMT), leading to a disto-proximal sequential fusion of CRs. The altered migration of CNC cells along the OFT would entail an extension of the endocardial territory affected by the EMT induction, increasing the number of activated endocardial cells that results in an excessive fusion of the CRs. This mechanism is consistent with the study of Plein et al., 2015 [43], who demonstrated that migration of CNC and EMT are interconnected morphogenetic processes. They found that a specific population of CNC cells induces EMT of the lining endocardium in early stages of cardiogenesis, promoting the cellularization of the endocardial cushions. Endocardial cells in turn regulate CNC migration through the endocardial cushion mesenchyme to form the two cell condensations of the aortic-pulmonary septation complex. The regulation of this event implies the establishment of a crosstalk between the CNC and the endocardium. Class 3 semaphorin (SEMA3C), an autocrine factor secreted by CNC cells, and neuropilin 1 (NRP1), a SEMA3C receptor located in the endocardial cell membrane, are key molecular elements involved in this crosstalk system [43]. Considering that CNC-derived SEMA3C and endocardial NRP1 are required for both, endocardial EMT and CNC migration, it seems plausible that this intercellular communication system is also responsible for the whole process of conotruncal septation, including fusion of CRs by EMT and formation of the CS by migrating CNC cells from the cell condensations. Indeed, this pathway has been demonstrated to be required specifically for the migration of CNC cells to the fusion area of CRs [43]. Another molecular pathway possibly involved is DLL4-JAG1-NOTCH1, which is required in the endocardium to ensure proper valve morphogenesis and cardiac septation through regulation of EMT, cell migration, proliferation, and extracellular matrix remodeling [44, 45]. The alteration of this signaling pathway gives rise to different types of BAV, including R-L BAV [44]. Given that NOTCH signaling is necessary for CNC patterning in the OFT [46], MacGrogan et al. (2016) [44] proposed that the endocardium is likely responsible for the regulation of the CNC migration through the NOTCH pathway. It would be of interest to delve into whether any of the signaling pathways described are also involved in the EMT process underlying the fusion of CRs.

Animals

Animals were handled in accordance with the European guidelines (Directive 2010/63/EU) and Spanish regulations (R.D. 53/2013; B.O.E. 01.02.2013) for the protection of experimental animals. The protocol was approved by the Ethics Committee of Animal Experiments of the University of Málaga (CEUMA; Ethics authorization number: 25/08/2021/118).

A total of 83 hamster embryos aged between 11 and 12 days postcoitum (ED11-ED12) were analyzed. Forty of these embryos belonged to an inbred strain with around 40% incidence of non-syndromic isolated BAV type A (R/L fusion type, including fused BAV, 2 sinus BAV and partial fusion BAV, according to [16]). The characteristics of this strain (T strain) have been previously reported [8, 12, 19, 21–24]. The remaining 43 embryos belonged to an outbred strain (RjHan:AURA; Janvier, France; control strain) with null incidence of BAV, which has been previously used as a control for the T strain [24].

The embryonic stage was determined by counting the days from the moment of the copula, which was directly visualized and designated as stage ED0. The embryos were harvested at the required stage by laparotomy and hysterectomy from the pregnant female and sacrificed by immersion in ice-cold phosphate buffer saline (PBS) 0.02M.

Histochemistry

ED11 and ED12 embryos (T strain: n = 40; control strain: n = 25) were fixed by immersion in 4% paraformaldehyde diluted in PBS overnight. Then, the embryos were dehydrated in a graded series of ethanol and embedded in paraffin (Histosec, Merck, Germany). The embryos were orientated to obtain eight µm thick transversal or frontal serial sections of the heart in a Leitz 1512 microtome. Some of these sections were dewaxed, hydrated, and stained with Delafield haematoxylin-eosin. Samples were visualized and photographed with a Leica DMSL light microscope equipped with a Leica DFC 500 camera or in an Olympus VS120 automated slide scanner.

TUNEL Assay

The terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling (TUNEL) assay was used to detect apoptosis in a total of 15 ED11 and ED12 embryos (T strain: n = 7; control strain: n = 8). A commercial kit (Roche, Switzerland) was employed following the manufacturer’s instructions. Omission of the transferase enzyme was used as a negative control. Finally, the sections were stained with 4’,6-diamidino-2-phenylindole dihydrochloride (DAPI, Sigma-Aldrich, USA) diluted 1/1000 in PBS to verify the nuclear location of the TUNEL signals. The sections were observed and photographed with a Leica Stellaris confocal microscope.

Single and double immunofluorescence

Sections from 32 ED11 and ED12 embryos belonging to the T (n = 15) and the control (n = 17) strains were deparaffinated and washed in Tris-PBS (TPBS, pH 7.8). Non-specific binding sites were saturated for 1h with 10% bovine serum, 1% bovine serum albumin and 0.1% Triton X-100 in TPBS (BST). The Avidin-Biotin blocking kit from Vector was used to block endogenous biotin. Then, sections were incubated with the primary antibody (Anti-Caspase 3, Anti-CD34, Anti-VE-Cadherin; details are provided in Supplemental Table 1) at 4°C overnight, rinsed in TPBS and incubated with an anti-rabbit IgG biotin conjugate (Sigma-Aldrich, USA, REF: B8895) for 1.5h at RT. After washing with TPBS, sections were incubated with FITC- or Cy3-conjugated extravidin (Sigma-Aldrich, USA) diluted 1/250 in TPBS for 1h at RT. Negative control samples were incubated with BST in the absence of the primary antibody. For CD34 and VE-Cadherin immunofluorescence, sodium citrate buffer (10mM Sodium Citrate, 0.05% Tween 20, pH 6.0) and EDTA buffer (1mM EDTA, 0.05% Tween 20, pH 8.0) were respectively required for antigen retrieval (boiling in microwave for 15 min), prior to blocking the non-specific bindings. After retrieval, samples were cooled at RT and rinsed with TPBS. For double immunofluorescence, Cy3-conjugated anti-α-actin smooth muscle (mouse monoclonal, Sigma-Aldrich, USA, REF: C6198) diluted 1/300 in BST was incubated for 1.5h at RT. Nuclei were counterstained with DAPI diluted 1:1000 in TPBS. Slides were observed and photographed with a Leica Stellaris confocal microscope.

Semithin sections

Eighteen ED11 embryos (T strain: n = 9; control strain: n = 9) were immersion fixed with a mix of 1% paraformaldehyde, 2% glutaraldehyde and 0.05M cacodylate buffer for 1h, washed in cacodylate buffer for 30 min and post-fixed by immersion in 1% osmium tetroxide diluted in 0.05M cacodylate buffer for 1.5h. Then, samples were rinsed in distilled water several times until complete removal of fixative. After that, they were dehydrated in graded acetone concentrations, washed in propylene oxide (Sigma-Aldrich, USA), and embedded in Araldite epoxy resin (Electron Microscopy Sciences, Hatfield, PA, US). Embryos were oriented to provide frontal sections of the cardiac OFT. One µm serial sections were obtained in a Leica UC7/FC7 ultramicrotome, stained with Toluidine Blue-Basic Fuchsin (Sigma-Aldrich, USA) and photographed in an Olympus VS120 automated slide scanner.

1. Anderson, R. H., et al. Normal and abnormal development of the intrapericardial arterial trunks in humans and mice. Cardiovasc. Res. 95, 108-115 (2012).

2. Anderson, R. H., Webb, S., Brown, N. A., Lamers, W. & Moorman, A. Development of the heart: (3) formation of the ventricular outflow tracts, arterial valves, and intrapericardial arterial trunks. Heart 89, 1110-1118 (2003).

3. Gittenberger-de Groot, A. C., Bartelings, M. M., DeRuiter, M. C. & Poelmann, R. E. Basics of cardiac development for the understanding of congenital heart malformations. Pediatr. Res. 57, 169-176 (2005).

4. Henderson, D. J., Eley, L. & Chaudhry. B. New concepts in the development and malformation of the arterial valves. J. Cardiovasc. Dev. Dis. 7, 38 (2020).

5. Kirby, M. L. & Waldo, K. Molecular embryogenesis of the heart. Pediatr. Devel. Pathol. 5, 516-543 (2002).

6. Spicer, D. E., Bridgeman, J. M., Brown, N. A., Mohun, T. J. & Anderson, R. H. The anatomy and development of the cardiac valves. Cardiol. Young 24, 1008-1022 (2014).

7. Waldo, K. L., Lo, C. W. & Kirby, M. L. Connexin 43 expression reflects neural crest patterns during cardiovascular development. Dev. Biol. 208, 307-323 (1999).

8. Soto-Navarrete, M. T., López-Unzu, M. A., Durán, A. C. & Fernández, B. Embryonic development of bicuspid aortic valves. Prog. Cardiovasc. Dis. 63, 407-418 (2020).

9. Peterson, J. C. et al. Bicuspid aortic valve formation: Nos3 mutation leads to abnormal lineage patterning of neural crest cells and the second heart field. Dis. Mod. Mech. 11, dmm034637 (2018).

10. Hinton, R. B. & Yutzey, K, E, Heart valve structure and function in development and disease. Annu. Rev. Physiol. 73, 29-46 (2011).

11. Hurlé, J. M., Colvee, E. & Blanco, A. M. Development of mouse semilunar valves. Anat. Embryol. 160, 83-91 (1980).

12. Sans-Coma, V. et al. Fusion of valve cushions as a key factor in the formation of congenital bicuspid aortic valves in Syrian hamsters. Anat. Rec. 244, 490–498 (1996).

13. Fedak, P. W. M. et al. Clinical and pathophysiological implications of a bicuspid aortic valve. Circulation 106, 900–4 (2002).

14. Masri, A., Svensson, L. G., Griffin, B. P. & Desai, M. Y. Contemporary natural history of bicuspid aortic valve disease: a systematic review. Heart 103, 1323-1330 (2017).

15. Roberts, W. C. The congenitally bicuspid aortic valve. A study of 85 autopsy cases. Am. J. Cardiol. 26, 72-83 (1970).

16. Michelena, H. I. et al. International consensus statement on nomenclature and classification of the congenital bicuspid aortic valve and its aortopathy, for clinical, surgical, interventional and research purposes. Eur. J. Cardiothorac. Surg. 60, 448–476 (2021).

17. Basso, C. et a An echocardiographic survey of primary school children for bicuspid aortic valve. Am. J. Cardiol. 93, 661-663 (2004).

18. Russo CF, et al. Is aortic wall degeneration related to bicuspid aortic valve anatomy in patients with valvular disease? J. Thorac. Cardiovasc. Surg. 136, 937-942 (2008).

19. Fernández, B. et al. Bicuspid aortic valves with different spatial orientations of the leaflets are distinct etiological entities. J. Am. Coll. Cardiol. 54, 2312-2318 (2009).

20. Waldo, K. L., Miyagawa-Tomita, S., Kumiski, D. & Kirby, M. L.Cardiac neural crest cells provide new insight into septation of the outflow tract: aortic sac to ventricular septal closure. Dev. Biol. 196, 129-144 (1998).

21. Fernández, B. et al. Bicuspid aortic valve in 2 model species and review of the literature. Vet. Pathol. 57, 321-331 (2020).

22. Sans-Coma, V. et al. Evidence for a quantitative genetic influence on the formation of aortic valves with two leaflets in the Syrian hamster. Cardiol. Young 3, 132–140 (1993).

23. Sans-Coma, V. et al. Genetically alike Syrian hamsters display both bifoliate and trifoliate aortic valves: Aortic valves in isogenic hamsters. J. Anat. 220, 92–101 (2012).

24. Soto-Navarrete, M. T. et al. Experimental evidence of the genetic hypothesis on the etiology of bicuspid aortic valve aortopathy in the hamster model. Front. Cardiovasc. Med. 9, 928362; 10.3389/fcvm.2022.928362 (2022).

25. Cripe, L., Andelfinger, G., Martin, L. J., Shooner, K. & Benson, D. W. Bicuspid aortic valve is heritable. J. Am. Coll. Cardiol. 44, 138–43 (2004).

26. Ray, H. J. & Niswander, L. Mechanisms of tissue fusion during development. Development 139, 1701-1711 (2012).

27. Houyel, L. & Meilhac, S. M. Heart development and congenital structural heart defects. Annu. Rev. Genom. Hum. G. 22, 257-284 (2021).

28. Punovuori, K., Malaguti, M. & Lowell, S. Cadherins in early neural development. Cell. Mol. Life. Sci. 78, 4435-4450 (2021).29. Martınez-Alvarez, C. et al. Medial edge epithelial cell fate during palatal fusion. Dev. Biol. 220, 343-357 (2000).

30. Kovacic, J. C., Mercader, N., Torres, M., Boehm, M., Fuster, V. Epithelial-to-mesenchymal and endothelial-to-mesenchymal transition: from cardiovascular development to disease. Circulation 125, 1795-1808 (2012).

31. Dickman, E. D., Rogers, R. & Conway, S. J. Abnormal skeletogenesis occurs coincident with increased apoptosis in the Splotch (Sp2H) mutant: putative roles for Pax3 and PDGFRα in rib patterning. Anat. Rec. 255, 353-361 (1999).

32. Poelmann, R. E., Mikawa, T. & Gittenberger-de Groot, A. C. Neural crest cells in outflow tract septation of the embryonic chicken heart: differentiation and apoptosis. Dev. Dyn. 212, 373-384 (1998).

33. Kurn, H. & Daly, D. T. Histology, Epithelial Cell. In: StatPearls [Internet]. StatPearls Publishing, Treasure Island (FL) https://www.ncbi.nlm.nih.gov/books/NBK559063/ (2023).

34. Nakaya, Y. & Sheng, G. Epithelial to mesenchymal transition during gastrulation: an embryological view. Dev. Growth. Differ. 50, 755-766 (2008).

35. Hay, E. D. Organization and fine structure of epithelium and mesenchyme in the developing chick embryo. In Epithelial-Mesenchymal Interactions; 18th Hahnemann Symposium, 1968. Williams & Wilkins (1968).

36. Abdulla, T., Luna-Zurita, L., De La Pompa, J. L., Schleich, J. M. & Summers, R. Epithelial to mesenchymal transition—The roles of cell morphology, labile adhesion and junctional coupling. Comput. Meth. Prog. Bio. 111, 435-446 (2013).

37. Yang, J. et al. Guidelines and definitions for research on epithelial–mesenchymal transition. Nat. Rev. Mol. Cell. Biol. 21, 341–352 (2020).

38. Duband, J. L. Diversity in the molecular and cellular strategies of epithelium-to-mesenchyme transitions: Insights from the neural crest. Cell Adhes. Migr. 4, 458-482 (2010).

39. Acloque, H., Adams, M. S., Fishwick, K., Bronner-Fraser, M. & Nieto, M. A. Epithelial-mesenchymal transitions: the importance of changing cell state in development and disease. J. Clin. Invest. 119, 1438-49 (2009).

40. Schiffmacher, A. T., Adomako-Ankomah, A., Xie, V. & Taneyhill, L. A. Cadherin-6B proteolytic N-terminal fragments promote chick cranial neural crest cell delamination by regulating extracellular matrix degradation. Dev. Biol. 444, S237-S251 (2018).

41. Freiholtz, D. et al. Bicuspid aortic valve aortopathy is characterized by embryonic epithelial to mesenchymal transition and endothelial instability. J. Mol. Med. 101, 801-811 (2023).

42. Jain, R. et al. Cardiac neural crest orchestrates remodeling and functional maturation of mouse semilunar valves. J. Clin. Invest. 121, 422–430 (2011).

43. Plein, A., Fantin, A. & Ruhrberg, C. Neural crest cells in cardiovascular development. Curr. Top. Dev. Biol. 111, 183–200 (2015).

44. MacGrogan, D. et al. Sequential ligand-dependent notch signaling activation regulates valve primordium formation and morphogenesis. Circ. Res. 118, 1480–1497 (2016).

45. Torregrosa-Carrión, R. et al. NOTCH Activation Promotes Valve Formation by Regulating the Endocardial Secretome. Mol. Cell. Proteomics. 18, 1782-1795 (2019).

46. High, F. A. et al. Murine Jagged1/Notch signaling in the second heart field orchestrates Fgf8 expression and tissue-tissue interactions during outflow tract development. J. Clin. Invest. 119, 1986–1996 (2009).

Acknowledgements

We acknowledge Prof. Valentín Sans Coma for his valuable suggestions, Luis Vida for technical assistance, Francisco David Navas for assistance in confocal microscopy, and Cristina Lucena for assistance in ultramicrotome sectioning.

Authors contribution

MTS-N, BP-V, ML-U, LM-C: acquisition of data; All authors: analysis and interpretation of data; MTS-N, BF: conception and design of experimental strategy; MTS-N, BF: original draft preparation; All authors: review and editing of the manuscript.

Funding statement

This work was supported by Junta de Andalucía, Consejería de Salud y Consumo (PI-0530-2019), Consejería de Economía y Conocimiento (UMA20-FEDERJA-041), Consejería de Universidad, Investigación e Innovación (PROYEXCEL_01009), Ministerio de Ciencia e Innovación (grants CGL2017-85090-P, fellowship PRE2018-083176 to MTS-N, FJC2021 047055-I to MAL-U), II Plan Propio de Investigación, Transferencia y Divulgación Científica de la Universidad de Málaga (Contrato Puente to MTS-N), and FEDER funds.

Data availability

The authors declare that the data supporting the findings of this study are available within the article, and in the supplementary information files.

Competing Interests Statement

All authors declare no conflicts of interest.

Supplemental Table 1. Antibodies used for immunofluorescence.

Antibody	Host species	Epitope recognition	Working dilution	Commercial info
Anti-Caspase 3	Rabbit	Polyclonal	1/500	Sigma-Aldrich, REF: C8487
Anti-CD34	Rabbit	Polyclonal	1/500	Abcam, REF: ab81289
Anti-VE-Cadherin	Rabbit	Polyclonal	1/500	Abcam, REF: ab33168

No competing interests reported.

SuppFigure.tif
Supplemental Figure 1. Transversal histological section (A) immunostained with CD34 (green) and α-actin (red) with nuclei stained with DAPI (blue), and frontal semithin sections (B) stained with Toluidine Blue-Basic Fuchsin of the OFT of an ED11 hamster embryo. Yellow lines indicate the corresponding position of the frontal semithin sections in the transversal view. Black lines delimited the aortic-pulmonary septation complex (AP). The regions with morphological signs of EMT are delimited by blue lines. No signs of EMT were found in the margins of the CRs, whereas the central portion of the CRs contained a higher concentration of cells undergoing EMT. Ao: aorta artery; CR: conotruncal ridge; Pul: pulmonary artery.

Download PDF

Reviews received at journal
20 Nov, 2024
Reviewers agreed at journal
09 Nov, 2024
Reviewers agreed at journal
13 Sep, 2024
Reviews received at journal
12 Sep, 2024
Reviewers agreed at journal
11 Sep, 2024
Reviewers invited by journal
11 Sep, 2024
Editor assigned by journal
11 Sep, 2024
Editor invited by journal
05 Sep, 2024
Submission checks completed at journal
05 Sep, 2024
First submitted to journal
29 Aug, 2024

You are reading this latest preprint version

Endocardial-mesenchymal transition underlies fusion of the conotruncal ridges during normal and bicuspid aortic valve development

Status:

Version 1

Abstract

Figures

Introduction

Results

Normal conotruncal septation process in the Syrian hamster

Mechanism of fusion of the conotruncal ridges

Fusion of the conotruncal ridges during bicuspid aortic valve formation

Discussion

Methods

Animals

Histochemistry

TUNEL Assay

Single and double immunofluorescence

Semithin sections

References

Declarations

Supplementary Table

Additional Declarations

Supplementary Files

Status:

Version 1