Neural progenitors of the mouse forebrain can be propagated in vitro as neurospheres in the presence of bFGF and EGF. However, less is understood whether regional characteristics or developmental stage properties of these cells are maintained in neurosphere cultures. Here we show that the original cell fate is lost in neurosphere cultures. We isolated neural progenitors from the dorsal telencephalon of D6-GFP mice and cultured them in vitro. The expression profile was specifically changed in cultured cells in just three passages. Markers of the dorsal forebrain were downregulated and several ventrally-expressed genes were induced. The altered gene expression led to a profound phenotypic change of cultured cells. D6-GFP positive cortical progenitors produce excitatory neurons in the cortex and few astrocytes in vivo but after culture in vitro, these cells differentiate into many astrocytes and also oligodendrocytes and inhibitory neurons. Wnt signaling in cultured neurospheres was downregulated in the same manner as other dorsal markers but dominant active Wnt signaling slowed down the loss of the dorsal identity in neurospheres.

Wnt signaling is involved in numerous processes during vertebrate CNS development. In this study, we used conditional Cre/loxP system in mouse to ablate or activate ß-catenin in the telencephalon in two time windows: before and after the onset of neurogenesis. We show that β-catenin mediated Wnt signals are required to maintain the molecular identity of the pallium. Inactivation of ß-catenin in the telencephalon before neurogenesis results in downregulated expression of dorsal markers Emx1, Emx2 and Ngn2, and in ectopic up-regulation of ventral markers Gsh2, Mash1 and Dlx2 in the pallium. In contrast, ablation of ß-catenin after the onset of cortical neurogenesis (E11.5) does not result in a dorso-ventral fate shift. In addition, activation of canonical Wnt signaling in the subpallium leads to a repression of ventral telencephalic cell identities as shown by the down-regulation of subpallial markers Dlx2, Nkx2.1, Gsh2, Olig2 and Mash1. This was accompanied with an expansion of dorsal identities ventrally as shown by the expanded expression domains of pallial markers Pax6 and Ngn2. Thus, our data suggest that canonical Wnt signals are involved in maintaining the identity of the pallium by controlling expression of dorsal markers and by suppressing ventral programs from being activated in pallial progenitor cells.

In the present work, the expression patterns of the Wnt antagonists of the Dickkopf (Dkk) family were characterized in the developing mouse forebrain. In situ hybridisation on sections from E12 embryos showed an expression of dkk2 in the thalamus and dkk3 in the cortical hem and thalamus. At later developmental stages (E15.5, E17.5, and P0), little or no expression of dkk1, dkk2, and dkk4 was found in the forebrain, while dkk3 expression was detected in the ventricular zone (VZ) of the lateral and III ventricles, cortical neurons, migrating cells of the primary and secondary dentate migration, and the neuroblastic layer of the eye. In the adult forebrain, dkk3 expression was detected in the lateral VZ, pyramidal neurons of the hippocampus, and cortical neurons. We also provide evidence indicating that only dkk1 and dkk4, along with two other Wnt antagonists axin2 and wif1, but not dkk2 and dkk3, are involved in a feedback mechanism to restrain Wnt signalling in transgenic mice carrying a conditional augmentation of beta-catenin in the forebrain.

Mouse Dach1 is a nuclear factor that is expressed during development in restricted areas of the central nervous system, neural crest, and limb buds. Its Drosophila homologue dachshund plays a role in differentiation of the eye imaginal disc, in leg morphogenesis, and in controlling neural differentiation in the mushroom bodies of the insect brain. Mouse Dach1 null homozygous survive pregnancy but become cyanotic after birth and subsequently die within 24 hr. In this report, the brain of Dach1 mutants was analyzed. Examination of mRNA expression of the central neuropeptides oxytocin, vasopressin, thyrotropin‐releasing hormone, growth hormone releasing hormone, and somatostatin revealed no difference between wild‐type and mutant newborn brains. Furthermore, no significant difference in cell proliferation as well as in the distribution of neurons, glia, radial glia, and neuronal progenitors was detected in the developing forebrain. Dach1‐positive cells, which were visualized with Enhanced Green Fluorescent Protein (EGFP), show similar distribution and axonal projections in the cortex and hippocampus in mutants and wild‐type controls. Neural stem cells derived from mutant and wild‐type newborn brains display similar growth kinetics when cultivated in vitro. © 2002 Wiley‐Liss, Inc.

β-Catenin plays a pivotal role in Wnt signaling during embryogenesis and is a component of adherens junctions. Since targeted disruption of the β-catenin gene is lethal at gastrulation we have used a D6-Cre mouse line for conditional inactivation of β-catenin in the mouse cerebral cortex and hippocampus after embryonic day (E) 10.5. In D6-Cre floxed β-catenin mice, hippocampal CA1–CA2 fields are disrupted in similar manner as in Wnt-3a and LEF-1 mutants. The cortex of D6-Cre floxed β-catenin mutants is strongly affected which contrasts with the normal cortex observed in Wnt-3a and LEF-1 mutants. Severe abnormalities in the organization of the neuroepithelium are observed that include disrupted interkinetic nuclear migration, loss of adherens junctions, impaired radial migration of neurons toward superficial layers and decreased cell proliferation after E15.5. At newborn stage, a premature disassembly of the radial glial scaffold and increased numbers of astrocytes are found in the cortex.

Drosophila dachshund is involved in development of eye and limbs and in the development of mushroom bodies, a brain structure required for learning and memory in flies. Its mouse homologue mDach1 is expressed in various embryonic tissues, including limbs, the eye, the dorsal spinal cord and the forebrain. We have isolated a forebrain-specific 2.5-kb enhancer element termed D6 from the mouse mDach1 gene and created D6-LacZ and D6-green fluorescent protein (GFP) reporter gene mouse lines. In embryonic stages, the D6 enhancer activity is first detected at embryonic day 10.5 in scattered cells of the outbuldging cortical vesicles. By embryonic day 12.5, D6 activity expands throughout the developing neocortex and the hippocampus. In the adult mouse brain, D6 enhancer is active in neurons of the cortical plate, in the CA1 layer of the hippocampus and in cells of the subventricular zone and the ventricular ependymal zone. Adult mice also show D6 activity in the olfactory bulb and in the mamillary nucleus. Cultured D6-positive cells, which were derived from embryonic and postnatal brains, show characteristics of neural stem cells. They form primary and secondary neurospheres that differentiate into neurons and astrocytes as examined by cell-specific markers.Our results show that D6 enhancer exerts highly tissue-specific activity in the neurons of the neocortex and hippocampus and in neural stem cells. Moreover, the fluorescence cell sorting of D6-GFP cells from embryonic and postnatal stages allows specific selection of primary neural progenitors and their analysis.

We have created a transgenic mouse line that expresses Cre recombinase under the control of the novel mouse promoter/enhancer D6. We describe the expression pattern of D6-Cre in a Gtrosa26 reporter background as assayed by LacZ activity. The enhancer activity starts at 10.5 days post-coitum in the telencephalon and is at the later embryonic stages highly restricted to the hippocampus and the neocortex. In adult mice D6-derived cells are found in cortical layers II-VI, in the granular cells of the dentate gyrus and in hippocampal fields CA1-CA3. D6-Cre activity is also detected in the ependymal and subependymal zone of the lateral ventricles which is known to harbor neural stem cells.

Using a yeast two hybrid system and pull-down assays we demonstrate that mouse Dac (mDac) specifically binds to mouse ubiquitin-conjugating enzyme mUbc9. In contrast to a direct interaction between Drosophila dachshund (dac) and eyes absent (eya)gene products, we cannot detect by the same methods that mDac binds to mEya2, a functional mouse homologue of the Drosophila Eya. Immunostaining of various cell lines that were transfected with mDac reveals that mDac protein is found predominantly in the nucleus but translocates to the cytoplasm and condensates along the nuclear membrane in a cell-cycle dependent manner. Deletion analysis of mDac show the intracellular localization and protein stability correlates with the binding to mUbc9. The C-terminal half of mDac, which associates with mUbc9, remains cytoplasmic and is degraded in proteasome whereas the non-interacting N-terminus is exclusively nuclear and more stable than the full-length mDac or its C-terminal portion. In situ hybridization on whole-mount embryos or tissue sections detects mUbc9 transcripts in complementary and overlapping areas with mDac expression, particularly in the proliferation zone of the limb buds, the spinal cord and forebrain. Mouse embryos stained with an anti-mDac antibody document that mDac is localized both in the nucleus and the cytoplasm with a cytoplasmic predominance in migrating neural crest cells. In the proliferation zone, visible nuclear envelopes are not formed and mDac is detected throughout the cells.