Roles of exogenous and endogenous FGF-2 in animal models of depression.

PURPOSE: Several members of the fibroblast growth factor (FGF) family have been shown to be dysregulated in individuals with major depression, and treatment with antidepressants has been reported to increase FGF-2 mRNA levels in the forebrain. METHODS: We have used the tail suspension test (TST), and olfactory bulbectomy (OBX), and FGF-2 deficient mice to investigate putative roles of FGF-2 as an antidepressant and mediator of antidepressive drug actions. RESULTS: FGF-2 applied intraventricularly generated antidepressant-like effects in the TST. FGF-2, similar to the antidepressant amitriptyline, attenuated neuron demise in the piriform cortex and posterolateral cortical nucleus of the amygdala following OBX. Moreover, OBX induced reduction in hippocampal neurogenesis could be ameliorated by subsequent treatment with either amitriptyline or FGF-2. Furthermore, FGF-2 was effective in reversing depressive-like behavior induced by OBX, monitored in the locomotor activity and the passive avoidance test. In bulbectomized FGF-2 deficient mice, treatment with amitriptyline protected neurons, but failed to reverse behavioral alterations. CONCLUSIONS: Together, these results suggest that FGF-2 constitutes both a potential target for antidepressive treatments and an important growth factor in the cytokine network underlying the actions of antidepressive drugs. The results further suggest a requirement of endogenous FGF-2 for mediating behavioral, but not neuroprotective actions of amitriptyline.

Repeated co-treatment with imipramine and amantadine induces hippocampal brain-derived neurotrophic factor gene expression in rats.

The problem of drug-resistant depression indicates a strong need for alternative antidepressant therapies. In our earlier papers we described synergistic, antidepressant-like effects of a combination of imipramine (IMI) and amantadine (AMA) in the forced swimming test in rats, an animal model of depression. Moreover, preliminary clinical data showed that the above-mentioned combination had beneficial effects in treatment-resistant patients. In addition, a number of studies predicted a role of the brain-derived neurotrophic factor (BDNF) in the mechanism of action of antidepressant drugs (ADs). Since the most potent effect of ADs on BDNF gene expression was found after prolonged treatment, in the present study we investigated the influence of repeated treatment with IMI (5 or 10 mg/kg) and AMA (10 mg/kg), given separately or jointly (twice daily for 14 day), on mRNA level (the Northern blot) in the hippocampus and cerebral cortex. The experiment was carried out on male Wistar rats. The tissue for biochemical assays was dissected 24 h after the last dose of IMI and AMA. We also studied the effect of repeated treatment with IMI and AMA on the action of 5-HT(1A)- and 5-HT(2A) receptor agonists (8-OH-DPAT and (+/-)DOI, respectively) in behavioral tests. The obtained results showed that in the hippocampus IMI (10 mg/kg), and in the cerebral cortex IMI (5 and 10 mg/kg) and AMA (10 mg/kg) significantly elevated BDNF mRNA level. Joint administration of IMI (5 or 10 mg/kg) and AMA (10 mg/kg) induced a more potent increase BDNF gene expression in the hippocampus (but not in cerebral cortex) and either inhibited the behavioral syndrome induced by (+/-)DOI or did not change the action of 8-OH-DPAT (compared to treatment with either drug alone). The obtained results suggest that the enhancement of BDNF gene expression may be essential for the therapeutic effect of co-administration of IMI and AMA to drug-resistant depressed patients, and that among other mechanisms, 5-HT(2A) receptors possibly play some role in this effect.

Antidepressant-mediated reversal of abnormal behavior and neurodegeneration in mice following olfactory bulbectomy.

Olfactory bulbectomy is an established animal model of depression that has been mostly investigated in rats. As in human major depression, bulbectomy induces behavioral alterations that can be ameliorated by chronic antidepressant treatment. Furthermore, bulbectomy in rats is known to induce neurodegeneration in some brain areas. The aim of the present study was to evaluate patterns of behavioral alterations and neurodegeneration, and their drug-induced reversibility, in bulbectomized mice. Our results reveal that in mice bulbectomy increases locomotor activity and induces deficits in the passive avoidance test, comparable to the effects seen in rats. Bulbectomy also induced neuronal degeneration, visualized by incorporation of Fluoro-Jade B, in the piriform cortex and the posterolateral cortical amygdaloid nucleus (PLCo). Chronic treatment with the antidepressant amitriptyline protected neurons in both areas and efficiently reversed abnormal locomotor activity induced by bulbectomy. Treatment with citalopram was effective in reversing the behavioral deficits induced by bulbectomy, but did not protect against neurodegeneration. Our data indicate significant overlap between mice and rats in terms of behavioral alterations and neurodegeneration following olfactory bulbectomy. However, there are differences in drug-mediated reversibility of neurodegeneration suggesting that neurodegeneration and protection may be "second-order" features in this animal model of depression. This may also be relevant in the context of studies using genetically altered mice.

Repeated treatment with mirtazepine induces brain-derived neurotrophic factor gene expression in rats.

Recent studies indicate a role of the brain-derived neurotrophic factor (BDNF) in the pathophysiology of depression, as well as in the mechanism of action of antidepressant drugs (ADs). It has been shown that serum BDNF levels are decreased in depressed patients. Moreover, antidepressant treatment increases serum BDNF levels and it is positively correlated with medication response. In addition, repeated administration of ADs induces an increase in rat hippocampal or cortical BDNF gene expression. Since the most potent effect of ADs on BDNF gene expression was found after prolonged treatment, in the present study we investigated the influence of repeated treatment (twice daily for 14 days) of the new AD mirtazapine (5 or 10 mg/kg) on BDNF mRNA level (the Northern blot) in rat hippocampus and cerebral cortex. Imipramine was used as a reference compound. The experiment was carried out on male Wistar rats. The tissue for biochemical assays was collected 24 h after the last doses of mirtazapine and imipramine. We also studied the effect of repeated mirtazapine on the action of the 5-HT2A receptor agonist (+/-)DOI in the behavioral test (head twitches induced by (+/-)DOI) in rats. The obtained results showed that, like imipramine (10 mg/kg), mirtazapine (10 mg/kg) increased BDNF gene expression in both the examined brain regions: in the hippocampus by 24.0 and 26.5%, in the cerebral cortex by 29.9 and 41.5%, respectively, compared with the vehicle-treated control. Neither mirtazapine nor imipramine administered repeatedly at a lower dose (5 mg/kg) significantly changed BDNF mRNA levels in the hippocampus and cerebral cortex. Repeated treatment with mirtazapine (10, but not 5 mg/kg) inhibited the behavioral syndrome induced by (+/-)DOI. This study provides first conclusive evidence that repeated mirtazapine administration increases BDNF mRNA levels; moreover, it indicates that the enhancement of BDNF gene expression may be essential for the clinical effect of mirtazapine.

Current perspectives on the development of non-biogenic amine-based antidepressants.

Compounds that inhibit the re-uptake and/or metabolism of biogenic amines (i.e. serotonin, norepinephrine, and dopamine) have been used to treat depression for more than 40 years. Selective re-uptake inhibitors, currently the most widely prescribed class of biogenic amine-based agents, are certainly safe and relatively easy to use, but do not exhibit either a faster onset of action or greater efficacy than their predecessors. An approach to overcome the limitations that may be inherent to these 'conventional' therapies is to circumvent the monoaminergic synapse. In this review, two potential antidepressant strategies are discussed that may converge with intracellular pathways impacted by chronic treatment with biogenic amine-based agents. Drugs emerging from these strategies may offer significant advantages over currently used antidepressants.

Regulation of BDNF expression in primary neuron culture by LY392098, a novel AMPA receptor potentiator.

The effects of a novel AMPA receptor potentiator (LY392098) on the expression of brain-derived neurotrophic factor (BDNF) were examined in primary neuron culture. The addition of either AMPA or LY392098 to cortical neurons elicited a time and concentration dependent increase in mRNA encoding BDNF. Moreover, co-addition of subeffective concentrations of AMPA (1 microM) and LY392098 (1 microM) resulted in dramatic increases in both BDNF mRNA (>25-fold) and protein ( approximately 7-fold) levels, whilst no changes in either NT-3 or NT-4 mRNA were detected. More modest ( approximately 1.5-2.5-fold) elevations in BDNF mRNA and protein expression were also produced by combinations of AMPA and LY392098 in cerebellar granule cell neurons. In contrast, AMPA and LY392098, either alone or in combination, did not elevate BDNF mRNA levels in primary astroglial cultures. Maximum elevations in BDNF mRNA and protein were produced by 6-12h of AMPA receptor activation 1-3h of AMPA receptor activation were required to elevate BDNF mRNA levels. AMPA receptor-mediated increases in BDNF mRNA and protein were abolished by the AMPA antagonist, NBQX, but were unaffected by the NMDA antagonist, MK-801. In cortical neuron cultures, activation of both L-type Ca(+2) channels and mitogen-activated protein (MAP) kinases contribute to AMPA receptor-mediated increases in BDNF mRNA. The ability of LY392098 to increase the expression of BDNF in primary neuron culture indicates this and related biarylpropylsulfonamides may be useful in the treatment of neuropsychiatric disorders.

[3H]ATPA: a high affinity ligand for GluR5 kainate receptors.

The pharmacological properties of [3H]ATPA ((RS)-2-amino-3(3-hydroxy-5-tert-butylisoxazol-4-yl)propanoic acid) are described. ATPA is a tert-butyl analogue of AMPA (alpha-amino-3-hydroxy-5-methyl-4-isoxazoleproprionic acid) that has been shown to possess high affinity for the GluR5 subunit of kainate receptors. [3H]ATPA exhibits saturable, high affinity binding to membranes expressing human GluR5 (GluR5) kainate receptors (Kd approximately 13 nM). No specific binding was observed in membranes expressing GluR2 and GluR6 receptors. Several compounds known to interact with the GluR5 kainate receptor inhibited [3H]ATPA binding with potencies similar to those obtained for competition of [3H]kainate binding to GluR5. Saturable, high affinity [3H]ATPA binding (Kd approximately 4 nM) was also observed in DRG neuron (DRG) membranes isolated from neonatal rats. The rank order potency of compounds to inhibit [3H]ATPA binding in rat DRG and GluR5 membranes were in agreement. These finding demonstrate that [3H]ATPA can be used as a radioligand to examine the pharmacological properties of GluR5 containing kainate receptors.

Amyloid beta peptide is not a candidate for the neurotrophic activities released from chromaffin cells.

We have previously shown that chromaffin cells, the neuron-like cells of the adrenal medulla, release proteins, which promote in vitro survival of a large number of peripheral and central nervous system neurons (cf. Lachmund, A., Gehrke, D., Krieglstein, K. and Unsicker, K, Trophic factors from chromaffin granules promote survival of peripheral and central nervous system neurons. Neuroscience, 1994, 62, 361-370). In a search for the active molecules we are testing compounds that are known to be synthesized and released by chromaffin cells. Amyloid precursor protein (beta APP) is one of these factors (Bieger, S., Klafki, H.-W. and Unsicker, K., Synthesis and release of the beta-amyloid precursor protein by bovine chromaffin cells. Neurosci. Lett., 1993, 162, 173-175). In the present study we have investigated the possibility that amyloid beta peptide (A beta P) generated from beta APP may have survival supporting effects for neurons from embryonic chick ciliary (CG) and dorsal root ganglia (DRG). Embryonic rat hippocampal neurons, for which promotion of short-term survival by A beta P has been reported (Whitson, J. S., Selkoe, D. J. and Cotman, C. W., Amyloid beta protein enhances the survival of hippocampal neurons in vitro. Science, 1989, 243, 1488-1490), were employed as a reference. A beta P fragment 1-40, administered over a wide range of concentrations (1.5-100 micrograms/ml) did not promote the survival of CG and DRG neurons isolated from embryonic day (E) 8 chick embryos. The peptide also failed to toxically suppress survival of these neuron populations in the presence of survival promoting factors. In confirmation of previous reports, the 1-40 peptide, in comparison to the reverse 40-1 peptide, significantly enhanced survival of hippocampal neurons. These results suggest that A beta P is not a trophic factor for the peripheral neuron populations tested and most likely, not a neurotrophic component among the neurotrophic factors released by chromaffin cells.

Adaptation of cortical NMDA receptors by chronic treatment with specific serotonin reuptake inhibitors.

Glycine displaces [3H]CGP-39653 ([3H]D,L-(E)-2-amino-4-propyl-5-phosphono-3-pentenoic acid) binding to the glutamate recognition site with both high and low affinity. We reported previously that chronic treatment with antidepressants reduced the proportion of high to low affinity sites, or, even eliminated the high affinity sites in case of citalopram. Here, we compared the effects of citalopram with another serotonin specific reuptake inhibitor, fluoxetine on this measure. Chronic administration of citalopram or fluoxetine eliminated high affinity glycine-displaceable [3H]CGP-39653 binding to the mouse cortex in 78 and 56% of animals, respectively, indicating that selective serotonin reuptake inhibitors produce qualitatively similar adaptive changes at NMDA receptors, that differ from other antidepressants in this neurochemical measure.

Neuroactive steroids modulate in vitro the Mg(2+)-dependent Ca(2+)-ATPase activity in cultured rat neurons.

1. The in vitro effect of neuroactive steroids on the Mg(2+)-dependent Ca(2+)-ATPase activity in neuronal membranes isolated from primary cell culture of rat cortex was examined. 2. A 1-hr treatment of neuronal cell culture with 17-beta-estradiol (10 pM) and pregnenolone sulfate (1 microM) resulted in an increase in the enzyme activity of as much as 130% and 160%, respectively. 3. Neuroactive steroids moderately decreased the stimulation of the Mg(2+)-dependent Ca(2+)-ATPase activity by 72 nM calmodulin, by 20-30%. 4. The effects of hormones on the ATPase activity were irreversible after extensive washing of the membranes. 5. These results suggest that 17-beta-estradiol and pregnenolone sulfate at physiological concentrations could participate in the regulation of neuronal calcium homeostasis at a membrane level.

The enhancement and the inhibition of noradrenaline-induced cyclic AMP accumulation in rat brain by stimulation of metabotropic glutamate receptors.

1. The actions of several metabotropic glutamate receptor and antagonists on noradrenaline (NA)-stimulated [3H]-cyclic AMP accumulation were investigated in rat cerebral cortical slices. 2. Quisqualate (QUIS), L-2-amino-3-phosphonopropionic acid (L-AP3) and glutamate (GLU) elicited concentration-dependent inhibition of (NA)-stimulated [3H]-cyclic AMP accumulation, with IC50 values of 105 +/- 29, 275 +/- 36 and 944 +/- 150 microM respectively. In contrast (Rs)-alpha-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid (AMPA) (0.5 mM) and N-methyl-D-aspartic acid (NMDA) (0.5 mM) had no effect. 3. (2S,3S,4S)-alpha-(Carboxycyclopropyl)glycine (L-CCGI), 1-Aminocyclo-pentane-1S,3R-dicarbo-xylate (1S,3R-ACPD), ibotenate (IBO) and (RS)-4-carboxy-3-hydroxy-phenylglycine (CHPG)elicited a concentration-dependent enhancement of NA-stimulated [3H]-cyclic AMP accumulation, with EC50 values of 2.5 +/- 0.11, 42 +/- 1.3, 97.8 +/- 2.1 and 157 +/- 13.4 microM, respectively. 4. (S)-3-carboxy-4-hydroxyphenylglycine (3C4HPG) and (S)-4-carboxy-3-hydroxyphenyl-glycine (4C3HPG) produced a biphasic effect, at concentrations up to 100 and 500 microM, respectively, they significantly enhanced the action of NA (100 microM), at 1mM concentration both compounds as well as alpha-methyl-4-carboxyphenylglycine (MCPG) produced a significant inhibition of NA-stimulated cyclic AMP accumulation. 5. A putative mGluR antagonist-L-AP3, inhibited the 1S,3R-ACPD (100 microM) induced enhancement of the action of NA (100 microM) on [3H]-cyclic AMP accumulation in a biphasic manner with an IC50 of 4.5 microM for the high affinity site, which represented 65% of the total and an IC50 of 283 microM for the low affinity site. 6. beta-adrenoceptor antagonist propranolol inhibited the interaction between 1S,3R-ACPD (100 microM) and NA (100 microM) on [3H]-cyclic AMP accumulation by about 80%, with an IC50 of 0.52 +/- 0.011 microM, to the level observed after 1S,3R-ACPD alone. Prazosin, an alpha 1-adrenoceptor antagonist was more potent (IC50 of 0.091 +/- 0.012 microM) but less efficacious (60% inhibition) as an inhibitor of the interaction either between NA and 1S,3R-ACPD while yohimbine, na alpha 2-adrenoceptor antagonist (up to 1 microM) had no effect. 7. Neither the protein kinase C inhibitor - staurosporine (10 microM) nor thapsigargin (1 microM), which depletes IP3 sensitive calcium stores, inhibited significantly the 1S,3R-ACPD (100 microM)-induced enhancement of the action of NA (100 microM) on [3H]-cyclic AMP accumulation. 8. Adenosine deaminase (0.5 U/ml) abolished both the 1S,3R-ACPD (100 microM)-induced [3H]-cyclic AMP accumulation and the synergistic interaction of this compound with NA (100 microM). 9. These results indicate the existence of different subtypes of metabotropic glutamate receptors in rat brain which either inhibit or enhance the NA-stimulated [3H]-cyclic AMP accumulation. The enhancement in cerebral cortical slices is mediated via receptors which are blocked with high affinity by L-AP3 and occurs via interactions with endogenous adenosine; the inhibition is mediated by receptors sensitive to quisqualate, L-AP3 and glutamate and may represent a predominant interaction between NA and excitatory amino acids (EAA), which in cerebral cortical slices is masked by excitatory effects.

The enhancement and the inhibition of noradrenaline-induced cyclic AMP accumulation in rat brain by stimulation of metabotropic glutamate receptors.

The actions of several metabotropic glutamate receptor agonists and antagonists on noradrenaline (NA)-stimulated [3H]-cyclic AMP accumulation were investigated in rat cerebral cortical slices. Quisqualate (QUIS), L-2-amino-3-phosphonopropionic acid (L-AP3) and glutamate (GLU) elicited concentration-dependent inhibition of (NA)-stimulated [3H]-cyclic AMP accumulation. In contrast (2S,3S,4S)-alpha-(Carboxy-cyclopropyl)glycine (L-CCGI), 1-Aminocyclo-pentane-1S,3R-dicarboxylate (1S,3R-ACPD), ibotenate (IBO) and (RS)-4-carboxy-3-hydroxy-phenylglycine (CHPG) elicited a concentration-dependent enhancement of NA-stimulated [3H]-cyclic AMP accumulation. A putative mGluR antagonist-L-AP3, inhibited the 1S,3R-ACPD-induced enhancement of the action of NA on [3H]-cyclic AMP accumulation in a biphasic manner with an IC50 of 4.5 microM for the high affinity site, which represented 65% of the total and an IC50 of 283 microM for the low affinity site.

The influence of prolonged antidepressant treatment on the changes in cyclic AMP accumulation induced by excitatory amino acids in rat cerebral cortical slices.

We investigated the effect of prolonged electroconvulsive shock (ECS) or imipramine treatment on cyclic AMP accumulation induced by ibotenate and glutamate in rat cerebral cortical slices. Prolonged imipramine or electroconvulsive shock treatment attenuated the ibotenate-induced increase in cyclic AMP accumulation and inhibited the synergistic interaction between ibotenate and noradrenaline; the glutamate-mediated inhibition of forskolin-stimulated cyclic AMP accumulation was not modified. Our results indicate that multiple effects of excitatory amino acids on cyclic AMP accumulation are modified differently by antidepressant treatment.

Antidepressant treatment influences cyclic AMP accumulation induced by excitatory amino acids in rat brain.

The prolonged imipramine or electroconvulsive shock treatment induced a decrease in ibotenate stimulated increase in cyclic AMP accumulation as well as it inhibited the synergistic interaction between ibotenate and noradrenaline; the glutamate-mediated inhibition of forskolin-stimulated cyclic AMP accumulation was not modified. Our results indicate that certain subtypes of metabotropic glutamate receptor are differently modified by prolonged antidepressant treatment.

The effect of interaction between (1S,3R)-1-aminocyclopentane-1,3-dicarboxylic acid (1S,3R)-ACPD and noradrenaline on cyclic AMP accumulation: different actions in brain slices, primary glial and neuronal cell cultures.

The existence of multiple metabotropic receptors for excitatory amino acids is firmly established by recent cloning experiments. Several subtypes of those receptors are coupled to adenylate cyclase. (1S,3R)-1-aminocyclopentane-1,3-dicarboxylic acid (1S,3R)-ACPD) is a selective agonist for metabotropic glutamate receptors, it influences cyclic AMP accumulation in brain and is able to enhance the effect of other substances which stimulate cyclic AMP accumulation. Here we present data showing that (1S,3R)-1-aminocyclopentane-1,3-dicarboxylic acid enhanced the action of noradrenaline on cyclic AMP accumulation in rat brain cortical slices (EC50 40 microM). On the contrary, it inhibited the action of noradrenaline in primary glial cell cultures (EC50 50 microM), being without any effect on noradrenaline-induced cyclic AMP accumulation in primary neuronal cell cultures. The differences may reflect stimulation of different types of metabotropic glutamate receptors in various preparations from rat brain. The nature of interaction between noradrenaline and (1S,3R)-ACPD remains to be studied.

Trans-(+-)-1-amino-1,3-cyclopentanedicarboxylate (trans-ACPD) stimulates cAMP accumulation in rat cerebral cortical slices but not in glial or neuronal cultures.

Recent cloning experiments indicate that multiple metabotropic receptors for excitatory amino acids (EAAs) exist, which are coupled to adenylate cyclase. Trans-(+-)-1-amino-1,3-cyclopentanedicarboxylate (trans-ACPD) is a selective agonist of metabotropic receptors for EAAs. One of the effects of trans-ACPD is stimulation of cAMP accumulation. In the present experiments, cAMP accumulation was measured using a [3H]adenine-prelabelling technique. It has been found that trans-AC-PD was able to induce significant stimulation of cAMP accumulation in rat cerebral cortical slices, with ED50 of 47.8 microM, which value is similar to that described earlier for hippocampal slices. However, trans-ACPD had no effect on cAMP accumulation either in primary neuronal or glial cell cultures. The reason for the lack of effects of trans-ACPD on cAMP accumulation in primary cultures from glial cells and neurons is discussed.

The development of neuropeptide Y-immunoreactive neurons in a model of pure cortical culture.

The development of neuropeptide Y-immunoreactive neurons in the rat brain cerebral cortex was studied in a model of a pure cortical culture. In this model, development of neurons devoid of any afferents from other brain structures could be observed. Since mutual interactions between neuropeptide Y and catecholamines have been postulated, such a pure cortical culture offers a possibility of studying the development of neuropeptide Y neurons devoid of any brainstem monoaminergic afferents. A tissue dissected from 16-day-old rat fetuses and cultivated in a dissociated culture for 14 days was examined immunohistochemically for the presence of neuropeptide Y-immunoreactive neurons. Three main types of neuropeptide Y-immunoreactive neurons were found: unipolar, bipolar and multipolar. Cell processes and terminal varicose fibres were also observed. The results obtained indicate that neuropeptide Y-immunoreactive neurons and fibres may develop in a pure culture of the rat cerebral cortex without the influence of any other structures.

Haloperidol-induced increase in neuropeptide Y immunoreactivity in the locus coeruleus of the rat brain.

The effect of haloperidol, a dopamine (preferably D2) receptor blocking agent on neuropeptide Y immunoreactivity was studied immunohistochemically in neurons of the locus coeruleus and striatum of rat brain. It was found that haloperidol given four times (5 and 2.5 mg/kg, i.p.) induced, after 24 h, a significant increase in the level of neuropeptide Y immunoreactivity in the locus coeruleus but not in the striatum. No changes in neuropeptide Y immunoreactivity in studied structures were observed after alpha-adrenergic receptor blocking agent phenoxybenzamine or serotonin-synthesis inhibitor D,L-p-chlorophenylalanine. The results suggest that the content of neuropeptide Y-immunoreactive material in nerve cell bodies of the locus coeruleus is inhibitorally controlled by monoaminergic (may be dopaminergic D2) receptors.