Explore the vast range of titles available.

Vascular development and maintenance are controlled by a complex transcriptional program, which integrates both extracellular and intracellular signals in endothelial cells. Here we study the roles of three closely related SoxF family transcription factors-Sox7, Sox17, and Sox18 -in the developing and mature mouse vasculature using targeted gene deletion on a mixed C57/129/CD1 genetic background. In the retinal vasculature, each SoxF gene exhibits a distinctive pattern of expression in different classes of blood vessels. On a mixed genetic background, vascular endothelial-specific deletion of individual SoxF genes has little or no effect on vascular architecture or differentiation, a result that can be explained by overlapping function and by reciprocal regulation of gene expression between Sox7 and Sox17. By contrast, combined deletion of Sox7, Sox17, and Sox18 at the onset of retinal angiogenesis leads to a dense capillary plexus with a nearly complete loss of radial arteries and veins, whereas the presence of a single Sox17 allele largely restores arterial identity, as determined by vascular smooth muscle cell coverage. In the developing retina, expression of all three SoxF genes is reduced in the absence of Norrin/Frizzled4-mediated canonical Wnt signaling, but SoxF gene expression is unaffected by reduced VEGF signaling in response to deletion of Neuropilin1 (Npn1). In adulthood, Sox7, Sox17, and Sox18 act in a largely redundant manner to maintain blood vessel function, as adult onset vascular endothelial-specific deletion of all three SoxF genes leads to massive edema despite nearly normal vascular architecture. These data reveal critical and partially redundant roles for Sox7, Sox17 and Sox18 in vascular growth, differentiation, and maintenance.

Canonical WNT signaling is required for proper vascularization of the CNS during embryonic development. Here, we used mice with targeted mutations in genes encoding canonical WNT pathway members to evaluate the exact contribution of these components in CNS vascular development and in specification of the blood-brain barrier (BBB) and blood-retina barrier (BRB). We determined that vasculature in various CNS regions is differentially sensitive to perturbations in canonical WNT signaling. The closely related WNT signaling coreceptors LDL receptor-related protein 5 (LRP5) and LRP6 had redundant functions in brain vascular development and barrier maintenance; however, loss of LRP5 alone dramatically altered development of the retinal vasculature. The BBB in the cerebellum and pons/interpeduncular nuclei was highly sensitive to decrements in canonical WNT signaling, and WNT signaling was required to maintain plasticity of barrier properties in mature CNS vasculature. Brain and retinal vascular defects resulting from ablation of Norrin/Frizzled4 signaling were ameliorated by stabilizing β-catenin, while inhibition of β-catenin-dependent transcription recapitulated the vascular development and barrier defects associated with loss of receptor, coreceptor, or ligand, indicating that Norrin/Frizzled4 signaling acts predominantly through β-catenin-dependent transcriptional regulation. Together, these data strongly support a model in which identical or nearly identical canonical WNT signaling mechanisms mediate neural tube and retinal vascularization and maintain the BBB and BRB.

The molecular basis of Wnt-Frizzled specificity is a central question in developmental biology. Reck, a multi-domain and multi-functional glycosylphosphatidylinositol-anchored protein, specifically enhances beta-catenin signaling by Wnt7a and Wnt7b in cooperation with the 7-transmembrane protein Gpr124. Among amino acids that distinguish Wnt7a and Wnt7b from other Wnts, two clusters are essential for signaling in a Reck- and Gpr124-dependent manner. Both clusters are far from the site of Frizzled binding: one resides at the amino terminus and the second resides in a protruding loop. Within Reck, the fourth of five tandem repeats of an unusual domain with six-cysteines (the CC domain) is essential for Wnt7a stimulation: substitutions P256A and W261A in CC4 eliminate this activity without changing protein abundance or surface localization. Mouse embryos carrying ReckP256A,W261A have severe defects in forebrain angiogenesis, providing the strongest evidence to date that Reck promotes CNS angiogenesis by specifically stimulating Wnt7a and Wnt7b signaling.

Beta-catenin (i.e., canonical Wnt) signaling controls CNS angiogenesis and the blood-brain and blood-retina barriers. To explore the role of the Discs large/membrane-associated guanylate kinase (Dlg/MAGUK) family of scaffolding proteins in beta-catenin signaling, we studied vascular endothelial cell (EC)-specific knockout of Dlg1/SAP97. EC-specific loss of Dlg1 produces a retinal vascular phenotype that closely matches the phenotype associated with reduced beta-catenin signaling, synergizes with genetically-directed reductions in beta-catenin signaling components, and can be rescued by stabilizing beta-catenin in ECs. In reporter cells with CRISPR/Cas9-mediated inactivation of Dlg1, transfection of Dlg1 enhances beta-catenin signaling ~4 fold. Surprisingly, Frizzled4, which contains a C-terminal PDZ-binding motif that can bind to Dlg1 PDZ domains, appears to function independently of Dlg1 in vivo. These data expand the repertoire of Dlg/MAGUK family functions to include a role in beta-catenin signaling, and they suggest that proteins other than Frizzled receptors interact with Dlg1 to enhance beta-catenin signaling.

The Amyloid Precursor Protein (APP) and Contactin (CNTN) families of cell-surface proteins have been intensively studied in the context of neural development and neuropsychiatric diseases. Earlier studies demonstrated both genetic and biochemical interactions between the extracellular domains of APP and CNTN3, but their precise binding interfaces were not defined. In the present study, we have used binding assays between APP-alkaline phosphatase (AP) fusion proteins and CNTN-Fc fusion proteins, together with alanine substitution mutagenesis, to show that: (i) the second Fibronectin domain (Fn(2)) in CNTN3 mediates APP binding; (ii) the copper binding domain (CuBD) in APP mediates CNTN3 binding; and (iii) the most important amino acids for APP-CNTN3 binding reside on one face of CNTN3-Fn(2) and on one face of APP-CuBD. These experiments define the regions of direct contact that mediate the binding interaction between APP and CNTN3.

Planar cell polarity (PCP) signaling controls the global orientation of surface structures, such as hairs and bristles, in both vertebrates and invertebrates. In Frizzled6(-/-) (Fz6(-/-)) mice, hair follicle orientations on the head and back are nearly random at birth, but reorient during early postnatal development to eventually generate a nearly parallel anterior-to-posterior array. We report the identification of a naturally occurring exon 5 deletion in Astrotactin2 (Astn2) that acts as a recessive genetic modifier of the Fz6(-/-) hair patterning phenotype. A genetically engineered Astn2 exon 5 deletion recapitulates the modifier phenotype. In Fz6(-/-);Astn2(ex5del/del) mice, hair orientation on the back is subtly biased from posterior-to-anterior, leading to a 180-degree orientation reversal in mature mice. These experiments suggest that Astn2, an endosomal membrane protein, modulates PCP signaling.

Endothelin signaling is required for neural crest migration and homeostatic regulation of blood pressure. Here, we report that constitutive overexpression of Endothelin-2 (Edn2) in the mouse retina perturbs vascular development by inhibiting endothelial cell migration across the retinal surface and subsequent endothelial cell invasion into the retina. Developing endothelial cells exist in one of two states: tip cells at the growing front and stalk cells in the vascular plexus behind the front. This division of endothelial cell states is one of the central organizing principles of angiogenesis. In the developing retina, Edn2 overexpression leads to overproduction of endothelial tip cells by both morphologic and molecular criteria. Spatially localized overexpression of Edn2 produces a correspondingly localized endothelial response. Edn2 overexpression in the early embryo inhibits vascular development at midgestation, but Edn2 overexpression in developing skin and brain has no discernible effect on vascular structure. Inhibition of retinal angiogenesis by Edn2 requires expression of Endothelin receptor A but not Endothelin receptor B in the neural retina. Taken together, these observations imply that the neural retina responds to Edn2 by synthesizing one or more factors that promote the endothelial tip cell state and inhibit angiogenesis. The response to Edn2 is sufficiently potent that it overrides the activities of other homeostatic regulators of angiogenesis, such as Vegf.

Pharmacologic control of Cre-mediated recombination using tamoxifen-dependent activation of a Cre-estrogen receptor ligand binding domain fusion protein [CreER(T)] is widely used to modify and/or visualize cells in the mouse. We describe here two new mouse lines, constructed by gene targeting to the Rosa26 locus to facilitate Cre-mediated cell modification. These lines should prove particularly useful in the context of sparse labeling experiments. The R26rtTACreER line provides ubiquitous expression of CreER under transcriptional control by the tetracycline reverse transactivator (rtTA); dual control by doxycycline and tamoxifen provides an extended dynamic range of Cre-mediated recombination activity. The R26IAP line provides high efficiency Cre-mediated activation of human placental alkaline phosphatase (hPLAP), complementing the widely used, but low efficiency, Z/AP line. By crossing with mouse lines that direct cell-type specific CreER expression, the R26IAP line has been used to produce atlases of labeled cholinergic and catecholaminergic neurons in the mouse brain. The R26IAP line has also been used to visualize the full morphologies of retinal dopaminergic amacrine cells, among the largest neurons in the mammalian retina. The two new mouse lines described here expand the repertoire of genetically engineered mice available for controlled in vivo recombination and cell labeling using the Cre-lox system.

In mammals, genetically-directed cell labeling technologies have not yet been applied to the morphologic analysis of neurons with very large and complex arbors, an application that requires extremely sparse labeling and that is only rendered practical by limiting the labeled population to one or a few predetermined neuronal subtypes. In the present study we have addressed this application by using CreER technology to non-invasively label very small numbers of neurons so that their morphologies can be fully visualized. Four lines of IRES-CreER knock-in mice were constructed to permit labeling selectively in cholinergic or catecholaminergic neurons [choline acetyltransferase (ChAT)-IRES-CreER or tyrosine hydroxylase (TH)-IRES-CreER], predominantly in projection neurons [neurofilament light chain (NFL)-IRES-CreER], or broadly in neurons and some glia [vesicle-associated membrane protein2 (VAMP2)-IRES-CreER]. When crossed to the Z/AP reporter and exposed to 4-hydroxytamoxifen in the early postnatal period, the number of neurons expressing the human placental alkaline phosphatase reporter can be reproducibly lowered to fewer than 50 per brain. Sparse Cre-mediated recombination in ChAT-IRES-CreER;Z/AP mice shows the full axonal and dendritic arbors of individual forebrain cholinergic neurons, the first time that the complete morphologies of these very large neurons have been revealed in any species. Sparse genetically-directed, cell type-specific neuronal labeling with IRES-creER lines should prove useful for studying a wide variety of questions in neuronal development and disease.

Disruption of the Frizzled3 (Fz3) gene leads to defects in axonal growth in the VIIth and XIIth cranial motor nerves, the phrenic nerve, and the dorsal motor nerve in fore- and hindlimbs. In Fz3−/− limbs, dorsal axons stall at a precise location in the nerve plexus, and, in contrast to the phenotypes of several other axon path-finding mutants, Fz3−/− dorsal axons do not reroute to other trajectories. Affected motor neurons undergo cell death 2 days prior to the normal wave of developmental cell death that coincides with innervation of muscle targets, providing in vivo evidence for the idea that developing neurons with long-range axons are programmed to die unless their axons arrive at intermediate targets on schedule. These experiments implicate planar cell polarity (PCP) signaling in motor axon growth and they highlight the question of how PCP proteins, which form cell–cell complexes in epithelia, function in the dynamic context of axonal growth. For the nervous system to become wired up correctly, neurons within the developing embryo must project over long distances to form connections with remote targets. They do this by lengthening their axons—the ‘cables’ along which electrical signals flow—and some axons in adult humans can grow to be more than 1 metre long. This type of long-range pathfinding activity is particularly common for neurons that control movement, as many of these neurons must establish connections with muscles that are some distance away from the brain. For example, motor neurons in the brainstem form connections with muscles in the face to control facial expressions, while motor neurons in parts of the spinal cord project to muscles in the limbs. Multiple signaling pathways tell the developing axons which direction to grow en route to their final targets. Now, Hua et al. have shown that an evolutionarily conserved protein called Frizzled3 is also involved in this process. In mouse embryos that lacked Frizzled3, the motor nerves that control breathing and limb movements were thinner than those in normal mice. In the mutant animals, many motor axons failed to form connections with their targets. Instead, these axons came to an abrupt halt midway along their intended paths and the neurons from which they originated died soon afterwards. These experiments support the idea that developing neurons are programmed to die unless their axons progress on the appropriate schedule. As well as increasing our knowledge of the networks of connections that form within the developing mammalian nervous system, the work of Hua et al. provides new insights into some of the molecular mechanisms by which these connections are established.