Alberto Ambesi-Impiombato

How to infer gene networks from expression profiles

Inferring, or ‘reverse‐engineering’, gene networks can be defined as the process of identifying gene interactions from experimental data through computational analysis. Gene expression data from microarrays are typically used for this purpose. Here we compared different reverse‐engineering algorithms for which ready‐to‐use software was available and that had been tested on experimental data sets. We show that reverse‐engineering algorithms are indeed able to correctly infer regulatory interactions among genes, at least when one performs perturbation experiments complying with the algorithm requirements. These algorithms are superior to classic clustering algorithms for the purpose of finding regulatory interactions among genes, and, although further improvements are needed, have reached a discreet performance for being practically useful.

Permanent Focal Brain Ischemia Induces Isoform-Dependent Changes in the Pattern of Na+/Ca2+ Exchanger Gene Expression in the Ischemic Core, Periinfarct Area, and Intact Brain Regions

Dysregulation of sodium [Na+]i and calcium [Ca2+]i homeostasis plays a pivotal role in the pathophysiology of cerebral ischemia. Three gene products of the sodium–calcium exchanger family NCX1, NCX2, and NCX3 couple, in a bidirectional way, the movement of these ions across the cell membrane during cerebral ischemia. Each isoform displays a selective distribution in the rat brain. To determine whether NCX gene expression can be regulated after cerebral ischemia, we used NCX isoform-specific antisense radiolabeled probes to analyze, by radioactive in situ hybridization histochemistry, the pattern of NCX1, NCX2, and NCX3 transcripts in the ischemic core, periinfarct area, as well as in nonischemic brain regions, after 6 and 24 h of permanent middle cerebral artery occlusion (pMCAO) in rats. We found that in the focal region, comprising divisions of the prefrontal, somatosensory, and insular cortices, all three NCX transcripts were downregulated. In the periinfarct area, comprising part of the motor cortex and the lateral compartments of the caudate-putamen, NCX2 messenger ribonucleic acid (mRNA) was downregulated, whereas NCX3 mRNA was significantly upregulated. In remote nonischemic brain regions such as the prelimbic and infralimbic cortices, and tenia tecta, both NCX1 and NCX3 transcripts were upregulated, whereas in the medial caudate-putamen only NCX3 transcripts increased. In all these intact regions, NCX2 signal strongly decreased. These results indicate that NCX gene expression is regulated after pMCAO in a differential manner, depending on the exchanger isoform and region involved in the insult. These data may provide a better understanding of each NCX subtypes pathophysiologic role and may allow researchers to design appropriate pharmacological strategies to treat brain ischemia.

Homer 1a Gene Expression Modulation by Antipsychotic Drugs: Involvement of the Glutamate Metabotropic System and Effects of D-Cycloserine

N-methyl-D-aspartate receptor hypofunction has been suggested to play a role in the pathophysiology of schizophrenia. New glutamatergic mechanisms involving metabotropic receptors have been recently proposed to further expand this hypothesis. “Homer” is a family of postsynaptic density proteins functionally and physically attached to glutamate metabotropic receptors. We investigated the activation of the early gene form of Homer after acute treatment with typical or atypical antipsychotic drugs alone or with the adjunction of D-cycloserine. This activation was compared with that of c-fos, considered a putative molecular marker of brain regions activated by antipsychotics. Male Sprague-Dawley rats were treated intraperitoneally with haloperidol (0.8 mg/Kg) or clozapine (15 mg/Kg) alone or with the adjunction of D-cycloserine (20 mg/Kg). Rats were sacrificed ninety minutes after injection and the brains were processed for quantitative in situ hybridization histochemistry. Haloperidol induced a statistically significant increase of Homer both in caudate-putamen and nucleus accumbens compared with controls; clozapine induced Homer significantly only in the accumbens. The adjunction of D-cycloserine attenuated the haloperidol-induced increase of Homer expression in caudate-putamen and nucleus accumbens and attenuated the clozapine-induced increase in the accumbens. The c-fos gene expression was potently induced by haloperidol in caudate-putamen and nucleus accumbens, and by clozapine only in the accumbens. The adjunction of D-cycloserine enhanced c-fos expression only for clozapine in both regions of the forebrain. These results demonstrate a differential involvement of glutamatergic metabotropic system in gene expression modulation induced by typical or atypical antipsychotic drugs and may suggest new molecular basis for the augmentation strategy by a glycine site partial agonist.

Acute administration of antipsychotics modulates Homer striatal gene expression differentially

Typical and atypical antipsychotics, the mainstay of schizophrenia pharmacotherapy, have been demonstrated to affect differently neuronal gene expression in several preclinical paradigms. Here we report the differential gene expression of the glutamatergic post-synaptic density proteins Homer and PSD-95 in rat forebrain following acute haloperidol or olanzapine treatment. Moreover, considering the extensive interactions between dopaminergic and opioidergic systems we also measured striatal preproenkephalin mRNA. Male Sprague-Dawley rats were treated with haloperidol 1 mg/kg or olanzapine 0.5 mg/kg or vehicle, i.p. and sacrificed 3 h after the injection. Homer gene expression was significantly increased in caudate putamen and nucleus accumbens of rats treated with haloperidol and in the core of accumbens of rats treated with olanzapine. No changes were detected for Homer in prefrontal and parietal cortex in any of the experimental groups. PSD-95 gene expression was not modulated in our paradigm by administration of either typical or atypical antipsychotics. These results (1) suggest a differential modulation of Homer by typical and atypical antipsychotics; (2) confirm that Homer can be induced as an early gene with putative direct effect on neuronal plasticity and (3) demonstrate different response to antipsychotics by different classes of postsynaptic density proteins at glutamatergic synapses.

Dopamine receptor subtypes contribution to Homer1a induction: insights into antipsychotic molecular action.

The inducible gene Homer1a has been considered a candidate gene for schizophrenia. Drugs efficacious in schizophrenia and acting as dopamine receptor antagonists induce Homer1a expression, although the specific role of the different dopamine receptors in its induction is not completely known. In this study, we explored Homer1a expression induced by selective antagonists at dopamine receptors (SCH-23390, D(1) receptor selective antagonist, 0.5 mg/kg; L-741,626, D(2) receptor selective antagonist, 2 mg/kg; U-99194, D(3) receptor selective antagonist, 5 mg/kg; L-745,870, D(4) receptor selective antagonist, 3 mg/kg), haloperidol (0.8 mg/kg), and terguride (0.5 mg/kg), a partial agonist at D(2) receptors. Moreover, we evaluated the expression of two Homer1a-related genes which play essential roles in synaptic plasticity: mGluR5 and Homer1b. Gene expression was analyzed in brain regions relevant for schizophrenia pathophysiology and therapy, namely the striatum, the cortex, and the hippocampus. In striatum, Homer1a was induced by D(2) receptor antagonists and, with a different distribution, by SCH-23390. In the cortex, Homer1a was differentially induced by D(1), D(2), and D(3) receptors antagonists, while haloperidol and terguride did not affect or reduced its expression. Homer1b expression was reduced by L-741,626, L-745,870, terguride, and haloperidol in the ventral caudate-putamen, in the nucleus accumbens and in the cortex, while SCH-23390 increased the expression in the core of the accumbens. mGluR5 expression was increased by SCH-23390 in the dorsomedial putamen, the core of the accumbens, and in some hippocampal subregions. A reduction of gene expression by terguride and an increase by L-745,870 was observed in the dorsomedial putamen. The changes in expression suggest that these gene transcripts are differentially regulated by antagonism at different dopamine receptors.

Method for quantitative in situ hybridization histochemistry and image analysis applied for Homer1a gene expression in rat brain.

Here we describe the detailed method for quantitative in situ hybridization histochemistry adopted in our previously published short communication on differential gene expression of the postsynaptic density proteins Homer and PSD-95 in rat forebrain after acute treatment with antipsychotic drugs [de Bartolomeis et al., Mol. Brain Res. 98 (2002) 124-129]. Specific 35S radiolabeled oligodeoxyribonucleotides were used to hybridize rat forebrain sections and data analysis was carried out on digitalized images acquired by means of a CCD camera. Special emphasis has been posed on data preprocessing options applied to a dataset obtained using a transparency scanner as an alternative image capturing method.

Pattern of acute induction of Homer1a gene is preserved after chronic treatment with first- and second-generation antipsychotics: effect of short-term drug discontinuation and comparison with Homer1a-interacting genes

Homer1a is a glutamate-related gene whose expression is induced by antipsychotics acutely (i.e. 90 min after treatment). Acute Homer1a expression is preserved after prolonged antipsychotic treatments, while the effects of short-term discontinuation after chronic antipsychotic treatment have not yet been assessed. Here, we studied early and long-term effects on gene expression by antipsychotics for Homer1a and other components of glutamatergic synapses. In the first paradigm, we evaluated Homer1a acute expression by single administration of antipsychotics (haloperidol 0.8 mg/kg, ziprasidone 10 and 4 mg/kg, clozapine 15 mg/kg). Haloperidol and ziprasidone induced Homer1a in the striatum. Induction by ziprasidone was dose-dependent. These results suggest that acute Homer1a expression correlates with dopaminergic affinity and motor side effects of antipsychotics. In the second paradigm, we studied antipsychotic-mediated long-term changes in Homer1a and glutamate-related genes. Rats were treated (21 days) with haloperidol 0.8 mg/kg, ziprasidone 4 mg/kg, or vehicle, and then sacrificed at 90 min (early time-point) or 24 h (delayed time-point) after last injection. Gene expression at these two time-points was compared. Homer1a preserved its pattern of expression at the early but not at the delayed time-point. Significant changes were also observed for PSD-95. The results suggest that Homer1a preserves its expression profile after chronic antipsychotics.

Ketamine-related expression of glutamatergic postsynaptic density genes: possible implications in psychosis.

Systemic administration of ketamine, a non-competitive antagonist of the N-methyl-d-aspartate receptor (NMDA-R), produces a condition of NMDA-R hypofunction, which is considered one of the putative molecular mechanisms involved in psychosis. In this study, we evaluated the effect of ketamine on glutamatergic markers of the postsynaptic density (PSD), a pivotal site for dopamine-glutamate interaction. We assessed gene expression of Homer1a, alpha and betaCaMKII, and dopamine transporter (DAT) by two different doses of ketamine. These genes were chosen because of their impact on signal transduction and dopamine-glutamate interplay in postsynaptic density. Moreover, Homer1a is modulated by antipsychotics and represents a candidate gene for schizophrenia. Male Sprague-Dawley rats were injected with saline, 12mg/kg ketamine or 50mg/kg ketamine, and sacrificed 90 minutes after injections. In situ hybridization histochemistry was used to quantitate the rate of gene expression in rat forebrain. Homer1a was induced by 50mg/kg ketamine in ventral striatum and by both 50 and 12mg/kg ketamine in nucleus accumbens, whereas gene expression was not affected in dorsal striatum. alphaCaMKII was increased by 12mg/kg ketamine against saline in almost all subregions assessed. betaCaMKII was not affected by ketamine. DAT was increased by both doses of ketamine in the ventro-tegmental area and substantia nigra pars compacta. We suggest that these changes may represent molecular adaptations to the perturbation in glutamatergic transmission induced by ketamine blockade of NMDA receptors and may be implicated in molecular alterations occurring in schizophrenia.

Decreased gene expression of calretinin and ryanodine receptor type 1 in tottering mice

Tottering mice are a spontaneously occurring animal model of human absence epilepsy. They carry a mutation in the P/Q-type calcium channel alpha1A subunit gene which is highly expressed by cerebellar Purkinje cells. In this study, we investigated the role of calretinin and ryanodine receptor type 1 (RyR1) gene expression in the cerebellum of tottering mice. Cerebellar tissue specimens from four experimental groups were processed for in situ hybridization histochemistry (ISHH): (1) wild-type (+/+); (2) heterozygous (tg/+) and two homozygous groups; either (3) without occurrence of an episode of paroxysmal dyskinesia (tg/tg-N); or (4) after an episode of paroxysmal dyskinesia (tg/tg-P) that lasted about 45 min on average. Quantitative analysis showed a statistically significant decrease (p = 0.0001, ANOVA) of calretinin gene expression at the level of the simple lobule of the cerebellum in both homozygous groups compared to the wild-type and heterozygous groups. RyR1 was decreased in the flocculus of the cerebellum in both the tg/tg-N and tg/tg-P groups compared to wild type (p = 0.0174, ANOVA). These results suggest that calretinin gene expression, as well as other genes involved in regulation of calcium homeostasis, such as RyR1, may play a role in the biochemical functional alterations present in tottering mice.