Search

During the past years considerable progress has been made in understanding the generation of cell diversity in the neural crest (NC). Sympathoadrenal (SA) cells constitute a major lineage among NC derivatives; they give rise to sympathetic neurons, neuroendocrine chromaffin cells, and the intermediate small intensely fluorescent (SIF) cells. The classic perception of how this diversification is achieved implies that (i) there is a common progenitor cell for sympathetic neurons and chromaffin cells, (ii) NC cells are instructed to a SA cell fate by signals derived from the wall of the dorsal aorta, especially bone morphogenetic proteins (BMP), and (iii) the local environments of secondary sympathetic ganglia and adrenal gland, respectively, are crucial for inducing differentiation of SA cells into sympathetic neurons and adrenal chromaffin cells. However, recent studies have suggested that the adrenal cortex is dispensable for the acquisition of a chromaffin cell fate. This review summarizes the current understanding of the development of SA cells. It covers the specification of SA cells from multipotent NC crest cells, the role of transcription factors during their development, the classic model of their subsequent diversification as well as alternative views for explaining the generation of endocrine versus neuronal SA derivatives.

The endothelin/endothelin receptor system plays a critical role in the differentiation and terminal migration of particular neural crest cell subpopulations. Targeted deletion of the G-protein-coupled endothelin receptors ET(A) and ET(B) was shown to result in characteristic developmental defects of derivatives of cephalic and cardiac neural crest and of neural crest-derived melanocytes and enteric neurons, respectively. Since both endothelin receptors are coupled to G-proteins of the G(q)/G(11)- and G(12)/G(13)-families, we generated mouse lines lacking Galpha(q)/Galpha(11) or Galpha(12)/Galpha(13) in neural crest cells to study their roles in neural crest development. Mice lacking Galpha(q)/Galpha(11) in a neural crest cell-specific manner had craniofacial defects similar to those observed in mice lacking the ET(A) receptor or endothelin-1 (ET-1). However, in contrast to ET-1/ET(A) mutant animals, cardiac outflow tract morphology was intact. Surprisingly, neither Galpha(q)/Galpha(11)- nor Galpha(12)/Galpha(13)-deficient mice showed developmental defects seen in animals lacking either the ET(B) receptor or its ligand endothelin-3 (ET-3). Interestingly, Galpha(12)/Galpha(13) deficiency in neural crest cell-derived cardiac cells resulted in characteristic cardiac malformations. Our data show that G(q)/G(11)- but not G(12)/G(13)-mediated signaling processes mediate ET-1/ET(A)-dependent development of the cephalic neural crest. In contrast, ET-3/ET(B)-mediated development of neural crest-derived melanocytes and enteric neurons appears to involve G-proteins different from G(q)/G(11)/G(12)/G(13).

Phox2B, a homeodomain transcription factor closely related to Phox2A, is expressed in peripheral and central noradrenergic neurons. In neural crest (NC) derivatives Phox2B is restricted to sympathetic and parasympathetic ganglia, enteric neurons, and adrenal and extraadrenal chromaffin cells. Similar to MASH-1, Phox2B has been implicated in synchronizing pan-neuronal and catecholaminergic phenotype-specific aspects of neurogenesis. The role of Phox2B for the differentiation of the neuroendocrine NC derivatives, the adrenal medullary chromaffin cells, has not been explored. We have previously reported that in MASH-1-deficient mice most chromaffin cells are arrested at the early neuroblast stage and lack catecholaminergic differentiation. We show now that in Phox2B knockout/lacZ knockin mice the maturation of presumptive chromaffin cells is arrested at an even earlier stage of development. The cells lack the catecholaminergic marker enzyme TH and fail to form a centrally located medulla. In contrast to MASH-1 (−/−) mice they do not express dHand, Phox2A, c-ret, neurofilament, neuron-specific tubulin, and NCAM and appear ultrastructurally more immature. Many of these cells die by apoptosis. Despite the complete lack of differentiation, few lacZ-positive adrenal cells can still be found at E16.5. We conclude that Phox2B regulates very early events in the differentiation of adrenal chromaffin cells distinct to steps, which essentially require MASH-1.

The development of sympathetic neurons and chromaffin cells is differentially controlled at distinct stages by various extrinsic and intrinsic signals. Here we use conditional deletion of Dicer1 in neural crest cells and noradrenergic neuroblasts to identify stage specific functions in sympathoadrenal lineages. Conditional Dicer1 knockout in neural crest cells of Dicer1Wnt1Cre mice results in a rapid reduction in the size of developing sympathetic ganglia and adrenal medulla. In contrast, Dicer1 elimination in noradrenergic neuroblasts of Dicer1DbhiCre animals affects sympathetic neuron survival starting at late embryonic stages and chromaffin cells persist at least until postnatal week 1. A differential function of Dicer1 signaling for the development of embryonic noradrenergic and cholinergic sympathetic neurons is demonstrated by the selective increase in the expression of Tlx3 and the cholinergic marker genes VAChT and ChAT at E16.5. The number of Dbh, Th and TrkA expressing noradrenergic neurons is strongly decreased in Dicer1-deficient sympathetic ganglia at birth, whereas Tlx3+/ Ret+ cholinergic neurons cells are spared from cell death. The postnatal death of chromaffin cells is preceded by the loss of Ascl1, mir-375 and Pnmt and an increase in the markers Ret and NF-M, which suggests that Dicer1 is required for the maintenance of chromaffin cell differentiation and survival. Taken together, these findings demonstrate distinct stage and lineage specific functions of Dicer1 signaling in differentiation and survival of sympathetic neurons and adrenal chromaffin cells.

The endothelin/endothelin receptor system plays a critical role in the differentiation and terminal migration of particular neural crest cell subpopulations. Targeted deletion of the G-protein-coupled endothelin receptors ET(A) and ET(B) was shown to result in characteristic developmental defects of derivatives of cephalic and cardiac neural crest and of neural crest-derived melanocytes and enteric neurons, respectively. Since both endothelin receptors are coupled to G-proteins of the G(q)/G(11)- and G(12)/G(13)-families, we generated mouse lines lacking Galpha(q)/Galpha(11) or Galpha(12)/Galpha(13) in neural crest cells to study their roles in neural crest development. Mice lacking Galpha(q)/Galpha(11) in a neural crest cell-specific manner had craniofacial defects similar to those observed in mice lacking the ET(A) receptor or endothelin-1 (ET-1). However, in contrast to ET-1/ET(A) mutant animals, cardiac outflow tract morphology was intact. Surprisingly, neither Galpha(q)/Galpha(11)- nor Galpha(12)/Galpha(13)-deficient mice showed developmental defects seen in animals lacking either the ET(B) receptor or its ligand endothelin-3 (ET-3). Interestingly, Galpha(12)/Galpha(13) deficiency in neural crest cell-derived cardiac cells resulted in characteristic cardiac malformations. Our data show that G(q)/G(11)- but not G(12)/G(13)-mediated signaling processes mediate ET-1/ET(A)-dependent development of the cephalic neural crest. In contrast, ET-3/ET(B)-mediated development of neural crest-derived melanocytes and enteric neurons appears to involve G-proteins different from G(q)/G(11)/G(12)/G(13).