Effect of early decrease in the lesion size on late brain tissue loss, synaptophysin expression and functionality after a focal brain lesion in rats.

The purpose of the present study is to determine the effects of early decrease in the lesion size on late brain tissue loss, synaptogenesis and functionality after a focal brain lesion in rats. The lesion was induced either to the cortex using the photothrombotic ischemic stroke or to the striatum using the malonate poisoning model. The cortical and striatal lesions amounted to 66-80 mm(3) at day 1 post-lesion and were reduced by 50% after the acute administration of dipyridyl (a liposoluble iron chelator) and aminoguanidine (an inhibitor of the inducible nitric oxide synthase), respectively. Loss of histologically intact tissue and synaptophysin expression as an indicator of synaptogenesis were examined at day 35 post-lesion. Both types of lesion resulted in synaptophysin upregulation in contralateral and ipsilateral cortical areas. On the contrary, brain tissue loss was greater after the striatal (-17%) than the cortical lesion (-5%). Synaptophysin expression and tissue loss were not different between drug- and vehicle-treated rats. Moreover, a set of standard neurological tests revealed a difference in deficit between the both types of lesion, yet only in the acute post-lesion stage. However, it did not distinguish between vehicle- and drug-treated rats whatever the lesion location. Our results indicate that late histological endpoints measurements are not recommended to probe the potential neuroprotective properties of a drug administered within the acute post-lesion stage. They also suggest that inhibition of cytotoxic mechanisms involved in lesion growth is of no clinical interest when it cannot lead to a long-term histological protection and/or increased synaptogenesis.

Potential role of the neuropeptide CGRP in the induction of differentiation of rat hepatic portal vein wall.

The media of the rat hepatic portal vein is composed of an internal circular muscular layer (CL) and an external longitudinal muscular layer (LL). These two perpendicular layers differentiate progressively from mesenchymal cells within the first month after birth. In this paper, we studied the development of calcitonin gene-related peptide (CGRP) innervation during post-natal differentiation of the vessel. We show that CGRP innervation is already present around the vessel at birth in the future adventitia but far from the lumen of the vessel. Progressively, CGRP immunoreactive fibers reached first LL then CL. CL by itself become only innervated at day 14 after birth. This corresponds to the time at which thick filaments (myosin) are visible in electron microscopy and desmin visualisable by immunocytochemistry. Furthermore, we provide evidence by autoradiography, that binding sites for CGRP are transiently expressed on the portal vein media at day 1 and 14 after birth. Vascular smooth muscle cells were transfected with constructs containing promoters for desmin or smooth muscle myosin heavy chain (smMHC). CGRP treatment of the cells significantly increased the expression of smMHC. Overall these results suggest that CGRP can potentially influence the differentiation of smooth muscle cells from the vessel wall.

Biphasic striatal acetylcholine release during and after transient cerebral ischemia in gerbils.

Acetylcholine (ACh) release into the extracellular space was measured by HPLC with electrochemical detection after in vivo intracerebral microdialysis in the striatum of gerbils subjected to 15 min of bilateral carotid artery occlusion followed by 5 h of recirculation. Tissue ACh and choline (Ch) contents were also determined during ischemia and after 5, 30, 60, and 120 min of reflow. Fifteen minutes of ischemia led to a significant transient increase in extracellular ACh concentration (threefold after 7.5 min of ischemia) concomitant with a reduced endogenous ACh level (-62%) and increased tissue Ch content (ninefold). Recirculation significantly reduced the ACh release during the early period of reflow (-50% vs. basal level), followed by a significant increase in ACh release between 1 and 3 h of reflow (45-55% vs. basal level) and subsequent normalization. Simultaneously, a "rebound" of tissue ACh level occurred in the early period of reflow (fourfold vs. ischemic value), followed by gradual normalization after 2 h of reperfusion, whereas a rapid decrease in tissue Ch levels was found after 30 min of reflow. These findings represent the first demonstration of a biphasic release of ACh during ischemia and reperfusion, as assessed by intracerebral microdialysis in gerbils.

Effect of intracellular iron loading on lipid peroxidation of brain slices.

The effect of artificially elevated cell iron content on oxygen-derived free radical production was assessed in brain slices by use of an iron ligand, 8-hydroxyquinoline (HQ). The iron complex Fe(3+)-HQ exhibited a high lipid solubility evidenced by n-octanol/water partition coefficient and was avidely taken up by brain slices. The catalytically active form of Fe3+ within the complex was evidenced by measuring the rate of ascorbate oxidation. Lipid peroxidation was assessed by measuring the thiobarbituric acid-reactive substances (TBARS) in brain homogenates or slices exposed to two doses of Fe(3+)-HQ (10 microM/20 microM, 100 microM/200 microM) or Fe(3+)-citrate (10 microM, 100 microM). Addition of the iron complexes to homogenates or slices resulted in a dose-dependent increase in lipid peroxidation. In homogenates, the effects were grossly similar with both complexes, whereas in slices the effects of Fe-HQ were significantly higher than those of Fe-citrate. Lipid peroxidation persisted in washed slices preexposed to Fe-HQ, but not in slices preexposed to the hydrophilic iron complex Fe-citrate. Fe-HQ-induced lipid peroxidation in slices was enhanced in the presence of H2O2, an effect that was not seen using Fe-citrate. Addition of Fe-HQ to brain homogenates in the presence of salicylic acid resulted in the production of 2,3-dihydroxybenzoic acid and the effect was potentiated in the presence of H2O2. This model of iron cell loading may be useful for evaluating the efficacy of antioxidant drugs.

Brain fixation for acetylcholine measurements.

Brain fixation using a commercially available microwave oven (power output: 750 W) has been investigated as a means for enzyme inactivation preventing post-mortem changes in brain acetylcholine (ACh) and choline (Ch) levels. Rats and mice were decapitated, and the severed heads immediately irradiated for 5.5 and 3 s, respectively, resulting in a complete inactivation of brain acetylcholine esterase (AChE) and choline acetyltransferase (ChAT). The ACh and Ch contents measured in various brain regions of rat and mouse were: (in rat) striatum 60.5 and 32.4 nmol/g, hippocampus 20.4 and 30.9 nmol/g, cortex 24.2 and 19.6 nmol/g; (in mouse) striatum 70 and 47.2 nmol/g, hippocampus 22. 1 and 30.2 nmol/g, cortex 22.9 and 27.9 nmol/g. These values were found in accordance with those reported in the literature by irradiating whole animals in instruments of higher power capabilities. Thus, the procedure described in the present work may be a simple and valuable means of brain fixation for neurochemical studies of brain ACh in small animal species.

Preischemic blood glucose supply to the brain modulates HSP(72) synthesis and neuronal damage in gerbils.

Preischemic hyperglycemia is known to aggravate brain damage caused by transient forebrain ischemia. Because heat shock proteins (HSPs) 72 have been proposed to play a protective role against ischemic neuronal injury, we studied the HSP(72) mRNA expression and protein synthesis in gerbils subjected to 10 min bilateral carotid occlusion under normoglycemic, hyperglycemic and fasting conditions. HSP(72) mRNA expression and HSP(72) synthesis were studied using in situ hybridization and immunostaining, respectively. After 8 h of blood recirculation, HSP(72) mRNAs were expressed in all the hippocampal subfields of the three different groups, with higher expression in the hyperglycemic gerbils. After 48 h of reperfusion, HSP(72) mRNAs had almost completely disappeared in the hyper- and normoglycemic groups, and were more strongly expressed in the CA(1) neurons of the fasted group. At this time, fasted gerbils exhibited intense HSP(72) immunoreactivity in the CA(1), whereas an absence of immunoreactivity was observed in that area in the other groups. Finally, ischemia was also associated with marked astrocytic activation, as evidenced by GFAP immunostaining. Overall results indicate that preischemic differences in blood glucose supply to the brain are related to HSP(72) mRNA expression (in terms of duration) and to HSP(72) protein induction (in terms of intensity) in the vulnerable CA(1) neurons of the hippocampus. Ability of CA(1) neurons to synthesize HSP(72) proteins was associated with higher neuronal survival in the fasted group after 48 h of reflow, suggesting a protective role of HSP(72), even though evaluation of neuronal damage at 7 days indicated that neuronal death was mainly delayed in the time.

Stress response to hypoxia in gerbil brain: HO-1 and Mn SOD expression and glial activation.

Hypoxic preconditioning has been shown to induce neuroprotection against a subsequent damaging insult. In order to study the underlying molecular and cellular mechanisms of hypoxic preconditioning, we investigated, in gerbil hippocampus, the effects in vivo of transient exposure to hypoxia (4% O(2) for 6 min followed by either 48 h or 7 days of reoxygenation) (i) on the induction of 72 kDa heat shock protein (HSP72), heme oxygenase-1 (HO-1) and manganese superoxide dismutase (Mn SOD) as assessed by Western immunoblotting and (ii) on the astroglial and microglial activation as detected by both immunohistochemistry and Western immunoblotting for GFAP, and histochemistry for isolectin B4, respectively. Our data show that, although hypoxia and subsequent reoxygenation led to neither neuronal damage nor HSP72 induction in gerbil hippocampus, it induced a progressive and sustained expression of HO-1 and Mn SOD. As expected from the absence of neuronal death, hypoxia was not associated with microglial activation but led to a significant astrocytic activation. These findings demonstrate that transient hypoxia enhances the antioxidative enzymatic defenses of the brain, which are susceptible to increased tolerance against a subsequent damaging insult.

Differential MnSOD and HO-1 expression in cerebral endothelial cells in response to sublethal oxidative stress.

The two inducible enzymes, manganese superoxide dismutase (MnSOD) and heme-oxygenase-1 (HO-1) may participate in the cellular defense of brain endothelium against oxidative stress. The time-dependent expression of MnSOD and HO-1 mRNAs and proteins was investigated in vitro in rat cerebral endothelial cells (CEC) subjected to sublethal mild or moderate hydroxyl radical-induced oxidative stress. Mild oxidative stress induced increases in both MnSOD and HO-1 mRNA and protein expression. Moderate oxidative stress resulted in a significant reduction in HO-1 mRNA and protein expression, whereas MnSOD expression pattern was similar to that observed after mild oxidative stress. A profound protein loss of both MnSOD and HO-1 was detected 24 h after exposure of CEC to a moderate oxidative stress. The data indicate that cerebral endothelial cells respond by increasing the expression of antioxidant defense enzymes in a manner dependent on the oxidative stress intensity.

Neurochemical stimulation of the rat substantia innominata increases cerebral blood flow (but not glucose use) through the parallel activation of cholinergic and non-cholinergic pathways.

Neurochemical activation of the substantia innominata (SI) in the rat, through the direct injection of the cholinergic agonist carbachol, has been reported to induce large increases in cerebral blood flow (CBF) throughout cortical and subcortical projection regions. The present study aimed to determine whether the vasomotor responses to cholinergic stimulation of the SI were, or were not, the consequence of an increase in metabolic activity. To this end, coupled measurements of CBF and cerebral glucose use (CGU) were undertaken during carbachol-elicited stimulation of the SI. Infusion of carbachol into the basal forebrain induced significant CBF increases in several ipsilateral cortical and subcortical areas including the amygdala. In contrast, CGU increased only in the ipsilateral amygdala and SI. Thus, we tested the hypothesis of a direct neurogenic, rather than metabolic, contribution of the basalocortical system. In this respect, carbachol-elicited stimulation resulted in significant increases in extracellular acetylcholine concentrations in the ipsilateral parietal cortex; systemic pretreatment with the muscarinic receptor antagonist scopolamine completely abolished the increase in cortical CBF elicited by cholinergic stimulation of the SI in the ipsilateral frontoparietal motor cortex while it failed to affect the increase observed in the ipsilateral temporal cortex. Several conclusions can be drawn from the present study. The stimulation of the SI by carbachol induces an increase in CBF that can be dissociated from changes in underlying glucose metabolism. Secondly, these induced changes in cortical CBF are paralleled by an increase in acetylcholine release. Lastly, the failure of scopolamine to block the flow response in all cortical regions would suggest that SI stimulation will evoke the release of vasodilatatory neurotransmitter(s) as well as acetylcholine itself.

Effects of iNOS-related NO on hearts exposed to liposoluble iron.

Inducible nitric oxide synthase (iNOS) protects heart against ischemia/reperfusion injury. However, it is unknown whether the beneficial effects of iNOS are mediated by the interaction of NO with radical oxygen species (ROS). To address this issue, we examined the effects of liposoluble iron-induced ROS generation in isolated perfused hearts from rats treated with lipopolysaccharide (LPS). LPS administration (10 mg/kg, i.p., 6 h before heart removal) induced iNOS expression and increased NO production as indicated by a 3-fold elevation of nitrite level in coronary effluents relative to control hearts. An enhanced expression of hemeoxygenase 1 protein was also observed in septic hearts compared to control. Iron-induced perfusion and contractile deficits were ameliorated by LPS with more important coronary than myocardial benefits. In iron-loaded hearts, oxidative stress as measured by the 2,3 dihydroxybenzoic acid/salicylic acid concentration ratio in cardiac tissue was 23% lower in septic than in control heart although the difference did not reach significance. In addition, the presence of the NO synthase inhibitor N-nitro-L-arginine in the perfusion medium totally blocked NO production but did not reverse the protective effects of LPS. The results indicate that LPS protects from iron-induced cardiac dysfunction by mechanisms independent on ex vivo NO production and suggest that NO acts as a trigger rather than a direct mediator of the cardioprotective effects of LPS in heart exposed to iron.

Effects of a direct injection of liposoluble iron into rat striatum. Importance of the rate of iron delivery to cells.

For a better understanding of the role of iron imbalance in neuropathology, a liposoluble iron complex (ferric hydroxyquinoline, FHQ) was injected into striatum of rats. The effects of two modalities of iron injections on brain damage, hydroxyl radical (*OH) production (assessed by the salicylate method coupled to microdialysis) and tissue reactive iron level (evaluated ex vivo by the propensity of the injected structure for lipid peroxidation) were examined. Rapid injection of FHQ (10 nmoles of 5 mM FHQ pH 3 solution over 1-min period) but not that of corresponding vehicle led to extensive damage associated with increased tissue free iron level in the injected region. Conversely, neither lesion nor free iron accumulation was observed after slow FHQ injection (10 nmoles of a 100 microM FHQ pH 7 solution over 1-h period) as compared to corresponding vehicle injection. Production of *OH was induced by slow FHQ injection but not by rapid FHQ injection, probably as a result of in vivo abolition of iron-induced *OH formation by acid pH. Indeed, rapid injection of FAC pH 7 (ferric ammonium citrate, 5 mM in saline) was associated with *OH formation whereas rapid injection of FAC pH 3 did not. Our results identify the rate of iron delivery to cells as an important determinant of iron toxicity and do not support a major role for extracellular *OH in damage associated with intracerebral iron injection.

Chemical preconditioning with 3-nitropropionic acid: lack of induction of neuronal tolerance in gerbil hippocampus subjected to transient forebrain ischemia.

Chemical preconditioning using the mitochondrial toxin, 3-nitropropionic acid (3-NP) has been reported to induce neuroprotection against subsequent global ischemia. To investigate the underlying mechanisms, Mongolian gerbils were pretreated with either vehicle or 3-NP at the dose of 3 or 10 mg/kg, intraperitoneal, 3 days prior to a 5-min bilateral carotid artery occlusion followed by either 48 h or 7 days of blood recirculation. Neuronal damage was assessed by a cresyl violet/fuchsin acid staining. Induction of heat shock protein 72 (HSP72) and manganese superoxide dismutase (MnSOD) expression was evaluated by Western blotting. Astroglial and microglial activation was detected by immunohistochemistry (glial fibrillary acid protein) and by histochemistry (isolectin B4 staining), respectively. Present data show that the hippocampal neuronal damage induced by ischemia were of similar extent between the vehicle- and 3-NP-treated gerbils, whatever the dose tested, indicating that 3-NP did not induce tolerance to transient forebrain ischemia under our experimental conditions. The lack of difference in the post-ischemic level of HSP72 and MnSOD protein expression and in the intensity of astroglial and microglial activation represents further indirect indications of the absence of 3-NP preconditioning effect. In conclusion, although chemical preconditioning with 3-NP is a well-established phenomenon at least in vitro and in models of focal ischemia, the relevance of 3-NP as a preconditioning molecule towards global brain ischemia remains an open question.

Dopamine and norepinephrine turnover in various regions of the rat brain after chronic manganese chloride administration.

Rats were treated with MnCl2 X 4H2O (1 mg/100 g/day, i.p.) for a period of 4 months. The turnover of dopamine (DA) and norepinephrine (NE) was measured in several brain regions (brain stem, hypothalamus, corpus striatum and "rest of the brain") by the decay in endogenous DA and NE after inhibition of tyrosine hydroxylase by alpha-methylparatyrosine. Monoamine oxidase (MAO) activity and manganese levels were also estimated. Manganese treatment produced a decrease in DA level and turnover in the corpus striatum but not in the rest of the brain. An increase in contents of NE was observed both in the brain stem and hypothalamus. NE turnover was found to be increased in the brain stem, decreased in the hypothalamus and unaltered in the rest of the brain. MAO activity was not significantly altered in all the brain regions studied. These results which show that chronic administration of manganese may cause regionally different changes in catecholamine turnover were discussed in relation to the accumulation of manganese in the brain regions and to other metabolic changes associated with manganese toxicity.

Importance of iron location in iron-induced hydroxyl radical production by brain slices.

Iron imbalance has been implicated in oxidative injury associated with many brain diseases. The present study investigated the importance of iron location in hydroxyl radical (.OH) generation and the link between .OH production evaluated by the salicylate method and lipid peroxidation monitored by thiobarbituric acid-reactive substances assay. Brain slices were exposed to increasing doses (2, 10 and 50 microM) of Fe(III) that was complexed either to a lipophilic (8-hydroxyquinoline, HQ) or to a hydrophilic (ammoniacal citrate) ligand. Both iron complexes resulted in an increased salicylate hydroxylation and lipid peroxidation, these effects being significantly more potent in presence of Fe(III)-HQ. Salicylate hydroxylation was linearly correlated to the intensity of TBARS formation but the slope of the curve was found to be higher with Fe(III)-HQ. The present results demonstrate that 1) cell-associated reactive iron is more prone than extracellular iron to induce .OH generation, 2) the level of lipid peroxidation depending on the site of .OH production, cannot be used as an index of the level of total .OH formation, 3) the salicylate method is a convenient method to detect .OH formed intracellularly, at least in vitro.

Effect of oxotremorine, physostigmine, and scopolamine on brain acetylcholine synthesis: a study using HPLC.

The synthesis rate of brain acetylcholine (ACh) was estimated in mice following i.v. administration of [3H]choline (Ch). The measurements were performed 1 min after the tracer injection, using the [3H]ACh/[3H]Ch specific radioactivity ratio as an index of ACh synthesis rate. Endogenous and labeled Ch and ACh were quantified using HPLC methodology. Oxotremorine and physostigmine (0.5 mg/kg, i.p.) increased the steady state concentration of brain ACh by + 130% and 84%, respectively and of Ch by + 60% (oxotremorine); they decreased ACh synthesis by 62 and 55%, respectively. By contrast, scopolamine (0.7 mg/kg, i.p.) decreased the cerebral content of Ch by - 26% and of ACh by - 23% without enhancing the synthesis of ACh. The results show the utility of HPLC methodology in the investigation of ACh turnover.

Time course of decline of radiolabeled acetylcholine formed following intracerebroventricular administration of tritiated choline: effects of oxotremorine and scopolamine.

Rats were injected intracerebroventricularly with 5 microCi of [methyl-3H]choline. The time course of decline of the radiolabeled acetylcholine (ACh) formed was estimated in the ipsilateral cerebral cortex and striatum. The [3H]ACh levels declined biphasically from the cerebral tissue. The initial decline proceeded rapidly, after which labeled ACh declined more slowly. Scopolamine (1 mg/kg, i.v.) caused a significant increase in the rate of [3H]ACh disappearance, which can be interpreted as an enhancement of ACh release. By contrast, oxotremorine (0.8 mg/kg, i.v.) markedly reduced the [3H]ACh disappearance. The results show that drug-induced changes in cholinergic neuronal activities can be estimated from the disappearance of radioactive ACh after labeling the endogenous transmitter through intracerebroventricular administration of labeled choline.