Pierre Mattar

Phosphorylation of Neurogenin2 Specifies the Migration Properties and the Dendritic Morphology of Pyramidal Neurons in the Neocortex

The molecular mechanisms specifying the dendritic morphology of different neuronal subtypes are poorly understood. Here we demonstrate that the bHLH transcription factor Neurogenin2 (Ngn2) is both necessary and sufficient for specifying the dendritic morphology of pyramidal neurons in vivo by specifying the polarity of its leading process during the initiation of radial migration. The ability of Ngn2 to promote a polarized leading process outgrowth requires the phosphorylation of a single tyrosine residue at position 241, an event that is neither involved in Ngn2 direct transactivation properties nor its proneural function. Interestingly, the migration defect observed in the Ngn2 knockout mouse and in progenitors expressing the Ngn2(Y241F) mutation can be rescued by inhibiting the activity of the small-GTPase RhoA in cortical progenitors. Our results demonstrate that Ngn2 coordinates the acquisition of the radial migration properties and the unipolar dendritic morphology characterizing pyramidal neurons through molecular mechanisms distinct from those mediating its proneural activity.

Basic Helix-Loop-Helix Transcription Factors Cooperate To Specify a Cortical Projection Neuron Identity

ABSTRACT Several transcription factors are essential determinants of a cortical projection neuron identity, but their mode of action (instructive versus permissive) and downstream genetic cascades remain poorly defined. Here, we demonstrate that the proneural basic helix-loop-helix (bHLH) gene Ngn2 instructs a partial cortical identity when misexpressed in ventral telencephalic progenitors, inducing ectopic marker expression in a defined temporal sequence, including early (24 h; Nscl2), intermediate (48 h; BhlhB5), and late (72 h; NeuroD, NeuroD2, Math2, and Tbr1) target genes. Strikingly, cortical gene expression was much more rapidly induced by Ngn2 in the dorsal telencephalon (within 12 to 24 h). We identify the bHLH gene Math3 as a dorsally restricted Ngn2 transcriptional target and cofactor, which synergizes with Ngn2 to accelerate target gene transcription in the cortex. Using a novel in vivo luciferase assay, we show that Ngn2 generates only ∼60% of the transcriptional drive in ventral versus dorsal telencephalic domains, an activity that is augmented by Math3, providing a mechanistic basis for regional differences in Ngn2 function. Cortical bHLH genes thus cooperate to control transcriptional strength, thereby temporally coordinating downstream gene expression.

Validating In Utero Electroporation for the Rapid Analysis of Gene Regulatory Elements in the Murine Telencephalon

With the ultimate goal of understanding how genetic modules have evolved in the telencephalon, we set out to modernize the functional analysis of cross‐species cis‐regulatory elements in mouse. In utero electroporation is rapidly replacing transgenesis as the method of choice for gain‐ and loss‐of‐function studies in the murine telencephalon, but the application of this technique to the analysis of transcriptional regulation has yet to be fully explored and exploited. To empirically define the developmental stages required to target specific populations of neurons in the dorsal telencephalon, or pallium, which gives rise to the neocortex in mouse, we performed a temporal and spatial analysis of the migratory properties of electroporated versus birth‐dated cells. Next, we compared the activities of two known Ngn2 enhancers via transgenesis and in utero electroporation, demonstrating that the latter technique more faithfully reports the endogenous telencephalic expression pattern observed in an Ngn2lacZ knock‐in line. Finally, we used this approach to test the telencephalic activities of a series of deletion constructs comprised of the zebrafish ER81 upstream regulatory region, allowing us to identify a previously uncharacterized enhancer that displays cross‐species activity in the murine piriform cortex and lateral neocortex, yet not in more medial domains of the forebrain. Taken together, our data supports the contention that in utero technology can be exploited to rapidly examine the architecture and evolution of pallial‐specific cis‐regulatory elements. Developmental Dynamics 236:1273–1286, 2007.

GSK3 temporally regulates neurogenin 2 proneural activity in the neocortex.

The neocortex is comprised of six neuronal layers that are generated in a defined temporal sequence. While extrinsic and intrinsic cues are known to regulate the sequential production of neocortical neurons, how these factors interact and function in a coordinated manner is poorly understood. The proneural gene Neurog2 is expressed in progenitors throughout corticogenesis, but is only required to specify early-born, deep-layer neuronal identities. Here, we examined how neuronal differentiation in general and Neurog2 function in particular are temporally controlled during murine neocortical development. We found that Neurog2 proneural activity declines in late corticogenesis, correlating with its phosphorylation by GSK3 kinase. Accordingly, GSK3 activity, which is negatively regulated by canonical Wnt signaling, increases over developmental time, while Wnt signaling correspondingly decreases. When ectopically activated, GSK3 inhibits Neurog2-mediated transcription in cultured cells and Neurog2 proneural activities in vivo. Conversely, a reduction in GSK3 activity promotes the precocious differentiation of later stage cortical progenitors without influencing laminar fate specification. Mechanistically, we show that GSK3 suppresses Neurog2 activity by influencing its choice of dimerization partner, promoting heterodimeric interactions with E47 (Tcfe2a), as opposed to Neurog2–Neurog2 homodimer formation, which occurs when GSK3 activity levels are low. At the functional level, Neurog2–E47 heterodimers have a reduced ability to transactivate neuronal differentiation genes compared with Neurog2–Neurog2 homodimers, both in vitro and in vivo. We thus conclude that the temporal regulation of Neurog2–E47 heterodimerization by GSK3 is a central component of the neuronal differentiation “clock” that coordinates the timing and tempo of neocortical neurogenesis in mouse.

RAS/ERK signaling controls proneural genetic programs in cortical development and gliomagenesis.

Neural cell fate specification is well understood in the embryonic cerebral cortex, where the proneural genes Neurog2 and Ascl1 are key cell fate determinants. What is less well understood is how cellular diversity is generated in brain tumors. Gliomas and glioneuronal tumors, which are often localized in the cerebrum, are both characterized by a neoplastic glial component, but glioneuronal tumors also have an intermixed neuronal component. A core abnormality in both tumor groups is overactive RAS/ERK signaling, a pro-proliferative signal whose contributions to cell differentiation in oncogenesis are largely unexplored. We found that RAS/ERK activation levels differ in two distinct human tumors associated with constitutively active BRAF. Pilocytic astrocytomas, which contain abnormal glial cells, have higher ERK activation levels than gangliogliomas, which contain abnormal neuronal and glial cells. Using in vivo gain of function and loss of function in the mouse embryonic neocortex, we found that RAS/ERK signals control a proneural genetic switch, inhibiting Neurog2 expression while inducing Ascl1, a competing lineage determinant. Furthermore, we found that RAS/ERK levels control Ascl1s fate specification properties in murine cortical progenitors–at higher RAS/ERK levels, Ascl1+ progenitors are biased toward proliferative glial programs, initiating astrocytomas, while at moderate RAS/ERK levels, Ascl1 promotes GABAergic neuronal and less glial differentiation, generating glioneuronal tumors. Mechanistically, Ascl1 is phosphorylated by ERK, and ERK phosphoacceptor sites are necessary for Ascl1s GABAergic neuronal and gliogenic potential. RAS/ERK signaling thus acts as a rheostat to influence neural cell fate selection in both normal cortical development and gliomagenesis, controlling Neurog2-Ascl1 expression and Ascl1 function.

A Conserved Regulatory Logic Controls Temporal Identity in Mouse Neural Progenitors

Neural progenitors alter their output over time to generate different types of neurons and glia in specific chronological sequences, but this process remains poorly understood in vertebrates. Here we show that Casz1, the vertebrate ortholog of the Drosophila temporal identity factor castor, controls the production of mid-/late-born neurons in the murine retina. Casz1 is expressed from mid/late stages in retinal progenitor cells (RPCs), and conditional deletion of Casz1 increases production of early-born retinal neurons at the expense of later-born fates, whereas precocious misexpression of Casz1 has the opposite effect. In both cases, cell proliferation is unaffected, indicating that Casz1 does not control the timing of cell birth but instead biases RPC output directly. Just as Drosophila castor lies downstream of the early temporal identity factor hunchback, we find that the hunchback ortholog Ikzf1 represses Casz1. These results uncover a conserved strategy regulating temporal identity transitions from flies to mammals.

Numb is required for the production of terminal asymmetric cell divisions in the developing mouse retina.

In the developing nervous system, cell diversification depends on the ability of neural progenitor cells to divide asymmetrically to generate daughter cells that acquire different identities. While much work has recently focused on the mechanisms controlling self-renewing asymmetric divisions producing a differentiating daughter and a progenitor, little is known about mechanisms regulating how distinct differentiating cell types are produced at terminal divisions. Here we study the role of the endocytic adaptor protein Numb in the developing mouse retina. Using clonal numb inactivation in retinal progenitor cells (RPCs), we show that Numb is required for normal cell-cycle progression at early stages, but is dispensable for the production of self-renewing asymmetric cell divisions. At late stages, however, Numb is no longer required for cell-cycle progression, but is critical for the production of terminal asymmetric cell divisions. In the absence of Numb, asymmetric terminal divisions that generate a photoreceptor and a non-photoreceptor cell are decreased in favor of symmetric terminal divisions generating two photoreceptors. Using live imaging in retinal explants, we show that a Numb fusion protein is asymmetrically inherited by the daughter cells of some late RPC divisions. Together with our finding that Numb antagonizes Notch signaling in late-stage RPCs, and that blocking Notch signaling in late RPCs almost completely abolishes the generation of terminal asymmetric divisions, these results suggest a model in which asymmetric inheritance of Numb in sister cells of terminal divisions might create unequal Notch activity, which in turn drives the production of terminal asymmetric divisions.

Neurog2 Simultaneously Activates and Represses Alternative Gene Expression Programs in the Developing Neocortex

Progenitor cells undergo a series of stable identity transitions on their way to becoming fully differentiated cells with unique identities. Each cellular transition requires that new sets of genes are expressed, while alternative genetic programs are concurrently repressed. Here, we investigated how the proneural gene Neurog2 simultaneously activates and represses alternative gene expression programs in the developing neocortex. By comparing the activities of transcriptional activator (Neurog2-VP16) and repressor (Neurog2-EnR) fusions to wild-type Neurog2, we first demonstrate that Neurog2 functions as an activator to both extinguish Pax6 expression in radial glial cells and initiate Tbr2 expression in intermediate neuronal progenitors. Similarly, we show that Neurog2 functions as an activator to promote the differentiation of neurons with a dorsal telencephalic (i.e., neocortical) identity and to block a ventral fate, identifying 2 Neurog2-regulated transcriptional programs involved in the latter. First, we show that the Neurog2-transcriptional target Tbr2 is a direct transcriptional repressor of the ventral gene Ebf1. Secondly, we demonstrate that Neurog2 indirectly turns off Etv1 expression, which in turn indirectly regulates the expression of the ventral proneural gene Ascl1. Neurog2 thus activates several genetic off-switches, each with distinct transcriptional targets, revealing an unappreciated level of specificity for how Neurog2 prevents inappropriate gene expression during neocortical development.

Ascl1 Participates in Cajal–Retzius Cell Development in the Neocortex

Cajal-Retzius cells are essential pioneer neurons that guide neuronal migration in the developing neocortex. During development, Cajal-Retzius cells arise from distinct progenitor domains that line the margins of the dorsal telencephalon, or pallium. Here, we show that the proneural gene Ascl1 is expressed in Cajal-Retzius cell progenitors in the pallial septum, ventral pallium, and cortical hem. Using a short-term lineage trace, we demonstrate that it is primarily the Ascl1-expressing progenitors in the pallial septum and ventral pallium that differentiate into Cajal-Retzius cells. Accordingly, we found a small, albeit significant reduction in the number of Reelin(+) and Trp73(+) Cajal-Retzius cells in the Ascl1(-/-) neocortex. Conversely, using a gain-of-function approach, we found that Ascl1 induces the expression of both Reelin, a Cajal-Retzius marker, and Tbr1, a marker of pallial-derived neurons, in a subset of early-stage pallial progenitors, an activity that declines over developmental time. Taken together, our data indicate that the proneural gene Ascl1 is required and sufficient to promote the differentiation of a subset of Cajal-Retzius neurons during early neocortical development. Notably, this is the first study that reports a function for Ascl1 in the pallium, as this gene is best known for its role in specifying subpallial neuronal identities.

Neural stem cell self-renewal requires the Mrj co-chaperone.

The Mrj co‐chaperone is expressed throughout the mouse conceptus, yet its requirement for placental development has prohibited a full understanding of its embryonic function. Here, we show that Mrj−/− embryos exhibit neural tube defects independent of the placenta phenotype, including exencephaly and thin‐walled neural tubes. Molecular analyses revealed fewer proliferating cells and a down‐regulation of early neural progenitor (Pax6, Olig2, Hes5) and neuronal (Nscl2, SCG10) cell markers in Mrj−/− neuroepithelial cells. Furthermore, Mrj−/− neurospheres are significantly smaller and form fewer secondary neurospheres indicating that Mrj is necessary for self‐renewal of neural stem cells. However, the molecular function of Mrj in this context remains elusive because Mrj does not colocalize with Bmi‐1, a self‐renewal protein. Furthermore, unlike in Mrj−/− placentas, intermediate filament‐containing aggregates do not accumulate in Mrj−/− neuroepithelium, ruling out nestin as a substrate for Mrj. Regardless, Mrj plays an important role in neural stem cell self‐renewal. Developmental Dynamics 238:2564–2574, 2009.