Generality of vertebrate developmental patterns: evidence for a dermomyotome in fish.

The somitic compartment that gives rise to trunk muscle and dermis in amniotes is an epithelial sheet on the external surface of the somite, and is known as the dermomyotome. However, despite its central role in the development of the trunk and limbs, the evolutionary history of the dermomyotome and its role in nonamniotes is poorly understood. We have tested whether a tissue with the morphological and molecular characteristics of a dermomyotome exists in nonamniotes. We show that representatives of the agnathans and of all major clades of gnathostomes each have a layer of cells on the surface of the somite, external to the embryonic myotome. These external cells do not show any signs of terminal myogenic or dermogenic differentiation. Moreover, in the embryos of bony fishes as diverse as sturgeons (Chondrostei) and zebrafish (Teleostei) this layer of cells expresses the pax3 and pax7 genes that mark myogenic precursors. Some of the pax7-expressing cells also express the differentiation-promoting myogenic regulatory factor Myogenin and appear to enter into the myotome. We therefore suggest that the dermomyotome is an ancient and conserved structure that evolved prior to the last common ancestor of all vertebrates. The identification of a dermomyotome in fish makes it possible to apply the powerful cellular and genetic approaches available in zebrafish to the understanding of this key developmental structure.

Distinct mechanisms regulate slow-muscle development.

Vertebrate muscle development begins with the patterning of the paraxial mesoderm by inductive signals from midline tissues [1, 2]. Subsequent myotome growth occurs by the addition of new muscle fibers. We show that in zebrafish new slow-muscle fibers are first added at the end of the segmentation period in growth zones near the dorsal and ventral extremes of the myotome, and this muscle growth continues into larval life. In marine teleosts, this mechanism of growth has been termed stratified hyperplasia [3]. We have tested whether these added fibers require an embryonic architecture of muscle fibers to support their development and whether their fate is regulated by the same mechanisms that regulate embryonic muscle fates. Although Hedgehog signaling is required for the specification of adaxial-derived slow-muscle fibers in the embryo [4, 5], we show that in the absence of Hh signaling, stratified hyperplastic growth of slow muscle occurs at the correct time and place, despite the complete absence of embryonic slow-muscle fibers to serve as a scaffold for addition of these new slow-muscle fibers. We conclude that slow-muscle-stratified hyperplasia begins after the segmentation period during embryonic development and continues during the larval period. Furthermore, the mechanisms specifying the identity of these new slow-muscle fibers are different from those specifying the identity of adaxial-derived embryonic slow-muscle fibers. We propose that the independence of early, embryonic patterning mechanisms from later patterning mechanisms may be necessary for growth.

Somite development in zebrafish.

A full understanding of somite development requires knowledge of the molecular genetic pathways for cell determination as well as the cellular behaviors that underlie segmentation, somite epithelialization, and somite patterning. The zebrafish has long been recognized as an ideal organism for cellular and histological studies of somite patterning. In recent years, genetics has proven to be a very powerful complementary approach to these embryological studies, as genetic screens for zebrafish mutants defective in somitogenesis have identified over 50 genes that are necessary for normal somite development. Zebrafish is thus an ideal system in which to analyze the role of specific gene products in regulating the cell behaviors that underlie somite development. We review what is currently known about zebrafish somite development and compare it where appropriate to somite development in chick and mouse. We discuss the processes of segmentation and somite epithelialization, and then review the patterning of cell types within the somite. We show directly, for the first time, that muscle cell and sclerotome migrations occur at the same time. We end with a look at the many questions about somitogenesis that are still unanswered.

The zebrafish slow-muscle-omitted gene product is required for Hedgehog signal transduction and the development of slow muscle identity.

Hedgehog proteins mediate many of the inductive interactions that determine cell fate during embryonic development. Hedgehog signaling has been shown to regulate slow muscle fiber type development. We report here that mutations in the zebrafish slow-muscle-omitted (smu) gene disrupt many developmental processes involving Hedgehog signaling. smu(-/-) embryos have a 99% reduction in the number of slow muscle fibers and a complete loss of Engrailed-expressing muscle pioneers. In addition, mutant embryos have partial cyclopia, and defects in jaw cartilage, circulation and fin growth. The smu(-/-) phenotype is phenocopied by treatment of wild-type embryos with forskolin, which inhibits the response of cells to Hedgehog signaling by indirect activation of cAMP-dependent protein kinase (PKA). Overexpression of Sonic hedgehog (Shh) or dominant negative PKA (dnPKA) in wild-type embryos causes all somitic cells to develop into slow muscle fibers. Overexpression of Shh does not rescue slow muscle fiber development in smu(-/-) embryos, whereas overexpression of dnPKA does. Cell transplantation experiments confirm that smu function is required cell-autonomously within the muscle precursors: wild-type muscle cells rescue slow muscle fiber development in smu(-/-) embryos, whereas mutant muscle cells cannot develop into slow muscle fibers in wild-type embryos. Slow muscle fiber development in smu mutant embryos is also rescued by expression of rat Smoothened. Therefore, Hedgehog signaling through Slow-muscle-omitted is necessary for slow muscle fiber type development. We propose that smu encodes a vital component in the Hedgehog response pathway.

Positive and negative regulation of muscle cell identity by members of the hedgehog and TGF-beta gene families.

We have examined whether the development of embryonic muscle fiber type is regulated by competing influences between Hedgehog and TGF-beta signals, as previously shown for development of neuronal cell identity in the neural tube. We found that ectopic expression of Hedgehogs or inhibition of protein kinase A in zebrafish embryos induces slow muscle precursors throughout the somite but muscle pioneer cells only in the middle of the somite. Ectopic expression in the notochord of Dorsalin-1, a member of the TGF-beta superfamily, inhibits the formation of muscle pioneer cells, demonstrating that TGF-beta signals can antagonize the induction of muscle pioneer cells by Hedgehog. We propose that a Hedgehog signal first induces the formation of slow muscle precursor cells, and subsequent Hedgehog and TGF-beta signals exert competing positive and negative influences on the development of muscle pioneer cells.

Identification of separate slow and fast muscle precursor cells in vivo, prior to somite formation.

We have examined the development of specific muscle fiber types in zebrafish axial muscle by labeling myogenic precursor cells with vital fluorescent dyes and following their subsequent differentiation and fate. Two populations of muscle precursors, medial and lateral, can be distinguished in the segmental plate by position, morphology and gene expression. The medial cells, known as adaxial cells, are large, cuboidal cells adjacent to the notochord that express myoD. Surprisingly, after somite formation, they migrate radially away from the notochord, becoming a superficial layer of muscle cells. A subset of adaxial cells develop into engrailed-expressing muscle pioneers. Adaxial cells differentiate into slow muscle fibers of the adult fish. We have named the lateral population of cells in the segmental plate, lateral presomitic cells. They are smaller, more irregularly shaped and separated from the notochord by adaxial cells; they do not express myoD until after somite formation. Lateral presomitic cells remain deep in the myotome and they differentiate into fast muscle fibers. Thus, slow and fast muscle fiber types in zebrafish axial muscle arise from distinct populations of cells in the segmental plate that develop in different cellular environments and display distinct behaviors.

Interactions of the p107 and Rb proteins with E2F during the cell proliferation response.

The E2F transcription factor is found in complexes with a variety of cellular proteins including the retinoblastoma tumor suppressor protein. Various assays have demonstrated a tight correlation between the functional capacity of Rb as a growth suppressor and its ability to bind to E2F. Moreover, only the underphosphorylated form of Rb, which appears to be the active species, interacts with E2F. Despite the fact that the majority of Rb becomes hyperphosphorylated at the end of G1, we now show that the E2F-Rb interaction persists through the G1/S transition and into S phase. A distinct E2F complex does appear to be regulated in relation to the transition from G1 to S phase. We now demonstrate that this complex contains the Rb-related p107 protein. Moreover, like the Rb protein, p107 inhibits E2F-dependent transcription in a co-transfection assay. This result, together with the observation that free, uncomplexed E2F accumulates as cells leave G1 and enter S phase, suggests that the p107 protein may regulate E2F-dependent transcription during G1. In contrast, although Rb does regulate the transcriptional activity of E2F, this association does not coincide with the G1 to S phase transition.

A cyclin A-protein kinase complex possesses sequence-specific DNA binding activity: p33cdk2 is a component of the E2F-cyclin A complex.

The E2F transcription factor has been found in association with the cyclin A protein, and this complex accumulates during the S phase of the cell cycle, suggesting that E2F may play a role in cell cycle control. In independent studies, cyclin A has been shown to be associated with two other proteins, the Rb-related p107 protein and the cdc2-related p33 cdk2 protein kinase. Through an analysis of the E2F-cyclin A complex, we now find that both the p107 protein and the cdc2-related p33cdk2 kinase are components of the previously described complex. Moreover, the complex possesses H1 kinase activity. These results thus define a cyclin A-cdk2 kinase complex that possesses sequence-specific DNA binding activity. This suggests that the cdk2 kinase may phosphorylate other DNA-bound substrates, and that one role of the E2F factor may be to localize this protein kinase to the DNA.

Domains of the adenovirus E1A protein required for oncogenic activity are also required for dissociation of E2F transcription factor complexes.

Recent experiments have shown that the cellular E2F transcription factor is found in complexes with cellular proteins and that one such complex contains the cyclin-A protein. Isolation of a cellular activity, which we term E2F-BF, can reconstitute the E2F-cyclin-A complex and has permitted a more detailed analysis of the mechanism of E1A dissociation. Through the analysis of a series of E1A mutants, we find that sequences in conserved region 1 (CR1) and conserved region 2 (CR2) are important for dissociation of the E2F complex, whereas amino-terminal sequences are not required. In contrast to the requirements for dissociation, only the CR1 sequences are required to block formation of the complex if E1A is added when the components are combined. We have also identified an activity, termed E2F-I, that inhibits E2F binding to DNA, again apparently through the formation of a complex with E2F. This inhibitory activity is also blocked by E1A, dependent on the same elements of the E1A protein that disrupt the interaction with E2F-BF. Because the E1A sequences that are important for releasing E2F from these interactions are also sequences necessary for oncogenesis, we suggest that this activity may be a critical component of the transforming activity of E1A.

Cell cycle regulation of the E2F transcription factor involves an interaction with cyclin A.

We have examined E2F binding activity in extracts of synchronized NIH 3T3 cells. During the G0 to G1 transition, there is a marked increase in the level of active E2F. Subsequently, there are changes in the nature of E2F-containing complexes. A G1-specific complex increases in abundance, disappears, and is then replaced by another complex as S phase begins. Analysis of extracts of thymidine-blocked cells confirms that the complexes are cell cycle regulated. We also show that the cyclin A protein is a component of the S phase complex. Each complex can be dissociated by the adenovirus E1A 12S product, releasing free E2F. The release of E2F from the cyclin A complex coincides with the stimulation of an E2F-dependent promoter. We suggest that these interactions control the activity of E2F and that disruption of the complexes by E1A contributes to a loss of cellular proliferation control.

Adenovirus 12S E1A gene represses differentiation of F9 teratocarcinoma cells.

The F9 teratocarcinoma cell line differentiates in vitro after treatment with retinoic acid and cAMP and has been a widely used model system for the study of the molecular events that are responsible for cellular commitment and differentiation during early development. Previous experiments have suggested intriguing parallels between the control of gene expression during F9 cell differentiation and the regulation of gene expression by adenovirus E1A. Transfection of a 12S E1A-expressing plasmid into terminally differentiated, nonproliferating F9 cells generates, at high frequency, colonies of dividing cells, each of which expresses E1A. Cell lines established from these colonies proliferate in the presence of retinoic acid and have lost the fully differentiated phenotype as characterized by the absence of expression of a series of differentiation-specific genes. We conclude that expression of the viral 12S E1A gene product interferes with retinoic acid-induced F9 cell differentiation. Moreover, the results suggest that the differentiation process, as defined by markers of terminal differentiation, may not be a permanent event but can be reversed by E1A expression.

Neuronal growth cone migration.

The neuronal growth cone is a semi-autonomous portion of the developing neuron that is highly specialized for motile activity. Migrating neurons may share some features with neuronal growth cones. I review some of what has been learned about growth cone initiation, the differentiation of axons and dendrites, the role of the cytoskeleton in motility, the movements of membrane vesicles, the factors regulating the rate and direction of growth cone movement, and the further differentiation of growth cones as they enter the target area and initiate synaptogenesis. Where appropriate, I draw comparisons to what is known about the migration of neurons.

Expression of the growth cone specific epitope CDA 1 and the synaptic vesicle protein SVP38 in the developing mammalian cerebral cortex.

CDA 1 is a novel antigen that within the brain is present specifically in neuronal growth cones. Electron microscope immunohistochemistry and subcellular fractionation showed the CDA 1 epitope to be on a cytosolic molecule. In cultured neurons, it is abundant in growth cones and not detectable in neurites or cell bodies. The development of the rat cerebral cortex was investigated by using the monoclonal antibody to CDA 1 and an antibody to SVP38, the synaptic vesicle glycoprotein. CDA 1 immunoreactivity in the rat cerebral cortex peaks just before birth and disappears by postnatal day 12, a few days before the major increase in the number of mature synapses. In contrast, SVP38 is expressed in parallel with the appearance of mature synapses. CDA 1 and SVP38 thus are markers of growth cones and synapses, respectively. Their expression during development reflects some of the structural and functional changes that occur during synapse formation.