Institute of Cell Biology and Neurobiology
During development, several populations of progenitor cells residing in the ventricular and subventricular zones in the dorsal telencephalon generate a large variety of neurons. These neurons acquire distinct morphologies, specific connections and physiological properties and serve distinct functions in the mammalian cerebral cortex.
We are interested in the cellular and molecular mechanisms underlying cell fate specification in the mouse cerebral cortex. We focus on the mechanisms controlling the generation, migration and axonal guidance of neurons of different cortical layers and areas. Our methods combine genetic, molecular and cell biological approaches to identify and characterize genes that control cortical development.
The fate of cortical progenitors, which progressively generate neurons and glial cells during development, is determined by temporally and spatially regulated signaling mechanisms.
A single layer of neuroepithelial cells lining the embryonic neural tube gives rise to the entire repertoire of neurons, astrocytes and oligodendrocytes of the adult central nervous system (CNS). Early in development some progenitors of the cortical ventricular zone (VZ) are multipotent and can generate both projection neurons and astrocytes. Later, precursor cells tend to become more lineage-restricted. In mice, the generation of cortical projection neurons and glia starts at E11.5 and follows a temporally specified sequence of cell commitment and differentiation in vivo and ex vivo. Following an inside first - outside last pattern of cortical lamination, neurons of deep layers 6 and 5 are generated prior to the molecularly and functionally distinct neurons of upper layers 4, 3 and 2. The time of birth of a neuron can be used to accurately predict its laminar fate, however the molecular determinants of this process are not clearly understood.
We discovered that young neurons can signal back to neuronal precursors and by doing so regulate the fate of uncommitted precursors and ensure the production of appropriate numbers of different types of neurons and glia. We established the Sip1 transcriptional repressor as a key mediator of this mechanism through its negative regulation of the expression of secreted factors in postmitotic neurons.

Figure 1: Role of Sip1 in neocortical cell fate decisions
(a) In the control cerebral cortex, postmitotic neurons secrete neurothrophin 3 (NT3). The production of NT3 is controlled by Sip1, which keeps it at low levels during early development (E12.5). When enough postmitotic neurons are generated (E14.5), the NT3 levels cross a threshold, signal back to the progenitors, block production of deep layer neurons and stimulate production of upper layer neurons. At E17.5, a similar mechanism induces the neurogenesis-gliogenesis switch. At this stage, Sip1 controls the level of Fgf9 production, which reaches its threshold only when neurogenesis is completed and Sip1 expression is downregulated. Fgf9 then induces proliferation of glial precursors. (b) In the Sip1 mutant cerebral cortex, NT3 and Fgf9 gene expression is no longer repressed, such that these two secreted factors reach threshold levels earlier. This in turn induces precocious generation of upper layer neurons during early corticogenesis, and glial precursors during later development.
Cortical projection neurons are arranged in six cortical layers. Neurons within each layer are generated at similar times and share similar morphology and patterns of connectivity. Neurons of upper layers 2 and 3, as well as some layer 5 neurons interconnect different cortical areas ipsilaterally or contralaterally via the corpus callosum. Layer 4 is the major input layer, while most neurons of deep layers 5 and 6 send their axons to subcortical targets in the spinal cord, pons, tectum, and thalamus.
We identified a transcription factor, named Satb2, which controls postmitotic specification of neurons in superficial layers of the neocortex. Satb2 regulates the expression of genes responsible for proper connectivity of cortico-cortical neurons. When Satb2 is inactivated these neurons start expressing the genetic programs of deep layer neurons. As a result these neurons project to subcortical targets.

Figure 3: Molecular mechanism of Satb2 function
Satb2 is required to assemble NURD chromatin remodeling complex on Ctip2 locus. This induces deacetylation of histones and inactivation of Ctip2 expression. MARs are so-called Matrix Attachment Regions on a chromosome that can be recognized by Satb2 protein.

Project Leader
Charité - Universitätsmedizin Berlin
CCM: Campus Charité Mitte
CC 2: Basic Sciences (Year 1)
Institute of Cell Biology and Neurobiology
Postal address:
Charitéplatz 1
10117 Berlin
Internal address:
Waldeyer-Haus
t: +49 30 450 528418, -628418
f: +49 30 450 528902