Radmila Manev

Antidepressants alter cell proliferation in the adult brain in vivo and in neural cultures in vitro.

The action of antidepressants on cell proliferation (bromodeoxyuridine (BrdU) or [3H]thymidine incorporation) was studied in the adult rat hippocampus in vivo and in neural precursors (immature rat cerebellar granule cells) in vitro. In vivo, prolonged (21 days) but not acute (single) intraperitoneal treatment with fluoxetine (5 mg/kg) resulted in a 3.4-fold increase of bromodeoxyuridine-positive cells in the subgranular zone of the dentate gyrus. In cell cultures, at 1 and 10 days in vitro, 48-h fluoxetine exposure (1 microM, which is comparable to therapeutic plasma concentrations) reduced thymidine incorporation when initiated at 1 day in vitro, but increased cell proliferation when initiated at 10 days in vitro. Clomipramine and imipramine produced similar action in vitro; desipramine was ineffective.

The phenotypic characteristics of heterozygous reeler mouse

Histological and behavioral traits are associated with reelin (Reln) haplo-insufficiency in heterozygous reeler mouse (rl+/-). These phenotypic traits are an approximately 50% decrease of brain Reln mRNA and Reln protein, an accumulation of nicotinamide-adenine dinucleotide phosphate-diaphorase (NADPH-d)-positive neurons in subcortical white matter, an age-dependent decrease in prepulse inhibition of startle (PPI), and neophobic behavior on the elevated plus-maze. Possible analogies between these rl+/- phenotypic traits and signs of psychosis vulnerability are discussed.

Fluoxetine increases the content of neurotrophic protein S100β in the rat hippocampus

Recent studies indicate that a protracted daily administration of the antidepressant fluoxetine to adult rats increases cell proliferation/neurogenesis in the hippocampus. It has been hypothesized that this action of fluoxetine might be mediated by neurotrophic factors. We hypothesized that glial S100beta could be such a factor, and using quantitative Western immunoblotting, we investigated the effect of a 21-day treatment of rats with fluoxetine (5 mg/kg), and found that fluoxetine increases the content of hippocampal S100beta.

Valproate administration to mice increases histone acetylation and 5-lipoxygenase content in the hippocampus.

Gene expression can be regulated by chromatin remodeling induced by the opposing actions of histone acetyltransferases and histone deacetylases (HDAC). HDAC inhibitors are considered putative anti-cancer drugs, but may also alter gene expression in the brain. Valproic acid (valproate; VPA), a drug used for treatment of bipolar disorder, has been characterized as a HDAC inhibitor. In neuronal cultures, VPA increases the expression of 5-lipoxygenase (5-LOX). Here we show that in vivo treatment of mice with intraperitoneal VPA injections increases the acetylation of histone H3 and the content of 5-LOX immunoreactive protein in the hippocampus. Since the extent of 5-LOX expression may alter mouse behavior, we suggest that VPA-altered chromatin remodeling and 5-LOX expression in the brain may be functionally important.

Trans-azetidine-2,4-dicarboxylic acid activates neuronal metabotropic receptors.

The expression of metabotropic glutamate receptors (mGluRs) in primary cultures of cerebellar granule neurones can be: (i) modulated by the degree of depolarization during the culture period, rendering neurones differently sensitive to agonist-stimulated inositol phosphate (IP) hydrolysis; (ii) down-regulated by specific mGluR agonists. In this culture the new rigid glutamate analogue, (+/-)-trans-azetidine-2,4-dicarboxylic acid (t-ADA) and the known mGluR agonist 1S,3R-aminocyclopentane-1,3-dicarboxylic acid (1S,3R-ACPD) stimulated IP formation in line with the depolarization-modified expression of mGluR1. However, the two compounds caused different patterns of mGluR down-regulation. The effects of t-ADA and 1S,3R-ACPD were also tested on transformed human embryonic kidney 293 cells transfected with mGluR1. Only 1S,3R-ACPD, but not t-ADA, stimulated IP hydrolysis, suggesting that t-ADA acts on a subtype of metabotropic receptors different from mGluR1. Hence, t-ADA might prove useful in differentiating the function of various mGluR subtypes.

Glia as a putative target for antidepressant treatments

BACKGROUND Since the early 1950s, various molecular mechanisms have been invoked to explain how antidepressants work. The most recent suggests that pharmacologically stimulated adult neurogenesis might be involved. Surprisingly, in the adult brain, an important source of new neurons and possibly mediators of neurogenesis appears to be glia, i.e., astrocytes. We have recently shown that protracted administration of the antidepressant fluoxetine to adult rats upregulated the astrocytic protein S100beta content and increased neurogenesis in the hippocampus. METHODS Rats were treated with fluoxetine for 21 days; before sacrifice bromodeoxyuridine (BrdU) was injected to label the proliferating cells. Immunofluorescence was used to identify proliferating BrdU-positive cells, and cells immunopositive for S100beta and its receptor receptor for advanced glycation end products (RAGE). RESULTS Typically, S100beta-positive cells were observed in the vicinity of BrdU-positive cells. On the other hand, we observed colocalization of RAGE receptors and BrdU immunoreactivities, suggesting that some proliferating cells express these receptors for S100beta. RAGE expression by neuronal cells or neuronal precursors and its activation by S100beta may promote their survival. LIMITATIONS The anatomical localization of hippocampal S100beta, its receptor RAGE, and BrdU-positive cells that we describe in this study is only indicative of a putative role for glia in antidepressant-stimulated neurogenesis. Functional in vitro and in vivo studies are needed to directly investigate this role; quantitative assays and time-course studies are also warranted. CONCLUSION We propose that a better understanding of glia functioning could establish its role as a target for novel antidepressant treatments.

5-Lipoxygenase is required for proliferation of immature cerebellar granule neurons in vitro

Primary cultures of rat cerebellar granule neurons express 5-lipoxygenase, an enzyme from the inflammatory pathway of arachidonic acid. Outside the central nervous system (CNS) 5-lipoxygenase participates in cell proliferation. We hypothesized that 5-lipoxygenase is needed for proliferation of immature cerebellar granule neurons. Using cultures prepared from 7-day-old rat pups, we confirmed in vitro neurogenesis by immunocytolabeling with 2-bromo-5-deoxyuridine and beta-tubulin isotype III and quantified the rate of cell proliferation by assaying [3H]thymidine incorporation. We found that immature cerebellar granule neurons express large amounts of 5-lipoxygenase, and that treatment with a 5-lipoxygenase antisense, to reduce expression of this gene, decreased significantly (by 60%) the content of 5-lipoxygenase protein and effectively reduced cell proliferation. [3H]thymidine incorporation was significantly reduced by each of the three 5-lipoxygenase inhibitors we tested: AA-861 [2-(12-hydroxydodeca-5, 10-diynyl)-3,5,6-trimethyl-1,4-benzoquinone], MK-886 (C(27)H(33)ClNO(2)S.Na), and L-655,238 [alpha-penyl-3-(2-quinolinylmethoxy)-benzenemethanol]. Their anti-proliferative effect was reversible. We propose that neuronal expression of 5-lipoxygenase is crucial for neurogenesis in vitro, and possibly also in vivo.

Cyclooxygenases and 5-lipoxygenase in Alzheimer's disease

Typically, cyclooxygenases (COXs) and 5-lipoxygenase (5-LOX), enzymes that generate biologically active lipid molecules termed eicosanoids, are considered inflammatory. Hence, their putative role in Alzheimers disease (AD) has been explored in the framework of possible inflammatory mechanisms of AD pathobiology. More recent data indicate that these enzymes and the biologically active lipid molecules they generate could influence the functioning of the central nervous system and the pathobiology of neurodegenerative disorders such as AD via mechanisms different from classical inflammation. These mechanisms include the cell-specific localization of COXs and 5-LOX in the brain, the type of lipid molecules generated by the activity of these enzymes, the type and the localization of receptors selective for a type of lipid molecule, and the putative interactions of the COXs and 5-LOX pathways with intracellular components relevant for AD such as the gamma-secretase complex. Considering the importance of these multiple and not necessarily inflammatory mechanisms may help us delineate the exact nature of the involvement of the brain COXs and 5-LOX in AD and would reinvigorate the search for novel targets for AD therapy.

Pathological Phosphorylation Causes Neuronal Death: Effect of Okadaic Acid in Primary Culture of Cerebellar Granule Cells

Abstract: We have investigated the role of protracted phosphatase inhibition and the consecutive protracted protein phosphorylation on neuronal viability. We found that in primary cultures of cerebellar granule neurons, the protracted (24‐h) inhibition of the serine/threonine protein phosphatases 1 and 2A (EC 3.1.3.16) by treatment of the cultures with okadaic acid (OKA; 5–20 nM) caused neurotoxicity that could be inhibited by the protein kinase inhibitor l‐(5‐isoquinolinylsulfonyl)‐2‐methylpiperazine (H7) or by the previous down‐regulation of the neuronal protein kinase C (PKC; ATP:protein phosphotransferase; EC 2.7.1.37). PKC was down‐regulated by exposure of the cultures for 24 h to 100 nM phorbol 12‐myristate 13‐acetate (TPA). The effect of the drugs used in the viability studies on the pattern of protein phosphorylation was measured by quantitative autoradiography. In particular, the 50‐ and 80‐kDa protein bands showed dramatic changes in the degree of phosphorylation: increase by OKA and brief TPA treatment; decrease by H7 or 24 h of TPA treatment; and inhibition of the OKA‐induced increase by H7 or 24 h of TPA treatment. The results suggest that the protracted phosphorylation, in particular that mediated by PKC, may lead to neuronal death and are in line with our previous suggestion that prolonged PKC translocation is operative in glutamate neurotoxicity.

Minocycline increases phosphorylation and membrane insertion of neuronal GluR1 receptors.

The tetracycline antibiotic minocycline beneficially affects neuronal functioning and also inhibits the enzyme 5-lipoxygenase (5-LOX). We hypothesized that similar to 5-LOX inhibitors, minocycline may increase phosphorylation and membrane insertion of the glutamate receptor GluR1. The experiments were performed in primary cultures of mouse striatal neurons and in the prefrontal cortex and striatum of minocycline-treated mice. In vitro, low micromolar minocycline concentrations increased GluR1 phosphorylation at Ser845 and Ser831 and increased the surface content of GluR1. Minocycline also increased GluR1 phosphorylation in vivo. Increased GluR1 phosphorylation and minocycline treatment have been associated with antidepressant and memory-enhancing activities. Direct consequences of minocycline-increased GluR1 phosphorylation are yet to be established.