Alessandro Guidotti - US grants

The mRNAs for the various gamma-aminobutyric acid type A (GABA-A) receptor subunits and their subtypes show significantly different distribution patterns within the brain, a finding that is consistent with pharmacological and biochemical evidence for polymorphism in the allosteric modulatory sites of this receptor. The mammalian brain, including that of human, has the ability to synthesize steroids--herein called neurosteroids--that function as positive or negative endogenous allosteric modulators of GABA-A receptor function; these steroids appear to be synthesized on glial cells. Thus, modulation of GABA-A receptors by drugs is determined by both the polymorphism of GABA-A receptor subtypes and the ability of the central nervous system (CNS) to produce neurosteroids. Studies with recombinant GABA-A receptors expressed in the 293 kidney tumor cell line have indicated that structural diversity of the allosteric modulatory sites associated with the extracellular domain of the receptor determines the pharmacological profiles of several classes of benzodiazepine (BZD) ligands. However, the pharmacological specificity of certain BZDs and their congeners may be attributable to their ability to bind to BZD recognition sites located in the outer mitochondrial membrane of glial cells and to modulate the rate or quality of mitochondrial neurosteroidogenesis. In turn, neurosteroids may modulate GABA-A receptor function by acting at the allosteric center associated with the gating mechanisms of the receptor-operated Cl- channel. The first objective of our proposal is to examine the functional and structural heterogeneity of native GABA-A receptors in different areas of the CNS. The second objective is to determine the pharmacological profiles of BZDs and their congeners that act on the extracellular domain of various recombinant GABA-A receptors as reconstituted in transfected tumor cell lines. The third goal will be to relate the data pertaining to objectives 1 and 2 to the acute in vivo pharmacological effects of BZDs after injection into rats. In the fourth objective, we will address the role of neurosteroids in drug-induced positive and negative modulation of GABA-A receptor function. Together, these studies will help to develop both new criteria for drug classification and new strategies for the development of drugs with preferential action at distinct allosteric modulatory sites of GABA-A receptor subtypes. This approach may help to identify drugs that are both devoid of side effects and highly selective for the treatment of anxiety disorders--such as generalized anxiety or panic disorders--sleep disorders, and epilepsy.

gas chromatography mass spectrometry; high performance liquid chromatography; hormone regulation /control mechanism; progesterone analog

DESCRIPTION: (from applicant's abstract): The overall goal of our research project is to utilize the diversity of GABAalpha receptor structure and function to develop via new strategies safer and more effective neuroactive drugs acting as modulators of GABAalpha receptor function capable of anticonvulsant, anxiolytic, and antipanic action without eliciting tolerance or physical dependence. The present proposal is the continuation of an ongoing project aimed at understanding and further defining the role of "neurosteroids" in drug- induced positive and negative allosteric modulation of GABA action at GABAalpha receptors. The hypothesis of the proposed studies is that an increase of the GABAalpha receptor-active neurosteroid allopregnanolone (ALLO) elicited by fluoxetine or other SSRIs may be associated with the antidysphoric, anxiolytic, and anticonvulsant actions of this class of drugs. Because ALLO and related steroids acting at GABAalpha receptors play a putative role in modulating anxiety, mood, and cognitive behavior, to validate our hypothesis we need to establish whether the effect of SSRIs on neurosteroids is class-specific or whether other major classes of psychotherapeutic agents that act on mood and cognition (i.e., antidepressants, typical and atypical antipsychotics, benzodiazepines, cognition enhancers) alter neurosteroid biosynthesis and release. The focus on fluoxetine (Prozac), other serotonin reuptake inhibitors (SSRIs), and neurosteroids stems from our original observation that fluoxetine and other SSRIs increase the brain content of allopregnanolone (ALLO), which in nanomolar concentrations with nongenomic action positively modulates GABAalpha receptor function, and also decreases the brain content of 5alpha-dihydroprogesterone (5alpha- DHP), which also in nanomolar concentrations with genomic action (via progesterone receptors) may control GABAalpha receptor subunit gene expression. We propose a systematic investigation of the action of fluoxetine, other SSRIs, and other antidepressant and psychotherapeutic agents on brain neurosteroid biosynthesis and release with the following Specific Aims: (1) establish the onset and the duration of changes in ALLO and 5alpha- DHP content in various rat brain regions and microdialysates following administration of fluoxetine or other SSRIs; (2) determine if the increase of ALLO elicited by SSRIs is specific for this class of drugs; (3) establish if there is a correlation between the increase in brain and microdialysate ALLO content and the behavioral effects that reveal the strength of the underlying GABAergic transmission; (4) determine if the 33alpha-hydroxysteroid oxidoreductase enzymes are the target for fluoxetinee's action on neurosteroid treatment; and (5) study if neurosteroids alter GABAalpha receptor subunit expression following long-term fluoxetine treatment. Additional evidence that neurosteroid alterations are causally associated with SSRI drug treatment would enable a completely new insight into the development of safer and more efficacious antidepressants by focusing on their neurosteroidal actions.

DESCRIPTION: (Adapted from the Investigator's Abstract) Telencephalic brain structures are known to follow an unusually complex course of embryological developmental in which RELN plays a pivotal signaling role. In this project, we want to establish whether in postnatal and adult mice genetic RELN deficiency (RELN haploinsufficiency) leads to the expression of progressive and region-specific brain abnormalities and whether some of the abnormalities resemble caricatures of the alterations observed in the brain of schizophrenic patients. We have selected the heterozygous reeler mutant (rl +/-) mouse as a model of RELN haploinsufficiency. In preliminary experiments we have shown that in adult rl +/- mice, in addition to a lower than normal RELN expression there are a number of microanatomical and neurophysiological abnormalities reminiscent of schizophrenia. Thus, we propose to 1) compare whether RELN expression and neuronal location are similar in rl +/- and wild-type (wt) mice; 2) determine whether RELN haploinsufficiency elicits changes in the relationship existing between RELN haploinsufficiency and altered cortical distribution of GABAergic interneurons, decrease of GAD67 expression and decrease in neuropil density; 3) study if there is a progression of these changes with age, or if these changes are restricted to cortex or are present in other brain areas that are also affected in schizophrenia; and 4) examine if RELN haploinsufficiency is a vulnerability factor that predisposes peripubertal mice to an incidence of behavioral and microanatomical abnormalities when exposed to noxious insults such as social isolation stress. Although the putative interaction of RELN haploinsufficiency with epigenetic factors occurring in schizophrenic patients probably differs from the factors operative in socially isolated mice, study of the murine model may reveal those mechanisms of neurodevelopment that are likely to be disrupted by RELN deficiency in the brain of schizophrenic patients. The understanding of RELN expression regulation in adult brain may be of great importance in the design of pharmacological tools to correct RELN brain insufficiency and to reduce the behavioral and microanatomical abnormalities associated to this deficiency.

DESCRIPTION (provided by applicant): The finding that schizophrenia (SZ) is a disorder characterized by a decrease of reelin and GAD67 (Guidotti et al., 2000), an increase of DNMT1 mRNA expression in cortical GABAergic interneurons (Costa et al., 2002a), and a hypermethylation of the reelin promoter CpG islands (Grayson, personal communication) encouraged us to consider that hypermethylation of promoter CpG islands is a mechanism operative in the dysfunction of GABAergic neurons in SZ. This project's overarching objective is to develop animal models of epigenetic reelin and GAD67 expression downregulation. We hypothesize that the downregulation of reelin and GAD67 in cortical GABAergic neurons of SZ brains can be replicated in mouse telencephalic GABAergic neurons with protracted administration of L-methionine in doses that increase: i) the content of the methyl donor S-adenosyl-methionine;ii) DNA-(cytosine-5)-methyltransferase (DNMT1) activity;and iii) covalent cytosine residues methylated in position 5 on the CpG-rich promoter regions of reelin and GADe? genes. One of the regulatory mechanisms involved in the process of control of gene activity by DNMT1 is its accessibility to target DNA segments. This accessibility may be regulated by the acetylated or deacetylated status of the nucleosomal core histones, which is governed by the balance of the activities of histone acetyltransferases (HAT) and histone deacetylases (HDAC). Studies in the field of cancer suggest that the increased activity of DNMTs observed in tumor cells can be downregulated by reducing HDAC activities with specific inhibitors. Hence, we have focused our attention on the action of HDAC inhibitors (i.e., valproate and benzamides) as putative drugs that may, by increasing core histone tail acetylation at nucleosomal sites, normalize in nuclei of telencephalic GABAergic neurons-reelin and GAD67 expression downregulation induced by hypermethylation of reelin or GAD67 promoter CpG islands. Recent reports suggest that typical and atypical antipsychotics are more potent, more efficacious, and less toxic if they are co-administered with valproate (VPA). The beneficial effects in the treatment of SZ obtained with the weak HDAC inhibitor VPA suggest that more potent HDAC inhibitors may represent a new opportunity for pharmacological interventions of putative therapeutic value in mitigating vulnerability to SZ among high risk individuals.

DESCRIPTION (provided by applicant): Treatment of schizophrenia (SZ) patients with monoaminergic antagonist drugs (typical or atypical neuroleptics) has an antipsychotic effect, but negative symptoms and cognition are not significantly improved. Hence, there is an urgent need to find new molecular targets for the development of pharmacological agents active on cognitive deficits and negative symptoms. In this regard, nicotine receptor agonists are some of the most promising treatment options currently under investigation (Freedman et al, 2008). Postmortem examination of the brain of SZ patients reveals: a) a decrease of high and low affinity nicotinic acetylcholine receptor (nAChR) subtypes in telencephalic GABAergic neurons, which is possibly alleviated by nicotine abuse, and b) a neuropathology of cortical, hippocampal and striatal GABAergic neurons that includes decreased expression of GAD67, the 67 kDa isoform of the glutamic acid decarboxylase enzyme that synthesizes GABA, and decreased expression of other GABAergic proteins including reelin and the NR2A subunit of the NMDA receptor (Guidotti et al, 2005; Lewis et al, 2005; Lisman et al, 2008; Woo et al, 2004; 2008). Evidence suggests that this GABAergic neuropathology of SZ brain is accompanied by the overexpression of DNA-methyltransferases 1 and 3a (DNMT1 and DNMT3a) in GABAergic neurons (Costa et al, 2007; Ruzicka et al, 2007; Veldic et al, 2005; 2007; Zhubi et al, 2009). DNMTs are the enzymes that catalyze the methylation of the cytosine carbon atom in position 5' of CpG dinucleotides of several gene promoters. To model this neuropathology in mice, we will use offspring of mothers subjected to restraint stress during pregnancy. These mice are characterized by increased levels of DNMT1 and 3a and decreased GAD67. Our preliminary experiments in normal mice (Satta et al, 2008) suggest that nicotine, acting at central nAChRs (1422, 17) present in GABAergic neurons, reduces the expression of DNMT1 and elicits GAD67 promoter demethylation, thereby upregulating the expression of GAD67. Hence, the use of synthetic nAChR ligands to selectively downregulate DNMT in GABAergic neurons may be an innovative attempt to control the downregulation of GAD67 and other genes operative in selected populations of telencephalic GABAergic neurons of SZ patients, while leaving the function of DNMT in cells that do not express nAChRs intact. To be investigated is whether selective 1422 nAChR agonists (A-85380, ABT-594), partial agonists (AMOP-H-OH), or 17 nAChR agonists (PN -282987), partial agonists (GTS-21) or positive allosteric modulators (galantamine) are better suited to downregulate DNMT and upregulate GAD67 expression in telencephalic GABAergic neurons. Thus, the proposed research is aimed at studying the action of nAChR ligands on the epigenetic regulation of GABAergic neurons to better understand the pathogenesis and pharmacological treatment of sensory/cognitive components of SZ, bipolar disorder and other diseases with psychiatric or cholinergic components.

DESCRIPTION (provided by applicant): The goal of this proposed research is to determine whether differences in the expression of epigenetic related biomarkers are present in lymphocytes of people with genetic high risk and/or prodromal symptoms of schizophrenia (HRSK subjects) as previously found in lymphocytes of chronic schizophrenia (SZ) patients. We will also assess whether the presence of these biomarkers predicts which HRSK subjects will progressively develop more psychotic symptoms over a two year period and who will convert from HRSK status to first episode schizophrenia or another diagnosable psychotic disorder. Investigators on this proposal, have been collaborating with a group of investigators in Hunan, China for several years (Wu et al., 2008; 2012). The China group is currently conducting a multi-center study supported by the Chinese Ministry of Health that involves 5000 HRSK and 2000 control subjects. The research we are proposing would be an addendum to this larger study. DNA methylation and demethylation are key epigenetic components involved in orchestrating transcription of specific genes in post mitotic neurons (Guidotti et al., 2011). It has been suggested that the DNA methylation/demethylation process is perturbed in neuropsychiatric disorders such as schizophrenia (SZ), bipolar disorder (BP), and autism. We have shown that abnormalities in some of these epigenetic marks in the brain of SZ patients are also present in their lymphocytes. Lymphocytes of SZ patients have higher DNMT1 mRNA expression than non-psychotic controls (Zhubi et al 2009), higher levels of other methylating and demethylating enzymes (growth arrest and DNA damage-inducible protein [GADD45] and ten-eleven translocation protein [TET]), and lower levels of glucocorticoid receptor and GAD67 mRNA (preliminary data). If differences in these potential epigenetic biomarkers can be confirmed in lymphocytes of HRSK subjects, some of the underlying biochemical developmental pathology leading to SZ might be uncovered. This result could provide epigenetic biomarkers useful in predicting which HRSK subjects are more likely to develop full symptoms of SZ or a related psychosis. Identifying potential biomarkers would have implications for prognosis and identifying patients who would be most likely to benefit from intensive early intervention and specialized treatment. This background leads us to pursue three Specific Aims for the current proposed study: AIM1: Measure the expression level of DNA methylation related genes in lymphocytes of HRSK subjects at baseline compared to controls and first episode and chronic SZ patients. AIM 2: Determine whether epigenetic chromatin status and the expression of SZ candidate genes is regulated by the binding of proteins that are constituents of the DNA- methylation/demethylation pathways in lymphocytes of HRSK subjects at baseline compared to controls and first episode and chronic SZ patients. AIM 3 Determine whether HRSK subjects who also have epigenetic lymphocyte abnormalities at baseline are more likely to progress to more severe psychotic-like symptoms or convert to diagnosable psychotic status during a two year follow up. This proposed research is potentially a major step toward the identification of objective biomarkers capable of predicting conversion to psychosis in HRSK subjects.

Emerging evidence suggests that an epigenetic misregulation of the genome, including brain region-specific altered DNA promoter methylation, is associated with the neuropathological manifestations of alcohol abuse and dependence. DNA (cytosine) methylation generally has been regarded as a highly stable epigenetic mark that ensures and maintains neuronal phenotype identity. However, the epigentic DNA marking is highly dynamic and requires the action of a family of DNA-methyltransferases (DNMT) and an active DNA-demethylation pathway (base excision repair [BER] pathway) that includes 5methyl cytosine (5MC) hydroxylation by a ten-eleven-translocation (TET) and deamination by an apolipoprotein B mRNA editing enzyme (Apobec). Our preliminary studies in bipolar (BP) disorder and major depressed (MD) patients who are also chronic alcoholics show decreased DNMT and a marked increase of TET-1 expression in the prefrontal cortex compared to controls. In BP and MD patients with comorbid alcoholism, there is also increased 5-hydroxyMC at GAD67 and BDNF promoters. The objective of our study is to investigate the expression and promoter binding of components of the DNA-methylation and DNA-demethylation network in corticolimbic structures of chronic uncomplicated alcoholics (no developmental disorders, no other psychiatric and neurological disorders) who consumed greater than 80 g of ethanol/day, obtained from the New South Wales Tissue Resource Center (see attached letter). Our working hypothesis is that DNA-methylation/demethylation dynamics may be involved in behavioral aspects of alcohol exposure in part through epigenetically mediated disruption of the balance between glutamatergic /GABAergic transmission and synaptic plasticity at corticolimbic circuit neurons. Testing this hypothesis in a cohort of uncomplicated chronic alcoholic subjects will reveal novel alcohol sensitive targets with obvious important therapeutic and public health implications, protein networks in human brain has obvious therapeutic and public health implications.

Project Summary Environmental factors, including alcohol abuse and stress, cause long-lasting changes in the regulation of gene expression in the brain via epigenetic mechanisms, such as DNA methylation. Similar to stress, alcohol stimulates glucocorticoid release that bind to specific receptors, i.e., the glucocorticoid receptor (encoded by NR3C1). As of today, little is known on the role of epigenetic DNA modifications in regulating the transcriptome in the human prefrontal cortex (PFC, BA10) and rat PFC during chronic alcohol exposure and withdrawal. The goal of research component #4 is to interrogate genome-wide changes in DNA methylation of novel gene networks, including the NR3C1 gene network in AUD patients. Additional goals are to study whether and how altered DNA methylation and/or hydroxymethylation marks underlie the pathophysiology of AUD. The genome- wide DNA methylation approach (Infinium MethylationEPICBeadChip, Illumina) will be used in prefrontal cortex samples obtained from 30 pairs of controls and AUD subjects from the New South Wales Tissue Resource Centre (University of Sydney, Australia). In preliminary studies we identified a differential pattern of total DNA methylation in AUD for 5,254 genes. However, this technique does not differentiate 5-methyl-cytosine (5mC) from 5-hydroxymethyl-cytosine (5hmC). Here, we propose to investigate the genome?wide distribution of 5mC and 5hmC using the TET bisulfite conversion method followed by the Human MethylationEPIC BeadChip assay. Hence, we will examine the enrichment of 5hmC/5mC in association with changes in novel gene expression measured by RNA-seq. Chromatin accessibility in association with previously identified epigenetic marks will be assessed by Assay for Transposase-Accessible Chromatin sequencing (ATAC-seq). Integration of different whole-genome approaches, i.e. genome-wide DNA methylation, RNA-seq and ATAC-seq will allow an in-depth investigation of the status of the epigenome in AUD. Additionally, using a reverse-translational approach, we propose to mechanistically investigate downstream effects of DNA methylation on neuronal function in PFC and on anxiety-like behaviors and escalation of alcohol self-administration in rats treated chronically with alcohol or following a 24 h alcohol-withdrawal. Because we observed an increase of DNA methylation associated with a downregulation of TET expression (the enzyme that catalyzes the conversion of 5mC to 5hmC), we propose the use of a dCas9-Tet1-mediated protein approach to correct methylation deficits at the levels of NR3C1 and other gene promoters in the PFC of rats and determine their effect on gene expression, anxiety and drinking behaviors. The proposed study will help to identify in the human and rat brain novel epigenetic mechanisms that may provide new therapeutic targets for the treatment of AUD.