Selective transfection of microglia in the brain using an antibody-based non-viral vector.

There are currently few approaches to transiently manipulate the expression of specific proteins in microglia of the brain. An antibody directed against an extracellular epitope of scavenger receptor class B, type I (SR-BI) was found to be selectively taken up by these cells in the brain. Other antibodies tested were not internalised by microglia. A vector was produced by linking the SR-BI antibody to polyethyleneimine and binding a DNA plasmid encoding green fluorescent protein. Infusions of this vector into the hippocampus produced a widespread transfection of cells, more than 80% of which were immunoreactive for microglial/macrophage markers. Transfection was not detected in cells expressing markers for astrocytes or neurons. Reporter gene expression was most prominent near the infusion site but was seen in tissue up to 4mm away. DNA bound to polyethyleneimine alone or to a vector containing a different antibody did not produce transfection in the brain. Single injections of the vector containing the SR-BI antibody into the brain also resulted in transfection of microglia, albeit with lower efficiency. Vector modifications to promote lysis of endosomes or entry of DNA into the nucleus did not increase efficiency. The findings clearly demonstrate the capacity of the SR-BI antibody to selectively target brain microglia. This approach offers considerable potential to deliver DNA and other molecules capable of modifying the function of these cells in vivo.

Connexin 43 regulates astrocytic migration and proliferation in response to injury.

Marked alterations in astrocyte function are a universal response to disease or injury in the central nervous system. Affected astrocytes develop characteristic morphological changes known as "reactive astrogliosis", characterized by increased expression of the intermediate filament proteins, glial fibrillary acidic protein and vimentin. Reactive astrocytes also display alterations in other proteins including a rapid up-regulation of the gap junction protein, Connexin 43. The present study tests whether Connexin 43 is directly involved in the astrocytic response to injury. We have down regulated Connexin 43 expression using a microRNA generating plasmid and investigated the functional consequences of this treatment on the response of astrocytes in primary culture to a well-characterized scratch wound injury. The treatment resulted in more than 70% transfection efficiency and a near complete depletion of Connexin 43 in transfected cells. Compared to cells transfected with non-targeting microRNA, the cells depleted of Connexin 43 showed a slower wound closure and fewer transfected cells in the wound area. These changes were associated with decreased proliferation of the Connexin 43-depleted cells as well as shorter processes extending into the wound area suggesting a direct impairment of migration. The effects were independent of gap junction conductivity as exposure to the gap junction blocker carbenoxolone did not affect the rate of wound healing. The findings directly indicate a role for Connexin 43 in the astrocytic response to injury and suggest that modification of Connexin 43 expression might provide a therapeutic target to alter potentially deleterious astrocytic responses.

Astrocytic responses to DNA delivery using nucleofection.

Nucleofection is a powerful non-viral transfection technique that can deliver plasmid DNA with high efficiency to cells that are traditionally difficult to transfect. In this study, we demonstrate that nucleofection of astrocytes grown in primary cell culture resulted in 76 ± 9% transfected cells and low cytotoxicity. However, the nucleofected astrocytes showed a reduced re-attachment to the growth media when replated and subsequent impairment of proliferation. This led to substantially decreased cell densities during the initial 72 h following transfection. Furthermore, these cells were less efficient at producing wound closure in a scratch model of injury. Nucleofection also resulted in the generation of a small proportion of polynucleated cells. The findings demonstrate that nucleofection provides a valuable technique for delivering DNA to astrocytes in culture. However, considerable care is needed in designing and interpreting such studies because of long-lasting changes induced in key properties of these cells by the nucleofection process.

Increased substance P immunoreactivity and edema formation following reversible ischemic stroke.

Previous results from our laboratory have shown that neurogenic inflammation is associated with edema formation after traumatic brain injury (TBI). This neurogenic inflammation was characterized by increased substance P (SP) immunoreactivity and could be attenuated with administration of SP antagonists with a resultant decrease in edema formation. Few studies have examined whether neurogenic inflammation, as identified by increased SP immunoreactivity, occurs after stroke and its potential role in edema formation. The present study examines SP immunoreactivity and edema formation following stroke. Experimental stroke was induced in halothane anaesthetized male Sprague-Dawley rats using a reversible thread model of middle cerebral artery occlusion. Increased SP immunoreactivity at 24 hours relative to the non-infarcted hemisphere was observed in perivascular, neuronal, and glial tissue, and within the penumbra of the infarcted hemisphere. It was not as apparent in the infarct core. This increased SP immunoreactivity was associated with edema formation. We conclude that neurogenic inflammation, as reflected by increased SP immunoreactivity, occurs following experimental stroke, and that this may be associated with edema formation. As such, inhibition of neurogenic inflammation may represent a novel therapeutic target for the treatment of edema following reversible, ischemic stroke.

Mitochondrial glutathione: a modulator of brain cell death.

The small fraction of glutathione in mitochondria in nonneural tissues is an important contributor to cell survival under some conditions. However, there has been only limited characterization of the properties and function of mitochondrial glutathione in cells from the brain. In astrocytes in culture, highly selective depletion of this glutathione pool does not affect cell viability, at least in the first 24 h, but does greatly increase susceptibility to exposure to nitric oxide or peroxynitrite. In vivo, a selective partial loss of glutathione develops during focal cerebral ischemia and persists during reperfusion. The timing and distribution of glutathione loss shows an apparent association with the likelihood that tissue infarction will subsequently develop. Furthermore, infarct volume is greatly decreased by intracerebroventricular infusion of glutathione monoethylester, a compound that can increase mitochondrial glutathione. Together these recent findings indicate that alterations in mitochondrial glutathione are likely to contribute to the severity of tissue damage in stroke and possibly other neurological disorders. Thus, this antioxidant pool provides a potentially useful target for therapeutic intervention.

Brain mitochondrial responses to postischemic reperfusion: a role for calcium and hydrogen peroxide?

During early recirculation following global brain ischemia, mitochondria are exposed to markedly elevated Ca(2+) concentrations and a short-lived production of reactive oxygen species, including hydrogen peroxide (H(2)O(2)). A brief increase in mitochondrial Ca(2+) and a subsequent increase in mitochondrial glutathione content have been observed. In the present study, we have confirmed the increase in mitochondrial glutathione in a rat model of global forebrain ischemia. This change was not inhibited by treatment of the rats with FK506, contrasting with our previous finding that cyclosporin A partially blocked the increase. These results suggest that induction of the mitochondrial permeability transition may be necessary for the increase in glutathione content in these organelles. To further investigate possible mitochondrial responses during early postischemic reperfusion, mitochondria isolated from normal brain were exposed to Ca(2+) and H(2)O(2), under conditions similar to those in intact cells. Respiratory activity was substantially modified when the mitochondria were exposed to Ca(2+) and H(2)O(2) together. Two distinct and largely noninteracting mechanisms apparently accounted for the responses to these agents. The effects of Ca(2+), but not H(2)O(2), were inhibited by cyclosporin A, again implicating the permeability transition in some of the mitochondrial changes.

Impairment of brain mitochondrial function by hydrogen peroxide.

Hydrogen peroxide, at concentrations comparable to those observed under some pathological conditions, produced a concentration-dependent inhibition of state 3 (ADP-stimulated) and uncoupled mitochondrial respiratory activity. The ADP:O ratio was also substantially reduced. In contrast, the organic peroxide, t-butylhydroperoxide at the same concentrations produced no significant changes in respiratory activity. Intramitochondrial glutathione was oxidised to a similar extent in the presence of hydrogen peroxide or t-butylhydroperoxide. Thus, changes in this endogenous antioxidant apparently did not underlie the different responses to these peroxides. The effects of hydrogen peroxide were not altered by deferoxamine indicating that the extramitochondrial generation of hydroxyl radicals was not likely to be involved. However, modifications arising from the generation of hydroxyl radicals within the mitochondria remain a likely contributor to the observed deleterious effects on respiratory function. The inhibitory effects of hydrogen peroxide were greatest when pyruvate plus malate were present as respiratory substrates. Lesser inhibition was seen with glutamate plus malate and no significant inhibitory effects were detected in the presence of succinate. The findings suggest that mitochondrial components involved in pyruvate oxidation were particularly sensitive to the hydrogen peroxide treatment. However, no significant change was seen in activity of either the pyruvate dehydrogenase complex or NADH-ubiquinone oxidoreductase (complex I) when measured directly following treatment of the mitochondria with hydrogen peroxide.

Improved recovery of highly enriched mitochondrial fractions from small brain tissue samples.

The investigation of mitochondrial abnormalities in brain commonly requires isolation of these organelles from small tissue samples. We have modified a mitochondrial isolation procedure based on Percoll density gradient centrifugation to increase the proportion of the total mitochondrial pool recovered while reducing contamination with synaptosomes and related structures containing cytoplasm. Initially, myelin was removed by centrifugation in 12% Percoll in isotonic buffer. The pellet was resuspended, treated with digitonin to break up synaptosomes and similar structures and subjected to discontinuous Percoll density gradient centrifugation. The mitochondrial fraction obtained from this procedure was highly metabolically active and well coupled, exhibiting respiratory control ratios above 5. The recovery of mitochondrial markers using a single rat forebrain as starting material was approximately 18% to 21%. When small tissue samples (approximately 50 mg wet weight) were used as starting material the recovery of the mitochondrial marker was approximately 16%. The ratio of recovery of a mitochondrial marker to the cytoplasmic marker lactate dehydrogenase exceeded 200 in preparations from a single rat forebrain. This is substantially greater than values reported for previously published procedures reflecting both an improved yield of mitochondria and a reduction in cytoplasmic contamination.

Cyclosporin A-sensitive changes in mitochondrial glutathione are an early response to intrastiatal NMDA or forebrain ischemia in rats.

An intrastriatal injection of NMDA produced an increase in glutathione to 152% of control values in mitochondria isolated from striatum at 1 h later. Total tissue glutathione was not changed. The mitochondrial increase was largely reversed by 2 h. Glutathione content was not significantly affected in mitochondria from a part of the cerebral cortex that did not exhibit damage following intrastriatal NMDA. Glutathione was similarly increased in mitochondria from both cortex and striatum at 1 h after a short period of forebrain ischemia, confirming our previous findings. The increases in mitochondrial glutathione developed shortly after accumulations of mitochondrial calcium that have been observed previously. Intravenous injection of cyclosporin A immediately following either the NMDA treatment or reversal of the ischemic period partially inhibited the increases in glutathione in mitochondria from the affected brain subregions. These studies provide evidence that early changes sensitive to cyclosporin A develop in mitochondria under pathological conditions in the intact brain. These glutathione increases are consistent with an induction of the mitochondrial permeability transition in the affected tissue.

Mitochondrial respiratory function and cell death in focal cerebral ischemia.

Pyruvate-supported oxygen uptake was determined as a measure of the functional capacity of mitochondria obtained from rat brain during unilateral middle cerebral artery occlusion and reperfusion. During ischemia, substantial reductions developed in both ADP-stimulated and uncoupled respiration in tissue from the focus of the affected area in the striatum and cortex. A similar pattern of change but with lesser reductions was seen in the adjacent perifocal tissue. Succinate-supported respiration was more affected than that with pyruvate in perifocal tissue at 2 h of ischemia, suggesting additional alterations to mitochondrial components in this tissue. Mitochondrial respiratory activity recovered fully in samples from the cortex, but not the striatum, within the first hour of reperfusion following 2 h of ischemia and remained similar to control values at 3 h of reperfusion. In contrast, impairment of the functional capacity of mitochondria from all three regions was seen in the first 3 h of reperfusion following 3 h of ischemia. Extensive infarction generally affecting the cortical focal tissue with more variable involvement of the perifocal tissue developed following 2 h of focal ischemia. Thus, mitochondrial impairment during the first 3 h of reperfusion was apparently not essential for tissue infarction to develop. Nonetheless, the observed mitochondrial changes could contribute to the damage produced by permanent focal ischemia as well as the larger infarcts produced when reperfusion was initiated following 3 h of ischemia.

The antioxidant defences of brain mitochondria during short-term forebrain ischemia and recirculation in the rat.

This study evaluated changes in the antioxidant defences of mitochondria induced by 30 min of forebrain ischemia and recirculation up to 24 h in rats. Following treatment, mitochondria were isolated from two brain subregions: the dorsolateral striatum, an area in which there is loss of most neurons, and the paramedian cortex in which most neurons are resistant to damage. During ischemia and the first few hours of recirculation, the mitochondrial defences were largely preserved based on measurements of the activities of the enzymes, superoxide dismutase, glutathione peroxidase and glutathione reductase, as well as the response of the mitochondria to a subsequent exposure to H2O2 in vitro. However, some moderate changes were detected, particularly in the mitochondria from the dorsolateral striatum. A decrease of 30% in the activity of superoxide dismutase was seen at the conclusion of the ischemic period and a small increase in susceptibility to changes induced by H2O2 was detected during early recirculation. This latter change preceded and possibly contributed to the development of an impairment of respiratory function detected in mitochondria from the dorsolateral striatum at 3 h of recirculation. At 24 h of recirculation, larger changes were seen in the activities of all three of the enzymes in mitochondria from the dorsolateral striatum but not the paramedian cortex that was associated with progression to advanced neuronal damage in the former subregion.

The pyruvate dehydrogenase complex is partially inactivated during early recirculation following short-term forebrain ischemia in rats.

The mechanisms of selective neuronal loss after short-term global ischemia remain undefined, but processes including increased proteolytic activity, impaired protein synthesis, and oxidative damage have been proposed to contribute. A decrease in activity of the pyruvate dehydrogenase complex in the dorsolateral striatum, an ischemia-susceptible region, is one change apparently differentiating this region from ischemia-resistant areas during early recirculation. To provide an insight into processes contributing to postischemic cell damage, the changes in the pyruvate dehydrogenase complex during early recirculation have been further characterized. These studies provide clear confirmation that the activity of the pyruvate dehydrogenase complex is reduced in mitochondria from the dorsolateral striatum by 3 h of recirculation. The decrease in activity was not accompanied by a loss of antigenic sites or by changes in electrophoretic mobility of the components of the complex. A reduction in activity of the E1 component of the complex (39-42% decrease), but not the E2 and E3 components, was observed that was apparently sufficient to explain the decrease in activity of the whole complex. These results indicate that the changes in activity of the pyruvate dehydrogenase complex in the dorsolateral striatum are not due to loss or gross disruption of the constituent proteins but rather most likely reflect a selective inactivation of a specific component of the complex.

Reduced activity of the pyruvate dehydrogenase complex but not cytochrome c oxidase is associated with neuronal loss in the striatum following short-term forebrain ischemia.

Previous studies have identified changes in the activities of the pyruvate dehydrogenase complex (PDHC) and cytochrome c oxidase during early recirculation following short-term cerebral ischemia. However, the relationship of these changes to the delayed selective neuronal loss that develops as a result of short-term ischemia is incompletely defined. The effects of ischemia and recirculation on the activities of these enzymes in the dorsolateral striatum, a region containing many susceptible neurons, and the ischemia-resistant paramedian cortex have been compared. No significant loss of activity of cytochrome c oxidase was seen in either region during the first few hours of recirculation following 30 min of ischemia. A decrease (of 32%) was observed at 24 h in the dorsolateral striatum. However, this probably resulted from changes in the mitochondrial fraction due to advanced neuronal degeneration. By contrast, there was a significant decrease (by 24%) in activity of PDHC at 3 h following a 30-min, but not a 10-min, ischemic period. Only the 30-min ischemic period resulted in extensive delayed neuronal loss. In the paramedian cortex, there was no significant change in PDHC and no neuronal loss following either ischemic period. These results provide strong evidence for a close association between neuronal loss and changes in the activity of PDHC but not cytochrome c oxidase in the dorsolateral striatum.

Energy metabolism, oxidative stress and neuronal degeneration in Alzheimer's disease.

Both altered energy metabolism and oxidative stress have been proposed to contribute to tissue damage in neurogenerative diseases. Animal models and cell culture studies provide evidence for a role of these processes in several forms of neuronal death. Reductions in the activities of some key mitochondrial enzymes have been found in autopsied brain in Alzheimer's disease. However, results obtained with biopsied brain tissue as well as assessments of metabolic rates for glucose in vivo indicate that a reduced functional capacity of mitochondria is probably not a general feature in the brain in Alzheimer's disease. These studies do not address the possibility that short-lived changes in energy metabolism affecting a small number of cells at any one time could be contributing to cell death. Several findings point to a moderate increase in oxidative damage in those areas of brain which are most severely affected in this disease, probably resulting from an increase in production of reactive oxygen species. Whether this is a contributor to neurodegeneration or a consequence of it remains unresolved.

Alterations in the glutathione content of mitochondria following short-term forebrain ischemia in rats.

Total glutathione was measured in mitochondria isolated following 30 min of ischemia and recirculation periods up to 24 h. Mitochondria prepared from the dorsolateral striatum, a region containing many neurons susceptible to short ischemic periods, were compared with those from the paramedian cortex, an ischemia-resistant region. Parallel increases in glutathione content (to approximately 150% of pre-ischemic values) were seen in both regions during the first few hours of recirculation. By 24 h of recirculation, there was a decrease below pre-ischemic values in preparations from the dorsolateral striatum but not the paramedian cortex. The early increases in mitochondrial glutathione were not associated with comparable increases in total tissue glutathione. A shorter (10 min) ischemic period also produced an early increase in mitochondrial glutathione but this was reversed more rapidly to preischemic values. The observed changes indicate post-ischemic modifications of cellular oxidative defenses in the two brain regions studied.

Calcium, energy metabolism and the development of selective neuronal loss following short-term cerebral ischemia.

Short-term cerebral ischemia results in the delayed loss of specific neuronal subpopulations. This review discusses changes in energy metabolism and Ca2+ distribution during ischemia and recirculation and considers the possible contribution of these changes to the development of selective neuronal loss. Severe ischemia results in a rapid decline of ATP content and a subsequent large movement of Ca2+ from the extracellular to the intracellular space. Similar changes are seen in tissue subregions containing neurons destined to die and those areas largely resistant to short-term ischemia, although differences have been observed in Ca2+ uptake between individual neurons. The large accumulation of intracellular Ca2+ is widely considered as a critical initiating event in the development of of neuronal loss but, as yet, definitive evidence has not been obtained. the increased intracellular Ca2+ content activates a number of additional processes including lipolysis of phospholipids and degradation or inactivation of some specific proteins, all of which could contribute to altered function on restoration of blood flow to the brain. Reperfusion results in a rapid recovery of ATP production. Cytoplasmic Ca2+ concentration is also restored during early recirculation as a result of both removal to the extracellular space and uptake into mitochondria. Within a few hours of recirculation, subtle increases in intracellular Ca2+ and a reduced capacity for mitochondrial respiration have been detected in some ischemia-susceptible regions. Both of these changes could potentially contribute to the development of neuronal loss. More pronounced alterations in Ca2+ homeostasis, resulting in a second period of increased mitochondrial Ca2+, develop with further recirculation in ischemia-susceptible regions. The available evidence suggests that these increases in Ca2+, although developing late, are likely to precede the irreversible loss of neuronal function and may be a necessary contributor to the final stages of this process.

Biochemical changes associated with selective neuronal death following short-term cerebral ischaemia.

A brief interruption of blood flow to the brain results in the selective loss of specific subpopulations of neurons. Important advances have been made in recent years in defining the biochemical changes associated with cerebral ischaemia and reperfusion and in identifying physical and chemical interventions capable of modifying the extent of neuronal loss. Neuronal death is not irreversibly determined by the ischaemic period but develops during recirculation over a period of hours or even days in different susceptible neuronal populations. The onset of ischaemia produces a rapid decline in ATP production and an associated major redistribution of ions across the plasma membrane including a large intracellular accumulation of Ca2+ in many neurons. Alterations subsequently develop in many other metabolites. These include a marked and progressive release of neurotransmitters and a rapid accumulation of free fatty acids. Most of these alterations are reversed within the first 20 min to 1 hr of recirculation. The changes essential for initiating damage in neurons destined to die have not been definitively identified although there is some evidence suggesting roles for the intracellular Ca2+ accumulation, the release of the neurotransmitter glutamate and a brief burst of free radical production which occurs during early recirculation. During further recirculation, there are reductions in oxidative glucose metabolism and protein synthesis in many brain regions. Few changes have been detected which distinguish tissue containing ischaemia-susceptible neurons from ischaemia-resistant regions until the development of advanced degeneration and neuronal loss. Subtle changes in cytoplasmic Ca2+ content and a decrease in the respiratory capacity of mitochondria are two changes apparently selectively affecting ischaemia-susceptible regions which could contribute to neuronal loss. The mitochondrial change may be one indicator of a slowly developing post-ischaemic increase in susceptibility to oxidative damage in some cells.

The calcium content of mitochondria from brain subregions following short-term forebrain ischemia and recirculation in the rat.

A procedure was established for determining the calcium content of mitochondria isolated from rat brain subregions based on changes in fura-2 fluorescence after disruption of the organelles with Triton X-100 and sodium dodecyl sulfate. Mitochondria isolated from the forebrain of normal rats contained 2.5 +/- 0.9 nmol of calcium/mg of protein. A 30-min ischemic period produced an approximately twofold increase in the calcium content of mitochondria isolated from the dorsolateral striatum, a region in which most neurons die within 24 h after this period of ischemia. The calcium content of mitochondria from the paramedian cortex, a region in which there are few ischemia-susceptible neurons, tended to be similarly increased, although this difference was not statistically significant. Larger increases (to approximately five times control values) were seen in mitochondria isolated from both regions after 10 min of recirculation. By 1 h of recirculation, mitochondrial calcium had returned close to preischemic control values in both regions. Longer recirculation periods produced no further changes in the calcium content of mitochondria from the paramedian cortex. However, mitochondrial calcium was again increased in the dorsolateral striatum after 6 h (6.5 nmol of calcium/mg of protein) and 24 h (8.7 nmol of calcium/mg of protein) of recirculation. This regionally selective increase in calcium in the dorsolateral striatum preceded the period during which the majority of neurons in this region exhibit advanced degenerative changes. Thus, this increase may be an essential step, albeit a late one, in the development of neuronal loss.

Delayed treatment with 1,3-butanediol reduces loss of CA1 neurons in the hippocampus of rats following brief forebrain ischemia.

This study examined the effect of 1,3-butanediol on the selective loss of CA1 pyramidal neurons following a short period of near-complete forebrain ischemia. Injection of 55 mmol 1,3-butanediol/kg body weight at 24 h of recirculation and again at 36 h following 10 min of forebrain ischemia markedly reduced damage to CA1 neurons examined at 72 h of recirculation compared with that in saline-treated rats. Comparable treatment with ethanol did not cause significant protection. Neuronal loss was also not reduced by 1,3-butanediol treatment when the ischemic period was extended to 15 min or by single treatments at 24 h or 36 h following 10 min of ischemia. However, a single treatment 5 min after reversal of 10 min of ischemia was effective in ameliorating cell loss. The difference in effectiveness of 1,3-butanediol following 10 min and 15 min of ischemia is consistent with a number of previous studies, indicating that the processes leading to loss of CA1 neurons are modified when the ischemic period is extended. Previous findings that 1,3-butanediol reduced damage in other ischemia-susceptible neuronal subpopulations but not in CA1 neurons most likely reflected the longer period of ischemia which was used. The results of the present investigation demonstrate that administration of 1,3-butanediol offers a novel approach for interfering with post-ischemic loss of CA1 neurons following a brief ischemic period which is effective even when initiated after prolonged recirculation periods.

The stimulus-evoked release of glutamate and GABA from brain subregions following transient forebrain ischemia in the rat.

The release of glutamate and GABA in response to K+ depolarization was determined for tissue prisms prepared from brain subregions removed from rats following 30 min of forebrain ischemia or recirculation periods up to 24 h. There were statistically significant effects of this treatment on release of both amino acids from samples of the dorsolateral striatum, an area developing selective neuronal degeneration. However, for at least the first 3 h of recirculation the calcium-dependent and calcium-independent release of both amino acids in this region were similar to pre-ischemic values. Differences were observed under some conditions at longer recirculation times. In particular there was a decrease in calcium-dependent GABA release at 24 h of recirculation and a trend towards increased release of glutamate at 6 h of recirculation and beyond. No statistically significant differences were seen in samples from the paramedian neocortex, a region resistant to post-ischemic damage. These results suggest that changes in the ability to release glutamate and GABA in response to stimulation are not necessary for the development of neurodegeneration in the striatum but rather that release of these amino acids may be modified as a result of the degenerative process.