GFAP-immunoreactive Astrocytes Research Articles

Transient Cerebral Ischemia Induces Active Astrocytosis Without Distinct Neuronal Death in the Gerbil Main Olfactory Bulb: A Long-Term Analysis

In the present study, we examined ischemia-induced neuronal and glial changes in the gerbil MOB at various time points during 60 days after 5 min of transient cerebral ischemia. The number of neuronal neuclei-immunoreactive neurons was not changed after ischemia/reperfusion (I/R). Myelin basic protein immunoreaction was well preserved after I/R. Five days after I/R, reactive form of GFAP-immunoreactive astrocytes began to increase in the external plexiform layer and granule cell layer: These reactive astrocytes peaked 10 days after I/R, thereafter, they decreased with time after I/R. Iba-1-immunoreactive microglia were ubiquitously distributed in all layers of the MOB. After I/R, significant changes in their morphology and immunoreactivity were not detected. The results of western blot analyses for GFAP, Iba-1 and MBP were similar to the immunohistochemical data. In addition, 8-hydroxy-2'-deoxyguanosine (a marker for DNA damage) immunoreactivity and SOD1, an antioxidant, protein levels were not changed in the ischemic MOB. These results indicate that neurons in the MOB are resistant to ischemic insult, showing that astrocytes are activated late in the ischemic MOB.

Minocycline inhibits glial proliferation in the H-Tx rat model of congenital hydrocephalus

BackgroundReactive astrocytosis and microgliosis are important features of the pathophysiology of hydrocephalus, and persistent glial "scars" that form could exacerbate neuroinflammation, impair cerebral perfusion, impede neuronal regeneration, and alter biomechanical properties. The purpose of this study was to determine the efficacy of minocycline, an antibiotic known for its anti-inflammatory properties, to reduce gliosis in the H-Tx rat model of congenital hydrocephalus.MethodsMinocycline (45 mg/kg/day i.p. in 5% sucrose at a concentration of 5-10 mg/ml) was administered to hydrocephalic H-Tx rats from postnatal day 15 to day 21, when ventriculomegaly had reached moderate to severe stages. Treated animals were compared to age-matched non-hydrocephalic and untreated hydrocephalic littermates. The cerebral cortex (both gray matter laminae and white matter) was processed for immunohistochemistry (glial fibrillary acidic protein, GFAP, for astrocytes and ionized calcium binding adaptor molecule, Iba-1, for microglia) and analyzed by qualitative and quantitative light microscopy.ResultsThe mean number of GFAP-immunoreactive astrocytes was significantly higher in untreated hydrocephalic animals compared to both types of controls (p < 0.001). Minocycline treatment of hydrocephalic animals reduced the number of GFAP immunoreactive cells significantly (p < 0.001). Likewise, the mean number of Iba-1 immunoreactive microglia was significantly higher in untreated hydrocephalic animals compared to both types of controls (p < 0.001). Furthermore, no differences in the numbers of GFAP-positive astrocytes or Iba-1-positive microglia were noted between control animals receiving no minocycline and control animals receiving minocycline, suggesting that minocycline does not produce an effect under non-injury conditions. Additionally, in six out of nine regions sampled, hydrocephalic animals that received minocycline injections had significantly thicker cortices when compared to their untreated hydrocephalic littermates.ConclusionsOverall, these data suggest that minocycline treatment is effective in reducing the gliosis that accompanies hydrocephalus, and thus may provide an added benefit when used as a supplement to ventricular shunting.

Neuronal damage is much delayed and microgliosis is more severe in the aged hippocampus induced by transient cerebral ischemia compared to the adult hippocampus

Activation of astrocytes and microglia in the post-ischemic hippocampus has been investigated using ischemia models. The aim of this study was to investigate differences of delayed neuronal death and gliosis in the hippocampal CA1 region (CA1) between adult and aged gerbils. Delayed neuronal death in the CA1 was later in the aged gerbil than in the adult gerbil after ischemia/reperfusion (I/R). GFAP-immunoreactive ( +) astrocytes and Iba-1 + microglia were activated following neuronal damage in both adult and aged gerbils after I/R. Changes in GFAP immunoreactivity and protein levels were similar in both groups: they were distinctly increased from 3 days after I/R. Iba-1 immunoreactivity and protein levels in the aged sham gerbil were much higher than those in the adult sham gerbil. Activation of microglia in the CA1 of the aged group was slower, lower 4 days and much higher 7 days than that in the adult gerbil after I/R. These observations indicate that delayed neuronal death in the CA1 of the aged group is slower than that in the adult group after I/R. In addition, microglial activation, not astrocytes, in the aged ischemia group is slower and more intense than that in the adult ischemia group.

Effects of Increased Iron Intake During the Neonatal Period on the Brain of Adult AβPP/PS1 Transgenic Mice

The present study was aimed to investigate neuropathological changes in AbetaPP/PS1 transgenic mice (Tg), as a model of Alzheimer's disease, subjected to supplementary iron administration in a critical postnatal period, in order to reveal the interaction of genetic and environmental risk factors in the pathogenesis of the disease. Twelve Tg and 10 wild-type (Wt) littermates were administered iron between the 12th and 14th post-natal days (TgFe, WtFe); 11 Tg and 15 Wt received vehicle (sorbitol 5%) alone in the same period (TgSb, WtSb). Mice were killed at the age of six months and processed for morphological and biochemical studies. No modifications in amyloid-beta burden were seen in iron-treated and non-iron-treated AbetaPP/PS1 mice. No differences in microglial reactions were observed when comparing the four groups of mice. Yet increased astrocytosis, as revealed by densitometry of GFAP-immunoreactive astrocytes, and increased expression levels of GFAP, as revealed by gel electrophoresis and western blotting, were found in iron-treated mice (both Tg and Wt) when compared with TgSb and WtSb. This was accompanied by significant changes in brain fatty acid composition in AbetaPP/PS1 mice that led to a lower membrane peroxidizability index and to reduced protein oxidative damage, as revealed by reduced percentages of the oxidative stress markers: glutamic semialdehyde, aminoadipic semialdehyde, Nepsilon-carboxymethyl-lysine, Nepsilon-carboxyethyl-lysine, and Nepsilon-malondialdehyde-lysine. These findings demonstrate that transient dietary iron supplementation during the neonatal period is associated with cellular and metabolic imprinting in the brain in adult life, but it does not interfere with the appearance of amyloid plaques in AbetaPP/PS1 transgenic mice.

Effects of alcohol administration during adulthood on parvalbumin and glial fibrillary acidic protein immunoreactivity in the rat cerebral cortex

The pathology of brain atrophy mediated by alcohol was investigated in all parts of the cerebral cortex (the frontal, parietal, temporal lobes and occipital cortex) by using two markers: parvalbumin (PV) and glial fibrillary acidic protein (GFAP). Three-month old male Wistar rats were divided into control (C) and alcohol-exposed groups. The control group received distilled water, whereas the alcohol-exposed groups received either a low dose (2g/kg body wt) or a high dose (5g/kg) of ethanol for periods of 21 days, 3 or 6 months. The brains of the animals were processed for immunohistochemistry using anti-parvalbumin and anti-GFAP antibodies and the number of PV immunoreactive (PV-ir) neurons and GFAP immunoreactive (GFAP-ir) astrocytes were counted per unit area. Results showed that all groups exposed to ethanol had significantly reduced numbers of PV-ir neurons in all parts of the cerebral cortex compared to those of the control group (p<0.05). In contrast, the numbers of GFAP-ir astrocytes were increased in all parts of the cerebral cortex following the exposure to a high dose of ethanol after 21-days (but not a low dose) and both high and low doses of ethanol after 3-months or 6-months treatment compared to those of age-matched control groups (p<0.05). This indicated that in young rats (21-days), PV-ir neurons in all cerebral cortex areas seemed to be more sensitive to alcohol than GFAP-ir astrocytes. Moreover, the change in densities of both PV-ir neurons and GFAP-ir astrocytes became more apparent after exposure to prolonged and high doses of ethanol. The decrease of PV-ir neurons and the increase of GFAP-ir astrocytes indicated that alcohol may induce pathology in broad areas of the cerebral cortex. This may explain the underlying mechanism of brain atrophy and other impairments found in alcoholics. For investigations of the effects of alcohol on mediating brain pathology, we recommend the use of the two markers (PV and GFAP).

Effects of pre-emptive drug treatment on astrocyte activation in the cuneate nucleus following rat median nerve injury

In this study, we examined the relationship between astrocyte activation in the cuneate nucleus (CN) and behavioral hypersensitivity after chronic constriction injury (CCI) of the median nerve. In addition, we also examined the effects of pre-emptive treatment with a number of drugs on astrocyte activation and hypersensitivity development in this model. Using immunohistochemistry and immunoblotting, little glial fibrillary acidic protein (GFAP; an astrocyte marker) immunoreactivity was detected in the CN of the normal rats. As early as 3 days after CCI, there was a significant increase in GFAP immunoreactivity in the lesion side of CN, and this reached a maximum at 7 days, and was followed by a decline. Counting of GFAP-immunoreactive astrocytes revealed that astrocytic hypertrophy, but not proliferation, contributes to increased GFAP immunoreactivity. Furthermore, microinjection of the glial activation inhibitor, fluorocitrate, into the CN at 3 days after CCI attenuated injury-induced behavioral hypersensitivity in a dose-dependent manner. These results suggest that median nerve injury-induced astrocytic activation in the CN modulated the development of behavioral hypersensitivity. Animals received MK-801 (glutamate N-methyl- d-aspartate (NMDA) receptor antagonist), clonidine (α 2-adrenoreceptor agonist), tetrodotoxin (TTX, sodium channel blocker) or lidocaine (local anesthetic) 30 min prior to median nerve CCI. Pre-treatment with MK-801, TTX, and 2% lidocaine, but not clonidine, attenuated GFAP immunoreactivity and behavioral hypersensitivity following median nerve injury. In conclusion, suppressing reactions to injury, such as the generation of ectopic discharges and activation of NMDA receptors, can decrease astrocyte activation in the CN and attenuate neuropathic pain sensations.

Alcohol administration during adulthood induces alterations of parvalbumin and glial fibrillary acidic protein immunoreactivity in rat hippocampus and cingulate cortex

Alcohol induces impairment of cognition, learning and memory. Neurotoxic effects of alcohol on the pathology of the hippocampus and the cingulate cortex were investigated in experimental rats. Parvalbumin (PV), a calcium-binding protein, is a crucial component of GABAergic neurons and glial fibrillary acidic protein immunoreactive (GFAP-ir) astrocytes have been used as markers. We investigated the effects of ethanol exposure during adulthood on the PV-ir neurons and GFAP-ir astrocytes in the hippocampus and the cingulate cortex of 3-month-old male Wistar rats. The rats were divided into 2 groups: control (C) and alcohol-exposed groups. The control group received distilled water whereas the alcohol-exposed groups received either a low dose (20%w/v, LD) or high dose (40%w/v, HD) of ethanol for periods of 21 days, 3 or 6 months. The brains of the animals were processed for immunohistochemistry using anti-parvalbumin and anti-GFAP antibodies and the numbers of PV immunoreactive (PV-ir) neurons and GFAP-ir astrocytes were counted/unit area. For each period of administration, the number of PV-ir neurons was significantly reduced for groups exposed to both the low and the high doses of ethanol compared to those of control groups in both the hippocampus and the cingulate cortex (p<0.01). In addition, the number of PV-ir neurons was progressively reduced after prolonged ethanol exposure. In contrast, there was a significantly increased number of GFAP-ir astrocytes observed in the hippocampus and the cingulate cortex in all groups exposed to ethanol and this was a function of both the duration and the dose of ethanol exposure, indicating that PV-ir neurons are as sensitive as the GFAP-ir astrocytes to ethanol exposure. Our data indicate that alcohol exposure induced a reduction of PV-ir neurons and an increase of GFAP-ir astrocytes in the hippocampus and the cingulate cortex and this may be associated with the impairment of cognition, learning and memory after chronic alcohol administration.

Juvenile separation stress induces rapid region- and layer-specific changes in S100ß- and glial fibrillary acidic protein–immunoreactivity in astrocytes of the rodent medial prefrontal cortex

The impact of juvenile stress exposure on astrocyte plasticity was assessed in the precocious rodent Octodon degus. Astrocytes expressing S100ß and glial fibrillary acidic protein (GFAP) were quantified in the limbic medial prefrontal cortex (mPFC), including the anterior cingulate (ACd), precentral medial (PrCm), infra- (IL) and prelimbic (PL) cortex and in the “non-limbic” somatosensory cortex (SSC). At the age of 21 days we compared (i) controls (C), (ii) stressed animals (SSR: separation stress/short reunion), which were exposed to 6 h separation from the family, followed by 1 h reunion with the family and (iii) stressed animals (SER: separation stress/extended reunion), which were stressed like group SSR but exposed to 48 h reunion. The observed glia response was already measurable 7 h after the onset of the stress exposure. Compared to controls SER and SSR animals showed elevated densities of S100ß-IR astrocytes in layers II/III and V–VI of the ACd, IL and PrCm, whereas no significant group differences were observed in the PL and SSC. The SSR group showed significantly decreased density of GFAP-immunoreactive astrocytes in all mPFC subregions. Only in the ACd the stress-induced changes in glia density were still evident after 48 h reunion with the family. Compared to controls, the length of GFAP-IR processes and the number of ramification points were significantly reduced in all mPFC subregions and in the SSC of the SSR group. In the SSC the stress-evoked changes in GFAP-glia density were opposite compared to the changes seen in the medial prefrontal cortical subregions, whereas the changes in GFAP-labeled processes were comparable to those observed in the mPFC. In summary, these results demonstrate that a single stress episode induces rapid and quite complex region- and cell-specific changes in glial cells, reflected by an upregulation of cytoplasmic (S100ß) and downregulation of cytoskeletal (GFAP) glial protein.

Effects of distal nerve injuries on dorsal-horn neurons and glia: Relationships between lesion size and mechanical hyperalgesia

Penetrating limb injuries are common and usually heal without long-lasting effects, even when nerves are cut. However, rare nerve-injury patients develop prolonged and disabling chronic pain (neuralgia). When pain severity is disproportionate to severity of the inciting injury, physicians and insurers may suspect exaggeration and limit care or benefits, although the nature of the relationship between lesion-size and the development and persistence of neuralgia remains largely unknown. We compared cellular changes in the spinal dorsal-horn (the initial CNS pain-processing area) after partial or total tibial-nerve axotomies in male Sprague–Dawley rats to determine if these changes are proportional to the numbers of peripheral axons cut. Unoperated rats provided controls. Plantar hind-paw responses to touch, pin, and cold were quantitated bilaterally to identify hyperalgesic rats. We also compared data from nerve-injured rats with or without hyperalgesic responses to mechanical hind-paw stimulation to evaluate concordance between pain behaviors and dorsal-horn cellular changes. Hyperalgesia was no less prevalent or severe after partial than after total axotomy. L5 spinal-cord sections from rats killed 7 days postoperatively were labeled for markers of primary afferents (substance P calcitonin gene-related peptide isolectin B4, gamma aminobutyric acid, and glial fibrillary acidic protein), then labeled cells were stereologically quantitated in somatotopically defined dorsal-horn regions. Total axotomy reduced markers of primary afferents more than partial axotomy. In contrast, GABA-immunoreactive profiles were similarly reduced after both lesions, and in rats with sensory loss versus hyperalgesia. Numbers of GFAP-immunoreactive astrocytes increased independently of lesion size and pain status. Small nerve injuries can thus have magnified and disproportionate effects on dorsal-horn neurons and glia, perhaps providing a biological correlate for the disproportionate pain of post-traumatic neuralgias (including complex regional pain syndrome-I) that follow seemingly minor nerve injuries. However, the presence of similar dorsal-horn changes in rats without pain behaviors suggests that not all transcellular responses to axotomy are pain-specific.

Mammillary body alterations and spatial memory impairment in Wistar rats with thioacetamide-induced cirrhosis

Brain tissue of patients diagnosed with hepatic encephalopathy exhibits cellular morphological changes that could be associated with memory impairment. The mammillary nuclei, located in the posterior part of the hypothalamus, are important for spatial memory formation. This work aimed to assess spatial reference memory and cellular changes in the mammillary nuclei of cirrhotic rats. Spatial reference memory of Wistar rats with thioacetamide-induced cirrhosis was assessed in the Morris water maze. Total cell number of neurons and glial cells and volume of the mammillary nuclei were quantified by stereology. Neuronal and astrocytic nuclear volume in mammillary nuclei and CA1 dorsal hippocampal subfield were assessed by nucleator probe. Cirrhotic rats showed an impaired spatial reference memory in comparison with control animals. Total number of neurons and glial cells were unaltered. In the medial mammillary nucleus (MMn), glial fibrillary acidic protein-immunoreactive astrocytes decreased in the cirrhotic group while the lateral part was unaffected. The medial part of the MMn was larger in the cirrhotic group. The cirrhotic rats showed morphometric cellular changes characterised by an increased neuronal and astrocytic nuclear volume in all the mammillary nuclei and CA1 hippocampal region. These findings suggest that cirrhotic rats show spatial memory impairment that could be linked to astrocytes and neuronal impairment in mammillary nuclei and hippocampus.

Neurodegeneration and cellular stress in the retina and optic nerve in rat cerebral ischemia and hypoperfusion models

Experimental cerebral ischemia induces a stress response in neuronal and non-neuronal cells. In the present study we aimed to evaluate detailed cellular stress responses and neurodegenerative changes in the retinas in rat focal cerebral ischemia and hypoperfusion models involving invasive vascular manipulations. Independent groups of adult male Wistar rats were subjected to i) transient middle cerebral artery occlusion (tMCAO), ii) permanent middle cerebral artery occlusion (pMCAO), iii) cortical photothrombosis of the sensorimotor cortex using Rose Bengal dye or iv) bilateral common carotid artery occlusion (BCCAO). Rats were killed, and their eyes with the optic nerves enucleated and processed for histology, immunohistochemistry for neuronal nuclei (NeuN), glial fibrillary acidic protein (GFAP), hypoxia-inducible factor 1α (HIF-1α), c-fos, αB-crystallin, heat shock protein (HSP) 27, HSP60 and HSP70, and detection of DNA defragmentation. The total number of the retinal ganglion cell layer (RGCL) neurons and GFAP-immunoreactive astrocytes located in the nerve fiber layer were estimated using unbiased stereological counting. Our findings indicate that although permanent and transient MCAO does not cause detectable morphological alterations in the retina or optic nerve, it evokes ischemic stress as revealed by HIF-1α and HSPs expression in the RGCL neurons and reactive gliosis in the Müller cells. Severe neurodegenerative changes in the retina and optic nerve of the BCCAO rats are accompanied by a significant increase in immunoreactivities for the c-fos, HSP27 and HSP70 as compared with the sham-operated animals. The retinas from the ipsilateral side of the Rose Bengal model showed a significant decrease in the total number of NeuN-positive neurons in the RGCL as compared with the contralateral ones. However, these eyes did not differ between each other in the HSPs and HIF-1α expression or in the GFAP-immunoreactivity of the Müller cells. In conclusion, our data suggest differential expression of various HSPs in the retina and possibly their distinct roles in the cerebral ischemia-mediated stress response and neurodegeneration.

Effects of the PPARγ activator pioglitazone on p38 MAP kinase and IκBα in the spinal cord of a transgenic mouse model of amyotrophic lateral sclerosis

Emerging evidence suggests the involvement of programmed cell death and inflammation in amyotrophic lateral sclerosis (ALS). To assess molecular pathological effects of the anti-inflammatory peroxisome proliferator-activated receptor-gamma (PPARgamma) agonist pioglitazone in ALS, we verified changes in the population of neurons, astrocytes, and microglia in the ventral horns of spinal cord lumbar segments from the pioglitazone-treated and non-treated groups of mice carrying a transgene for G93A mutant human superoxide dismutase-1 (SOD1) (ALS mice) and non-transgenic littermates (control mice), performed immunohistochemical and immunoblot analyses of PPARgamma, active form of phosphorylated p38 mitogen-activated protein kinase (p-p38) and inhibitor of nuclear factor-kappaB (NF-kappaB)-alpha (IkappaBalpha) in the spinal cords, and compared the results between the different groups. Image analysis revealed that optical density of NeuN-immunoreactive neurons was significantly lower in the non-treated groups of presymptomatic and advanced ALS mice than in the non-treated groups of age-matched control mice and was recovered with pioglitazone treatment, and that optical densities of GFAP-immunoreactive astrocytes and Iba1-immunoreactive microglia were significantly higher in the non-treated group of advanced ALS mice than in the non-treated group of control mice and were recovered with pioglitazone treatment. Immunohistochemical analysis demonstrated that immunoreactivities for PPARgamma and p-p38 were mainly localized in neurons, and that IkappaBalpha immunoreactivity was mainly localized in astrocytes and microglia. Immunoblot analysis showed that pioglitazone treatment resulted in no significant change in nuclear PPARgamma-immunoreactive density, a significant decrease in cytosolic p-p38-immunoreactive density, and a significant increase in cytosolic IkappaBalpha-immunoreactive density. Our results suggest that pioglitazone protects motor neurons against p38-mediated neuronal death and NF-kappaB-mediated glial inflammation via a PPARgamma-independent mechanism.

ROSETTE-FORMING GLIONEURONAL TUMOR

Rosette-forming glioneuronal tumor is a newly described mixed glial and neuronal tumor. We describe two cases and review the literature to better characterize this entity. Patients were surgically treated, and tumors were diagnosed by light microscopy and immunohistochemistry using the avidin-biotin complex method. PubMed was searched for previously reported cases. Patient 1 was a 38-year-old woman who presented with headaches and no neurological abnormality. Magnetic resonance imaging showed a solid mass in the fourth ventricle. Subtotal excision of the mass caused transient gait ataxia. Patient 2 was a 51-year-old woman with dizziness who fell and sustained head trauma. Magnetic resonance imaging revealed a right paramedian cerebellar cystic and nodular mass and a separate nodule in the vermis, which were excised gross totally with no morbidity. Microscopic examination showed neuroepithelial tumors composed of neurocytic cells focally forming well-defined rosettes that were immunopositive for neuronal markers and of elongated, glial fibrillary acidic protein-immunoreactive astrocytes. No histological anaplasia was present. Both patients were well 18 and 8 months after surgery, respectively. Eighteen rosette-forming glioneuronal tumors were identified with the literature search. These are tumors of young adulthood (range, 12-59 yr) usually in or close to the fourth ventricle. Histologically, they are low-grade, although multiple foci or local extension may prevent total excision and account for some recurrences. On imaging, they are cystic, solid, or both, with minimal perilesional edema or mass effect. They are composed of neurocytic and glial elements, probably arising from a common progenitor in the subependymal plate, and need to be differentiated from a variety of glioneuronal tumors.

Alteration of Glial Fibrillary Acidic Proteins Immunoreactivity in Astrocytes of the Spinal Cord Diabetic Rats

Diabetes affects retinal and nervous glial cells, especially the astrocytes. A key indicator of this response is the alteration in the level of intermediate filament glial fibrillary acidic protein (GFAP) and number of GFAP-immunoreactive astrocytes. To date, no study has investigated the effect of diabetes on the distribution of GFAP-immunoreactive astrocytes in the spinal cord. Therefore, the present study investigated the effect of diabetes on the number of GFAP-immunoreactive astrocytes in the gray matter of the spinal cord of streptozotocin-induced diabetic Wistar rats. Animals were divided into six groups (n = 7); 6 weeks and 12 weeks diabetic duration groups and their respective age-matched normal control and sham control groups. Our results demonstrated a significant (P < 0.001) decrease in the number of GFAP-immunoreactive astrocytes in different areas of the spinal cord sections of the 6 weeks and 12 weeks long diabetic rats when compared with the spinal cord of normal and sham control groups of comparable age. The mean percentage in total number of GFAP-immunoreactive astrocytes in the whole gray matter areas of the spinal cord of the 6 and 12 weeks diabetic groups were approximately 28% and 41% less than control groups. Furthermore, the 12 weeks diabetic group showed a significant (P < 0.001) reduction in the number of GFAP-immunoreactive astrocytes when compared with the 6 weeks diabetic animals. These results suggest that the induction of diabetes is associated with a reduction in GFAP-positive astrocytes in the spinal cord, which may affect the functional support and role of astrocytic cells in the nervous tissue. This in turn may contribute to the pathological changes associated with diabetic state in the central nervous system.

Consumption of green tea alters glial fibriliary acidic protein immunoreactivity in the spinal cord astrocytes of STZ-diabetic rats

We examined the effect of green tea consumption on glial fibriliary acidic protein (GFAP) expression in spinal cord of streptozotocin (STZ) treated rats. Three groups (n = 10) were used in this study: (i) controls; (ii) STZ-induced diabetic rats given tap water; and (iii) an STZ-induced diabetic group given green tea. Immunohistochemistry showed a significant (P < 0.001) decrease in the number of GFAP immunoreactive astrocytes in spinal cord sections of diabetic rats compared to non-diabetic controls. Diabetic rats treated with green tea showed a significant (P < 0.01) increase in the number GFAP-immunoreactive astrocytes in all the spinal cord gray areas as compared to water-drinking diabetic rats. Immunoblotting confirmed that the diabetic spinal cord tissue expressed 71.0 ± 7.0% less GFAP compared to non-diabetic controls and that the GFAP content in diabetic rats increased up to 86.34 ± 18.74% compared to non-diabetic controls after 12 weeks of green tea consumption. In conclusion, consumption of green tea may represent an achievable adjunct therapy for improving changes seen in diabetic spinal cord.

Role of reactive nitrogen and reactive oxygen species against MPTP neurotoxicity in mice

There is growing evidence indicating that reactive nitrogen species (RNS) and reactive oxygen species (ROS) are a major contributor to the pathogenesis and progression of Parkinson's disease. Here we investigated whether edaravone (free radical scavenger), minocycline (inducible nitric oxide synthase, iNOS inhibitor), 7-nitroindazole (neuronal NOS, nNOS inhibitor), fluvastatin (endothelial NOS, eNOS activator) and pitavastatin (eNOS activator) can protect against MPTP neurotoxicity in mice under the same condition. The present study showed that 7-nitroindazole could protect dose-dependently against the striatal dopamine depletions in mice 5 days after MPTP treatment. In contrast, edaravone, minocycline, fluvastatin and pitavastatin did not show the neuroprotective effect on MPTP-induced striatal dopamine depletion. Our immunohistochemical study showed that TH (tyrosine hydroxylase) and DAT (dopamine transporter) immunoreactivity was decreased significantly in the striatum and substantia nigra 5 days after MPTP treatment. The administration of 7-nitroindazole showed a protective effect against the severe reductions in levels of TH and DAT immunoreactivity in the striatum and substantia nigra 5 days after MPTP treatment. Furthermore, our Western blot analyses study showed the remarkable loss of TH protein levels in the striatum 5 days after MPTP treatment. In contrast, 7-nitroindazole prevented a significant loss in TH protein levels in the striatum 5 days after MPTP treatment. On the other hand, GFAP (glial fibrillary acidic protein) immunoreactivity increased significantly in the striatum and substantia nigra, 5 days after MPTP treatment. 7-Nitroindazole ameliorated severe increases in number of GFAP immunoreactive astrocytes in the striatum and substantia nigra 5 days after MPTP treatment. Furthermore, our Western blot analyses study showed the increase of GFAP protein levels in the striatum 5 days after MPTP treatment. 7-Nitroindazole prevented a significant increase in the GFAP protein levels in the striatum 5 days after MPTP treatment. The present results indicate that 7-nitroindazole can protect dose-dependently against the striatal dopamine depletions in mice 5 days after MPTP treatment. In contrast, edaravone, minocycline, fluvastatin and pitavastatin did not show the neuroprotective effect on MPTP-induced striatal dopamine depletions. These findings demonstrate that the overexpression of nNOS may play a major role in the neurotoxic processes of MPTP, as compared to the production of ROS, the overexpression of iNOS and the modulation of eNOS. Thus, our findings provide strong evidence for neuroprotective properties of nNOS inhibitor in this animal model of Parkinson's disease.

Changes in Glial Fibrillary Acidic Protein Immunoreactivity in the Detate Gyrus and Hippocampus Proper of Adult and Aged Dogs

Astrocytes perform neuron-supportive tasks, repair and scarring process in the central nervous system. In this study, we observed glial fibrillary acidic protein (GFAP), a marker for astrocytes, immunoreactivity in the dentate gyrus and hippocampus proper (CA1-3 region) of adult (2-3 years of age) and aged (10-12 years of age) dogs. In the adult group, GFAP immunoreactive astrocytes were distributed in all layers of the dentate gyrus and CA1-3 region, except in the stratum pyramidale of the CA1-3 region. In the aged group, GFAP immunoreactivity decreased markedly in the molecular layer of the dentate gyrus. However, GFAP immunoreactivity in the CA1-3 region increased in all layers, and the cytoplasm of GFAP immunoreactive astrocytes was hypertrophied. GFAP protein levels in the aged dentate gyrus decreased; however, GFAP levels in the CA1-3 region increased. These results suggest that the morphology of astrocytes and GFAP protein levels in the hippocampal dentate gyrus and CA1 region are changed, respectively, with age.

Differential Immunoreactivity of Microglial and Astrocytic Marker Protein in the Hippocampus of the Seizure Resistant and Sensitive Gerbils

In the present study, we compared differences in ionized calcium-binding adapter molecule 1 (Iba-1) and glial fibrillary acidic protein (GFAP) immunoreactivities for microglia and astrocytes, respectively, in the hippocampus of the seizure-resistant (SR) and seizure-sensitive (SS) gerbils. The density of Iba-1 immunoreactive microglia in the hippocampal CA1 region (CA1) and dentate gyrus (DG) of the SS gerbil was higher than that in the SR gerbil, and many Iba-1 immunoreactive microglia in the SS gerbil were hypertrophied in morphology. In contrast, we could not find significant difference in the density of GFAP immunoreactive astrocytes between the SR and SS gerbils. This result indicates that Iba-1 immunoreactive microglia in CA1 and DG of the SS gerbil are activated compared to those in the SR gerbil.

Microvascular P-glycoprotein expression at the blood–brain barrier following focal astrocyte loss and at the fenestrated vasculature of the area postrema

The multidrug transporter, P-glycoprotein, expressed at the blood–brain barrier is thought to be important for limiting access of toxic agents to the brain, but its relationship to astrocyte expression is unclear. We have studied P-glycoprotein expression in the inferior colliculus after a temporary loss of blood–brain barrier integrity following chemically induced astrocyte loss and at the fenestrated vascular endothelium of the area postrema. Male Fisher F344 rats given 3-chloropropanediol showed astrocyte loss from 12 to 24 h until the lesion was repopulated 8–28 days later. In non-dosed tissue, P-glycoprotein expression was seen the entire length of platelet endothelial cell adhesion molecule immunoreactive vessels. Within 6 h of dosing, a significant ( p < 0.05) reduction in the total length of P-glycoprotein immunoreactive vasculature was evident. By 48 h, P-glycoprotein immunoreactivity was heavily fragmented. The total length of P-glycoprotein immunoreactive vessels became minimal at 4 days ( p < 0.001) but was still present in many vessels. From 6 to 28 days, P-glycoprotein immunoreactivity returned across the inferior colliculus, in parallel with astrocytic repopulation of the lesion, and by 28 days resembled that seen in control tissue. The area postrema showed GFAP immunoreactive astrocytes but which made limited contact with the vasculature, while the platelet endothelial cell adhesion molecule immunoreactive vasculature showed no expression of P-glycoprotein. These findings provide evidence supporting a link between GFAP-astrocyte and P-glycoprotein expression in the mature brain vasculature in vivo.

Aquaporin 1 and aquaporin 4 overexpression in bovine spongiform encephalopathy in a transgenic murine model and in cattle field cases

Aquaporins (AQP) are a family of transmembrane proteins that act as water selective channels. AQP1 and AQP4 are widely expressed in the central nervous system where they play several roles. Overexpression of AQP has been reported in some human and animal transmissible spongiform encephalopathies, but information is scanty about their distribution in the central nervous system in bovine spongiform encephalopathy (BSE). Double immunohistochemistry for AQP1, AQP4 and GFAP was developed in a transgenic mouse line overexpressing the bovine cellular prion protein (BoTg110), intracerebrally infected with cattle BSE. Western blot for AQP1 and AQP4, and immunohistochemistry for both AQP and GFAP were carried out in cases of BSE-diagnosed cattle as part of surveillance plan in Catalonia (Spain). A marked increase in AQP1 and AQP4 was observed in mice at the terminal stage of the disease, when they had a wide range of clinical signs, whereas no increase could be observed in the early stage before the onset of the clinical signs. In cattle which did not show evidence of clinical signs, both AQP already showed a great increase. The AQP overexpression correlated with GFAP-immunoreactive astrocytes and PrPres deposition in both cases. The results of this study suggest that AQP overexpression in glial cells could lead to an imbalance in water and ion homeostasis which could contribute to triggering the typical histopathological changes of BSE.