A transcriptional regulatory element of the gene encoding the 67,000-M(r) form of human glutamate decarboxylase is similar to a Drosophila regulatory element

Odorant Sensory Input Modulates DNA Secondary Structure Formation and Heterogeneous Ribonucleoprotein Recruitment on the Tyrosine Hydroxylase and Glutamic Acid Decarboxylase 1 Promoters in the Olfactory Bulb

Adaptation of neural circuits to changes in sensory input can modify several cellular processes within neurons, including neurotransmitter biosynthesis levels. For a subset of olfactory bulb interneurons, activity-dependent changes in GABA are reflected by corresponding changes in Glutamate decarboxylase 1 (Gad1) expression levels. Mechanisms regulating Gad1 promoter activity are poorly understood, but here we show that a conserved G:C-rich region in the mouse Gad1 proximal promoter region both recruits heterogeneous nuclear ribonucleoproteins (hnRNPs) that facilitate transcription and forms single-stranded DNA secondary structures associated with transcriptional repression. This promoter architecture and function is shared with Tyrosine hydroxylase (Th), which is also modulated by odorant-dependent activity in the olfactory bulb. This study shows that the balance between DNA secondary structure formation and hnRNP binding on the mouse Th and Gad1 promoters in the olfactory bulb is responsive to changes in odorant-dependent sensory input. These findings reveal that Th and Gad1 share a novel transcription regulatory mechanism that facilitates sensory input-dependent regulation of dopamine and GABA expression.SIGNIFICANCE STATEMENT Adaptation of neural circuits to changes in sensory input can modify several cellular processes within neurons, including neurotransmitter biosynthesis levels. This study shows that transcription of genes encoding rate-limiting enzymes for GABA and dopamine biosynthesis (Gad1 and Th, respectively) in the mammalian olfactory bulb is regulated by G:C-rich regions that both recruit heterogeneous nuclear ribonucleoproteins (hnRNPs) to facilitate transcription and form single-stranded DNA secondary structures associated with repression. hnRNP binding and formation of DNA secondary structure on the Th and Gad1 promoters are mutually exclusive, and odorant sensory input levels regulate the balance between these regulatory features. These findings reveal that Th and Gad1 share a transcription regulatory mechanism that facilitates odorant-dependent regulation of dopamine and GABA expression levels.

Analysis of the GAD1 promoter: trans-acting factors and DNA methylation converge on the 5' untranslated region

GAD67 corresponds to one of two enzymes that decarboxylates glutamate to produce γ-aminobutyric acid, the main inhibitory neurotransmitter in the mammalian central nervous system, hence defining the cellular phenotype of a diverse set of inhibitory interneurons of the brain. Reduced cortical GAD67 mRNA levels have consistently been reported in schizophrenia and bipolar disorder with psychosis. The human gene encoding GAD67, GAD1, is located on chromosome 2q31.1 and the transcriptional start site resides within a large CpG island that spans a region extending from upstream through the first exon. We have analyzed the GAD1 promoter using transient transfection analysis of upstream and downstream sequences in NT2 cells, a human neuroprogenitor cell line. Interestingly, results from these studies show that cis-acting regulatory elements are located downstream of the RNA start site and are in the region corresponding to the first exon. Trans-acting factors such as Pitx2 and the Dlx family of transcription factors are active in promoting downstream reporter expression even when all of the 5' flanking sequences are removed. However, those constructs that contain an internal deletion from +66 to +173 bp fail to support expression even when these factors are provided in trans. We have previously shown that the Class I histone deacetylase inhibitor MS-275 potently activates GAD1 mRNA expression in NT2 cells suggesting the possibility that the promoter is sensitive to drugs that induce chromatin remodeling. Using methyl DNA immuneprecipitation of MS-275-treated NT2 cells, we provide data showing that Class I HDAC inhibition mediated an increase in GAD1 expression and that this was accompanied by decreased GAD1 promoter methylation. Moreover, the reduced levels of GAD1 DNA methylation are highest in those regions proximal to the location of the in vitro defined cis-acting regulatory elements. Our data suggest that changes in promoter methylation associated with gene regulation are not random but overlap the locations of proximal cis-acting elements. This article is part of a Special Issue entitled 'Trends in neuropharmacology: in memory of Erminio Costa'.

Mutational screening and association study of glutamate decarboxylase 1 as a candidate susceptibility gene for bipolar affective disorder and schizophrenia

Recent evidence from postmortem studies suggests that GAD1 encoding the gamma-aminobutyric acid (GABA) synthetic enzyme GAD67 is a functional candidate susceptibility gene for both bipolar affective disorder (BPAD) and schizophrenia. Previous studies suggest linkage between D2S326 near GAD1 and BPAD. We systematically screened GAD1 exons, flanking intronic sequences, and the promoter sequence for polymorphisms in 16 BPAD patients and five controls from Denmark. We identified eight single nucleotide polymorphisms (SNPs) including two in the promoter sequence. An association study of SNPs covering GAD1 was performed in a Danish sample of 82 BPAD subjects and 120 controls and in a Scottish sample of 197 individuals with schizophrenia, 200 BPAD subjects and 199 controls. Linkage disequilibrium (LD) and haplotype frequencies were estimated from genotype data from eight SNPs. Strong pairwise LD was observed among all pairs of neighboring markers. In the Danish sample, we found weak association between BPAD and two promoter SNPs spaced 1 kb apart. Furthermore, one, two, and three loci haplotype analysis showed weak association with BPAD in the Danish sample. The results from the association studies indicate that promoter variants are of importance for the Danish BPAD cases and we cannot reject the hypothesis of GAD1 as a functional candidate gene for BPAD. No association was observed between BPAD or schizophrenia and any of the investigated SNPs in the Scottish sample set. Thus the results obtained from the Scottish sample suggest that the GAD1 gene variants do not play a major role in the predisposition to schizophrenia.

Characterization of the rat GAD67 gene promoter reveals elements important for basal transcription and glucose responsiveness

GAD65 and GAD67 are two isoforms of the enzyme glutamic acid decarboxylase which catalyze the production of GABA from glutamate, primarily in the brain. However, GAD and GABA also prevail in the retina, testes and islets of Langerhans. The main function of GABA is in neurotransmission, and it is involved in paracrine signalling in islets, but has also been suggested to play a role as a trophic factor in synaptogenesis and to be an important metabolite feeding into the tricarboxylic acid cycle via the GABA-shunt. Both GAD isoforms are subject to regulation, e.g. by synaptic activity. GAD65 is regulated at the level of enzyme activity by association and dissociation from its cofactor, PLP, whereas GAD67 is controlled at the level of its mRNA. To study this process in further detail, we have isolated and characterized the 5'-flanking region of the rat GAD67 gene. We report the transcriptional initiation sites and promoter sequences important for expression in islet beta-cells and C6 glioma cells, and demonstrate that the GAD67 promoter harbors elements that are responsive to glucose in primary islet cells.

Regulation of cell-type specific expression of lacZ by the 5'-flanking region of mouse GAD67 gene in the central nervous system of transgenic mice

The transcriptional regulation of the murine gene encoding the 67-kDa form of glutamic acid decarboxylase (GAD67) was studied by beta-galactosidase histochemistry in transgenic mice carrying fusion genes between progressively longer portions of the 5'-upstream regulatory region of GAD67 and E. coli lacZ. No expression was detected in brains of mice carrying 1.3 kb of upstream sequences including a housekeeping and two conventional promoters, and two negative regulatory elements with homology to known silencers. In mice carrying the same portion of the promoter region plus the first intron, lacZ expression in the adult central nervous system was found in few, exclusively neuronal sites. The number of correctly stained GABAergic centres increased dramatically with increasing the length of the 5'-upstream region included in the construct which suggests that multiple putative spatial enhancers are located in this region. Their action is influenced by epigenetic mechanisms that may be due to site-of-integration and transgene copy-number effects. Additional cis-acting elements are needed to obtain fully correct expression in all GABAergic neurons of the adult central nervous system.

Expression of glutamic acid decarboxylase during human neuronal differentiation: studies using the NTera-2 culture system

Human NTera-2N neurons, but not the parental NTera-2 teratocarcinoma line, decarboxylate [2-(15)N]glutamine to form gamma-[15N]aminobutyric acid (GABA). The reverse transcriptase-polymerase chain reaction (RT-PCR) followed by Southern blotting showed that NTera-2N neurons transcribe the glutamic acid decarboxylase p67 (GAD67) gene, and also demonstrated that there is developmentally regulated alternative splicing of GAD67 mRNA in NTera-2N neurons. As in rat central nervous system (CNS), this mRNA processing generates two RNA transcripts, owing to the inclusion or exclusion of an approximately 80 bp coding region insert. In embryonic day 16 (E16) rat brain, the larger of the two GAD67 mRNAs, which encodes a truncated, inactive apoenzyme, reaches a concentration almost equal to that of the smaller transcript, which encodes functional GAD67. In developing NTera-2N neurons, however, the larger transcript is barely detectable by RT-PCR. RT-PCR also revealed that rat CNS of all ages examined contains GAD65 mRNA, and that GAD65 mRNA is below the detectable range in NTera-2N neurons.

Structure and alternative promoters of the mouse glutamic acid decarboxylase 67 gene

gamma-Aminobutyric acid is synthesized by glutamic acid decarboxylase (GAD), which has two forms, GAD65 and GAD67. Genomic clones coding mouse GAD67 (mGAD67) have been isolated. The restriction map of the overlapping clones covers a region of more than 45 kb of genomic DNA. The mGAD67 gene contains 16 translated exons in addition to an exon which is preferentially expressed in foetal brain. The rapid amplification of 5'-cDNA ends showed that mGAD67 gene transcripts have two different 5'-untranslated regions. Analysis of the genomic clones encompassing the 5'-exons revealed that the two transcripts arose from a single gene by alternative splicing using two different donor sites and a common acceptor. The exons were found 1.5 and 0.6 kb upstream of exon 1. The corresponding promoter regions of these exons have a number of putative regulatory elements, including Sp1- and Krox-24-binding sites. Analysis of mGAD67 transcripts demonstrated that each of the 5'-untranslated exons was expressed in mouse brain. In contrast, exon 0A, but not exon 0B, was expressed in mouse testis and pancreas. These results suggest that these transcripts may be regulated under the control of independent promoters.

Structure and the promoter region of the mouse gene encoding the 67-kD form of glutamic acid decarboxylase

We have cloned and determined the complete structure of the murine gene encoding the 67-kD form of glutamic acid decarboxylase (GAD67), the gamma-aminobutyric acid synthetic enzyme. Its coding region comprises 18 exons spanning 42 kb of genomic DNA. Exon 1 together with 64 bp of exon 2 defines the 5' untranslated region of GAD67 mRNA. Exon 18 specifies the protein's carboxyl terminal and the entire 3' untranslated region. Exons 7/A and 7/B are solely contained in the coding regions of two alternatively spliced bicistronic embryonic mRNAs, which code for the truncated embryonic GAD forms. The promoter region (P1) corresponding to the main group of transcription initiation sites is devoid of TATA and CAAT boxes but has putative binding sites for the transcription factor SP1 and is embedded in a large G + C-rich domain of a CpG island, features shared by the promoters of constitutively expressed housekeeping genes. Primer extension data suggests the existence of additional transcription start sites at 130 bp and 295 bp upstream from the major initiation site that are utilized less frequently in adult brain. The tentative distal promoters (P2 and P3) that correspond to the minor start sites resemble tissue-specific promoters with TATA and CAAT-like boxes. In 1.3 kb of the 5'-upstream region, we identified several putative transcription factor binding sites such as AP2, Hox, E-box, egr-1, and NF-kappaB and putative neuronal-specific regulatory elements, including the neuronal-restrictive silencer element, which may have functional significance in the developmental and tissue-specific expression of the GAD67 gene.

Two glutamate decarboxylase forms corresponding to the mammalian GAD65 and GAD67 are expressed during development of the chick telencephalon

The gamma-aminobutyric acid (GABA)-synthesizing enzyme glutamate decarboxylase (GAD) was studied during development of the chick telencephalon. By means of reverse-phase HPLC analysis, we showed that GABA indeed accumulates during embryogenesis, whereas the levels of glutamate, the substrate for GAD, are more or less unchanged up to later developmental stages. The enzyme activity increased approximately 25-fold from embryonic day 3 to embryonic day 17. Immunoblotting data revealed that two GAD proteins, of approximately 65 and 67 kDa, were present during the period investigated. Furthermore, Northern blot analysis with probes obtained from rat cDNA sequences, as well as a chicken-specific probe for GAD65 generated by means of reverse transcriptase-polymerase chain reaction (RT-PCR), strengthened the interpretation that the chick embryo expresses genes corresponding to GAD65 and GAD67. The rat probes recognized transcript sizes of 3.9 kb (GAD65) and 5.6 kb (GAD67), sizes which are different from those of the rat brain (Erlander et al., Neuron, 7, 91-100, 1991). Sequencing of the RT-PCR products revealed a high level of homology (82% at the nucleotide level) between the mammalian and chick GAD65 genes. Taken together, these findings suggest that the chick embryo expresses two GAD genes during embryogenesis. The functional properties of each gene product remain to be investigated.

Molecular genetic analysis of the role of GABAergic systems in the behavioral and cellular actions of alcohol

Recent studies implicate the inhibitory neurotransmitter gamma-aminobutyric acid (GABA) in many neurochemical actions of ethanol and a variety of behavioral responses to acute and chronic ethanol treatment. However, the molecular mechanisms responsible for genetic differences in initial neurochemical or behavioral sensitivity to ethanol, and adaptation following chronic or repeated ethanol administration, remain to be elucidated. Pharmacogenetic research will increasingly move toward mapping, cloning, identification, and functional analysis of the genes underlying the actions of ethanol. The approaches discussed here permit molecular analysis of both known and previously unknown genes regulating behavioral sensitivity to ethanol. The synthesis of molecular methods and behavioral genetics offers immediate hope for delineating the role of the GABA(A) receptor complex, and other determinants of GABAergic neurotransmission, in determining genetic variation in behavioral responses to ethanol.

Estrogen modulation of mRNA levels for the two forms of glutamic acid decarboxylase (GAD) in female rat brain

Two separate forms of glutamic acid decarboxylase, now termed GAD65 and GAD67, are the rate limiting enzymes for synthesis of gamma-aminobutyric acid (GABA). Because of the significance of GABA to neuroendocrine processes, numerous attempts have been made to determine the impact of gonadal steroids on enzyme functioning with inconclusive results. Therefore, we attempted to determine the impact of estradiol on mRNA levels for each form of GAD by quantitative in situ hybridization histochemistry in various brain regions. Ovariectomized rats were treated with estradiol benzoate or oil vehicle on 2 consecutive days and the brains collected on the third day. DNA probes specific for GAD65 and GAD67 were radiolabeled with CTP32 using asymmetric polymerase chain reaction. Results of in situ hybridizations for each probe on alternate sections from the same animals were analyzed for magnocellular preoptic area (McPOA), dorsal medial nucleus of the hypothalamus (DMN), zona incerta (ZI), and midbrain central gray (MCG). In the McPOA, estradiol exerted opposite effects on the frequency distribution of pixels per cell for two GAD mRNA probes, significantly increasing GAD65 (P < .05) and decreasing GAD67 (P < .01; Kolmogorov-Smirnov). In the DMN, estradiol treatment significantly increased GAD67 by 60% (P < .05; two-way ANOVA) but decreased GAD65 mRNA by 73% (P < .01). Note the direction of effects are opposite between McPOA and DMN. In MCG, analysis showed no estradiol effect on GAD mRNA levels/cells, but the proportion of cells expressing detectable levels of GAD65 or GAD67 increased by 33-40% in estradiol-treated rats (chi 2, P < .001).

Transient expression of glutamate decarboxylase and gamma-amino butyric acid in embryonic lens fibers of the rat

We have determined the localization and developmental expression of glutamate decarboxylase (GAD67) in the rat lens. Immunofluorescence experiments showed that GAD67 was transiently expressed in the nuclear fiber cells of the lens between embryonic days (E) 15 and 20, with maximal immunostaining occurring on E17 and E18. gamma-amino butyric acid (GABA) co-localized with GAD67 in the embryonic nuclear fiber cells. Reverse transcription-polymerase chain reaction (RT-PCR) tests showed that at least three alternatively spliced forms of GAD67 mRNA, including mRNAs with and without the I80 and the I86 insert, were transiently co-expressed with GAD67 in the embryonic lens. The major GAD67 protein in the lens was 67 kDa. We conclude that enzymatically active GAD67 is transiently expressed in the lens nuclear fiber cells of the embryonic rat. The transient expression is regulated by transcriptional and/or posttranscriptional processes. We speculate on the basis of possible common gene regulatory elements for glutamate and ornithine decarboxylases and the involvement of these enzymes with polyamine synthesis, that the transient expression of GAD67 may be connected to nuclear and/or DNA breakdown during lens fiber cell differentiation.

Dopaminergic and glutamatergic blocking drugs differentially regulate glutamic acid decarboxylase mRNA in mouse brain

Dopaminergic and glutamatergic inputs play an important role in regulating the activity of GABAergic neurons in basal ganglia. To understand more fully the biochemical interactions between these neurotransmitter systems, the effects of blocking dopamine and glutamate (N-methyl-D-aspartate) (NMDA) receptors on the expression of glutamic acid decarboxylase (GAD) mRNA were examined. Persistent blockade of dopamine receptors was achieved by daily injections of EEDQ, a relatively non-selective irreversible D1 and D2 dopamine receptor antagonist, or FNM, a relatively selective irreversible D2 dopamine receptor antagonist. Persistent blockade of NMDA receptors was achieved by continuously infusing dizocilpine (MK-801), a non-competitive NMDA receptor antagonist. The levels of GAD mRNA in mouse brain were measured by in situ hybridization histochemistry following treatment with these agents. Repeated administration of EEDQ increased the levels of GAD mRNA in corpus striatum and frontal and parietal cortex; the first significant effects were seen after 4 days of treatment. Treatment with FNM elicited effects similar to those produced by EEDQ, except FNM also significantly increased GAD mRNA in nucleus accumbens. Neither EEDQ nor FNM produced significant effects on GAD mRNA in olfactory tubercle or septum. Infusion of MK-801 produced a rapid and marked decrease in the levels of GAD mRNA in corpus striatum, nucleus accumbens, olfactory tubercle, septum and frontal and parietal cortex; significant changes were seen as early as 2 days of treatment. No significant effects were seen in globus pallidus. Cellular analysis of emulsion autoradiograms from corpus striatum revealed that MK-801 reduced the amount of GAD mRNA in individual cells as well as the proportion of cells expressing high levels of GAD mRNA. These results suggest that dopamine, though its interaction with D2 dopamine receptors, exerts an inhibitory effect on the expression of GAD mRNA, and that glutamate, though its interaction with NMDA receptors, exerts a stimulatory effect on GAD mRNA expression. They show further that the regulation of gene expression by dopamine receptors or NMDA receptors is different in different regions of the brain.

Transcriptional organization of a Drosophila glutamic acid decarboxylase gene

We previously described the sequence and expression pattern of a Drosophila mRMA (Gad) that encodes the major soluble form of glutamic acid decarboxylase (GAD). We now report the transcriptional organization of the Drosophila Gad gene. Based on a combination of DNA sequence, RNase protection, primer extension, and polymerase chain reaction analyses, we conclude that the transcription unit for a 3.1-kb Gad mRNA is composed of eight exons that span approximately 17-kb genomic interval. By this analysis, the site of Gad transcript initiation overlaps with a recognition sequence that confers binding of the zeste transcription factor to other promoter elements. We emphasize that our analysis of the Gad transcription unit provides no evidence for alternative RNA splicing as a mechanism for the generation of GAD isoforms. Thus, the several GAD-immunoreactive proteins (putative GAD isoforms) that can be detected in Drosophila extracts are probably encoded by distinct genes.