Over-activation of BMP signaling in neural crest cells precipitates heart outflow tract septation

Background

The mammalian outflow tract (OFT) is a transient embryonic structure that gives rise to the roots of the aorta and the pulmonary artery and plays a crucial role in adjusting the pressure of blood coming from the heart. It is composed of several cell types, such as the second heart field (SHF) mesodermal cells, endothelial cells (EC), and cardiac neural crest cells (cNCC). Their cross-talk is essential to form the correct structure of the OFT (1). Among the several cell types involved, the cNCC are known to play a predominant role in OFT development. Due to their invasion of the cardiac cushion and the rupture of the endocardium, they are eventually responsible for the formation of the septum that divides the two arteries (2).

The BMP pathway consists of different secreted cytokines and transmembrane receptors that are able to phosphorylate intracellular transcription factors (Smads). In mouse, different loss-of-function mutations of BMP-related genes lead to failure in OFT septation (3). Nevertheless, the molecular mechanism with which BMP is regulating the division of the arteries has not been fully understood.

Darrigrand et al in this study show an alternative approach to study the role of BMP in cNCC, taking advantage of the tissue-specific mutation of a perinuclear phosphatase, Dullard. This protein is able to dephosphorylate Smad1/5/8, leading to a down-regulation of the downstream transcriptional cascade (4). Mice carrying Dullard mutation under the control of Pax3 or Wnt1 promoters (early cNCC markers) show an upregulation of BMP signaling and a severe acceleration of the OFT septation.

Important results

The authors, in order to characterize the activation of BMP in the OFT, show a gradient of phosphorylated (thus, active) Smad1/5/8 decreasing towards the distal end of the OFT axis. Interestingly, Dullard deletion in the cNCC does not prevent the persistence of this gradient, although it increases the phosphorylation of Smad1/5/8 along the OFT wall, as expected.  Given the homogeneous expression of the ligands along all the myocardium, Darrigrand and colleagues speculate the presence of a fine-tuned intracellular inhibition of the signaling rather than a gradient of the ligands.

Despite the smaller diameter of the OFT in Dullard^fl/fl embryos, no difference in cNCC number was observed, suggesting higher compaction of these cells. Indeed the authors observed an increased condensation of cNCC close to the endocardium, compared to the wild types.

Subsequently, Darrigrand and colleagues performed a detailed morphological analysis of the OFT of wild-type and mutant embryos taking advantage of light-sheet microscopy (Fig. 1 A, B). Thanks to this 3D rendering, they could observe an obstruction of the pulmonary artery (Fig. 1B) in the mutants, probably the cause of death of these embryos after E12.5 due to impaired blood dynamics.

 

 

Furthermore, in order to perform an unbiased search for candidate genes responsible for this premature septation of the OFT, the authors performed a single cell RT-PCR for 44 genes involved in epithelial-to-mesenchymal transition (EMT) on sorted cNCC and EC from the OFT (Fig 1, C). An unsupervised hierarchical clustering highlighted 6 populations and showed that mutant cells properly maintain their endocardial and epicardial fate, while they segregate in distinct subpopulations of cNCCs.  In fact, almost no mutant cells were found in the population characterized by a high expression of mature SMC markers (Myh11, Acta2). Conversely, Dullard^fl/fl cells were enriched in the two populations with low expression of mesenchymal genes (Snai2, Twist1) and high expression of epithelial markers (Cdh5). This shows that Dullard deletion affects the transcriptome of cNCC, shifting their fate more to an epithelial-like state, in line with their increased condensation.

Among the genes highly expressed in mutant cells, a remarkable one was Sema3c, a gene involved in cell migration and convergence (5). This result was confirmed by in situ hybridizations, showing a gradient along the OFT very similar to the BMP activity and colocalizing with areas of high cNCC compaction.

What I like about this preprint

The embryonic development of the OFT has been understudied for many years. This structure, in fact, being at the transition between the ventricle and a big artery, stayed for years in the shadow even though defects in its development give rise to 30% of congenital heart diseases.

In addition, I appreciate that the authors wanted to shed light on a pathway like the BMP that is involved in many different processes during cardiac development. This pathway, due to its context- and tissue-dependency, hasn’t been fully understood yet and many more in vivo studies are required to fully characterize its important roles.

Moreover, I like how the authors took advantage of 3D imaging techniques together with a preliminary analysis of single-cell transcriptomics. Given the many cell types necessary for the correct development of the OFT, this in vivo analyses fills an important gap in our understanding about how the OFT is formed.

Open questions

1. Considering the often intertwined roles of the TGF-beta and BMP pathways, have you considered a possible alteration of the signaling downstream of TGF-beta in Dullard mutants? A paper from Hayata and colleagues (6) shows that Dullard is able to affect the phosphorylation state of Smad2/3 as well. It would be nice to know how these two pathways are interacting in cNCC and if the phenotype you observe has any contribution from the TGF-beta pathway.
2. EMT is a very important process that has been proved to be regulated by TGF-beta and BMP pathways. Do you think that, in the context of your study, the EMT process in the cardiac cushions is inhibited? Or is it then reversed after the early formation of the cushions?
3. Do you think that BMP signaling through the Smads is directly regulating Sema3c expression or is it relying on an indirect effect? Is there any ChIP data available to show Sema3c as a possible direct target?

References

1. Neeb Z, Lajiness JD, Bolanis E, Conway SJ. Cardiac outflow tract anomalies. Wiley Interdiscip Rev Dev Biol. 2013 Jul;2(4):499-530
2. Keyte A, Hutson MR. The Neural Crest in Cardiac Congenital Anomalies. Differ Res Biol Divers. 2012 Jul;84(1):25–40
3. Jia Q, McDill BW, Li S-Z, Deng C, Chang C-P, Chen F. Smad signaling in the neural crest regulates cardiac outflow tract remodeling through cell autonomous and non-cell autonomous effects. Dev Biol. 2007 Nov 1;311(1):172–84.
4. Sakaguchi M, Sharmin S, Taguchi A, Ohmori T, Fujimura S, Abe T, et al. The phosphatase Dullard negatively regulates BMP signalling and is essential for nephron maintenance after birth. Nat Commun. 2013 Jan 29;4:1398.
5. Plein A, Calmont A, Fantin A, Denti L, Anderson NA, Scambler PJ, et al. Neural crest–derived SEMA3C activates endothelial NRP1 for cardiac outflow tract septation. J Clin Invest. 2015 Jul 1;125(7):2661-76
6. Hayata T, Ezura Y, Asashima M, Nishinakamura R, Noda M. Dullard/Ctdnep1 regulates endochondral ossification via suppression of TGF-β J Bone Miner Res.2015 Feb;30(2):318-29

Tags: bmp, development, heart, light sheet, neural crest, outflow tract, single cell

doi: https://doi.org/10.1242/prelights.9679

Read preprint (No Ratings Yet)