Search PubMed⌕ Search

Biomedical subjects

N H Diemer

Publications and source records attributed to N H Diemer.

At least 19 recordsLinked to original sources

In vivo cellular uptake of glutamate is impaired in the rat hippocampus during and after transient cerebral ischemia: a microdialysis extraction study.

Using microdialysis in CA1 of the rat hippocampus, we studied the effect of transient cerebral ischemia on in vivo uptake and on extracellular levels of glutamate during, and at different time points after ischemia. (3)H-D-aspartate (test substance), and (14)C-mannitol (reference substance), were added to the dialysis perfusate, and the cellular extraction of (3)H-D-aspartate was calculated from scintillation analysis of fractionated dialysate samples. The extraction of (3)H-D-aspartate was studied both in a tracer like condition with a perfusate concentration of 0.2 microM, and in a condition of high saturation level, with 1.0 mM D-aspartate added to the perfusate. In between radioisotope perfusions, dialysate was sampled for analysis of amino acid content by HPLC. During ischemia, extraction of (3)H-D-aspartate (0.2 microM) declined to a maximum reduction of 68%. In the hours after ischemia, extraction of (3)H-D-aspartate (0.2 microM) was decreased by 32%. In the days after ischemia, there was a progressive decline in extraction of (3)H-D-aspartate (1.0 mM), reaching a reduction of 89% on Day 4 after ischemia. Extracellular glutamate remained at control levels at all time points after ischemia. The present study is the first to investigate uptake of glutamate in the intact rat brain in relation to cerebral ischemia. Evidence is provided that uptake of Glu is restrained during ischemia, and that in the hours after ischemia, the extracellular turnover of glutamate is decreased. In the course of the days after ischemia, degeneration of CA1 pyramidal cells occurs concomitantly with a progressive decline in glutamate transport ability, possibly of pathogenetic importance to CA1 pyramidal cell loss.

Animals↗

Changes in gene expression following induction of ischemic tolerance in rat brain: detection and verification.

Tolerance against ischemic insults can be elicited in the CA1 region of rat hippocampus by inducing a short ischemic period 2-3 days prior to the ischemic insult. To detect genes whose expression changes following induction of ischemic tolerance (IT), we applied a differential display technique called restriction fragment differential display-PCR (RFDD-PCR). RFDD-PCR displays the coding region of mRNA and allows detection of differentially expressed mRNA. Double-stranded cDNA generated using a T25V primer is digested by the endonuclease TaqI, and adapters are ligated onto the cDNA fragments. When amplifying the adapter-containing cDNA fragments under high-stringency conditions, reproducible PCR profiles are obtained. By comparing these profiles from naïve and ischemia-tolerant rat brains statistically, significant expression changes of 20 fragments were identified. To verify the observed changes, quantitative PCR and in situ hybridization were performed for three fragments representing proteins with quite different functions (GluR2-flop, SC1, and p68 RNA helicase). Quantitative PCR displayed the same degree of regulation as RFDD-PCR, but in situ hybridization did not display any regulation. As the applied PCR-based techniques detect only polyadenylated mRNA, whereas in situ hybridization detects both nonadenylated and adenylated mRNA, changes in the polyadenylation state of the mRNA, rather than inconsistent changes in the total amount of mRNA, probably explain this discrepancy. Thus, our results show that the expression of genes hitherto not related to IT changes with the induction of IT and that the degree of regulation displayed by RFDD-PCR can be verified by quantitative PCR.

Activated-Leukocyte Cell Adhesion Molecule↗

Is the AMPA receptor subunit GluR2 mRNA an early indicator of cell fate after ischemia? A quantitative single cell RT-PCR study.

After a moderate global cerebral ischemia, two hypothetical populations of pyramidal neurons are present among the hippocampal CA1 pyramidal neurons: one that will die and another one that will survive. Prior analysis of dissected hippocampal CA1 regions has shown a reduction of the GluR1-3 mRNA following ischemia. In order to identify these changes in single neurons, quantitative single cell RT-PCR was used to analyze the expression of GluR1-4 mRNA in rats 24 h after ischemia and also in rats after tolerance inducing ischemia. Control CA1 cells had a median copy-number of 290, 247, 207 and 16 GluR1-4, respectively. The tolerant cells showed small significant up-regulations of GluR1, 3 and 4 mRNA, while the GluR2 mRNA showed a more than 4-fold up-regulation compared to control cells. All the cells from ischemic animals displayed down-regulations of GluR1-3 mRNA. The GluR4 mRNA was not detectable in the ischemic animals. Our results thus show that the CA1 neurons react uniformly 24 h after a moderate ischemia independent of the fate of the neuron: thus two neuron populations with different GluR2 profiles cannot be identified in post-ischemic animals at 24 h. It seems however that an increased level of GluR2 can be used as an indicator of tolerance to ischemia.

Animals↗

Focal cerebral ischemia induces increased myelin basic protein and growth-associated protein-43 gene transcription in peri-infarct areas in the rat brain.

Although oligodendrocytes are vulnerable to focal cerebral ischemia, remyelination of denuded or regenerating axons in the peri-infarct area has been observed in the central nervous system. We studied the expression of myelin basic protein (MBP), a major component of central nervous system myelin, in peri-infarct areas in adult rat brain after transient middle cerebral artery occlusion (MCAO) and correlated it to the expression of the growth-associated protein-43 (GAP-43), a marker for axonal regeneration and sprouting, using non-radioactive in situ hybridization techniques. Within the infarct, MBP messenger RNA (mRNA) had disappeared by 24 h, whereas myelin protein, identified by MBP and myelin oligodendrocyte glycoprotein (MOG) immunohistochemistry, appeared structurally intact until day 3. Peri-infarct oligodendrocytes increased their expression of MBP mRNA from 24 h to maximal levels at day 7, corresponding to the appearance of process-bearing MBP and occasional MOG-immunoreactive oligodendrocytes in parallel sections. Quantitative analysis revealed significant increases in the density of oligodendrocytes (up to 7.6-fold) and in the level of MBP mRNA expressed by individual cells. Parallel sections showed that increased expression of GAP-43 mRNA in neurons was concomitant to MBP mRNA upregulation in oligodendrocytes. While the mechanisms regulating oligodendrocyte survival and myelination signals are not clear at this point, axonal sprouting could putatively serve as a stimulus for the upregulation of oligodendrocyte cell numbers, differentiation state, and/or active myelination in the peri-infarct areas.

Animals↗

Intracerebroventricular metformin attenuates salt-induced hypertension in spontaneously hypertensive rats.

The aim of this study was to examine the effects of long-term continuous intracerebroventricular (icv) infusion of metformin on blood pressure (BP) in spontaneously hypertensive rats (SHR). To accelerate the development of hypertension, SHR were fed a 8% NaCl diet during the 3-week study period. Metformin was given in the following doses: 0 (isotonic saline; n = 7), 25 (n = 8), 50 (n = 6), 100 (n = 6), and 200 microg/day icv (n = 5). Mean arterial pressure (MAP) and heart rate (HR) were measured by radiotelemetry, and as a measure of the contribution of sympathetic nerve activity to BP, the decrease in MAP in response to ganglionic blockade with hexamethonium, 30 mg/kg iv, was determined once weekly. In vehicle treated rats, MAP increased by 27+/-4 mm Hg, whereas in rats treated with a low dose of metformin (25 microg/day), MAP increased only by 7+/-3 mm Hg (P < .01). The hypotensive response to hexamethonium was attenuated by all doses of metformin suggesting that chronic icv metformin decreased central sympathetic outflow. The highest doses of metformin (100 and 200 microg/day) also prevented development of hypertension, but these doses were highly neurotoxic as demonstrated by histologic evaluation post mortem. Fast-Fourier transformation of MAP revealed increased variability within the 0.15 to 0.6 Hz frequency range in rats treated with neurotoxic doses of metformin, suggesting impaired sympathetic control of BP in these animals. In conclusion, long-term icv infusion with apparently nontoxic doses of metformin attenuates hypertension and decreases the hypotensive responses to ganglionic blockade in SHR, suggesting a centrally elicited sympathoinhibitory action.

Animals↗

GABAA receptor antisense epilepsy: histological changes following infusion of antisense oligodeoxynucleotide to GABAA receptor gamma 2 subunit into rat hippocampus.

A deficiency of neuronal inhibition mediated by gamma-aminobutyric acid (GABA) via the GABAA receptor complex has been hypothesised to be a central factor in epileptogenesis. Intrahippocampal infusion of antisense oligodeoxynucleotide (ODN) to the GABAA receptor gamma 2 subunit in rats leads to electrographic limbic status epilepticus. In this model, epileptic phenomena are accompanied by loss of hippocampal neurones. The purpose of the present study was to investigate the time-course of morphological changes following hippocampal antisense 'knockdown' of the GABAA receptor gamma 2 subunit. gamma 2 subunit antisense ODN was infused continuously into the right hippocampus for periods between 1 and 5 days. After about 4 days of infusion, pronounced neurodegenerative changes were consistently observed within the ipsilateral hippocampus. In general, marked loss of CA3 pyramidal cells was found. The notion that the histological changes induced by the antisense ODN were specific to the applied ODN sequence was supported by the finding that a mismatch control ODN did not induce neurodegenerative changes, except for a small lesion in the immediate vicinity of the infusion site. Extensive ipsilateral hippocampal infiltration with monocytes and macrophages was a feature of antisense ODN infusion, but was considerably less pronounced after the infusion of control ODN. Immunocytochemistry using an antibody labeling glial fibrillary acidic protein (GFAP), revealed marked astroglial hypertrophy/proliferation after 4 days of antisense treatment, i.e., coincident with the development of neurodegeneration, in the ipsilateral hippocampus. At this time GFAP-immunoreactivity was also evident in the contralateral hippocampus, indicating contralateral spread of seizure activity.

Animals↗

Ischemic tolerance affects the adenylation state of GluR2 mRNA.

To analyse GluR2 regulations in the rat hippocampal CA1 region following global and tolerance-inducing ischemia in situ hybridization (ISH) and quantitative PCR (Q-PCR) was applied. In addition, cDNA was synthesised from two different primer combinations in order to elucidate possible differences in the adenylation state of GluR2 mRNA. Following global ischemia, ISH and Q-PCR both showed reductions to half of control levels of GluR2 mRNA in consent with previously published results. Following tolerance induction, ISH showed no changes, whereas PCR analysis showed up-regulation to 228% of control value for the general cDNA synthesis, and no change for the specific cDNA synthesis. This indicates that tolerance-inducing ischemia does not increase the amount of GluR2 mRNA; instead polyadenylation of the existing GluR2 mRNA pool takes place.

Animals↗

Neuroprotective effect of 8-OH-DPAT in global cerebral ischemia assessed by stereological cell counting.

The neuroprotective effect of the 5-HT(1A) receptor agonist 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT) was tested in a 2-vessel occlusion model in rats. The post-ischemic core temperature was carefully monitored for 24 h. After 7 days of survival, the viable CA1 neurons were counted in an 8-OH-DPAT (125 microg/kg/h) and vehicle-treated group using the optical fractionator method. The vehicle-treated ischemic rats had a median number of dorsal CA1 neurons of 49,900 whereas the 8-OH-DPAT-treated ischemic rats had a significant lower median number of dorsal CA1 neurons 105,200 (P=0. 018). 8-OH-DPAT significantly lowered the core temperature compared to the vehicle-treated group during the 24-h post-ischemic period. Hypothermia is proposed as a possible explanation of the neuroprotective effect of 8-OH-DPAT.

8-Hydroxy-2-(di-n-propylamino)tetralin↗

Quantitative PCR analysis of AMPA receptor composition in two paradigms of global ischemia.

Quantitative PCR was used to analyse the expression of GluR1, GluR2, GluR2 flip, GluR2 flop and GluR3 mRNA in animals after ischemia and tolerance-inducing ischemia. The ischemic animals showed a decrease in the GluRs to approximately 30%, except for GluR2-flip, which decreased to 75%. The tolerance animals displayed regulation of GluR1 to 75%, GluR2 and GluR2-flop to 283% and 265% respectively. We did not find a correlation between GluR2 regulation and cell loss in the ischemic group. The selective upregulation of GluR2/GluR2 flop in tolerant animals indicates a possible mechanism for enhanced AMPA receptor desensitisation leading to tolerance to ischemia.

Animals↗

GluR2 protein synthesis and metabolism in rat hippocampus following transient ischemia and ischemic tolerance induction.

In this study we have determined the metabolic half-life, protein synthesis and expression of alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) receptor subunit GluR2 in the hippocampus of the living rat. Synthesized proteins were pulse labeled in vivo using intracarotid infusion or intrahippocampal injection of L-[(35)S] labeled amino acids, and the GluR2 protein immunoprecipitated in order to measure the tracer incorporation at different survival time-points. A limited time course study suggested a metabolic half-life of 144 and 108 h in the CA1 region in control animals following carotid artery infusion and intrahippocampal injection, respectively. Twenty-four hours following a moderate ischemic insult, GluR2 protein synthesis was decreased significantly in both the CA1 and DG/CA3 region, whereas the total protein synthesis was decreased significantly only in the CA1 region. Twenty-four hours following ischemic tolerance induction, a significant increase in GluR2 expression was found in the CA1 region using quantitative Western blotting, while no change was found in the dentate gyrus (DG)/CA3 or in expression of GluR1 protein. Data from labeling experiments did not reveal the reason for the increased amount of GluR2 in the CA1 region of the tolerant animals. This study shows that following global ischemia the GluR2 synthesis is decreased both in the CA1 and DG/CA3, which, together with the found GluR2 metabolic half-life, contradict a selective loss of GluR2 protein as a triggering mechanism for the delayed CA1 pyramidal cell death. Twenty-four hours following tolerance induction, we found an increased GluR2 expression in the CA1 region, suggesting that GluR2 plays a role in the acquisition of ischemic tolerance. Our study suggests the ability of neurons to regulate the AMPA receptor subunit expression through changes in protein synthesis and stability.

Animals↗

Reduced [3H]IP3 binding but unchanged IP3 receptor levels in the rat hippocampus CA1 region following transient global ischemia and tolerance induction.

Changes in inositol (1,4,5)-trisphosphate (IP3) binding properties and the protein level of the IP3 receptor have been reported in different pathological conditions in the brain, e.g. cerebral ischemia, Alzheimer's disease, and Huntingtons disease. We used the 4-vessel occlusion model in rat brain to investigate the effect of transient ischemia insults on the IP3 receptor mRNA level, the IP3 receptor protein level and [3H]IP3 binding. Recirculation periods were limited (1-72 h) to avoid the development of delayed neuronal death. We found that the IP3 receptor mRNA levels were decreased after damage-inducing ischemia (9 min) in the hippocampus CA1 and CA3 regions. The mRNA levels were unaltered after tolerance-inducing ischemia (3 min). However, [3H]IP3 binding was significantly reduced after both damage- and tolerance-inducing ischemia in the hippocampus CA1 region. Furthermore, all investigated brain areas showed a decreased [3H]IP3 binding when tolerance-inducing ischemia was followed by a second ischemic insult (3 + 8.5 min ischemia). The IP3 receptor protein levels remained constant in all investigated brain areas. These results indicate that a reduced [3H]IP3 binding capability in the particularly vulnerable areas occurs as an early consequence of cerebral ischemia, before IP3 receptor protein levels are reduced in these areas. Structural or conformational changes altering IP3 binding may be of necessity on the pathway leading to down-regulation of IP3 receptor protein levels, as observed by others.

Adaptation, Physiological↗

Is the 5-HT2C receptor a therapeutic target in cerebral ischaemia?

This study examines the effect of a 5-HT2C agonist (RO 60-0175, (s)-2-(chloro-5-fluoro-indol-1-yl)-1-methylethylamine) and a 5-HT2C antagonist (RO 43-0440, benzofuran-2-carboxamidine) for neuroprotective activity in a rat model of global cerebral ischaemia. A mini-osmotic pump implanted subcutaneously delivered 0.25 mg/kg/hr. Seven days after ischaemia the rats were sacrificed and the damage in the CA1 pyramidal cell layer in hippocampus was estimated and the treated groups were compared with vehicle groups. Pretreatment with the 5-HT2C agonist RO 60-0175 significantly increased the damage, whereas the 5-HT2C antagonist RO 43-0440 had no effect on the cell damage. Measurement of the core temperature in a RO 60-0175-treated group of rats revealed no effect compared to a vehicle-treated group. Thus the aggravation of damage in the RO 60-0175-treated group cannot be explained by temperature effect. Our data do not indicate the 5-HT2C receptor as a therapeutic target in cerebral ischaemia.

Amidines↗

Secretory phospholipase A2 potentiates glutamate-induced rat striatal neuronal cell death in vivo.

The secretory phospholipases A2 (sPLA2) OS2 (10, 20 and 50 pmol) or OS1, (50 pmol) purified from taipan snake Oxyuranus scutellatus scutellatus venom, and the excitatory amino acid glutamate (Glu) (2.5 and 5.0 micromol) were injected into the right striatum of male Wistar rats. Injection of 10 and 20 pmol OS2 caused no neurological abnormalities or tissue damage. OS2 (50 pmol) caused apathy and circling towards the injection side. Histology revealed an infarct at the injection site. Injection of 50 pmol OS1 showed very little or no signs of neurotoxicity. Injection of 2.5 micromol Glu caused no tissue damage or neurological abnormality. After injection of 5.0 micromol Glu, the animals initially circled towards the side of injection, and gradually developed generalized clonic convulsions. These animals showed a well demarcated striatal infarct. When non-toxic concentrations of 20 pmol OS2 and 2.5 micromol Glu were co-injected, a synergistic neurotoxicity was observed. Extensive histological damage occurred in the entire right hemisphere, and in several rats comprising part of the contralateral hemisphere. These animals were apathetic in the immediate hours following injection, with circling towards the side of injection in the following days. Thus, OS2 greatly potentiates glutamate excitoxicity in vivo.

Animals↗

Characterization of a new AMPA receptor radioligand, [3H]2-amino-3-(3-carboxy-5-methyl-4-isoxazolyl)propionic acid.

(RS)-2-Amino-3-(3-carboxy-5-methyl-4-isoxazolyl)propionic acid (ACPA), which is a potent and selective agonist at (RS)-2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic acid (AMPA) receptors, has previously been shown to desensitize AMPA receptors to a much lower degree than AMPA itself. We now report the synthesis of [3H]ACPA (32.5 Ci/mmol), the neurochemical and pharmacological characterization of [3H]ACPA binding, and a comparison of the distribution of [3H]ACPA, [3H]AMPA, and [3H](S)-5-fluorowillardiine binding sites in rat brain. Under equilibrium conditions, [3H]ACPA was shown to bind to a single population of receptor sites on rat brain membranes. [3H]ACPA was shown to bind with single and similar affinities (15-45 nM) to cloned AMPA receptor subunits (GluR1-4), expressed in insect cells, whereas a K(D) value of 330 nM was determined for the binding of [3H]ACPA to cloned kainic acid preferring GluR5 subunits. Whereas Bmax and K(D) values for [3H]ACPA binding, determined using filtration techniques, were different from such obtained in centrifugation assays, Bmax and K(D) values as well as association and dissociation constants were not significantly affected by the addition of the chaotropic agent KSCN. K(D) values, determined under equilibrium conditions, were, however, markedly different from K(D) values derived from kinetic data. Furthermore, the results of analyses of these kinetic data were consistent with the existence of two different populations of [3H]ACPA binding sites. The pharmacology of [3H]ACPA binding sites was characterized using a series of AMPA receptor agonists and antagonists. Whereas addition of KSCN had little effect on the affinities of AMPA receptor agonists for [3H]ACPA binding, this chaotropic agent reduced the affinities of AMPA receptor antagonists structurally related to AMPA. Based on these and previously reported data, the AMPA receptor agonists, ACPA, AMPA and (S)-5-fluorowillardiine, seem to bind to and activate AMPA receptors in a nonidentical fashion, and these three agonists together may be useful tools for studies of AMPA receptor mechanisms.

Animals↗

Equilibrium binding characteristics of [3H]thiomuscimol.

The equilibrium binding characteristics of the tritiated GABAA agonist, 5-aminomethyl-3-isothiazolol (thiomuscimol) are described. Using the filtration technique to separate bound- from free-ligand, [3H]thiomuscimol was shown to bind to the GABA(A) receptor site(s) in a saturable manner with a Kd value of 28+/-6.0 nM and a Bmax value of 50+/-4.0 fmol/mg original tissue. In parallel binding experiments, the Kd and Bmax values for [3H]muscimol were determined to be 5.4+/-2.8 nM and 82+/-11 fmol/mg original tissue, respectively. In binding assays using the centrifugation technique, Kd and Bmax values for [3H]thiomuscimol were found to be 116+/-22 nM and 154 13 fmol/mg original tissue, respectively, whereas a Kd value of 16+/-1.8 nM and a Bmax value of 155+/-8.0 fmol/mg original tissue were determined for [3H]muscimol. In comparative inhibition studies using the GABA(A) antagonist SR 95531 and a series of specific GABAA agonists, the binding sites for [3H]thiomuscimol and [3H]muscimol were shown to exhibit similar pharmacological profiles. Autoradiographic studies disclosed similar regional distribution of [3H]thiomuscimol and [3H]muscimol binding sites in rat brain. Highest densities of binding sites were detected in cortex, hippocampus, and cerebellum, whereas low densities were measured in the midbrain structures of rat cortex. In conclusion, the equilibrium GABA(A) receptor binding characteristics of [3H]thiomuscimol are very similar to those of [3H]muscimol.

Animals↗

Double-tracer autoradiographic study of protein synthesis and glucose consumption in rats with focal cerebral ischemia.

A double-tracer autoradiographic method for simultaneous measurement of regional glucose utilization (rCMRglc) and regional protein synthesis (PS) in consecutive brain sections is described and applied to study the metabolism of the ischemic penumbra 2 h after occlusion of the middle cerebral artery (MCAO) in rats. In halothane anesthesia, the left middle cerebral artery was permanently occluded. Two hours after MCAO an i.v. bolus injection of 14C-deoxyglucose and 3H-leucine was given and circulated for 45 min. Two sets of brain sections were processed for quantitative autoradiography. Neighboring brain sections exposed an X-ray film (3H-insensitive), and a 3H-sensitive for determination of rCMRglc and PS, respectively. Sections for PS determination were washed in trichloroacetic acid (TCA) prior to film exposure in order to remove 14C-deoxyglucose and unincorporated 3H-leucine. Regional rates of PS and glucose utilization were measured by densitometric image analysis. Normal rates of metabolism were defined as mean +/- 2 SD of values in the non-ischemic cortex. The volumes of ischemic cortex displaying normal rates of PS and glucose utilization, respectively, were measured. The cortical volume with normal PS was significantly less than that of normal rCMRglc: 142 (127-147) mm3 vs. 203 (184-206) mm3. Treatment with the glutamate antagonists MK-801 (1 mg kg-1) and NBQX (30 mg kg-1 x 2) did not significantly change this, although MK-801 tended to reduce the size of the metabolic penumbra calculated as the difference between ischemic cortex with reduced PS and ischemic cortex with reduced rCMRglc.

Animals↗

Antisense oligonucleotide to GABA(A) receptor gamma2 subunit induces limbic status epilepticus.

Gamma-Aminobutyric acid (GABA) is the principal inhibitory neurotransmitter in the brain. A deficiency of GABAergic inhibition mediated via the GABAA receptor complex has for a long time been suspected to be a central factor in epileptogenesis. Status epilepticus is a condition of sustained and prolonged excitation of neuronal circuits, as detected by epileptiform discharges in the electroencephalogram (EEG). Reduction of GABAA receptor-mediated hippocampal inhibition has been implicated in the development of status epilepticus. The present study provides direct evidence of a link between the GABAA receptor and epilepsy. We show that selective inhibition of the expression of the GABAA receptor gamma2 subunit in the rat hippocampus by means of antisense oligonucleotides leads to spontaneous electrographic seizures that evolve into profound limbic status epilepticus, ultimately resulting in severe neurodegenerative changes. Concurrent treatment with diazepam prevents the development of status epilepticus and markedly reduces neuronal cell loss. These findings strongly support the hypothesis that the GABAA receptor is critically involved in the pathogenesis of seizures and status epilepticus.

Animals↗

Microglia and macrophages are major sources of locally produced transforming growth factor-beta1 after transient middle cerebral artery occlusion in rats.

The potentially neurotrophic cytokine transforming growth factor-beta1 (TGF-beta1) is locally expressed following human stroke and experimental ischemic lesions, but the cellular source(s) and profile of induction have so far not been established in experimental focal cerebral ischemia. This study presents the time course and a cellular localization of TGF-beta1 mRNA, visualized by in situ hybridization combined with immunohistochemical staining for microglia, macrophages, or astrocytes, on brain sections from adult spontaneously hypertensive rats subjected to transient proximal occlusion of their middle cerebral artery. Six hours after ischemia, an early and transient neuronal and microglial expression of TGF-beta1 mRNA was observed in the extraischemic cingulate and frontal cortices. Both early and protracted expression of TGF-beta1 mRNA in the caudate-putamen and neocortical infarcts and in the caudate-putamen penumbra colocalized with OX42/ED1-immunoreactive microglia and macrophages, whereas TGF-beta1 mRNA in the neocortical penumbra colocalized with OX42/ED1-immunoreactive cells of a microglial morphology. No astrocytes were double-labeled. The number of TGF-beta1 mRNA-expressing microglia and macrophages increased strongly during the first week. Thereafter, TGF-beta1 mRNA became increasingly restricted to the neocortical penumbra (3 weeks), and after 3 months it was confined to activated microglia in the anterior commissure. Our data establish activated microglia and macrophages as the major source of TGF-beta1 mRNA following experimental focal cerebral ischemia. Consequently, TGF-beta1-mediated functions may be exerted by microglia both in the early degenerative phase, and later in combination with blood-borne macrophages, in the remodeling and healing phase after focal cerebral ischemia.

Animals↗