Embryonic expression patterns of the neural cell adhesion molecule gene are regulated by homeodomain binding sites.

During development of the vertebrate nervous system, the neural cell adhesion molecule (N-CAM) is expressed in a defined spatiotemporal pattern. We have proposed that the expression of N-CAM is controlled, in part, by proteins encoded by homeobox genes. This hypothesis has been supported by previous in vitro experiments showing that products of homeobox genes can both bind to and transactivate the N-CAM promoter via two homeodomain binding sites, HBS-I and HBS-II. We have now tested the hypothesis that the N-CAM gene is a target of homeodomain proteins in vivo by using transgenic mice containing native and mutated N-CAM promoter constructs linked to a beta-galactosidase reporter gene. Segments of the 5' flanking region of the mouse N-CAM gene were sufficient to direct expression of the reporter gene in the central nervous system in a pattern consistent with that of the endogenous N-CAM gene. For example, at embryonic day (E) 11, beta-galactosidase staining was found in postmitotic neurons in dorsolateral and ventrolateral regions of the spinal cord; at E14.5, staining was seen in these neurons throughout the spinal cord. In contrast, mice carrying an N-CAM promoter-reporter construct with mutations in both homeodomain binding sites (HBS-I and HBS-II) showed altered expression patterns in the spinal cord. At E11, beta-galactosidase expression was seen in the ventrolateral spinal cord, but was absent in the dorsolateral areas, and at E 14.5, beta-galactosidase expression was no longer detected in any cells of the cord. Homeodomain binding sites found in the N-CAM promoter thus appear to be important in determining specific expression patterns of N-CAM along the dorsoventral axis in the developing spinal cord. These experiments suggest that the N-CAM gene is an in vivo target of homeobox gene products in vertebrates.

Function of the pre-T-cell receptor alpha chain in T-cell development and allelic exclusion at the T-cell receptor beta locus.

The pre-T-cell receptor, composed of the T-cell receptor (TCR) beta chain (TCRbeta), pre-Talpha (pTalpha) chain, and CD3 molecules, has been postulated to be a transducer of signals during the early stages of T-cell development. To examine the function of the transmembrane pTalpha chain during tbymocyte development, we generated pTalpha-/- embryonic stem cells and assayed their ability to differentiate into lymphoid cells in vivo after injection into recombination-activating gene (RAG)-2-deficient blastocysts. Thymocytes representing all stages of T-cell differentiation were detected in the thymus of pTalpha-/- chimeric mice, indicating that thymocyte development can occur without pTalpha. However, greatly reduced thymocyte numbers and substantially increased percentages of both CD4-CD8- thymocytes and TCRgammadelta+ thymocytes suggest that pTalpha plays a critical role in thymocyte expansion. To investigate the role of the pTalpha chain in allelic exclusion at the TCRbeta locus, a functionally rearranged TCRbeta minigene was introduced into pTalpha-/- and pTalpha+/- embryonic stem cells, which were subsequently assayed by RAG-2-deficient blastocyst complementation. In the absence of pTalpha, expression of the transgenic TCRbeta inhibited rearrangement of the endogenous TCRbeta locus to an extent similar to that seen in normal TCRbeta transgenic mice, suggesting that pTalpha may not be required for signaling allelic exclusion at the TCRbeta locus.

Mice deficient in the orphan receptor steroidogenic factor 1 lack adrenal glands and gonads but express P450 side-chain-cleavage enzyme in the placenta and have normal embryonic serum levels of corticosteroids.

The orphan nuclear receptor steroidogenic factor 1 (SF-1) is expressed in the adrenal cortex and gonads and regulates the expression of several P450 steroid hydroxylases in vitro. We examined the role of SF-1 in the adrenal glands and gonads in vivo by a targeted disruption of the mouse SF-1 gene. All SF-1-deficient mice died shortly after delivery. Their adrenal glands and gonads were absent, and persistent Mullerian structures were found in all genotypic males. While serum levels of corticosterone in SF-1-deficient mice were diminished, levels of adrenocorticotropic hormone (ACTH) were elevated, consistent with intact pituitary corticotrophs. Intrauterine survival of SF-1-deficient mice appeared normal, and they had normal serum level of corticosterone and ACTH, probably reflecting transplacental passage of maternal steroids. We tested whether SF-1 is required for P450 side-chain-cleavage enzyme (P450scc) expression in the placenta, which expresses both SF-1 and P450scc, and found that in contrast to its strong activation of the P450scc gene promoter in vitro, the absence of SF-1 had no effect on P450scc mRNA levels in vivo. Although the region targeted by our disruption is shared by SF-1 and by embryonal long terminal repeat-binding protein (ELP), a hypothesized alternatively spliced product, we believe that the observed phenotype reflects absent SF-1 alone, as PCR analysis failed to detect ELP transcripts in any mouse tissue, and sequences corresponding to ELP are not conserved across species. These results confirm that SF-1 is an important regulator of adrenal and gonadal development, but its regulation of steroid hydroxylase expression in vivo remains to be established.

Selective neuronal toxicity of cocaine in embryonic mouse brain cocultures.

Cocaine exposure in utero causes severe alterations in the development of the central nervous system. To study the basis of these teratogenic effects in vitro, we have used cocultures of neurons and glial cells from mouse embryonic brain. Cocaine selectively affected embryonic neuronal cells, causing first a dramatic reduction of both number and length of neurites and then extensive neuronal death. Scanning electron microscopy demonstrated a shift from a multipolar neuronal pattern towards bi- and unipolarity prior to the rounding up and eventual disappearance of the neurons. Selective toxicity of cocaine on neurons was paralleled by a concomitant decrease of the culture content in microtubule-associated protein 2 (MAP2), a neuronal marker measured by solid-phase immunoassay. These effects on neurons were reversible when cocaine was removed from the culture medium. In contrast, cocaine did not affect astroglial cells and their glial fibrillary acidic protein (GFAP) content. Thus, in embryonic neuronal-glial cell cocultures, cocaine induces major neurite perturbations followed by neuronal death without affecting the survival of glial cells. Provided similar neuronal alterations are produced in the developing human brain, they could account for the qualitative or quantitative defects in neuronal pathways that cause a major handicap in brain function following in utero exposure to cocaine.

Identification of a Drosophila muscle development gene with structural homology to mammalian early growth response transcription factors.

In Drosophila, stripe (sr) gene function is required for normal muscle development. Some mutations disrupt embryonic muscle development and are lethal. Other mutations cause total loss of only a single muscle in the adult. Molecular analysis shows that sr encodes a predicted protein containing a zinc finger motif. This motif is homologous to the DNA binding domains encoded by members of the early growth response (egr) gene family. In mammals, expression of egr genes is induced by intercellular signals, and there is evidence for their role in many developmental events. The identification of sr as an egr gene and its pattern of expression suggest that it functions in muscle development via intercellular communication.

Impairment of axonal development and of synaptogenesis in hippocampal neurons of synapsin I-deficient mice.

Synapsin I, the most abundant of all neuronal phosphoproteins, is enriched in synaptic vesicles. It has been hypothesized to regulate synaptogenesis and neurotransmitter release from adult nerve terminals. The evidence for such roles has been highly suggestive but not compelling. To evaluate the possible involvement of synapsin I in synaptogenesis and in the function of adult synapses, we have generated synapsin I-deficient mice by homologous recombination. We report herein that outgrowth of predendritic neurites and of axons was severely retarded in the hippocampal neurons of embryonic synapsin I mutant mice. Furthermore, synapse formation was significantly delayed in these mutant neurons. These results indicate that synapsin I plays a role in regulation of axonogenesis and synaptogenesis.

Exogenous vascular endothelial growth factor induces malformed and hyperfused vessels during embryonic neovascularization.

Vascular endothelial growth factor (VEGF) is a potent and specific endothelial mitogen that is able to induce angiogenesis in vivo [Leung, D. W., Cachianes, G., Kuang, W.-J., Goeddel, D. V. & Ferrara, N. (1989) Science 246 1306-1309]. To determine if VEGF also influences the behavior of primordial endothelial cells, we used an in vivo vascular assay based on the de novo formation of vessels. Japanese quail embryos injected with nanomolar quantities of the 165-residue form of VEGF at the onset of vasculogenesis exhibited profoundly altered vessel development. In fact, the overall patterning of the vascular network was abnormal in all VEGF-injected embryos. The malformations were attributable to two specific endothelial cell activities: (i) inappropriate neovascularization in normally avascular areas and (ii) the unregulated, excessive fusion of vessels. In the first instance, supernumerary vessels directly linked the inflow channel of the heart to the aortic outflow channel. The second aberrant activity led to the formation of vessels with abnormally large lumens. Ultimately, unregulated vessel fusion generated massive vascular sacs that obliterated the identity of individual vessels. These observations show that exogenous VEGF has an impact on the behavior of primordial endothelial cells engaged in vasculogenesis, and they strongly suggest that endogenous VEGF is important in vascular patterning and regulation of vessel size (lumen formation).

WW6: an embryonic stem cell line with an inert genetic marker that can be traced in chimeras.

Mutant mice produced by gene targeting in embryonic stem (ES) cells often have a complex or embryonic lethal phenotype. In these cases, it would be helpful to identify tissues and cell types first affected in mutant embryos by following the contribution to chimeras of ES cells homozygous for the mutant allele. Although a number of strategies for following ES cell development in vivo have been reported, each has limitations that preclude its general application. In this paper, we describe ES cell lines that can be tracked to every nucleated cell type in chimeras at all developmental stages. These lines were derived from blastocysts of mice that carry an 11-Mb beta-globin transgene on chromosome 3. The transgene is readily detected by DNA in situ hybridization, providing an inert, nuclear-localized marker whose presence is not affected by transcriptional or translational controls. The "WW" series of ES lines possess the essential features of previously described ES lines, including giving rise to a preponderance of male chimeras, all of which have to date exhibited germ-line transmission. In addition, clones selected for single or double targeting events form strong chimeras, demonstrating the feasibility of using WW6 cells to identify phenotypes associated with the creation of a null mutant.

Atrial-like phenotype is associated with embryonic ventricular failure in retinoid X receptor alpha -/- mice.

We have recently characterized a cardiac model of ventricular chamber defects in retinoid X receptor alpha (RXR alpha) homozygous mutant (-/-) gene-targeted mice. These mice display generalized edema, ventricular chamber hypoplasia, and muscular septal defects, and they die at embryonic day 15. To substantiate our hypothesis that the embryos are dying of cardiac pump failure, we have used digital bright-field and fluorescent video microscopy and in vivo microinjection of fluorescein-labeled albumin to analyze cardiac function. The affected embryos showed depressed ventricular function (average left ventricular area ejection fraction, 14%), ventricular septal defects, and various degrees of atrioventricular block not seen in the RXR alpha wild-type (+/+) and heterozygous (+/-) littermates (average left ventricular area ejection fraction, 50%). The molecular mechanisms involved in these ventricular defects were studied by evaluating expression of cardiac-specific genes known to be developmentally regulated. By in situ hybridization, aberrant, persistent expression of the atrial isoform of myosin light chain 2 was identified in the ventricles. We hypothesize that retinoic acid provides a critical signal mediated through the RXR alpha pathway that is required to allow progression of development of the ventricular region of the heart from its early atrial-like form to the thick-walled adult ventricle. The conduction system disturbances found in the RXR alpha -/- embryos may reflect a requirement of the developing conduction system for the RXR alpha signaling pathway, or it may be secondary to the failure of septal development.

A transgene coding for a human insulin analog has a mitogenic effect on murine embryonic beta cells.

We have investigated the mitogenic effect of three mutant forms of human insulin on insulin-producing beta cells of the developing pancreas. We examined transgenic embryonic and adult mice expressing (i) human [AspB10]-proinsulin/insulin ([AspB10]ProIN/IN), produced by replacement of histidine by aspartic acid at position 10 of the B chain and characterized by an increased affinity for the insulin receptor; (ii) human [LeuA3]insulin, produced by the substitution of leucine for valine in position 3 of the A chain, which exhibits decreased receptor binding affinity; and (iii) human [LeuA3, AspB10]insulin "double" mutation. During development, beta cells of AspB10 embryos were twice as abundant and had a 3 times higher rate of proliferation compared with beta cells of littermate controls. The mitogenic effect of [AspB10]ProIN/IN was specific for embryonic beta cells because the rate of proliferation of beta cells of adults and of glucagon (alpha) cells and adrenal chromaffin cells of embryos was similar in AspB10 mice and controls. In contrast to AspB10 embryos, the number of beta cells in the LeuA3 and "double" mutant lines was similar to the number in controls. These findings indicate that the [AspB10]ProIN/IN analog increased the rate of fetal beta-cell proliferation. The mechanism or mechanisms that mediate this mitogenic effect remain to be determined.

Expression of pax-6 during urodele eye development and lens regeneration.

Regeneration of eye tissues, such as lens, seen in some urodeles involves dedifferentiation of the dorsal pigmented epithelium and subsequent differentiation to lens cells. Such spatial regulation implies possible action of genes known to be specific for particular cell lineages and/or axis. Hox genes have been the best examples of genes for such actions. We have, therefore, investigated the possibility that such genes are expressed during lens regeneration in the newt. The pax-6 gene (a gene that contains a homeobox and a paired box) has been implicated in the development of the eye and lens determination in various species ranging from Drosophila to human and, because of these properties, could be instrumental in the regeneration of the urodele eye tissues as well. We present data showing that pax-6 transcripts are present in the developing and the regenerating eye tissues. Furthermore, expression in eye tissues, such as in retina, declines when a urodele not capable of lens regeneration (axolotl) surpasses the embryonic stages. Such a decline is not seen in adult newts capable of lens regeneration. This might indicate a vital role of pax-6 in newt lens regeneration.

Differential tubulogenic and branching morphogenetic activities of growth factors: implications for epithelial tissue development.

At least two kidney epithelial cell lines, the Madin-Darby canine kidney (MDCK) and the murine inner medullary collecting duct line mIMCD-3, can be induced to form branching tubular structures when cultured with hepatocyte growth factor (HGF) plus serum in collagen I gels. In our studies, whereas MDCK cells remained unable to form tubules in the presence of serum alone, mIMCD-3 cells formed impressive branching tubular structures with apparent lumens, suggesting the existence of specific factors in serum that are tubulogenic for mIMCD-3 cells but not for MDCK cells. Since normal serum does not contain enough HGF to induce tubulogenesis, these factors appeared to be substances other than HGF. This was also suggested by another observation: when MDCK cells or mIMCD-3 cells were cocultured under serum-free conditions with the embryonic kidney, both cell types formed branching tubular structures similar to those induced by HGF; however, only in the case of MDCK cells could this be inhibited by neutralizing antibodies against HGF. Thus, the embryonic kidney produces growth factors other than HGF capable of inducing tubule formation in the mIMCD-3 cells. Of a number of growth factors examined, transforming growth factor alpha (TGF-alpha) and epidermal growth factor (EGF) were found to be tubulogenic for mIMCD-3 cells. Whereas only HGF was a potent tubulogenic factor for MDCK cells, HGF, TGF-alpha, and EGF were potent tubulogenic factors for mIMCD-3 cells. Nevertheless, there were marked differences in the capacity of these tubulogenic factors to induce tubulation as well as branching events in those tubules that did form (HGF >> TGF-alpha > EGF). Thus, at least three different growth factors can induce tubulogenesis and branching in a specific epithelial cell in vitro (though to different degrees), and different epithelial cells that are capable of forming branching tubular structures demonstrate vastly different responses to tubulogenic growth factors. The results are discussed in the context of branching morphogenesis during epithelial tissue development.

The role of insulin-like growth factor II and its receptor in mouse preimplantation development

Insulin-like growth factor II (IGF-II) and its receptor, the IGF-II/mannose-6-phosphate (IGF-II/M6P) receptor, are first expressed from the zygotic genome at the two-cell stage of mouse development. However, their role is not clearly defined. Insulin-like growth factor II is believed to mediate growth through the heterologous type 1 IGF and insulin receptors, whereas the IGF-II/M6P receptor is believed to act as a negative regulator of somatic growth by limiting the availability of excess levels of IGF-II. These studies demonstrate that IGF-II does have a role in growth regulation in the early embryo through the IGF-II/M6P receptor. Insulin-like growth factor II stimulated cleavage rate in two-cell embryos in vitro. Moreover, this receptor is required for the glycaemic response of two-cell embryos to IGF-II and for normal progression of early embryos to the blastocyst stage. Improved development of embryos in crowded culture supports the concept of an endogenous embryonic paracrine activity that enhances cell proliferation. These responses indicate that the IGF-II/M6P receptor is functional and likely to participate in such a regulatory circuit. The functional role of IGF-II and its receptor is discussed with reference to regulation of early development.

Characterizing embryonic gene expression patterns in the mouse using nonredundant sequence-based selection

This article investigates the expression patterns of 160 genes that are expressed during early mouse development. The cDNAs were isolated from 7.5 d postcoitum (dpc) encloderm, a region that comprises visceral encloderm (VE), definitive encloderm, and the node-tissues that are required for the initial steps of axial specification and tissue patterning in the mouse. To avoid examining the same gene more than once, and to exclude potentially ubiquitously expressed housekeeping genes, cDNA sequence was derived from 1978 clones of the Endoderm library. These yielded 1440 distinct cDNAs, of which 123 proved to be novel in the mouse. In situ hybridization analysis was carried out on 160 of the cDNAs, and of these, 29 (18%) proved to have restricted expression patterns.