Regulation of neuronal gene expression is critical to central nervous system development. Here, we show that REST regulates the transitions from pluripotent to neural stem/progenitor cell and from progenitor to mature neuron. In the transition to progenitor cell, REST is degraded to levels just sufficient to maintain neuronal gene chromatin in an inactive state that is nonetheless poised for expression. As progenitors differentiate into neurons, REST and its co-repressors dissociate from the RE1 site, triggering activation of neuronal genes. In some genes, the level of expression is adjusted further in neurons by CoREST/MeCP2 repressor complexes that remain bound to a site of methylated DNA distinct from the RE1 site. Expression profiling based on this mechanism indicates that REST defines a gene set subject to plasticity in mature neurons. Thus, a multistage repressor mechanism controls the orderly expression of genes during development while still permitting fine tuning in response to specific stimuli.

No abstract

The suprachiasmatic nucleus (SCN) of the hypothalamus is a dominant circadian pacemaker in the mammalian brain controlling the rest-activity cycle and a series of physiological and endocrine functions to provide a foundation for the successful elaboration of adaptive sleep and waking behavior. The SCN is anatomically and functionally organized into two subdivisions: (1) a core that lies adjacent to the optic chiasm, comprises predominantly neurons producing vasoactive intestinal polypeptide (VIP) or gastrin-releasing peptide (GRP) colocalized with GABA and receives dense visual and midbrain raphe afferents, and (2) a shell that surrounds the core, contains a large population of arginine vasopressin (AVP)-producing neurons in its dorsomedial portion, and a smaller population of calretinin (CAR)-producing neurons dorsally and laterally, colocalized with GABA, and receives input from non-visual cortical and subcortical regions. In this paper, we present a detailed quantitative analysis of the organization of the SCN core and shell in the rat and place this in the context of the functional significance of the subdivisions in the circadian control of regulatory systems.

The retinal ganglion cells giving rise to retinohypothalamic projections in the rat were identified using retrograde transport of horseradish peroxidase (HRP) or FluoroGold injected into the suprachiasmatic nucleus (SCN), and using transneuronal transport of the Bartha strain of the swine herpesvirus (PRV-Bartha). When PRV-Bartha is injected into one eye, it is taken up by retinal ganglion cells, replicated, transported to axon terminals in the SCN, and released. There the virus may take one, or both, of two paths to retinal ganglion cells in the contralateral eye: 1) uptake by SCN neurons, replication, and release from the neurons with uptake and retrograde transport in retinal afferents originating in the contralateral retina; 2) transneuronal passage through axo-axonic appositions between retinal afferents in the SCN with subsequent retrograde transport of virus to the contralateral retina. The ganglion cells thus labeled are a homogeneous population of small neurons (mean diameter, 12.8 +/- 2.2 microns and mean area, 81.8 +/- 21.8 microns 2) with sparsely branching dendrites that are widely distributed over the retina. This population is best identified when virus labeling of retinal projections in areas beyond the hypothalamus is eliminated by lateral geniculate lesions that transect the optic tract at its entry into the geniculate complex. The same population is labeled with retrograde tracers but, with both HRP and FluoroGold, other ganglion cells are labeled, presumably from uptake by fibers of passage, indicating that the virus is a more reliable marker for ganglion cells giving rise to retinohypothalamic projections. The ganglion cells identified correspond to a subset of type III, or W, cells.

(1999) Proc. Natl. Acad. Sci. U. S. A. 96, 9873-9878). Here we show that the co-repressor mSin3A also interacts with REST. The REST-mSin3A association involves the NH 2 -terminal repressor domain of REST and the paired amphipathic helix 2 domain of mSin3A. REST forms complexes with endogenous mSin3A in mammalian cells, and both mSin3A and CoREST interact with REST in intact mammalian cells. REST repression is blocked in yeast lacking Sin3 and rescued in its presence. In mammalian cells, repression by REST is reduced when binding to mSin3A is inhibited. In mouse embryos, the distribution of mSin3A and REST transcripts is largely coincident. The pattern of CoREST gene expression is more restricted, suggesting that mSin3A is required constitutively for REST repression, whereas CoREST is recruited for more specialized repressor functions.Many genes essential for neuronal functioning, including the brain type II voltage-dependent sodium channel, neuronal growth factors, and neurotransmitter receptors, are repressed in non-neuronal cells by the transcriptional repressor, REST/ NRSF (2-5). Removal of the REST/NRSF binding site (RE1/ NRSE) from the regulatory region of transgenes (6) or expression of a dominant negative form of REST/NRSF in vivo (5) results in aberrant expression of the target genes in non-neural tissues. Deletion of the mouse REST/NRSF gene by homologous recombination results in embryonic lethality (5). Because of its importance in establishing and maintaining the expression pattern of a large number of genes required for neuronal functioning, it was of interest to identify the molecules involved in the repressor mechanism.Previous studies identified two distinct domains in the NH 2 and COOH termini of REST that were both necessary and sufficient to repress brain type II sodium channel reporter genes in transient transfection assays (7) and showed that repression by each of these two domains required distinct titratable factors (8). Recently, repression from the COOH-terminal domain was determined to be mediated by the co-repressor, CoREST (1). We sought to determine whether, in addition to CoREST, mSin3A, a co-repressor for several regulated repressor complexes, might also be involved in REST repression. We found that mSin3A is indeed a functional co-repressor for REST. mSin3A interacts with REST in vitro, in yeast and in intact mammalian cells, and interestingly, the binding site maps to the NH 2 -terminal repressor domain in REST. Furthermore, experiments both in yeast and mammalian cells showed that mSin3A is involved in repressor function. In vivo, the expression patterns of the co-repressors mSin3A and CoREST are distinct. Specifically, in early embryogenesis CoREST exhibits a much more restricted pattern of expression compared with REST and mSin3A, suggesting that the composition of the REST repressor complex during development is dynamic. EXPERIMENTAL PROCEDURES Plasmid ConstructionsYeast Two-hybrid Constructs-LexA-mSin3A was obtained by cloning full-length mSin3A into pBTM116 by standard PCR 1...

A detailed analysis of the cytoarchitecture, retinohypothalamic tract (RHT) projections, and immunohistochemical localization of major cell and fiber types within the hypothalamic suprachiasmatic nuclei (SCN) was conducted in five mammalian species: two species of opossum, the domestic cat, the guinea pig, and the house mouse. Cytoarchitectural and immunohistochemical studies were conducted in three additional species of marsupial mammals and in the domestic pig. The SCN in this diverse transect of mammalian taxonomy bear striking similarities. First, the SCN are similar in location, lying close to the third ventricle (3V) dorsal to the optic chiasm (OC), with a cytoarchitecture characterized by small, tightly packed neurons. Second, in all groups studied, the SCN receive bilateral retinal input. Third, the SCN contain immunohistochemically similar elements. These similarities suggest that the SCN developed characteristic features early in mammalian phylogeny. Some details of SCN organization vary among the species studied. In marsupials, vasopressin-like immunoreactive (VP-LI) and vasoactive intestinal polypeptide-like immunoreactive (VIP-LI) cells codistribute primarily in the dorsomedial aspects of the SCN, while in eutherians, VP-LI and VIP-LI cells are separated into SCN subnuclei. Furthermore, the marsupial RHT projects to the periventricular dorsomedial region, whereas the eutherian RHT projects more ventrally in the SCN into the zone that typically contains VIP-LI perikarya.

No abstract

No abstract