Search

Despite its long history, the central effects of progressive depletion of vitamin A in adult mice has not been previously described. An examination of vitamin-deprived animals revealed a progressive and ultimately profound impairment of hippocampal CA1 long-term potentiation and a virtual abolishment of long-term depression. Importantly, these losses are fully reversible by dietary vitamin A replenishment in vivo or direct application of all trans -retinoic acid to acute hippocampal slices. We find retinoid responsive transgenes to be highly active in the hippocampus, and by using dissected explants, we show the hippocampus to be a site of robust synthesis of bioactive retinoids. In aggregate, these results demonstrate that vitamin A and its active derivatives function as essential competence factors for long-term synaptic plasticity within the adult brain, and suggest that key genes required for long-term potentiation and long-term depression are retinoid dependent. These data suggest a major mental consequence for the hundreds of millions of adults and children who are vitamin A deficient.

Nurr1, a member of the nuclear hormone receptor superfamily, was recently demonstrated to be of critical importance in the developing central nervous system, where it is required for the generation of midbrain dopamine cells. Nuclear receptors encompass a transcriptional activation function (activation function 2; AF2) within their carboxyl-terminal domains important for ligand-induced transcriptional activation. Since a Nurr1 ligand remains to be identified, the role of the Nurr1 AF2 region in transcriptional activation is unclear. However, here we show that the Nurr1 AF2 contributes to constitutive activation independent of exogenously added ligands in human embryo kidney 293 cells and in neural cell lines. Extensive mutagenesis indicated a crucial role of the AF2 core region for transactivation but also identified unique features differing from previously characterized receptors. In addition, Nurr1 did not appear to interact with, and was not stimulated by, several previously identified coactivators such as the steroid receptor coactivator 1. In contrast, adenovirus protein E1A, stably expressed in 293 cells, was shown to contribute to AF2-dependent activation. Finally, while the AF2 core of RXR is required for ligand-induced transcriptional activation by Nurr1-RXR heterodimers, the functional integrity of Nurr1 AF2 core is not critical. These results establish that the ligand binding domain of Nurr1 has intrinsic capacity for transcriptional activation depending on cell type and mode of DNA binding. Furthermore, these results are consistent with the possibility that gene expression in the central nervous system can be modulated by an as yet unidentified ligand interacting with the ligand binding domain of Nurr1.

The orphan nuclear receptor NURR1 was previously demonstrated to be required for the generation of mesencephalic dopamine (DA) cells. However, even in the absence of NURR1, which is normally expressed as cells become postmitotic, neuronal differentiation is induced and expression of several genes detected in developing dopamine cells appears normal during early stages of development. These include the homeobox transcription factors engrailed and Ptx-3 as well as aldehyde dehydrogenase 2, here defined as the earliest marker identified in developing DA cells, expressed already in mitotic DA progenitors. We have used the expression of these dopaminergic markers, retrograde axonal tracing, and apoptosis analyses to study the fate of the DA progenitor cells in the absence of NURR1. We conclude that NURR1 plays a critical role in the maturation, migration, striatal target area innervation, and survival of differentiating mesencephalic DA cells.

The implementation of neural stem cell lines as a source material for brain tissue transplants is currently limited by the ability to induce specific neurochemical phenotypes in these cells. Here, we show that coordinated induction of a ventral mesencephalic dopaminergic phenotype in an immortalized multipotent neural stem cell line can be achieved in vitro. This process requires both the overexpression of the nuclear receptor Nurr1 and factors derived from local type 1 astrocytes. Over 80% of cells obtained by this method demonstrate a phenotype indistinguishable from that of endogenous dopaminergic neurons. Moreover, this procedure yields an unlimited number of cells that can engraft in vivo and that may constitute a useful source material for neuronal replacement in Parkinson's disease.

Retinoids regulate gene expression through the action of retinoic acid receptors (RARs) and retinoid-X receptors (RXRs), which both belong to the family of nuclear hormone receptors1,2. Retinoids are of fundamental importance during development2, but it has been difficult to assess the distribution of ligand-activated receptors in vivo. This is particularly the case for RXR, which is a critical unliganded auxiliary protein for several nuclear receptors, including RAR1, but its ligand-activated role in vivo remains uncertain. Here we describe an assay in transgenic mice, based on the expression of an effector fusion protein linking the ligand-binding domain of either RXR or RAR to the yeast Gal4 DNA-binding domain, and the in situ detection of ligand-activated effector proteins by using an inducible transgenic lacZ reporter gene. We detect receptor activation in the spinal cord in a pattern that indicates that the receptor functions in the maturation of limb-innervating motor neurons. Our results reveal a specific activation pattern of Gal4–RXR which indicates that RXR is a critical bona fide receptor in the developing spinal cord.

NRs function as molecular machines to transduce a hormonal signal into a transcriptional response. As sequence-specific DNA binding proteins, the action of the receptor primarily occurs at the site of the target gene. Quiescent genes could be viewed as being wrapped in a protective chromatin shield to fend off centroviral advances from an errant RNA polymerase. Dramatic effects of hormones on chromatin structure have long been recognized; whether this was causal to the transcriptional response or merely its consequence was unclear. In addition, how such changes in chromatin structure were directed by receptors was simply unknown. One popular model is that mere binding by receptor was sufficient to initiate a change in chromatin by altering nucleosome position. In contrast, the effort to identify the biochemical basis for transcriptional regulation has led in recent years to the consideration that NR cofactors might serve as active mediators of the regulatory effect. From the view of an insider, it has often seemed that those labs studying chromatin remodeling and those studying NR cofactors appeared to be on opposite sides of the arena taking unrelated approaches to an otherwise common quest. However, in the last year, these two fields, like lost allies, have been reintroduced to each other and are enjoying a new-found synergy. M. Beato (University of Marburg) outlined the general pathway by which the MMTV LTR utilizes a chromatin-based structure to control its transcription. In this case, nucleosomes are phased in a fashion that allows glucocorticoid receptor but not other transcription factors to bind to target DNA. Accordingly, hormone treatment leads to a rapid alteration in chromatin structure and at the same time promotes cooperative binding of other transcription factors. Presumably, it is the packaging of DNA into a repressive chromatin structure that restricts the accessibility of the template to the basal transcription machinery. Thus, as suggested by Beato, and supported by experiments reported by O. Wrange (Karolinska Institute) using in vitro–reconstituted nucleosomes, the role of the glucocorticoid receptor is to function, at least in part, to counteract chromatin-mediated repression. As discussed by L. Krause (UCSD) and J. Kadonaga (UCSD), the “ground” state or default status of an endogenous target gene would be transcriptionally inactive. The role of a transcription factor would be to initially counteract the chromatin effect leading to a derepressed template. This, in turn, would be followed by a “true activation” event equivalent to robust transcriptional initiation. Kadonaga described dissection of this putative multistep process by creating chromatin templates to directly assess the action of NRs in an in vitro–reconstituted system. While previous in vitro transcription studies have been described, this is the first with chromatin-based templates. The effects with the estrogen receptor (ER) are clear and dramatic. Fifteen- to fifty-fold inductions were seen, all in a hormone-dependent fashion. The estrogen receptor can be added either before or after chromatin assembly, and the entire process was shown to be dependent on the known activation domains in the ER. The improved regulation appears to be a consequence of chromatin suppressing basal activity of the promoter thus increasing the fold of induction. Interestingly, estrogen antagonists failed to activate although these complexes can be bound to the template. Continuing in this vein, A. Wolffe (NICHD) described the use of replication-dependent chromatin assembly to identify three regulatory steps in the regulation of transcription by the thyroid hormone receptor (TR). These three steps include: (1) the establishment of a repressive chromatin structure; (2) disruption of the chromatin template; and (3) transcriptional activation. As is apparent, this is similar to the events described by Kadonaga and Beato although with a completely different regulatory system. In this approach, cloned templates are microinjected into the Xenopus oocyte nucleus, which are then assembled in a natural chromatin structure. In this way, Wolffe has been able to address one of the more vexing problems in NR transcription by demonstrating that nonliganded RXR:TR heterodimers can not only bind to chromatin but also may facilitate the formation of a repressive chromatin structure. Conclusions that arise out of this important model system include the demonstration that the nonliganded TR:RXR heterodimers bind to nucleosome DNA, make use of chromatin to repress transcription, control nucleosome position, and can influence both nucleosome modificat

The receptor for 9-cis-retinoic acid, retinoid X receptor (RXR), forms heterodimers with several nuclear receptors, including the receptor for all-trans-retinoic acid, RAR. Previous studies have shown that retinoic acid receptor can be activated in RAR/RXR heterodimers, whereas RXR is believed to be a silent co-factor. In this report we show that efficient growth arrest and differentiation of the human monocytic cell line U-937 require activation of both RAR and RXR. Also, we demonstrate that the allosteric inhibition of RXR is not obligatory and that RXR can be activated in the RAR/RXR heterodimer in the presence of RAR ligands. Remarkably, RXR inhibition by RAR can also be relieved by an RAR antagonist. Moreover, the dose response of RXR agonists differ between RXR homodimers and RAR/RXR heterodimers, indicating that these complexes are pharmacologically distinct. Finally, the AF2 activation domain of both subunits contribute to activation even if only one of the receptors is associated with ligand. Our data emphasize the importance of signaling through both subunits of a heterodimer in the physiological response to retinoids and show that the activity of RXR is dependent on both the identity and the ligand binding state of its partner.

NGFI-B, Nurr1, and Nor1 are three closely related orphan members of the steroid/thyroid hormone receptor superfamily. These receptors can bind to DNA as monomers and exhibit constitutive transcriptional activity. Moreover, two of the receptors, NGFI-B and Nurr1, have previously been shown to form heterodimers with the retinoid X receptor (RXR). Such heterodimers as well as complexes formed between RXR and the all-trans retinoic acid receptor bind to DNA response elements composed of direct repeats spaced by five nucleotides (DR5). However, whereas retinoic acid receptor can inhibit ligand-dependent RXR activation, NGFI-B and Nurr1 allow efficient RXR activation through DR5 elements and thus define a distinct pathway for vitamin A signaling. In this study we demonstrate that the most recently identified member of the subfamily, Nor1, shows similar monomer DNA-binding and constitutive transactivation properties as NGFI-B and Nurr1. In contrast, however, Nor1 is unable to promote RXR signaling due to its inability to form heterodimers with RXR. To begin to understand the physiological implications of these functional differences we used in situ hybridization to compare the distribution of Nor1, NGFI-B, and Nurr1 messenger RNAs during different developmental stages. The receptors are expressed in both distinct and overlapping patterns, predominantly in the central nervous system. Notably, Nurr1 is expressed in the prenatal ventral midbrain in a region that gives rise to dopaminergic neurons. Nor1 is also expressed during embryonic development, and all three receptors show a complex distribution in the postnatal brain. Furthermore, Nor1 colocalizes with NGFI-B in the adrenal glands and thymus, two tissues in which NGFI-B has been suggested to be functionally important. These data may indicate redundancy between members of the NGFI-B/Nurr1/Nor1 subfamily and could explain why no phenotypic disturbances have yet been found in mice in which the NGFI-B gene has been inactivated.

Nurr1 and NGFI-B are closely related orphan members of the steroid-thyroid hormone receptor family involved in immediate early responses to stimuli such as growth factors. In-situ hybridization in the developing and adult mouse and rat demonstrated Nurr1 mRNA in several regions during early central nervous system (CNS) development. Expression persisted through the pre- and postnatal periods and was also found in several areas in the adult CNS. Positive areas include the olfactory bulb, parts of the cortex, the hippocampal formation and substantia nigra where Nurr1 and tyrosine hydroxylase mRNAs were co-expressed. 6-Hydroxydopamine-induced degeneration of mesencephalic dopamine neurons led to a corresponding loss of Nurr1 mRNA, demonstrating a link between Nurr1 and dopaminergic neurons. NGFI-B mRNA was not found in the prenatal CNS but was highly expressed in the adult brain in many areas including the olfactory bulb, cortex, basal ganglia and hippocampus. The spatiotemporal distribution of Nurr1 and NGFI-B mRNAs suggests that these transcription factors are involved in the development and maturation of specific sets of CNS neurons. The experimental data imply that one of these functions may be to control gene regulatory events important for development and function of those neurons that degenerate in patients with Parkinson's disease.

Nuclear receptor heterodimers recognize response elements composed of two direct repeats of the consensus sequence 5'-AGGTCA-3' separated by one to five base pairs. The 1.9 A crystal structure of the complex formed by the DNA-binding domains of the 9-cis retinoic acid receptor and thyroid hormone receptor bound to a thyroid-response element shows that the subunits interact through a DNA-supported interface involving the carboxy-terminal extension of the DNA-binding domain of the thyroid hormone receptor. The stereochemistry suggests a mechanism by which heterodimers recognize the inter-half-site spacing between direct repeats.

In addition to its role as a 9-cis retinoic acid receptor, RXR has an important role in the regulation of multiple hormonal pathways through heterodimerization with nuclear receptors. Here, we show that two orphan receptors, NGFI-B and NURR1, which have been shown previously to interact with DNA as monomers, also can heterodimerize with RXR. These heterodimers bind selectively to a class of retinoic acid response elements composed of direct repeats spaced by 5 nucleotides. In this respect they are similar to heterodimers formed between RXR and the receptor for all-trans retinoic acid, RAR. However, whereas RXR is inhibited in the RXR-RAR heterodimer, NGFI-B/NURR1 promote efficient activation in response to RXR ligands and therefore shift RXR from a silent to an active heterodimerization partner. These data show that NGFI-B and NURR1 can increase the potential of RXR to modulate gene expression in a ligand-dependent manner by allowing a distinct class of direct repeats to serve as specific RXR response elements. Because expression of both NGFI-B and NURR1 is rapidly induced by various growth factors, these findings also suggest a novel mechanism for convergence between vitamin A or retinoid and growth factor signaling pathways.

The recent characterization of the alcohol* dehydrogenase Aldh1a2 and the cytochrome P450 Cyp26, two enzymes involved in retinoid metabolism, has helped to explain how bioactive retinoids are made and catabolized. By the elegant definition of an Aldh1a2 null mutation as a dominant suppressor of a Cyp26 null mutation, it is now unequivocally demonstrated that the main function of Cyp26 is to degrade endogenous all-trans retinoic acid rather than to synthesize bioactive hydroxylated retinoids.

A DNA fragment containing glucocorticoid receptor binding sites in the mouse mammary tumor virus promoter was reconstituted in vitro with histones to form nucleosome cores, which become positioned on the DNA fragment in a sequence-specific manner. Glucocorticoid receptor binding to specific DNA sequences was analyzed by quantitative DNase I footprinting. The receptor interacted with surprisingly high affinity with one of the binding sites in the reconstituted promoter, although it was reduced by a factor of approximately 2 compared with the same site in protein-free DNA. By contrast, the affinity for random genomic nucleosomal sites was drastically reduced compared with histone-free DNA. Thus, reconstituting the promoter in vitro resulted in a 60- to 70-fold increase in binding specificity. Such an increase in selective binding may help to explain the ability of glucocorticoid receptor to effectively locate its target sites in chromatin.

Nature Genet. 31, 7–8(2002). Two errors were introduced while preparing this article for press. They were introduced after the author provided his corrections to the proofs, and so he did not have an opportunity to correct the errors. In two places, Aldh1a2, an aldehyde dehydrogenase, is referred toincorrectly as an alcohol dehydrogenase.

Steroid hormones exert profound effects on differentiation, development, and homeostasis in higher eukaryotes through interactions with nuclear receptors. We describe a novel orphan nuclear receptor, termed the pregnane X receptor (PXR), that is activated by naturally occurring steroids such as pregnenolone and progesterone, and synthetic glucocorticoids and antiglucocorticoids. PXR exists as two isoforms, PXR.1 and PXR.2, that are differentially activated by steroids. Notably, PXR.1 is efficaciously activated by pregnenolone 16α-carbonitrile, a glucocorticoid receptor antagonist that induces the expression of the CYP3A family of steroid hydroxylases and modulates sterol and bile acid biosynthesis in vivo. Our results provide evidence for the existence of a novel steroid hormone signaling pathway with potential implications in the regulation of steroid hormone and sterol homeostasis.

Summary The prospect of using cell replacement therapies has raised the key issue of whether elucidation of developmental pathways can facilitate the generation of therapeutically important cell types from stem cells. Here we show that the homeodomain proteins Lmx1a and Msx1 function as determinants of midbrain dopamine neurons, cells that degenerate in patients with Parkinson's disease. Lmx1a is sufficient and required to trigger dopamine cell differentiation. An early activity of Lmx1a is to induce the expression of Msx1, which complements Lmx1a by inducing the proneural protein Ngn2 and neuronal differentiation. Importantly, expression of Lmx1a in embryonic stem cells results in a robust generation of dopamine neurons with a "correct" midbrain identity. These data establish that Lmx1a and Msx1 are critical intrinsic dopamine-neuron determinants in vivo and suggest that they may be essential tools in cell replacement strategies in Parkinson's disease.

SummaryThe generation of specific types of neurons from stem cells offers important opportunities in regenerative medicine. However, future applications and proper verification of cell identities will require stringent ways to generate homogeneous neuronal cultures. Here we show that transcription factors like Lmx1a, Phox2b, Nkx2.2, and Olig2 can induce desired neuronal lineages from most expressing neural progenitor cells by a mechanism resembling developmental binary cell-fate switching. Such efficient selection of cell fate resulted in remarkable cellular enrichment that enabled global gene-expression validation of generated neurons and identification of previously unrecognized features in the studied cell lineages. Several sources of stem cells have a limited competence to differentiate into specific neuronal cell types; e.g., dopamine neurons. However, we show that the combination of factors that normally promote either regional or dedicated neuronal specification can overcome limitations in cellular competence and also promote efficient reprogramming in more remote neural contexts, including human neural progenitor cells.

We have reconstituted a nucleosome with core histones from rat liver using a restriction fragment containing a sequence from the mouse mammary tumour virus (MTV) long terminal repeat (LTR). This sequence harbours glucocorticoid responsive elements (GREs) which mediate glucocorticoid hormone induction of transcription from the MTV promoter via glucocorticoid receptor (GR) binding. Exonuclease III and DNase I footprinting demonstrated that the reconstituted nucleosome was specifically located between positions -219 and -76. A nucleosome was previously shown to be located at a similar or identical position in the MTV promoter in situ and to be structurally altered upon glucocorticoid hormone induction. We demonstrated, by DNase I footprinting, that GR is able to bind sequence specifically to the DNA in the in vitro assembled nucleosome. No evidence for unfolding of the nucleosome was obtained, but the DNase I footprinting pattern demonstrated GR induced local alterations in the DNA.

Transgenic mice lacking the nuclear orphan transcription factor Nur-related receptor 1 (Nurr1) fail to develop mesencephalic dopamine neurons. There is a highly homologous NURR1 gene in humans (formerly known as NOT) which therefore constitutes a good candidate gene for neurologic and psychiatric disorders with an involvement of the dopamine neuron system, such as Parkinson's disease, schizophrenia, and manic-depression. By direct sequencing of genomic DNA, we found two different missense mutations in the third exon of NURR1 in two schizophrenic patients and another missense mutation in the same exon in an individual with manic-depressive disorder. All three mutations caused a similar reduction of in vitro transcriptional activity of NURR1 dimers of about 30–40%. Neither of these amino acid changes, nor any sequence changes whatsoever, were found in patients with Parkinson's disease or control DNA material of normal populations. Am. J. Med. Genet. (Neuropsychiatr. Genet.) 96:808–813, 2000. © 2000 Wiley-Liss, Inc.