3D analysis of synaptic vesicle density and distribution after acute foot-shock stress by using serial section transmission electron microscopy.

Behavioural stress has shown to strongly affect neurotransmission within the neocortex. In this study, we analysed the effect of an acute stress model on density and distribution of neurotransmitter-containing vesicles within medial prefrontal cortex. Serial section transmission electron microscopy was employed to compare two groups of male rats: (1) rats subjected to foot-shock stress and (2) rats with sham stress as control group. Two-dimensional (2D) density measures are common in microscopic images and are estimated by following a 2D path in-section. However, this method ignores the slant of the active zone and thickness of the section. In fact, the active zone is a surface in three-dimension (3D) and the 2D measures do not accurately reflect the geometric configuration unless the active zone is perpendicular to the sectioning angle. We investigated synaptic vesicle density as a function of distance from the active zone in 3D. We reconstructed a 3D dataset by estimating the thickness of all sections and by registering all the image sections into a common coordinate system. Finally, we estimated the density as the average number of vesicles per area and volume and modelled the synaptic vesicle distribution by fitting a one-dimensional parametrized distribution that took into account the location uncertainty due to section thickness. Our results showed a clear structural difference in synaptic vesicle density and distribution between stressed and control group with improved separation by 3D measures in comparison to the 2D measures. Our results showed that acute foot-shock stress exposure significantly affected both the spatial distribution and density of the synaptic vesicles within the presynaptic terminal.

Stress and corticosterone increase the readily releasable pool of glutamate vesicles in synaptic terminals of prefrontal and frontal cortex.

Stress and glucocorticoids alter glutamatergic transmission, and the outcome of stress may range from plasticity enhancing effects to noxious, maladaptive changes. We have previously demonstrated that acute stress rapidly increases glutamate release in prefrontal and frontal cortex via glucocorticoid receptor and accumulation of presynaptic SNARE complex. Here we compared the ex vivo effects of acute stress on glutamate release with those of in vitro application of corticosterone, to analyze whether acute effect of stress on glutamatergic transmission is mediated by local synaptic action of corticosterone. We found that acute stress increases both the readily releasable pool (RRP) of vesicles and depolarization-evoked glutamate release, while application in vitro of corticosterone rapidly increases the RRP, an effect dependent on synaptic receptors for the hormone, but does not induce glutamate release for up to 20 min. These findings indicate that corticosterone mediates the enhancement of glutamate release induced by acute stress, and the rapid non-genomic action of the hormone is necessary but not sufficient for this effect.

Pharmacological characterization of BDNF promoters I, II and IV reveals that serotonin and norepinephrine input is sufficient for transcription activation.

Compelling evidence has shown that the effects of antidepressants, increasing extracellular serotonin and noradrenaline as a primary mechanism of action, involve neuroplastic and neurotrophic mechanisms. Brain-derived neurotrophic factor (BDNF) has been shown to play a key role in neuroplasticity and synaptic function, as well as in the pathophysiology of neuropsychiatric disorders and the mechanism of action of antidepressants. The expression of BDNF is mediated by the transcription of different mRNAs derived by the splicing of one of the eight 5' non-coding exons with the 3' coding exon (in rats). The transcription of each non-coding exon is driven by unique and different promoters. We generated a gene reporter system based on hippocampal and cortical neuronal cultures, in which the transcription of luciferase is regulated by BDNF promoters I, II, IV or by cAMP response element (CRE), to investigate the activation of selected promoters induced by monoaminergic antidepressants and by serotonin or noradrenaline agonists. We found that incubation with fluoxetine or reboxetine failed to induce any activation of BDNF promoters or CRE. On the other hand, the incubation of cultures with selective agonists of serotonin or noradrenaline receptors induced a specific and distinct profile of activation of BDNF promoters I, II, IV and CRE, suggesting that the monoaminergic input, absent in dissociated cultures, is essential for the modulation of BDNF expression. In summary, we applied a rapidly detectable and highly sensitive reporter gene assay to characterize the selective activation profile of BDNF and CRE promoters, through specific and different pharmacological stimuli.

Emotional memory impairments in a genetic rat model of depression: involvement of 5-HT/MEK/Arc signaling in restoration.

Cognitive dysfunctions are common in major depressive disorder, but have been difficult to recapitulate in animal models. This study shows that Flinders sensitive line (FSL) rats, a genetic rat model of depression, display a pronounced impairment of emotional memory function in the passive avoidance (PA) task, accompanied by reduced transcription of Arc in prefrontal cortex and hippocampus. At the cellular level, FSL rats have selective reductions in levels of NMDA receptor subunits, serotonin 5-HT(1A) receptors and MEK activity. Treatment with chronic escitalopram, but not with an antidepressant regimen of nortriptyline, restored memory performance and increased Arc transcription in FSL rats. Multiple pharmacological manipulations demonstrated that procognitive effects could also be achieved by either disinhibition of 5-HT(1A)R/MEK/Arc or stimulation of 5-HT4R/MEK/Arc signaling cascades. Taken together, studies of FSL rats in the PA task revealed reversible deficits in emotional memory processing, providing a potential model with predictive and construct validity for assessments of procognitive actions of antidepressant drug therapies.

Serine/threonine kinases as molecular targets of antidepressants: implications for pharmacological treatment and pathophysiology of affective disorders.

It is currently a widely accepted opinion that adaptive, plastic changes in the molecular and cellular components of neuronal signaling systems correlate with the effects on mood and cognition observed after long-term treatment with antidepressant drugs. Protein phosphorylation represents a key step for most signaling systems, and it is involved in the regulation of virtually all cellular functions. Two serine/threonine kinases, Ca2+ /calmodulin-dependent protein kinase II and cyclic AMP-dependent protein kinase, have been shown to be activated in the brain following antidepressant treatment. The changes in kinase activity are mirrored by changes in the phosphorylation of selected protein substrates in subcellular compartments (presynaptic terminals and microtubules), which, in turn, may contribute to the modulation of synaptic transmission observed with antidepressants. The molecular consequences of protein kinase activation may account for some of the alterations in neural function induced by antidepressants, and may suggest novel possible strategies of pharmacological intervention.

Long-term treatment with S-adenosylmethionine induces changes in presynaptic CaM kinase II and synapsin I.

BACKGROUND: According to current hypotheses, antidepressant drug action is the result of adaptive changes in neuronal signaling mechanisms rather than a primary effect on neurotransmitter transporters, receptors, or metabolic enzymes. Among the signaling mechanisms involved, protein kinases and phosphorylation have been shown to be modified by drug treatment. Presynaptic signaling (calcium/calmodulin-dependent protein kinase II [CaMKII]) and the protein machinery regulating transmitter release have been implicated in the action of these drugs. METHODS: We investigated the effect of S-adenosylmethionine (SAM), a compound with putative antidepressant activity, on presynaptic CaMKII and its synaptic vesicle substrate synapsin I. The activity of CaMKII was assayed in synaptic subcellular fractions prepared from hippocampus (HI), frontal cortex (FCX), striatum (STR), and parieto-temporal cortex. RESULTS: The kinase activity was increased after SAM treatment in the synaptic vesicle fraction of HI (31.7%), FCX (35.9%), and STR (18.4%). The protein level of CaMKII was also increased in synaptic vesicles of HI (40.4%). The synapsin I level was unchanged in synaptic vesicles but markedly increased in synaptic cytosol of HI (75.8%) and FCX (163.0%). No changes for both CaMKII and synapsin I level were found in homogenates, suggesting that synaptic protein changes are not explained by an increase in total level of proteins, but rather by translocation to nerve terminals. CONCLUSIONS: Similar to typical antidepressant drugs, SAM induces changes in CaMKII activity and increases synapsin I level in HI and FCX nerve terminals, suggesting a modulatory action on transmitter release.

Synaptotagmin is endogenously phosphorylated by Ca2+/calmodulin protein kinase II in synaptic vesicles.

The cytoplasmic domain of synaptotagmin (a synaptic vesicle-specific protein) has a high degree of homology with the Ca(2+)-phospholipid binding domain of protein kinase C. The Ca(2+)-phospholipid binding activity of synaptotagmin has been implicated in the docking and fusion of synaptic vesicles with the presynaptic membrane during Ca(2+)-induced exocytosis. The protein sequence contains potential phosphorylation sites for various protein kinases which could modulate its binding activity. At present there is no clear evidence that the protein is endogenously phosphorylated in intact vesicles. Here it is reported that phospho-synaptotagmin was immunoprecipitated from endogenously phosphorylated synaptic vesicles. The conditions used indicate that synaptotagmin, as synapsin I, is phosphorylated by Ca2+/calmodulin-dependent protein kinase II.

Modifications in brain CaM kinase II after long-term treatment with desmethylimipramine.

The present study investigated the effect of long-term (15 mg/kg for 15 days) and acute (15 mg/kg, single administration) treatment with desmethylimipramine, a tricyclic antidepressant drug, on calcium/calmodulin-dependent protein kinase II (CaMKII), a kinase implicated in the mechanism of antidepressant drug action. Similar to selective and non-selective serotonin reuptake inhibitors, long-term, but not acute, treatment with desmethylimipramine markedly increased the activity of CaMKII in the hippocampal synaptic vesicle fraction (+51.9%). The kinase activity was also increased in the same fraction of frontal cortex (+24.2%) and in the striatum (+45.9%), although in this last area the mechanism appeared to be different because the protein level of the kinase was also markedly increased (+43.7%). However, the effect of treatment was not restricted to the presynaptic kinase, because CaMKII activity was also increased in the total cellular cytosol in cortical areas. The autonomous (calcium-independent) activity of CaMKII was assayed for the first time after antidepressant treatment, and found to be increased in synaptic vesicles of all three areas. These results confirmed the involvement of CaMKII in antidepressant drug action and suggested that modulation of transmitter release is a primary component in the action of psychotropic drugs.

Effect of reboxetine treatment on brain cAMP- and calcium/calmodulin-dependent protein kinases.

Previous studies showed that the type II Ca(2+)/calmodulin- and cAMP-dependent protein kinases (CaMKII and PKA) are affected by long-term antidepressant treatment in presynaptic and somatodendritic compartments, respectively. This study describes the long-term effects of the selective noradrenaline reuptake inhibitor reboxetine on PKA and CaMKII, in both the microtubule and subsynaptosomal fractions of rat brain. Unlike other antidepressants, chronic reboxetine induced in the cerebrocortical soluble and microtubule fractions a decrease in the [(32)P]cAMP binding to the type II PKA regulatory subunit. No change in the cAMP-dependent endogenous phosphorylation of the protein substrate, microtubule-associated protein 2 was observed. In the hippocampal subsynaptosomal fractions (synaptic vesicles and synaptosomal membranes) reboxetine induced a robust increase in the activity but not in the expression of CaMKII. An increase in the calcium/calmodulin-dependent phosphorylation of presynaptic substrates was also detected. These findings showed that reboxetine modulates post-receptor signal transduction systems in rat brain.

Modifications in brain cAMP- and calcium/calmodulin-dependent protein kinases induced by treatment with S-adenosylmethionine.

Several lines of evidence suggest that the mechanism of action of antidepressant drugs (AD) involves adaptive changes occurring in intraneuronal post-receptor signal transduction cascades. Protein phosphorylation has a key role in signal transduction and was previously found to be a target in the action of AD (5-HT and/or NA reuptake blockers). Several studies showed that cAMP- and type II Ca2+/calmodulin-dependent protein kinases (PKA and CaMKII) are markedly affected by typical AD in two different and complementary cellular districts, respectively microtubules (a somatodendritic compartment) and synaptic vesicles (a presynaptic terminal compartment). In order to investigate whether the effect on protein kinases may be involved in the therapeutic action of drugs it is interesting to compare the effect of atypical AD with that of typical drugs. In this study the effect of the atypical AD S-adenosylmethionine (SAMe) was tested. Repeated (12 days) SAMe treatment induced in cerebrocortical microtubules an increase in the binding of cAMP to the RII PKA regulatory subunit and an increase in the endogenous phosphorylation of microtubule-associated protein 2, an effect resembling that of typical AD. In synaptic terminals the treatment induced an increase in the activity of CaMKII and in the endogenous phosphorylation of vesicular substrates. However, this modification was found in the cerebral cortex rather than in the hippocampus, where typical AD affect CaMKII. In addition the synapsin I level was decreased in the hippocampus and increased in the cerebral cortex, an effect not detected with typical AD.

Behavioural sensitization in 6-hydroxydopamine lesioned rats involves the dopamine signal transduction: changes in DARPP-32 phosphorylation.

"Priming" is a phenomenon of behavioural sensitization observed in unilaterally 6-hydroxydopamine lesioned rats following exposure to a dopamine agonist. After priming, a single dose of the D1 agonist SKF 38393 (3 mg/kg) produces contralateral turning, while the same dose is inactive in drug-naive, lesioned animals. The molecular mechanisms of "priming" were investigated here by studying the phosphorylation of dopamine and adenosine 3'-5' monophosphate regulated phosphoprotein (DARPP-32), a dopamine- and cyclic AMP-regulated phosphoprotein functionally linked to D1 receptors in striatum. Dephospho-form of DARPP-32 were measured by a back-phosphorylation assay. All assays were performed in striata from both lesioned and unlesioned sides. A significant decrease of dephospho-DARPP-32 (27%) was observed in the denervated striatum of primed rats, indicating an increased phosphorylation in vivo of DARPP-32 in response to the D1 agonist. The levels of DARPP-32 protein, as measured by quantitative immunoblotting, remained unchanged in all experimental groups. This study shows that priming is expressed as an increased transduction of the D1 receptor message.

Isolated p65 protein reproduces membrane binding activity of synaptic vesicles.

Purified synaptic vesicles are highly enriched with a protein which binds cell plasma membranes. The binding is selective for acidic phospholipids and sialoglycosphingolipids. In partition chromatography of vesicle proteins, the binding activity was co-eluted with a limited set of proteins. Among them the most abundant species were two vesicle-specific proteins: p65 and synaptophysin. In affinity chromatography of vesicle proteins, only p65 bound to a column of immobilized lysoganglioside. The same protein, purified by preparative electrophoresis, retained the binding activity and fully reproduced the hemagglutinating property and the selectivity for acidic lipids of whole vesicles. The results suggest that the (hemagglutinating) lipid binding properties of the vesicles are mainly if not exclusively due to p65.

Properties of a synaptic vesicle protein binding plasma membranes.

Synaptic vesicles contain a protein (agglutinin) which binds cell plasma membranes. The protein activity is titrated measuring the agglutination of rabbit trypsinized-fixed red blood cells induced by serial dilutions of purified vesicles. The protein is highly concentrated in vesicles: the specific activity is 150-fold higher in vesicles compared to the brain homogenate, and 20-fold higher compared to the crude synaptosomal membrane fraction. The binding is specifically inhibited by the four major brain sialoglycosphingolipids, with a higher affinity for polysialo- rather than for monosialoganglioside (GM1) IC50 are 3.9 x 10(-5) M (GT1b), 8.9 x 10(-5) M (GD1a), 2.8 x 10(-4) M (GD1b) and 4.9 x 10(-4) M (GM1). Cells which are not agglutinated by the vesicles (human red blood cells) can be activated by incubation with gangliosides: insertion of the glycolipids in the plasma membranes makes them sensitive to the vesicle haemagglutinin.

cAMP-dependent phosphorylation system after short and long-term administration of moclobemide.

Accumulating evidence suggested that signal transduction cascade including protein phosphorylation is implicated in the neurochemical action of antidepressant agents. Clinical data indicated that moclobemide, a short acting and reversible inhibitor of monoamino oxidase type. A, is an effective antidepressant medication. However, little is known about the intracellular effects of this compound. Thus, in the present study we assessed the binding of cAMP to cAMP-dependent protein kinase (PKA) in rat cerebral cortex following short and long-term administration of moclobemide. The results showed that 21 days of treatment with moclobemide significantly increased the specific [32P]-cAMP covalent binding into the soluble 52-54 kDa cAMP-receptor. This effect was not seen following 1, 5 and 12 days of treatment. These findings suggest that PKA could be implicated in the biochemical effects of moclobemide.

Effects of fluvoxamine on the protein phosphorylation system associated with rat neuronal microtubules.

We have studied the phosphorylation system associated with the rat cerebrocortical microtubule fraction after short- and long-term administration (15 mg/kg) of fluvoxamine, a selective serotonin reuptake inhibitor with antidepressant activity. Fluvoxamine administered for 5 days significantly enhanced the 32P incorporation stimulated by cAMP into MAP2, while it failed to produce this effect after 12 and 21 days. Moreover, in the same periods of treatment no changes were observed in basal phosphorylation and in the pattern of microtubule proteins. In conclusion, our results suggest that changes in the protein phosphorylation system associated with the microtubule fraction could represent an early neurochemical modification involved in the action of fluvoxamine.

Dopamine mediated responses in 6-hydroxydopamine lesioned rats involve changes of the signal transduction.

A single dose of the D1 agonist SKF 38393 (3 mg/kg) produces contralateral turning in unilaterally 6-hydroxydopamine lesioned rats only after a previous exposure of the animals to a dopamine agonist. This priming phenomenon is here investigated by studying the phosphorylation of DARPP-32, a dopamine- and cyclic AMP-regulated phosphoprotein functionally linked to D1 receptors in striatum. Dephospho-form of DARPP-32 in striatal tissue was measured by a back-phosphorylation assay. While the levels of DARPP-32 protein, as measured by quantitative immunoblotting, remained unchanged, a significant decrease of dephospho-DARPP-32 was observed in the denervated striatum of primed rats, indicating an increased phosphorylation in vivo of DARPP-32 in response to the D1 agonist. This study shows that an alteration of the dopamine-dependent signal transduction is related to the behavioral response to dopamine agents, suggesting a possible mechanism involved in the effects of these drugs in parkinsonian patients.

Modification of presynaptic CaM kinase II affinity for ATP in hippocampus after long term blockade of serotonin reuptake.

Ca2+/calmodulin-dependent protein kinase II (CaMKII) is markedly enriched at synapses, where it is involved in the control of synaptic transmission, transmitter release and synaptic plasticity. CaMKII has also been found to be involved in the long-term action of antidepressants on post-receptor signaling mechanisms, because monoamine reuptake inhibitors induced an increase in autophosphorylation and activity of the kinase in nerve terminals of hippocampus. To study whether changes in the amount of enzyme or kinetic changes, due to posttranslational modifications, are responsible for kinase activation in nerve terminals, alpha-CaMKII level and kinetic constants of the autophosphorylation reaction as a function of ATP concentration were measured in presynaptic cytosol from hippocampus. Treatment with two serotonin reuptake inhibitors did not change the level of presynaptic kinase or the Vmax of autophosphorylation reaction. Instead the Km of the kinase for ATP was decreased 2.8-fold with fluvoxamine and 3.5-fold with paroxetine, implying an increase in the affinity for ATP. This result represents the first finding of changes in kinetic constants of a major brain enzyme after treatment with antidepressant drugs.

Ca2+/phospholipid-binding and syntaxin-binding of native synaptotagmin I.

Synaptotagmin, a synaptic vesicle protein endowed with multiple properties, is the putative calcium sensor in neuroexocytosis. Ca2+/phospholipid binding and syntaxin binding activity of synaptotagmin were previously investigated using recombinant fusion proteins. In phospholipid binding the EC50 for calcium obtained was different when fusion proteins containing one (C2A) or both (C2A+C2B) binding domains were used. It was alternatively proposed that one or both synaptotagmin binding domains are important for calcium-sensing and triggering of transmitter release. In this study the binding activity of native full-length synaptotagmin, immobilized on beads, was investigated. We found the kinetic parameters of Ca2+/phospholipid binding to be compatible with the role of calcium sensor for synaptotagmin (EC50 for calcium = 72 +/- 7 microM), with the two C2 domains supporting separate and complementary calcium sensing properties. The binding of native syntaxin to synaptotagmin was measurable in the absence of calcium, but was markedly stimulated (2.2-fold) in the presence of mM calcium. It may be speculated that the two domains have a synergistic action in fast synchronous transmitter release, whereas C2B domain alone may support slow asynchronous release, working as a high affinity calcium sensor.