The thalamus acts as a central integrator for processing and relaying sensory and motor information to and from the cerebral cortex, and the habenula plays pivotal roles in emotive decision making by modulating dopaminergic and serotonergic circuits. These neural compartments are derived from a common developmental progenitor domain, called prosomere 2, in the caudal forebrain. Thalamic and habenular neurons exhibit distinct molecular profile, neurochemical identity, and axonal circuitry. However, the mechanisms of how their progenitors in prosomere 2 give rise to these two populations of neurons and contribute to the forebrain circuitry remains unclear. In this study, we discovered a previously unrecognized role for Tcf7l2, a transcription factor known as the canonical Wnt nuclear effector and diabetes risk-conferring gene, in establishing neuronal identity and circuits of the caudal forebrain. Using genetic and chemical axon tracers, we showed that efferent axons of the thalamus, known as the thalamocortical axons (TCAs), failed to elongate normally and strayed from their normal course to inappropriate locations in the absence of Tcf7l2. Further experiments with thalamic explants revealed that the pathfinding defects of Tcf7l2-deficient TCAs were associated at least in part with downregulation of guidance receptors Robo1 and Robo2 expression. Moreover, the fasciculus retroflexus, the main habenular output tract, was missing in embryos lacking Tcf7l2. These axonal defects may result from dysregulation of Nrp2 guidance receptor. Strikingly, loss of Tcf7l2 caused a post-mitotic identity switch between thalamic and habenular neurons. Despite normal acquisition of progenitor identity in prosomere 2, Tcf7l2-deficient thalamic neurons adopted a molecular profile of a neighboring forebrain derivative, the habenula. Conversely, habenular neurons failed to maintain their normal post-mitotic neuronal identity and acquired a subset of thalamic neuronal features in the absence of Tcf7l2. Our findings suggest a unique role for Tcf7l2 in generating distinct neuronal phenotypes from homogeneous progenitor population, and provide a better understanding of the mechanism underlying neuronal specification, differentiation, and connectivity of the developing caudal forebrain.

The mammalian thalamus is an essential diencephalic derivative that plays unique roles in processing and relaying sensory and motor information to and from the cerebral cortex. The profile of transcription factors and lineage tracing experiments revealed a spatiotemporal relationship between diencephalic progenitor domains and discrete differentiated neurons contributing to thalamic nuclei. However, the precise molecular mechanisms by which heterogeneous thalamic neurons become specified and assemble into distinct thalamic nuclei are still poorly understood. Here, we show that a combinatorial interaction between the bHLH transcription factors Ascl1 and Helt is required for acquiring thalamic progenitor identity. Surprisingly, in the combined absence of Ascl1 and Helt, rostral thalamic progenitors (TH-R) adopt a molecular profile of a more rostral diencephalic derivative, the prethalamus. Furthermore, we show that the prethalamic factors Dlxs upregulated by Ascl1/Helt deficiency play unique roles in regulating thalamic progenitor specification, and that derepression of Dlx2 and Dlx5 suppress generation of TH-R neurons. Taken together, our results suggest a model whereby the combined activity of two distinct bHLH factors plays a key role in the development of discrete classes of thalamic interneurons.

Six6, a sine oculis homeobox protein, plays a crucial and conserved role in the development of the forebrain and eye. To understand how the expression of Six6 is regulated during embryogenesis, we screened ~250 kb of genomic DNA encompassing the Six6 locus for cis-regulatory elements capable of directing reporter gene expression to sites of Six6 transcription in transgenic mouse embryos. Here, we describe two novel enhancer elements, that are highly conserved in vertebrate species and whose activities recapitulate Six6 expression in the ventral forebrain and eye, respectively. Cross-species comparisons of the Six6 forebrain enhancer sequences revealed highly conserved binding sites matching the consensus for homeodomain and SoxB1 transcription factors. Deletion of either of the binding sites resulted in loss of the forebrain enhancer activity in the ventral forebrain. Moreover, our studies show that members of the SoxB1 family, including Sox2 and Sox3, are expressed in the overlapping region of the ventral forebrain with Six6 and can bind to the Six6 forebrain enhancer. Loss of function of SoxB1 genes in vivo further emphasizes their role in regulating Six6 forebrain enhancer activity. Thus, our data strongly suggest that SoxB1 transcription factors are direct activators of Six6 expression in the ventral forebrain.

The SRY-related HMG box transcription factor Sox2 plays critical roles throughout embryogenesis. Haploinsufficiency for SOX2 results in human developmental defects including anophthalmia, microphthalmia and septo-optic dysplasia, a congenital forebrain defect. To understand how Sox2 plays a role in neurogenesis, we combined genomic and in vivo transgenic approaches to characterize genomic regions occupied by Sox2 in the developing forebrain. Six3, a homeobox gene associated with holoprosencephaly, a forebrain midline defect, was identified as a Sox2 transcriptional target. This study shows that Sox2 directly regulates a previously unidentified long-range forebrain enhancer to activate Six3 expression in the rostral diencephalon. Further biochemical and genetic evidences indicated a direct regulatory link between Sox2 and Six3 during forebrain development, providing a better understanding of a common molecular mechanism underlying these forebrain defects.

The sine oculis homeobox protein Six3 plays pivotal roles in the development of the brain and craniofacial structures. In humans, SIX3 haploinsufficiency results in holoprosencephaly, a defect in anterior midline formation. Although much is known about the evolutionarily conserved functions of Six3, the regulatory mechanism responsible for the expression pattern of Six3 remains relatively unexplored. To understand how the transcription of Six3 is controlled during embryogenesis, we screened ∼300 kb of genomic DNA encompassing the Six3 locus for cis-acting regulatory elements capable of directing reporter gene expression to sites of Six3 transcription in transgenic mouse embryos. We identified a novel enhancer element, whose activity recapitulates endogenous Six3 expression in the ventral midbrain, pretectum, and thalamus. Cross-species comparisons revealed that this Six3 brain enhancer is functionally conserved in other vertebrates. We also showed that normal Six3 transcription in the ventral midbrain and pretectum is dependent on Ascl1, a basic helix-loop-helix proneural factor. Moreover, loss of Ascl1 resulted in downregulation of the Six3 brain enhancer activity, emphasizing its unique role in regulating Six3 expression in the developing brain.