Multiple actions of stem cell factor in neural crest cell differentiation in vitro.

The neural crest is a transient tissue of the vertebrate embryo that gives rise to most primary sensory neurons and pigment cells in the adult organism, among other cell types and tissues. Many neural crest cells are pluripotent in the sense that their progeny can generate more than one phenotype. The presence of pluripotent neural crest cell-derived cells at sites of terminal differentiation suggests that location-specific cues from the embryonic environment, such as growth factors, are involved in directing their survival, proliferation, and cell type specification. We have therefore examined the influences of one pertinent growth factor, stem cell factor (SCF), on neural crest cell development by in vitro colony assay in a serum-free culture medium. SCF showed three major effects. (1) SCF is trophic for early neural crest cells, that is, either pluripotent cells and/or their more mature progeny. This effect occurs only if SCF is present throughout the culture period, and it is not observed when a neurotrophin is present in addition to SCF. (2) More colonies contain sensory neuron precursors in the presence of SCF. This effect is neutralized by NGF and neurotrophin-3 (NT-3), but not by brain-derived neurotrophic factor (BDNF). (3) The combination of SCF and any neurotrophin tested (NGF, BDNF, NT-3) is trophic for melanogenic cells, whereas SCF alone does not detectably affect melanogenesis. This suggests either that both types of factor are required for melanotrophic action or that melanogenic cells become dependent on neurotrophins after exposure to SCF. Our observation that SCF is required during the first half of the culture period only, and NGF during the second half only, indicates the latter possibility. Whereas coat color changes in the mouse mutants W (c-kit defect) and Steel (SCF defect) and several in vivo and in vitro studies by other investigators have shown previously that SCF is melanotrophic, they also indicated the requirement of an additional factor, or factors, in melanogenesis. Our data suggest that SCF affects neural crest cell development at multiple levels and that survival of melanogenic cells is mediated by a combination of SCF and a neurotropin, rather than by SCF alone.

Effect of the Steel gene product on melanogenesis in avian neural crest cell cultures.

Mutations at the Steel (Sl) and dominant white spotting (W) loci affect three embryonic lineages: primordial germ cells, hemopoietic stem cells and neural-crest-derived melanocytes. The gene products of these loci are a peptide growth factor, called here stem cell factor (SCF), and its tyrosine kinase receptor, the proto-oncogene c-kit. We have studied how chicken recombinant SCF affects the development of melanocytes from quail neural crest cells in secondary culture under defined conditions. We observed that the total number of neural crest cells, of melanocytes and of their precursors was higher in the presence than in the absence of SCF. Labelling with bromodeoxyuridine showed that SCF had a modest and transient mitogenic effect on the neural crest population. SCF also enhanced the differentiation rate of melanocyte precursors, recognized by the "melanocyte early marker" monoclonal antibody (MelEM MAb), and of melanocytes, since the proportion of both subpopulations significantly increased in the presence of SCF. Finally, SCF increased the survival of the neural crest population since in its presence the total number of cells remained stable while it gradually declined in control cultures. Our results support the notion that SCF sustains the survival of the neural crest population and stimulates the rate of the melanogenic differentiation process.

Stem cell factor is a neurotrophic factor for neural crest-derived chick sensory neurons.

We have found that stem cell factor (SCF) selectively enhances the survival of cultured embryonic chick dorsal root ganglia (DRG) neurons. Neurons grown in the presence of SCF expressed both neurofilament 150 kDa subunit and calcitonin-gene related peptide. SCF does not, however, enhance the survival of parasympathetic, placode-derived sensory or sympathetic neurons in culture. Combining SCF with brain-derived neurotrophic factor or neurotrophin-3, but not with NGF, maintains more neurons than either factor alone, suggesting that these factors have partially overlapping activities. SCF preferentially rescues small neurons from the DRG. Labeling studies with bromodeoxyuridine indicate that the neurons sustained by SCF are not differentiating from a dividing progenitor.

Evidence that enteric neurons may derive from the sympathoadrenal lineage.

The first neurons that differentiate in the embryonic foregut of mammals transiently express catecholamine biosynthetic enzymes and accumulate catecholamine. Since this transmitter is found predominantly in cells of the sympatho-adrenal (SA) lineage, it has been suggested that enteric and sympathetic neurons may derive from the same progenitor. Enteric neurons would then lose the catecholamine phenotype during further development, as the two lineages diverge. We have further investigated this possibility using the SA1 monoclonal antibody that binds selectively to SA progenitor cells in the embryonic rat. We find that SA1 binds to the tyrosine hydroxylase+, neurofilament+, and SCG10+ cells of the Embryonic Day 14.5 (E14.5) rat foregut. We also find that a marker for later neuronal differentiation in the SA lineage, B2, also appears in the myenteric plexus concomitant with the loss of SA1 staining. Thus, at least some enteric neuronal precursors may exhibit the SA1----B2 antigenic switch previously observed in developing sympathetic neurons at E14.5. SA1 staining in the foregut partially overlaps with staining for neuropeptide Y, vasoactive intestinal polypeptide, and serotonin. These results support the hypothesis that enteric and sympathetic neurons derive from a common progenitor and that as the markers for the SA lineage are down-regulated, the many types of enteric neurons begin to differentiate.

Antibody markers identify a common progenitor to sympathetic neurons and chromaffin cells in vivo and reveal the timing of commitment to neuronal differentiation in the sympathoadrenal lineage.

Using specific antibody markers and double-label immunofluorescence microscopy, we have followed the fate of progenitor cells in the sympathoadrenal (SA) sublineage of the neural crest in developing rat embryos. Such progenitors are first recognizable in the primordial sympathetic ganglia at embryonic day 11.5 (E11.5), when they express tyrosine hydroxylase. At this stage, the progenitors also coexpress neuronal markers such as SCG 10 and neurofilament, together with a series of chromaffin cell markers called SA 1-5 (Carnhan and Patterson, 1991 a). The observation of such doubly labeled cells is consistent with the hypothesis that these cells represent a common progenitor to sympathetic neurons and adrenal chromaffin cells. Subsequent to E 11.5, expression of the chromaffin markers is extinguished in the sympathetic ganglia but retained by cells within the adrenal gland. Concomitant with the loss of the SA 1-5 immunoreactivity in sympathetic ganglia, a later sympathetic neuron-specific marker, B2, appears. In dissociated cell suspensions, some B2+ cells that coexpress SA 1 are seen. This implies a switch in the antigenic phenotype of developing sympathetic neurons, rather than a replacement of one cell population by another. The SA 1----B2 transition does not occur for the majority of cells within the adrenal primordium. In vitro, most B2+ cells fail to differentiate into chromaffin cells in response to glucocorticoid. Instead, they continue to extend neurites and then die. Taken together, these data imply that the SA 1----B2 transition correlates with a loss of competence to respond to an inducer of chromaffin differentiation. Thus, the development of SA derivatives is controlled both by environmental signals and by changes in the ability of differentiating cells to respond to such signals.

The generation of monoclonal antibodies that bind preferentially to adrenal chromaffin cells and the cells of embryonic sympathetic ganglia.

Adrenal chromaffin cells, sympathetic neurons, and small intensely fluorescent (SIF) cells are each derived from the neural crest, produce catecholamines, and share certain morphological features. These cell types are also partially interconvertible in cell culture (Doupe et al., 1985a,b; Anderson and Axel, 1986). Thus, these cells are said to be members of the sympathoadrenal (SA) lineage and could share a common progenitor. To investigate the origins of this lineage further, we used the cyclophosphamide immuno-suppression method (Matthew and Patterson, 1983) to generate five monoclonal antibodies (SA1-5) that bind strongly to chromaffin cells, with little or no labeling of sympathetic neurons or SIF cells in frozen sections from adult rats. Competition experiments indicate that these antibodies bind to at least three distinct epitopes in tissue sections. The SA antibodies also label most of the cells of embryonic sympathetic ganglia and adrenal primordia. Labeling of sympathetic ganglia appears as the cells initially coalesce and express high levels of tyrosine hydroxylase (TH). Not all TH+ cells in the embryo are SA 1-5+, however; carotid body SIF cells, nodose ganglion TH+ cells, and the transiently TH+ cells in the dorsal root ganglia do not display detectable SA 1-5 labeling. Thus, the expression of these markers for the SA 1-5 lineage is selective. SA antigen expression is hormonally controlled; removal of glucocorticoid and addition of NGF to cultured adrenal chromaffin cells result in the loss of SA 1-5 labeling. These results suggest that the presumed precursors for sympathetic neurons and SIF cells initially express chromaffin cell markers.

Isolation of the progenitor cells of the sympathoadrenal lineage from embryonic sympathetic ganglia with the SA monoclonal antibodies.

Our previous articles in this series described the production of five monoclonal antibodies (SA1-5) that bind to adrenal chromaffin cells and to cells in embryonic sympathetic ganglia and adrenal primordia (Carnahan and Patterson, 1991), and the downregulation of the sympathoadrenal (SA) antigens in vivo as neuronal markers begin to be expressed (Anderson et al., 1991). These results support the hypothesis that sympathetic neurons and adrenal chromaffin cells are derived from a common embryonic progenitor that displays both neuron- and chromaffin cell-specific markers. We have taken advantage of the fact that at least some of the SA antigens are expressed on the cell surface to isolate SA+ cells from embryonic day 14.5 rat superior cervical, sympathetic ganglia by fluorescence-activated cell sorting. This population of cells is significantly enriched in the expression of markers (tyrosine hydroxylase and neurofilament) found in the putative progenitors in situ. Growth in glucocorticoid maintains the expression of the SA antigens in the sorted cells and induces the chromaffin cell marker enzyme phenylethanolamine N-methyl transferase. In contrast, growth of the sorted cells in basic fibroblast growth factor, NGF, and insulin results in the rapid loss of SA 1 expression and the outgrowth of neurites. The ability to manipulate the fate of the SA+ cells in vitro confirms the suggestion from the in vivo observations that the SA+ cells in the ganglia are at least bipotential progenitors, capable of differentiating along the chromaffin or neuronal pathways.