Temporal mechanisms of myogenic specification in human induced pluripotent stem cells

Understanding the mechanisms of myogenesis in human induced pluripotent stem cells (hiPSCs) is a prerequisite to achieving patient-specific therapy for diseases of skeletal muscle. hiPSCs of different origin show distinctive kinetics and ability to differentiate into myocytes. To address the unique cellular and temporal context of hiPSC differentiation, we perform a longitudinal comparison of the transcriptomic profiles of three hiPSC lines that display differential myogenic specification, one robust and two blunted. We detail temporal differences in mechanisms that lead to robust myogenic specification. We show gene expression signatures of putative cell subpopulations and extracellular matrix components that may support myogenesis. Furthermore, we show that targeted knockdown of ZIC3 at the outset of differentiation leads to improved myogenic specification in blunted hiPSC lines. Our study suggests that β-catenin transcriptional cofactors mediate cross-talk between multiple cellular processes and exogenous cues to facilitate specification of hiPSCs to mesoderm lineage, leading to robust myogenesis.

Isolation and characterization of Xenopus Hey-1: a downstream mediator of Notch signaling

Regulation of Notch signaling likely occurs, at least in part, at the level of basic helix-loop-helix (bHLH) transcription factors that function downstream of Suppressor of Hairless (Su(H)) in the Notch pathway. To begin to characterize modulation of Notch signaling during organogenesis, we examined the bHLH transcription factor, XHey-1 (hairy related-1) in early Xenopus laevis embryos. XHey-1 is expressed in numerous tissues during early development including the somites, head, embryonic kidneys, and heart. Importantly, the expression of XHey-1 was significantly altered in response to perturbation of Notch signaling by means of inducible constructs that served to either activate or suppress Notch signaling through Su(H) in a temporally controlled manner.

Left-right asymmetry determination in vertebrates

A distinctive and essential feature of the vertebrate body is a pronounced left-right asymmetry of internal organs and the central nervous system. Remarkably, the direction of left-right asymmetry is consistent among all normal individuals in a species and, for many organs, is also conserved across species, despite the normal health of individuals with mirror-image anatomy. The mechanisms that determine stereotypic left-right asymmetry have fascinated biologists for over a century. Only recently, however, has our understanding of the left-right patterning been pushed forward by links to specific genes and proteins. Here we examine the molecular biology of the three principal steps in left-right determination: breaking bilateral symmetry, propagation and reinforcement of pattern, and the translation of pattern into asymmetric organ morphogenesis.

Wnt antagonism initiates cardiogenesis in Xenopus laevis

Heart induction in Xenopus occurs in paired regions of the dorsoanterior mesoderm in response to signals from the Spemann organizer and underlying dorsoanterior endoderm. These tissues together are sufficient to induce heart formation in noncardiogenic ventral marginal zone mesoderm. Similarly, in avians the underlying definitive endoderm induces cardiogenesis in precardiac mesoderm. Heart-inducing factors in amphibians are not known, and although certain BMPs and FGFs can mimic aspects of cardiogenesis in avians, neither can induce the full range of activities elicited by the inducing tissues. Here we report that the Wnt antagonists Dkk-1 and Crescent can induce heart formation in explants of ventral marginal zone mesoderm. Other Wnt antagonists, including the frizzled domain-containing proteins Frzb and Szl, lacked this activity. Unlike Wnt antagonism, inhibition of BMP signaling did not promote cardiogenesis. Ectopic expression of GSK3beta, which inhibits beta-catenin-mediated Wnt signaling, also induced cardiogenesis in ventral mesoderm. Analysis of Wnt proteins expressed during gastrulation revealed that Wnt3A and Wnt8, but not Wnt5A or Wnt11, inhibited endogenous heart induction. These results indicate that diffusion of Dkk-1 and Crescent from the organizer initiate cardiogenesis in adjacent mesoderm by establishing a zone of low Wnt3A and Wnt8 activity.

TGF-beta superfamily signaling and left-right asymmetry

Despite an outwardly bilaterally symmetrical appearance, most internal organs of vertebrates display considerable left-right (LR) asymmetry in their anatomy and physiology. The orientation of LR asymmetry with respect to the dorsoventral and anteroposterior body axes is invariant such that fewer than 1 in 10,000 individuals exhibit organ reversals. The stereotypic orientation of LR asymmetry is ensured by distinct left- and right-side signal transduction pathways that are initiated by divergent members of the transforming growth factor-beta (TGF-beta) superfamily of secreted proteins. During early embryogenesis, the TGF-beta-like protein Nodal (or a Nodal-related ortholog) is expressed by the left lateral plate mesoderm and provides essential LR cues to the developing organs. In chick embryos at least, bone morphogenetic protein (BMP) signaling is active on the right side of the embryo and must be inhibited on the left in order for Nodal to be expressed. Thus, at a key point in the determination of LR asymmetry, left-sided signaling is mediated by the transcription factors Smad2 and Smad3 (regulated by Nodal), whereas signaling on the right depends on Smad1 and Smad5 (which are regulated by BMP). This review summarizes the considerable progress that has been made in recent years in understanding the complex network of feedback and feedforward circuitry that regulates both the left- and right-sided pathways. Also discussed is the problem of how signal transduction mediated by the Smad proteins can pattern LR asymmetry without interfering with coincident dorsoventral patterning, which relies on the same Smad proteins.

Notch regulates cell fate in the developing pronephros

The mechanisms that regulate cell fate within the pronephros are poorly understood but are important for the subsequent development of the urogenital system and show many similarities to nephrogenesis in the definitive kidney. Dynamic expression of Notch-1, Serrate-1, and Delta-1 in the developing Xenopus pronephros suggests a role for this pathway in cell fate segregation. Misactivation of Notch signaling using conditionally active forms of either Notch-1 or RBP-J/Su(H) proteins prevented normal duct formation and the proper expression of genetic markers of duct cell differentiation. Inhibition of endogenous Notch signaling elicited the opposite effect. Taken together with the mRNA expression patterns, these data suggest that endogenous Notch signaling functions to inhibit duct differentiation in the dorsoanterior region of the anlage where cells are normally fated to form tubules. In addition, elevated Notch signaling in the pronephric anlage both perturbed the characteristic pattern of the differentiated tubule network and increased the expression of early markers of pronephric precursor cells, Pax-2 and Wilms' tumor suppressor gene (Wt-1). We propose that Notch signaling plays a previously unrecognized role in the early selection of duct and tubule cell fates as well as functioning subsequently to control tubule cell patterning and development.

Serrate and Notch specify cell fates in the heart field by suppressing cardiomyogenesis

Notch signaling mediates numerous developmental cell fate decisions in organisms ranging from flies to humans, resulting in the generation of multiple cell types from equipotential precursors. In this paper, we present evidence that activation of Notch by its ligand Serrate apportions myogenic and non-myogenic cell fates within the early Xenopus heart field. The crescent-shaped field of heart mesoderm is specified initially as cardiomyogenic. While the ventral region of the field forms the myocardial tube, the dorsolateral portions lose myogenic potency and form the dorsal mesocardium and pericardial roof (Raffin, M., Leong, L. M., Rones, M. S., Sparrow, D., Mohun, T. and Mercola, M. (2000) Dev. Biol., 218, 326–340). The local interactions that establish or maintain the distinct myocardial and non-myocardial domains have never been described. Here we show that Xenopus Notch1 (Xotch) and Serrate1 are expressed in overlapping patterns in the early heart field. Conditional activation or inhibition of the Notch pathway with inducible dominant negative or active forms of the RBP-J/Suppressor of Hairless [Su(H)] transcription factor indicated that activation of Notch feeds back on Serrate1 gene expression to localize transcripts more dorsolaterally than those of Notch1, with overlap in the region of the developing mesocardium. Moreover, Notch pathway activation decreased myocardial gene expression and increased expression of a marker of the mesocardium and pericardial roof, whereas inhibition of Notch signaling had the opposite effect. Activation or inhibition of Notch also regulated contribution of individual cells to the myocardium. Importantly, expression of Nkx2. 5 and Gata4 remained largely unaffected, indicating that Notch signaling functions downstream of heart field specification. We conclude that Notch signaling through Su(H) suppresses cardiomyogenesis and that this activity is essential for the correct specification of myocardial and non-myocardial cell fates.

Expression of connexin 30 in Xenopus embryos and its involvement in hatching gland function

Connexins are a family of proteins that assemble to form gap junction channels. Cell-cell communication through gap junctions mediates many important events in embryogenesis, including limb patterning, lens physiology, neuronal function, left-right asymmetry, and secretion from gland tissue. We studied the expression of connexin 30 (Cx30) in the Xenopus embryo and find that it is expressed in the developing hatching gland and pronephros. To determine whether its expression plays a functional role in the activity of the hatching gland, we exposed pre-hatching embryos to drugs that block gap junctional communication. This resulted in a continuation of normal growth and development but specifically abolished hatching. The treatment did not affect Cx30 or Xenopus hatching enzyme transcription, suggesting a post-transcriptional effect on Cx30 gap junctions. We conclude that junctional communication, possibly mediated by Cx30, is involved in secretion of hatching enzyme in Xenopus.

Evolutionary conservation of mechanisms upstream of asymmetric Nodal expression: reconciling chick and Xenopus

Recent experiments have suggested a pathway of genes that regulate left-right asymmetry in vertebrate embryogenesis. The most downstream member of this cascade is nodal (XNR-1 in frogs), which is expressed in the left-side lateral mesoderm. Previous work in the chick [Levin, 1998] suggests that an inductive interaction by Shh (Sonic hedgehog) present at the midline was needed for the left-sided expression of nodal, which by default would not be expressed. Interestingly, it has been reported [Lohr et al., 1997] that in Xenopus, right-side mesoderm that is explanted at st. 15 and allowed to develop in culture, goes on to express nodal, suggesting that lateral mesoderm expresses this gene by default and that a repression of nodal by the midline is needed to achieve asymmetry. Such a contradiction raises interesting questions about the degree of conservation of the mechanisms upstream of nodal asymmetry and, in general, about the differences in the LR pathway among species. Thus we examined this issue directly. We show that in the chick, as in the frog, explanted mesoderm from both sides does, indeed, go on to express nodal, including both the medial and lateral expression domains. Ectopic nodal expression in the medial domain on the right side is not sufficient to induce an ectopic lateral domain. We also show that explanted lateral tissue regenerates node/notochord structures exhibiting Shh expression. Furthermore, we show that Xenopus explants done at st. 15 also regenerate notochord by the stage at which XNR-1 would be expressed. Thus explants are not isolated from the influence of the midline. In contrast to the midline repressor model previously suggested [Lohr et al., 1997] to explain the presence of nodal expression in explants, we propose that the expression is due to induction by signals secreted by regenerating node and notochord tissue (Shh in the chick). Thus our results are consistent with Shh being necessary for nodal induction in both species, and we provide an explanation for both sets of data in terms of a single conserved mechanism upstream of nodal expression.

Subdivision of the cardiac Nkx2.5 expression domain into myogenic and nonmyogenic compartments

Nkx2.5 is expressed in the cardiogenic mesoderm of avian, mouse, and amphibian embryos. To understand how various cardiac fates within this domain are apportioned, we fate mapped the mesodermal XNkx2.5 domain of neural tube stage Xenopus embryos. The lateral portions of the XNkx2.5 expression domain in the neural tube stage embryo (stage 22) form the dorsal mesocardium and roof of the pericardial cavity while the intervening ventral region closes to form the myocardial tube. XNkx2.5 expression is maintained throughout the period of heart tube morphogenesis and differentiation of myocardial, mesocardial, and pericardial tissues. A series of microsurgical experiments showed that myocardial differentiation in the lateral portion of the field is suppressed during normal development by signals from the prospective myocardium and by tissues located more dorsally in the embryo, in particular the neural tube. These signals combine to block myogenesis downstream of XNkx2.5 and at or above the level of contractile protein gene expression. We propose that the entire XNkx2.5/heart field is transiently specified as cardiomyogenic. Suppression of this program redirects lateral cells to adopt dorsal mesocardial and dorsal pericardial fates and subdivides the field into distinct myogenic and nonmyogenic compartments.

Gap junction-mediated transfer of left-right patterning signals in the early chick blastoderm is upstream of Shh asymmetry in the node

Invariant patterning of left-right asymmetry during embryogenesis depends upon a cascade of inductive and repressive interactions between asymmetrically expressed genes. Different cascades of asymmetric genes distinguish the left and right sides of the embryo and are maintained by a midline barrier. As such, the left and right sides of an embryo can be viewed as distinct and autonomous fields. Here we describe a series of experiments that indicate that the initiation of these programs requires communication between the two sides of the blastoderm. When deprived of either the left or the right lateral halves of the blastoderm, embryos are incapable of patterning normal left-right gene expression at Hensen's node. Not only are both flanks required, suggesting that there is no single signaling source for LR pattern, but the blastoderm must be intact. These results are consistent with our previously proposed model in which the orientation of LR asymmetry in the frog, Xenopus laevis, depends on large-scale partitioning of LR determinants through intercellular gap junction channels (M. Levin and M. Mercola (1998) Developmental Biology 203, 90–105). Here we evaluate whether gap junctional communication is required for the LR asymmetry in the chick, where it is possible to order early events relative to the well-characterized left and right hierarchies of gene expression. Treatment of cultured chick embryos with lindane, which diminishes gap junctional communication, frequently unbiased normal LR asymmetry of Shh and Nodal gene expression, causing the normally left-sided program to be recapitulated symmetrically on the right side of the embryo. A survey of early expression of connexin mRNAs revealed that Cx43 is present throughout the blastoderm at Hamburger-Hamilton stage 2–3, prior to known asymmetric gene expression. Application of antisense oligodeoxynucleotides or blocking antibody to cultured embryos also resulted in bilateral expression of Shh and Nodal transcripts. Importantly, the node and primitive streak at these stages lack Cx43 mRNA. This result, together with the requirement for an intact blastoderm, suggests that the path of communication through gap junction channels circumvents the node and streak. We propose that left-right information is transferred unidirectionally throughout the epiblast by gap junction channels in order to pattern left-sided Shh expression at Hensen's node.

Zebrafish narrowminded suggests a genetic link between formation of neural crest and primary sensory neurons

In the developing vertebrate nervous system, both neural crest and sensory neurons form at the boundary between non-neural ectoderm and the neural plate. From an in situ hybridization based expression analysis screen, we have identified a novel zebrafish mutation, narrowminded (nrd), which reduces the number of early neural crest cells and eliminates Rohon-Beard (RB) sensory neurons. Mosaic analysis has shown that the mutation acts cell autonomously suggesting that nrd is involved in either the reception or interpretation of signals at the lateral neural plate boundary. Characterization of the mutant phenotype indicates that nrd is required for a primary wave of neural crest cell formation during which progenitors generate both RB sensory neurons and neural crest cells. Moreover, the early deficit in neural crest cells in nrd homozygotes is compensated later in development. Thus, we propose that a later wave can compensate for the loss of early neural crest cells but, interestingly, not the RB sensory neurons. We discuss the implications of these findings for the possibility that RB sensory neurons and neural crest cells share a common evolutionary origin.

Cerberus regulates left-right asymmetry of the embryonic head and heart

BACKGROUND

Most of the molecules known to regulate left-right asymmetry in vertebrate embryos are expressed on the left side of the future trunk region of the embryo. Members of the protein family comprising Cerberus and the putative tumour suppressor Dan have not before been implicated in left-right asymmetry. In Xenopus, these proteins have been shown to antagonise members of the transforming growth factor beta (TGF-beta) and Wnt families of signalling proteins.

RESULTS

Chick Cerberus (cCer) was found to be expressed in the left head mesenchyme and in the left flank of the embryo. Expression on the left side of the head was controlled by Sonic hedgehog (Shh) acting through the TGF-beta family member Nodal; in the flank, cCer was also regulated by Shh, but independently of Nodal. Surprisingly, although no known targets of Cerberus are expressed asymmetrically on the right side of the embryo at these stages, misexpression of cCer on this side of the embryo led to upregulation of the transcription factor Pitx2 and reversal of the direction of heart and head turning, apparently as independent events. Consistent with the possibility that cCer may be acting on bilaterally expressed TGF-beta family members such as the bone morphogenetic proteins (BMPs), this result was mimicked by right-sided misexpression of the BMP antagonist, Noggin.

CONCLUSIONS

Our findings suggest that cCer maintains a delicate balance of different TGF-beta family members involved in laterality decisions, and reveal the existence of partially overlapping molecular pathways regulating left-right asymmetry in the head and trunk of the embryo.

Spatially distinct head and heart inducers within the Xenopus organizer region

BACKGROUND

The mouse anterior visceral endoderm, an extraembryonic tissue, expresses several genes essential for normal development of structures rostral to the anterior limit of the notochord and has been termed the head organizer. This tissue also has heart-inducing activity and expresses mCer1 which, like its Xenopus homolog cerberus, can induce markers of cardiac specification and anterior neural tissue when ectopically expressed. We investigated the relationship between head and heart induction in Xenopus embryos, which lack extraembryonic tissues.

RESULTS

We found three regions of gene expression in the Xenopus organizer: deep endoderm, which expressed cerberus; prechordal mesoderm, which showed overlapping but non-identical expression of genes characteristic of the murine head organizer, such as XHex and XANF-1; and leading-edge dorsoanterior endoderm, which expressed both cerberus and a subset of the genes expressed by the prechordal mesoderm. Microsurgical ablation of the cerberus-expressing endoderm decreased the incidence of heart, but not head, formation. Removal of prechordal mesoderm, in contrast, caused deficits of anterior head structures. Finally, although misexpression of cerberus induced ectopic heads, it was unable to induce genes thought to participate in head induction.

CONCLUSIONS

In Xenopus, the cerberus-expressing endoderm is required for heart, but not head, inducing activity. Therefore, this tissue is not the topological equivalent of the murine anterior visceral endoderm. We propose that, in Xenopus, cerberus is redundant to other bone morphogenetic protein (BMP) and Wnt antagonists located in prechordal mesoderm for head induction, but may be necessary for heart induction.

Embryological basis for cardiac left-right asymmetry

Asymmetric heart tube looping and chamber morphogenesis is a complex process that is just beginning to be understood at the genetic level. Rightward looping is the first embryological manifestation of consistently oriented, left-right asymmetric development of nearly all visceral organs. Intuitively, invariant anatomical asymmetry must derive from a novel mechanism capable of integrating dorsoventral and anteroposterior information. The details of this process are emerging for several vertebrates and reveal that overall left-right asymmetry, once polarized with respect to dorsoventral and anteroposterior axes, unfolds through distinct left- and right-sided programs of gene expression. These, in turn, regulate expression of cardiac and chamber-specific genes which guide heart morphogenesis and differentiation.

Gap junctions are involved in the early generation of left-right asymmetry

Invariant left-right asymmetry of the visceral organs is a fundamental feature of vertebrate embryogenesis. While a cascade of asymmetrically expressed genes has been described, the embryonic mechanism that orients the left-right axis relative to the dorsoventral and anteroposterior axes (a prerequisite for asymmetric gene expression) is unknown. We propose that this process involves dorsoventral differences in cell-cell communication through gap junctions composed of connexin proteins. Global modulation of gap junctional states in Xenopus embryos by pharmacological agents specifically induced heterotaxia involving mirror-image reversals of heart, gut, and gall bladder. Greatest sensitivity was observed between st. 5 and st. 12, well before the onset of organogenesis. Moreover, heterotaxia was also induced following microinjection of dominant negative and wild-type connexin mRNAs to modify the endogenous dorsoventral difference in junctional communication. Heterotaxia was induced by either blocking gap junction communication (GJC) dorsally or by introducing communication ventrally (but not the reverse). Both connexin misexpression and exposure to GJC-modifying drugs altered expression of the normally left-sided gene XNR-1, demonstrating that GJC functions upstream of XNR-1 in the pathway that patterns left-right asymmetry. Finally, lineage analysis to follow the progeny of microinjected cells indicated that they generally do not contribute the asymmetric organs. Together with the early sensitivity window, this suggests that GJC functions as part of a fundamental, early aspect of left-right patterning. In addition, we show that a potential regulatory mutation in Connexin43 is sufficient to cause heterotaxia. Despite uncertainty about the prevalence of the serine364 to proline substitution reported in human patients with laterality defects, the mutant protein is both a mild hypomorph and a potent antimorph as determined by the effect of its expression on left-right patterning. Taken together, our data suggest that endogenous dorsoventral differences in GJC within the early embryo are needed to consistently orient left-right asymmetry.

PDGF mediates cardiac microvascular communication

The diversity of cellular and tissue functions within organs requires that local communication circuits control distinct populations of cells. Recently, we reported that cardiac myocytes regulate the expression of both von Willebrand factor (vWF) and a transgene with elements of the vWF promoter in a subpopulation of cardiac microvascular endothelial cells (J. Cell Biol. 138:1117). The present study explores this communication. Histological examination of the cardiac microvasculature revealed colocalization of the vWF transgene with the PDGF alpha-receptor. Transcript analysis demonstrated that in vitro cardiac microvascular endothelial cells constitutively express PDGF-A, but not B. Cardiac myocytes induced endothelial expression of PDGF-B, resulting in PDGF-AB. Protein measurement and transcript analysis revealed that PDGF-AB, but not PDGF-AA, induced endothelial expression of vWF and its transgene. Antibody neutralization of PDGF-AB blocked the myocyte-mediated induction. Immunostaining demonstrated that vWF induction is confined to PDGF alpha-receptor-positive endothelial cells. Similar experiments revealed that the PDGF-AB/alpha-receptor communication also induces expression of vascular endothelial growth factor and Flk-1, critical components of angiogenesis. The existence of this communication pathway was confirmed in vivo. Injection of PDGF-AB neutralizing antibody into the amniotic fluid surrounding murine embryos extinguished expression of the transgene. In summary, these studies suggest that environmental induction of PDGF-AB/alpha-receptor interaction is central to the regulation of cardiac microvascular endothelial cell hemostatic and angiogenic activity.

Small-molecule control of insulin and PDGF receptor signaling and the role of membrane attachment

BACKGROUND

Receptor tyrosine kinases (RTKs) regulate the proliferation, differentiation and metabolism of cells, and play key roles in tissue repair, tumorigenesis and development. To facilitate the study of RTKs, we have made conditional alleles that encode monomeric forms of the normally heterotetrameric insulin receptor and monomeric platelet-derived growth factor (PDGF) beta receptors fused to the FK506-binding protein 12 (FKBP12). The chimeric receptors can be induced to undergo dimerization or oligomerization by a small synthetic molecule called FK1012, and the consequences were studied in cells and embryonic tissues.

RESULTS

When equipped with an amino-terminal plasma membrane localization sequence and expressed in HEK293 cells, these chimeric receptors could signal to downstream targets as indicated by the FK1012-dependent activation of p70 S6 kinase (p70(S6k)) and mitogen-activated protein (MAP) kinase. In Xenopus embryos, the engineered PDGF receptor protein induced the formation of mesoderm from animal-pole explants in an FK1012-dependent manner. A cytosolic variant of the protein underwent efficient transphosphorylation, yet failed to activate appreciably either p70(S6k) or MAP kinase following treatment with FK1012. These results provide evidence of a requirement for membrane localization of RTKs, consistent with current models of RTK signaling.

CONCLUSION

We have developed an approach using the small molecule FK1012 to conditionally activate chimeric proteins containing FKBP fused to the insulin receptor or to the PDGF beta receptor. Using this system, we were able to induce mesoderm formation in Xenopus animal-cap tissue and to demonstrate that membrane localization is required for RTK signaling in transfected cells. This system should allow the further dissection of RTK-mediated pathways.

Organizer induction determines left-right asymmetry in Xenopus

Vertebrates appear bilaterally symmetrical but have considerable left-right (LR) asymmetry in the anatomy and placement of internal organs such as the heart. Although a number of asymmetrically expressed genes are known to affect LR patterning, both the initial source of asymmetry and the mechanism that correctly orients the LR axis remain controversial. In this study, we show that the induction of dorsal organizing centers in the embryo can orient LR asymmetry. Ectopic organizing centers were induced by microinjection of mRNA encoding a variety of body axis duplicating proteins, including members of the Wnt signal transduction pathway. The ectopic and primary body axes form side-by-side conjoined twins, with the secondary axis developing as either the left or right sibling. In all cases, correct LR asymmetry was observed in the left twin, regardless of whether it was derived from the primary axis or induced de novo by injection of Xwnt-8, beta-catenin, or Siamois mRNA. In contrast, the right twin was generally unbiased, regardless of the origin of the left body axis, as seen in many instances of experimentally induced and spontaneous conjoined twins. An unanticipated exception was that right twins induced by beta-catenin and Siamois, two downstream effectors of Wnt signaling, exhibited predominately normal heart looping, even when they formed the right twin. Taken together, these results indicate that LR asymmetry is locally oriented as a consequence of Wnt signaling through beta-catenin and Siamois. We discuss the possibility that signals upstream of beta-catenin and Siamois might be required in order for a right sibling to be randomized.

Spina bifida occulta in homozygous Patch mouse embryos

In normal embryos, mRNA encoding platelet-derived growth factor A (PDGF A) and the platelet-derived growth factor receptor alpha (PDGFR alpha) are found within and adjacent to the site of vertebral development, the sclerotome. These patterns of expression are consistent with PDGF action on the developing sclerotome and dermis. Homozygous Patch (Ph) mutant mouse embryos lack the receptor gene (Pdgfra) due to an extensive deletion at that locus. Consistent with the spatial pattern of Pdgfra expression, striking deformities are found in the spine and ribcage of Ph/Ph embryos. In particular, we show that late-gestation Ph/Ph embryos have occult spina bifida involving the entire spinal column. We have analyzed the progression of the axial defects in homozygous Patch embryos in detail. By late gestation it appears that the components of the vertebrae are present, yet the neural arches of the spine are misshapen. We propose that PDGF A is required for proper positioning of the neural arch condensation at all axial levels. Furthermore, since the neural tube appears to close normally, we suggest that spina bifida in the Ph homozygote is caused primarily by a somitic mesoderm abnormality rather than a neural tube defect.

Distribution and functions of platelet-derived growth factors and their receptors during embryogenesis

Platelet-derived growth factors (PDGFs) are soluble proteins that mediate intercellular signaling via receptor tyrosine kinases. The patterns of PDGF and PDGF receptor expression during embryogenesis are complex and dynamic and suggest that signaling can be autocrine or paracrine, depending on the particular tissue and the stage of development. Mesenchymal cells throughout the embryo and within some developing organs produce PDGF receptors, whereas their ligands are often produced by adjacent epithelial or endothelial cells. Disruption of PDGF signaling in the embryo leads to morphogenetic defects and embryonic or perinatal lethality. Tissues that are particularly susceptible to the absence of PDGF signaling are migrating mesoderm cells during gastrulation, nonneuronal neural crest cell derivatives, and kidney mesangial cells. These tissues share the common feature of undergoing epithelial-mesenchymal transitions. We review current knowledge of the distribution of PDGF ligands and receptors and discuss how this distribution may relate to several roles for PDGF during embryogenesis, particularly the regulation of mesenchymal cell behavior.

Embryonic mesoderm cells spread in response to platelet-derived growth factor and signaling by phosphatidylinositol 3-kinase

Abnormal mesoderm movement, leading to defects in axial organization, is observed in mouse and Xenopus laevis embryos deprived of platelet-derived growth factor (PDGF) AA signaling. However, neither the cellular response to PDGF nor the signaling pathways involved are understood. Herein we describe an in vitro assay to examine the direct effect of PDGF AA on aggregates of Xenopus embryonic mesoderm cells. We find that PDGF AA stimulates aggregates to spread on fibronectin. This behavior is similar to that of migrating mesoderm cells in vivo that spread and form lamellipodia and filipodia on contact with fibronectin-rich extracellular matrix. We go on to show two lines of evidence that implicate phosphatidylinositol 3-kinase (PI3K) as an important component of PDGF-induced mesoderm cell spreading. (i) The fungal metabolite wortmannin, which inhibits signaling by PI3K, blocks mesoderm spreading in response to PDGF AA. (ii) Activation of a series of receptors with specific tyrosine-to-phenylalanine mutations revealed PDGF-induced spreading of mesoderm cells depends on PI3K but not on other signaling molecules that interact with PDGF receptors including phospholipase C gamma, Ras GTPase-activating protein, and phosphotyrosine phosphatase SHPTP2. These results indicate that a PDGF signal, medicated by PI3K, can facilitate embryonic mesoderm cell spreading on fibronectin. We propose that PDGF, produced by the ectoderm, influences the adhesive properties of the adjacent mesoderm cells during gastrulation.

Cloning and expression of Xenopus CCT gamma, a chaperonin subunit developmentally regulated in neural-derived and myogenic lineages

The chaperonin containing TCP-1 (CCT) is a eukaryotic cytoplasmic chaperonin, consisting of multiple distinct subunits in a double-toroid structure. In vitro, the CCT has been shown to assist in the folding of tubulin and actin into active conformations through an ATP-dependent mechanism. The function and distribution of these proteins in vivo are also not known. In this report, we show that the expression of two CCT subunits (alpha and gamma) are developmentally regulated in neural-derived and myogenic lineages. While expression in the central nervous system and muscle is consistent with a role in tubulin and actin conformation, we also detect robust expression in the developing cranial neural crest. Enrichment in the neural crest may represent the presence of a novel substrate for the CCT. We have also cloned the complete cDNA for the Xenopus ortholog of CCT gamma, which has 87% amino acid identity with the mouse protein. This remarkable evolutionary conservation suggests a conserved function for this protein among vertebrates, and possibly among all eukaryotes.

PDGF signalling is required for gastrulation of Xenopus laevis

During Xenopus gastrulation, platelet-derived growth factor (PDGF) receptor-alpha is expressed in involuting marginal zone cells which migrate over ectodermal cells expressing PDGF-A. To investigate the role of PDGF signalling during this process, we have generated a novel point mutant of PDGF receptor-alpha analogous to the W37 mutation of c-kit. This molecule is a specific, potent, dominant inhibitor of PDGF signalling in vivo. Injection of RNA encoding this protein into Xenopus embryos prevents closure of the blastopore, leads to abnormal gastrulation and a loss of anterior structures. Convergent extension is not inhibited in these embryos, but rather, involuting mesodermal cells fail to adhere to the overlying ectoderm. PDGF may therefore be required for mesodermal cell-substratum interaction.

Cyclopamine, a steroidal alkaloid, disrupts development of cranial neural crest cells in Xenopus

Cyclopamine is a steroidal alkaloid which causes limb and craniofacial defects in many vertebrate species. We have used Xenopus laevis as a model system to characterize the defects caused by cyclopamine at the cellular level. The most dramatic consequence of cyclopamine treatment in the Xenopus embryo is a defect in formation of craniofacial cartilage. Much of this cartilage is absent in treated animals. As in avian and mammalian species, Xenopus craniofacial cartilage is derived primarily from cells of the cranial neural crest. Grafting experiments show that development of the cartilaginous derivatives of the cranial neural crest is impaired after cyclopamine treatment, and this is at least partially due to a direct effect on presumptive crest cells. A culture system was used to determine the cellular response to the drug. Cyclopamine did not block the initial emigration of cells from a neural plate explant. However, cell death is seen in treated cultures after 4 days. Trunk neural crest cells and transformed cell lines are resistant to cyclopamine. We therefore conclude that cyclopamine specifically causes death of cranial neural crest cells and that lethality is likely to account for the teratogenic effects of this compound.

An inductive role for the endoderm in Xenopus cardiogenesis

Heart induction in Xenopus has been thought to be dependent primarily on the interaction of the heart primordia with the Spemann organizer. We demonstrate, however, that signals derived from the deep dorsoanterior endoderm during early gastrulation are also essential for heart formation. The presence of deep endoderm dramatically enhances heart formation in explants of heart primordia, both in the presence and absence of organizer. Likewise, extirpation of the entire endoderm can decrease the frequency of heart formation in embryos that retain organizer activity. Finally, we show that the combined presence of both endoderm and organizer is necessary and sufficient to induce heart in ventral mesoderm explants that would not otherwise form heart tissue. Xenopus heart induction, therefore, may be a multistep process requiring separate dorsalization and cardiogenic signalling events. This is the first demonstration of a heart-inducing role for the endoderm in Xenopus, indicating that the mechanism of heart formation may be similar in most vertebrates.

Localization of PDGF A and PDGFR alpha mRNA in Xenopus embryos suggests signalling from neural ectoderm and pharyngeal endoderm to neural crest cells

In situ hybridization analysis of Xenopus laevis embryos reveals that mRNA encoding the platelet-derived growth factor alpha receptor (PDGFR alpha) is expressed in cephalic neural crest masses prior to migration from the future neural tube and during their migration into the visceral arches. The analysis of fluorescently labeled neural crest tissue transplanted to unlabeled host embryos demonstrates that neural crest cells are the only detectable source of PDGFR alpha mRNA within visceral arches. Transcripts encoding PDGF A are present in neural ectoderm, otic vesicle and pharyngeal endoderm. Their location suggests that PDGF A provides a signal, first from the neural epithelium and later from the otic vesicle and pharyngeal endoderm, to cephalic neural crest cells during their migration in the arch region.

Morphological differences in Xenopus embryonic mesodermal cells are specified as an early response to distinct threshold concentrations of activin

The involution of presumptive mesoderm that occurs during amphibian gastrulation is a complex process requiring the coordinated action of a diverse range of cells. We show that cells with distinct morphologies, resembling each of those normally found in the involuting tissue of the Xenopus embryo, are induced in dispersed animal pole cells by different doses of the potent mesoderm-inducing factor activin. Each cell type is induced within a restricted dose range of activin concentrations, the boundaries of which are well demarcated shortly after activin treatment. In contrast, Brachyury and goosecoid, two genes thought to pattern the presumptive mesoderm, and the gene encoding platelet-derived growth factor receptor alpha, which is expressed in the mesoderm of gastrula stage embryos, are induced by broad, overlapping ranges of high activin concentrations at such early times. Similarly, the response of the gene encoding platelet-derived growth factor A, which is expressed normally in ectoderm of gastrula stage embryos, diminishes gradually as the activin concentration increases. Dose windows for the expression of these four genes narrow and become distinct from one another in cell aggregates after several hours in culture, suggesting that activin prompts a dynamic program of gene expression in induced mesoderm.

Xenopus laevis cellular retinoic acid-binding protein: temporal and spatial expression pattern during early embryogenesis

There is increasing evidence that retinoic acid (RA) has a role in establishing normal axial patterns during Xenopus laevis embryo-genesis. Several types of retinoid binding proteins are thought to mediate the effects of RA. We report the isolation of a cDNA, named xCRABP-b, which encodes a X. laevis cellular retinoic acid-binding protein (xCRABP). This cDNA hybridises to a transcript in gastrular stage embryos of approximately 3 kb, much larger than those CRABP transcripts expressed in mice. The expression of the xCRABP mRNA is generally restricted to tissues which are sensitive to the teratogenic effects of excess RA. It is likely, that during normal X. laevis embryogenesis, concentrations of RA in RA-responsive cells are modulated by the xCRABP gene product.

The Xenopus platelet-derived growth factor alpha receptor: cDNA cloning and demonstration that mesoderm induction establishes the lineage-specific pattern of ligand and receptor gene expression

We have cloned the Xenopus PDGF alpha receptor cDNA and have used this clone, along with cDNA encoding PDGF A, to examine their expression pattern in Xenopus embryos and to determine the factors responsible for lineage specificity. Recombinant Xenopus alpha receptor expressed in COS cells exhibits PDGF-A-dependent tyrosine kinase activity. We find that receptor mRNA is present in cultured marginal zone tissue explants and in animal cap tissue induced to form mesoderm either by grafting to vegetal tissue or by treatment with recombinant activin A. In contrast, PDGF A mRNA is expressed in cultured, untreated animal cap tissue and is suppressed by mesoderm induction. These results suggest that ectodermally produced PDGF A may act on the mesoderm during gastrulation and that mesoderm induction establishes the tissue pattern of ligand and receptor expression.

Expression of mouse PDGF-A and PDGF alpha-receptor genes during pre- and post-implantation development: evidence for a developmental shift from an autocrine to a paracrine mode of action

We examined the expression of platelet-derived growth factor (PDGF)-A and the PDGF alpha-receptor in pre-implantation and early post-implantation mouse embryos. At two-cell and blastocyst stages, all cells express mRNA and protein for both ligand and receptor. In contrast, early post-implantation embryos express PDGF-A chain mRNA in both embryonic ectoderm and in the ectoderm lining the ectoplacental cavity, while mRNA for PDGF alpha-receptor is localized to the mesoderm layers of both embryonic and extra-embryonic membranes. At days 3.5 and 7.5, receptors are demonstrably functional in response to exogenous PDGF-AA. We propose that chronic autostimulation of PDGF alpha-receptors occurs in pre-implantation embryos, whereas, following implantation, early mesoderm development is dependent on stimulation by ectodermally produced PDGF-A.

Rapid, complete and reversible transformation by v-sis precedes irreversible transformation

v-sis is the oncogene of simian sarcoma virus, but whether tumor growth is maintained by v-sis expression alone or requires additional changes is unknown. To distinguish these possibilities we studied a model of reversible transformation including tumorigenicity using NIH3T3 cells bearing a metallothionein promoter-v-sis construction. Cells subcultured from 10 out of 18 tumors from athymic mice, all less than 0.1 g and less than or equal to 21 days in age, reverted to a normal phenotype but exhibited transformation upon addition of zinc as judged by morphology, growth rate, saturation density and anchorage independence of growth. Thus, activation of v-sis alone is sufficient for initiation and early autocrine-based growth of tumors. However, the cells from the remaining and predominantly larger, 0.5 +/- 0.7 g, tumors did not revert and exhibited zinc-independent transformation as judged by the same criteria. Southern analysis and examination of the regulation of v-sis product expression in cells derived from these tumors showed no change in zinc-dependent and reversible regulation of v-sis sequences. These results suggest that subsequent tumor growth strongly favors acquisition of additional irreversible change(s) in the tumor cell genome at high frequency (44%). Thus an early event of a multistep process stimulated by v-sis-dependent transformation best accounts for the sum of results.

Platelet-derived growth factor receptor alpha-subunit gene (Pdgfra) is deleted in the mouse patch (Ph) mutation

Platelet-derived growth factor receptors are composed of two subunits (alpha and beta) that associate with one another to form three functionally active dimeric receptor species. The two subunits are encoded by separate loci in humans and other species. In this study, we used conventional interspecific backcross mapping and an analysis of a deletional mutation to establish close linkage between the alpha-subunit gene (Pdgfra) and the dominant spotting (W) locus on mouse chromosome 5. Further, by analyzing the restriction fragment length polymorphisms in interspecific F1 hybrids, we were able to demonstrate that the closely associated patch (Ph) locus carries a deletion in Pdgfra. This observation was confirmed by both DNA and RNA analysis of 10.5-day fetuses produced from crosses between Ph heterozygotes. Out of 16 fetuses analyzed, Pdgfra genomic sequences were absent and no mRNA for the receptor was detected in 6 fetuses that were developmentally abnormal (the presumptive Ph homozygotes). We also determined that the deletion associated with the Ph mutation does not extend into the coding sequences of the adjacent Kit gene, by analysis of the genomic DNA from both the interspecific F1 hybrids and the presumptive Ph homozygotes. The absence of Pdgfra genomic sequences and the lack of detectable message associated with the Ph mutation should make this mutant a valuable asset for understanding the role of the receptor alpha subunit during mammalian development.

Dominant-negative mutants of a platelet-derived growth factor gene

Using site-directed mutagenesis of a PDGF-A cDNA clone, we identify two domains that are required to generate stable, mitogenically active PDGF-AA homodimers. Alteration of the tetra-basic amino acid sequence (Arg84-Arg-Lys-Arg to Arg-Ser-Asn-Gly) results in the formation of stable pro-PDGF-A homodimers that lack mitogenic activity. Substitution of serine for Cys129 destabilizes PDGF-A subunits within the cell. Genes incorporating either the processing lesion or the cysteine substitution suppress wild-type PDGF-A gene expression in a trans-dominant fashion. Suppression occurs because the mutant PDGF subunits dimerize with wild-type subunits to form inactive or unstable heterodimers. Suppression is exerted across phylogenetic boundaries; thus, the mouse PDGF-A chain mutants inhibit the activity of the wild-type Xenopus PDGF-A. The cysteine mutant gene suppresses expression of PDGF-B (c-sis), as well as PDGF-A. The processing mutant gene, however, suppresses only PDGF-A. Dominant-negative mutations of PDGF and other growth factors which, like PDGF, function as dimers may prove useful for creating animals models of growth factor deficiency disease states and for revealing the function of growth factors during early embryonic development.

Selective expression of PDGF A and its receptor during early mouse embryogenesis

Murine homologs of the PDGF A, PDGF B, and PDGF receptor alpha subunit genes were cloned. These were used, together with a mouse PDGF receptor beta subunit cDNA clone, to monitor gene expression in early postimplantation mouse embryos and in F9 embryonal carcinoma cells. RNAse protection analysis shows that PDGF A chain, but not B chain, mRNA is expressed in 6.5- to 8.5-day embryonic and extraembryonic tissues. Both alpha and beta receptor subunit mRNAs are expressed in early embryos, however, alpha subunit mRNA appears earlier and is more abundant than beta subunit mRNA. Undifferentiated F9 embryonal carcinoma stem cells express abundant levels of A chain, but not B chain, mRNA. Neither of the PDGF receptor genes is expressed in stem cells. Treatment with retinoic acid stimulates expression of both PDGF receptor genes. As in postimplantation mouse embryos, alpha receptor subunit mRNA appears earlier and is substantially more abundant than beta subunit mRNA. Collectively, these data demonstrate that the genes encoding the two chains of PDGF and their receptors are regulated independently during development and suggest that the two systems have some nonoverlapping functions in vivo. PDGF A, but not PDGF B, may be particularly important in modulating early events in mouse embryonic development.

Platelet-derived growth factor A chain is maternally encoded in Xenopus embryos

Transcription of zygotic genes does not occur in early Xenopus embryos until the mid-blastula transition, 6 to 7 hours after fertilization. Before this time, development is directed by maternal proteins and messenger RNAs stored within the egg. Two different forms of the A chain of platelet-derived growth factor (PDGF) are shown here to be encoded by maternal messenger RNAs. The two forms closely resemble human PDGF; however, the long form contains a hydrophobic region near the carboxyl terminus. The presence of PDGF messenger RNA in the embryo supports the idea that endogenous growth factors act at the earliest stages of embryogenesis.

Growth factor superfamilies and mammalian embryogenesis

With the availability of amino acid and nucleotide sequence information has come the realization that growth factors can be clustered in to superfamilies. Several of these superfamilies contain molecules that were not initially identified because of growth-promoting activities; rather they were discovered through their ability to regulate other processes. Certain members of these superfamilies are present during early mammalian embryogenesis. However, until recently, it has been difficult to manipulate the developing mammalian embryo to observe directly the effects of inappropriate, excessive, or reduced expression of these molecules. Despite this limitation, at least some of these molecules have been implicated in the control of differentiation and morphogenesis, two actions unpredicted from the cell biology of most of the growth factors. Moreover, these actions are reflected in nonmammalian species where homologues of the mammalian growth factors control crucial steps in the choice of developmental fate. This review describes five growth factor superfamilies and the role these molecules may have in controlling proliferation, differentiation, and morphogenesis during mammalian development.

Strong transcriptional activation of translocated c-myc genes occurs without a strong nearby enhancer or promoter

We have studied the transcriptional activation of translocated c-myc genes in murine plasmacytomas in which the translocation juncture occurs within the first intron of c-myc and juxtaposes c-myc with the immunoglobulin C alpha gene segment. It has been widely suggested that a novel transcriptional enhancer element located near the C alpha gene segment might activate the translocated c-myc gene. We have carried out an extensive search for such an element and find no significant transcriptional enhancer activity in a 22 kb region encompassing the translocation junction, C alpha gene segment and regions 3' of C alpha. We also find that the cryptic promoter region of the translocated c-myc gene is a very weak promoter of transcription. Despite this evidence against the presence of strong transcriptional regulatory elements, the translocated c-myc gene locus is transcribed at high rates that are 25-greater than 100% of that measured for the highly active immunoglobulin genes in murine plasmacytomas. These data suggest the presence of a novel type of strong activator of transcription in the murine heavy chain locus.

Immunoglobulin heavy-chain enhancer requires one or more tissue-specific factors

Enhancer sequences are regulatory regions that greatly increase transcription of certain eukaryotic genes. An immunoglobulin heavy-chain variable gene segment is moved from a region lacking enhancer activity to a position adjacent to the known heavy-chain enhancer early in B-cell maturation. In lymphoid cells, the heavy-chain and SV40 enhancers bind a common factor essential for enhancer function. In contrast, fibroblast cells contain a functionally distinct factor that is used by the SV40 but not by the heavy-chain enhancer. The existence of different factors in these cells may explain the previously described lymphoid cell specificity of the heavy-chain enhancer.

Translocation affects normal c-myc promoter usage and activates fifteen cryptic c-myc transcription starts in plasmacytoma M603

Plasmacytoma M603 contains one normal, nontranslocated c-myc gene and one translocated c-myc gene in which c-myc exon 1 is juxtaposed with the immunoglobulin heavy chain enhancer and c-myc exons 2 and 3 are juxtaposed with C alpha. We find that steady-state c-myc RNA levels are 2-4 fold elevated in M603 relative to normal liver or spleen and that these transcripts originate predominantly if not exclusively from the translocated c-myc gene. Although both promoters on the nontranslocated c-myc gene are repressed, the proximal promoter, P1, is active on the translocated 5' c-myc region which is juxtaposed with the immunoglobulin heavy chain enhancer. The 3' portion of the translocated c-myc gene is transcribed from fifteen cryptic start sites and spliced at aberrant donor and acceptor splice sites, thereby generating a mixture of transcripts with different, abnormal 5' untranslated regions. Although the reason that translocation activates the cryptic c-myc starts in M603 is not completely understood, we show that truncation of the c-myc gene is not sufficient to activate cryptic transcription sites.

Transcriptional enhancer elements in the mouse immunoglobulin heavy chain locus

Two regions in the immunoglobulin heavy chain locus were tested for their ability to enhance transcription of the SV40 early promoter. A portion of the intervening sequence between the heavy chain joining region (Jh) and the constant region of the mu chain (Cmu) can enhance transcription when it is cloned either 5' or 3' to the SV40 early promoter. The region between C alpha and the alpha switch site, which occurs 5' to the translocated c-myc oncogene in many murine plasmacytomas, does not show transcriptional enhancer activity in this assay.

The interaction of Cibacron Blue F3GA with troponin and its subunits

Binding of troponin to Cibacron Blue F3GA-agarose column and its selective release from the gel in the presence of 0.5 M KCl provides the basis for a new purification method. The two-step procedure consists of isoelectric precipitation of tropomyosin and chromatography of the resultant crude troponin supernatant on Affi-Gel Blue column. Adsorption of troponin to the immobilized dye appears to occur through the troponin-T subunit. Troponin-I and troponin-C do not bind to the blue agarose column, whereas troponin-T binds to it very tightly. Binding of the dye to troponin-T prevents formation of troponin-T-troponin-C complex, but does not interfere with direct interaction of troponin-T with troponin-I. The activity of troponin in conferring calcium sensitivity on actomyosin ATPase is not affected by Cibacron Blue. Circular dichroism and difference absorption measurements of complexes of the blue dye with troponin and its subunits reveal the presence of a tight binding site on whole troponin and on troponin-T (KA greater than or equal to 10(6) M). The existence of weak binding sites for the dye on troponin and all of its subunits is deduced from difference absorption studies. Cibacron Blue appears to be a sensitive probe for subunit interactions in troponin.