Glial response to neuronal activity: GFAP-mRNA and protein levels are transiently increased in the hippocampus after seizures

We have recently demonstrated that electrically induced seizures lead to dramatic increases in mRNA for GFAP in areas in which seizures occur. The present study evaluates the time course of the changes in the GFAP-mRNA levels after seizures and the relationship between these changes and GFAP protein levels to understand the role of neuronal activity in regulating glial gene expression. GFA protein and mRNA levels were measured in hippocampi from rats in which seizures were induced by: (1) 50-Hz stimulus trains delivered 12 times over the course of 1 day via indwelling electrodes implanted chronically in the CA3 region of the hippocampus; and (2) intraperitoneal injections of pentylenetetrazol. In the case of the electrically induced seizures, we also compared the glial response in animals that had never experienced a seizure with the response in animals that previously had been kindled but had not experienced a seizure for 30 days. Electrically induced seizures led to rapid transient increases in GFAP-mRNA levels in the hippocampus ipsi- and contralateral to the stimulation. GFAP-mRNA increased about five-fold 1 day after the end of seizure activity and returned to near-control levels by 4 days. There were no detectable increases in GFA protein at 1 day but by 2 days GFA protein levels had increased about two-fold. GFA protein levels remained elevated until 4 days poststimulation and then began to decrease. The responses were similar when seizures were induced in kindled animals, except that the GFAP protein levels remained elevated for somewhat longer. Pentylenetetrazol-induced seizures also led to increases in GFAP-mRNA and GFA protein levels but the extent of the increases was not as great as after kindled seizures. These results suggest that gene expression in astrocytes in likely to be upregulated in any situation in which seizures occur. These changes may fundamentally alter the homeostatic activities of the affected astrocytes which, in turn, could have important consequences on the development of the epileptic state.     

Plasticity of astrocytes

It is becoming apparent that astrocytes carry out a large number of different functions in brain and are able to modify their characteristics throughout life, that is they exhibit a high degree of plasticity in their phenotype. For example, the morphology of astrocytes changes markedly during neuronal migration, maturation, and degeneration. It is conceivable that these cells must constantly adjust their abilities to meet changes in brain environment. Several examples of astrocytic plasticity are presented in this review. First, the ability of astrocytes to recognize neuronal signals can change qualitatively as well as quantitatively; evidence suggests that the expression of glial receptors may be developmentally regulated by both intrinsic and extrinsic signals. Second, the expression of adrenergic receptors by astrocytes in adult brain can change in response to neuronal degeneration. The up-regulation of β-adrenergic receptors in this case suggests that these receptors play a role in function of reactive astrocytes. Finally, glial morphology can be reciprocally regulated by neurotransmitters such as norepinephrine and glutamate. This reciprocal regulation may be significant since both ß-adrenergic receptors and glutamate transporters are found predominantly in astrocytes in the brain. The change in glial morphology may also affect neuronal activity by changing the volume of the extracellular space. © 1994 Wiley-Liss, Inc.     

In vitro protection of 3-nitropropionic acid-induced toxicity of astrocytes by basic fibroblast growth factor and thrombin

The protective action of basic fibroblast growth factor (bFGF) and thrombin on cultured cortical astrocytes in vitro to the toxic effects of 3-nitropropionic acid (3-NPA) was examined. 3-NPA produced concentration- and time-dependent astrocyte loss as indicated by decrease in the number of glial fibrillary acidic protein (GFAP) positive cells and increase in lactate dehydrogenase (LDH) level of the culture medium. The 3-NPA-induced cellular loss was apparent within 12 h and was maximal at 24 h. The presence of bFGF (10 ng/ml) by itself increased the number of astrocytes at various time intervals, and attenuated the 3-NPA-induced cell loss significantly at various concentrations (0.017–1.7 mM), in 24 and 48 h of exposure. Lower concentrations of thrombin (up to 1 nM) had no effect on the number of astrocytes but higher concentrations (10–100 nM) produced greater cell loss. Thrombin (1 nM) prevented the 3-NPA-induced decrease in GFAP positive cells at both 24 and 48 h intervals. This was further substantiated by the fact that thrombin as low as 0.01 nM attenuated the 3-NPA-induced increase of LDH activity at 24 and 48 h exposure times. But, with 1 nM of thrombin, the attenuation of the LDH activity was seen only at 24 h. The results indicate that 3-NPA produced acute astrocyte toxicity and was attenuated by bFGF and lower concentrations of thrombin.     

Gap junction disappearance in astrocytes and leptomeningeal cells as a consequence of protozoan infection

Trypanosoma cruzi and Toxoplasma gondii are protozoan parasites capable of causing infections of the nervous system. In order to determine effects of infection by these organisms on intercellular communication in the brain, dye coupling and connexin abundance and distribution were examined in leptomeningeal cells and astrocytes infected with T. cruzi or T. gondii. For both cell types infected with either type of protozoan parasite, intercellular diffusion of intracellularly injected Lucifer Yellow was dramatically reduced. Immunocytochemistry with antibodies specific for connexin43 (in astrocytes) or both connexin43 and connexin26 (for leptomeningeal cells) demonstrated that punctate gap junctional staining was much reduced in infected cells, although uninfected neighbors could display normal connexin abundance and distribution. Western blot analyses revealed that connexin43 abundance in both cell types infected with either parasite was similar to that in uninfected cells. Phosphorylation state of connexin43 (inferred from electrophoretic mobility of connexin43 isoforms) was not significantly affected by the infection process. Immunocytochemistry of whole brains from animals acutely infected with either parasite also showed a marked reduction in connexin43 expression. We conclude that infection of both types of brain cells with either protozoan parasite results in a loss of intercellular communication and organized gap junction plaques without affecting expression levels or posttranslational processing of gap junction proteins. Presumably, these changes in gap junction distribution result from altered targeting of the junctional protein to the plasma membrane, and/or from changes in assembly of subunits into functional channels.     

Necessity and redundancy of guidepost cells in the embryonic Drosophila CNS

Guidepost cells are specific cellular cues in the embryonic environment utilized by axonal growth cones in pathfinding decisions. In the embryonic Drosophila CNS the RP motor axons make stereotypic pathways choices involving distinct cellular contacts: (i) extension across the midline via contact with the axon and cell body of the homologous contralateral RP motoneuron, (ii) extension down the contralateral longitudinal connective (CLC) through contact with connective axons and longitudinal glia, and (iii) growth into the intersegmental nerve (ISN) through contact with ISN axons and the segmental boundary glial cell (SBC). We have now ablated putative guidepost cells in each of the CNS pathway subsections and uncovered their impact on subsequent RP motor axon pathfinding. Removal of the longitudinal glia or the SBC did not adversely affect pathfinding. This suggests that the motor axons either utilized the alternative axonal substrates, or could still make filopodial contact with the next pathway section's cues. In contrast, RP motor axons did require contact with the axon and soma of their contralateral RP homologue. Absence of this neuronal substrate frequently impeded RP axon outgrowth, suggesting that the next cues were beyond filopodial reach. Together these are the first direct ablations of putative guidepost cells in the CNS of this model system, and have uncovered both pathfinding robustness and susceptibility by RP axons in the absence of specific contacts.     

Voltage-gated currents expressed by rat microglia in culture.

Using the whole-cell patch-clamp technique, at least three types of voltage-gated currents expressed by cultured rat microglia were identified: an inward rectifier K+ current, a delayed rectifier K+ current (IK), and a Na+ current activated by depolarization. The inward rectifier conductance was activated by hyperpolarization to potentials more negative than -80 mV, depended on the external K+ concentration, and declined over time during whole cell recording, as the cell was internally dialyzed. The delayed rectifier current was activated by depolarization to potentials more positive than -40 mV and the rates of activation and deactivation showed a voltage-dependence similar to such currents seen in other preparations. An inward current possibly carried by Na+ was seen in a small percentage of cells. Recordings had been made from two morphological cell types, namely process-bearing ("ramified") and non-process-bearing ("ameboid"). Each of these currents was present in microglia of both morphological types. However, microglial morphology, which is thought to represent different states of activation, was significantly related to the types of combinations of currents expressed in a given cell.     

Axon-glia transfer of a protein and a carbohydrate

We have investigated the transfer of a fluorescent protein, the fluorescein isothiocyanate derivative of bovine serum albumin (FITC-BSA), and a fluorescent carbohydrate, FITC-dextran, from the crayfish medial giant axon (MGA) to the periaxonal glial cells. The dialyzed tracer was injected into one of the two MGAs, and, after a transfer period of 15-60 min, the tissue was fixed for histological examination of fluorescence distribution. With each tracer, the periaxonal sheath of the injected MGA was specifically labeled. Similar results were obtained with several different fixatives. During the transfer period, there was no appreciable change in the resting potential or conducted action potential of the MGA or in the resting potentials of the adaxonal glial cells. Polyacrylamide gel electrophoresis indicated that the axoplasmic and sheath fluorescence was produced by the intact tracers. These results suggest that "foreign" macromolecules can be exchanged from crayfish axons to glia under physiological conditions. Such transfers may indicate a substantial intercellular traffic of molecules or a means whereby neurons can eliminate waste materials.     

Effect of melatonin on d-amphetamine-induced neuroglial alterations in postnatal rat hippocampus and prefrontal cortex

Amphetamine is a psychostimulant drug that produces long-lasting neurotoxic effects on the central nervous system. Recent studies suggested that glia might contribute to amphetamine-induced neuropathy. Excessive activation of astrocytes can be deleterious to the neuron. Amphetamine-induced lesions during development have the potential to produce numerous permanent abnormalities in neural circuitry and function, including memory deficit. In the present study, postnatal rats were injected with either saline or d-amphetamine for 7 consecutive days, starting on postnatal day 4 (P4). Our results found that d-amphetamine caused a marked increase in glia fibrillary acidic protein (GFAP), an astroglia marker, expression that implicated astrogliosis in both hippocampus and prefrontal cortex. The effect of d-amphetamine on hippocampal and prefrontal cortex neurons was also investigated, and we detected a downregulation of βIII-tubulin, a marker of premature neuron expression. Furthermore, we found that pretreatment with melatonin, a major hormone secreted from the pineal gland, prevented glial cell activation and βIII-tubulin reduction, caused by d-amphetamine in both hippocampus and prefrontal cortex. The present study suggests that melatonin can attenuate the detrimental effect of d-amphetamine on glial and neuronal cells.     

Interictal infraslow activity in patients with epilepsy

ObjectiveTo evaluate if interictal infraslow activity (ISA), as obtained from a conventional EEG system, can contribute information about the epileptogenic process.MethodsThe entire long-term intracranial monitoring sessions of 12 consecutive patients were evaluated on an XLTEK system for ISA. Three additional patients had long-term scalp recordings.ResultsIn intracranial as well as scalp recordings, the ISA background was consistently higher in the waking state than during sleep. From this background emerged intermittently focal changes, which could achieve in intracranial recordings millivolt amplitudes, while they remained in the microvolt range in scalp recordings. Although they were mainly contiguous between adjacent channels, this was not necessarily the case and intermittent build-up could be seen distant from the epileptogenic zone or radiographic lesion.ConclusionsInterictal ISA can be detected in routine intracranial and scalp recordings, without the need for DC amplifiers, and can provide additional information.SignificanceSince ISA is a separate element of the electromagnetic spectrum, apparently non-neuronal in origin, its assessment should be included not only in the pre-surgical evaluation of epilepsy patients but also in patients with other neurologic disorders and normal volunteers.