Nuclear translocation of endonuclease G in degenerating neurons after permanent middle cerebral artery occlusion in mice.

Endonuclease G (EndoG) is a mitochondrial enzyme, known to be involved in caspase-independent cell death following translocation to the cellular nucleus. Nuclear translocation of EndoG has been observed in the ischemic area following transient occlusion of the middle cerebral artery (MCA) in mice, but not after permanent MCA occlusion. In this study we investigated the cellular and temporal expression of EndoG in infarcted cortex during the first 24 h after permanent MCA occlusion in mice, using immunohistochemistry, quantitative rt-PCR and cell specific immunoflourescence markers. EndoG translocated from the cytoplasm to the nucleus as early as 4 h and with a significant increase in the number of EndoG positive nuclei at 12 and 24 h after MCA occlusion. Nuclear translocation of EndoG was observed in degenerating NeuN positive neurons that were evenly distributed throughout the developing infarct. Translocation of EndoG was supported by unaltered EndoG mRNA levels. EndoG was neither expressed in GFAP positive astrocytes nor in CD11b positive microglia/macrophages. In contrast, CD11b positive microglia, but not infiltrating CD11b positive bone marrow-derived macrophages, were shown to express activated caspase-3. The translocation of EndoG to the nucleus of neurons in the infarct implicates EndoG in ischemic neuronal degeneration after permanent MCA occlusion in mice. Increased knowledge about EndoG involvement in ischemic neuronal cell death in mice might offer a promise to control processes involved in neuronal cell death pathways in stroke.

Axonal lesion-induced microglial proliferation and microglial cluster formation in the mouse.

Microglia are innate immune cells and form the first line of defense of the CNS. Proliferation is a key event in the activation of microglia in acute pathology, and has been extensively characterized in rats, but not in mice. In this study we investigated axonal-lesion-induced microglial proliferation and surface antigen expression in C57BL/6 mice. Transection of the entorhino-dentate perforant path projection results in an anterograde axonal and a dense terminal degeneration that induces a region-specific activation of microglia in the dentate gyrus. Time-course analysis showed activation of microglial cells within the first week post-lesion and cell counting demonstrated a significant 1.6-fold increase in microglial numbers 24 h post-lesion reaching a maximal 3.8-fold increase 3 days post-lesion compared with controls. Double staining for the microglial macrophage antigen-1 and the proliferation marker bromodeoxyuridine, injected 1 h prior to perfusion, showed that lesion-reactive microglia accounted for the vast majority of proliferating cells. Microglia proliferated as soon as 24 h after lesion and 25% of all microglial cells were proliferating 3 days post-lesion. Immunofluorescence double staining showed that most activated, proliferating microglia occurred in multicellular clusters and co-expressed the intercellular adhesion molecule-1 and the hematopoietic stem cell marker cluster of differentiation 34. In conclusion, this study extends observations of axonal lesion-induced microglial proliferation in rats to mice, and provides new information on early microglial proliferation and microglial cluster formation and surface antigen expression in the mouse.

Microglia and macrophages express tumor necrosis factor receptor p75 following middle cerebral artery occlusion in mice.

The proinflammatory and potential neurotoxic cytokine tumor necrosis factor (TNF) is produced by activated CNS resident microglia and infiltrating blood-borne macrophages in infarct and peri-infarct areas following induction of focal cerebral ischemia. Here, we investigated the expression of the TNF receptors, TNF-p55R and TNF-p75R, from 1 to 10 days following permanent occlusion of the middle cerebral artery in mice. Using quantitative polymerase chain reaction (PCR), we observed that the relative level of TNF-p55R mRNA was significantly increased at 1-2 days and TNF-p75R mRNA was significantly increased at 1-10 days following arterial occlusion, reaching peak values at 5 days, when microglial-macrophage CD11b mRNA expression was also increased. In comparison, the relative level of TNF mRNA was significantly increased from 1 to 5 days, with peak levels 1 day after arterial occlusion. In situ hybridization revealed mRNA expression of both receptors in predominantly microglial- and macrophage-like cells in the peri-infarct and subsequently in the infarct, and being most marked from 1 to 5 days. Using green fluorescent protein-bone marrow chimeric mice, we confirmed that TNF-p75R was expressed in resident microglia and blood-borne macrophages located in the peri-infarct and infarct 1 and 5 days after arterial occlusion, which was supported by Western blotting. The data show that increased expression of the TNF-p75 receptor following induction of focal cerebral ischemia in mice can be attributed to expression in activated microglial cells and blood-borne macrophages.

Effect of aging and Alzheimer's disease-like pathology on brain monoamines in mice.

Aging is the greatest single risk factor of the neurodegenerative disorder Alzheimer's disease (AD). The monoaminergic system, including serotonin (5-HT), dopamine (DA) and noradrenaline (NA) modulates cognition, which is affected in AD. Changes in monoamine levels have been observed in AD, but these can both be age- and/or disease-related. We examined whether brain monoamine levels change as part of physiological aging and/or AD-like disease in APPSWE/PS1ΔE9 (APP/PS1) transgenic mice. The neocortex, hippocampus, striatum, brainstem and cerebellum of 6-, 12-, 18- and 24-month-old B6C3 wild-type (WT) mice and of 18-month old APP/PS1 and WT mice were analysed for 5-HT, DA and NA contents by high pressure liquid chromatography (HPLC), along with neocortex from 14-month-old APP/PS1 and WT mice. While, we observed no aging effect in WT mice, we detected region-specific changes in the levels of all monoamines in 18-month-old transgenic compared with WT mice. This included reductions in 5-HT (-30%), DA (-47%) and NA (-32%) levels in the neocortex and increases of 5-HT in the brainstem (+18%). No changes were observed in any of the monoamines in the neocortex from 14-month-old APP/PS1 mice. In combination, these findings indicate that aging alone is not sufficient to affect brain monoamine levels, unlike the APPSWE/PS1ΔE9 genotype.

IFNgamma enhances microglial reactions to hippocampal axonal degeneration.

Glial reactivity is implicated in CNS repair and regenerative responses. Microglia, the cells responding earliest to axonal injury, produce tumor necrosis factor-alpha (TNFalpha), a cytokine with both cytopathic and neuroprotective effects. We have studied activation of hippocampal microglia to produce TNFalpha in response to transection of perforant path axons in SJL/J mice. TNFalpha mRNA was produced in a transient manner, peaking at 2 d and falling again by 5 d after lesioning. This was unlike other markers of glial reactivity, such as Mac-1 upregulation, which were sustained over longer time periods. Message for the immune cytokine interferon-gamma (IFNgamma) was undetectable, and glial reactivity to axonal lesions occurred as normal in IFNgamma-deficient mice. Microglial responses to lesion-induced neuronal injury were markedly enhanced in myelin basic protein promoter-driven transgenic mice, in which IFNgamma was endogenously produced in hippocampus. The kinetics of TNFalpha downregulation 5 d after lesion was not affected by transgenic IFNgamma, indicating that IFNgamma acts as an amplifier and not an inducer of response. These results are discussed in the context of a regenerative role for TNFalpha in the CNS, which is innately regulated and potentiated by IFNgamma.

A quantitative in situ hybridization and polymerase chain reaction study of microglial-macrophage expression of interleukin-1beta mRNA following permanent middle cerebral artery occlusion in mice.

Interleukin-1beta (IL-1beta) is known to play a central role in ischemia-induced brain damage in rodents. In comparison to the rat, however, the available data on the cellular synthesis of IL-1beta mRNA and protein in the mouse are very limited. Here, we report on the time profile, the topography and the quantitative, cellular expression of IL-1beta mRNA in mice subjected to permanent occlusion of the distal middle cerebral artery (MCA). The in situ hybridization analysis showed that IL-1beta mRNA was expressed during the first post-surgical hour in a small number of high-expressing macrophage-like cells, located in cortical layers I and II of the future infarct. At 2 h, a significant number of faintly labeled IL-1beta mRNA-expressing cells had appeared in the developing peri-infarct, and the number remained constant at 4 h and 6 h, when the hybridization signal began to distribute to the cellular processes. Quantitative PCR performed on whole hemispheres showed a significant 20-fold increase in the relative level of IL-1beta mRNA at 12 h and a highly significant 42-fold increase at 24 h, at which time single IL-1beta mRNA-expressing cells were supplemented by aggregates and perivascular infiltrates of intensely labeled IL-1beta mRNA-expressing cells. Immunohistochemistry and double immunohistochemical stainings in addition to combined in situ hybridization, confirmed that the intensely labeled IL-1beta mRNA-expressing and IL-1beta protein synthesizing cells predominantly were glial fibrillary acidic protein-immunonegative, macrophage associated antigen-1-immunopositive microglia-macrophages. By day 5 there was a dramatic decline in the relative level of IL-1beta mRNA in the ischemic hemisphere. In summary, the data provide evidence that permanent occlusion of the distal MCA in mice results in expression of IL-1beta mRNA and IL-1beta synthesis in spatially and temporally segregated subpopulations of microglia and macrophages.

No loss of hippocampal hilar somatostatinergic neurons after repeated electroconvulsive shock: a combined stereological and in situ hybridization study.

Electrically induced seizures with anesthesia and muscle relaxation (ECT) is commonly used in the therapy of psychotic depression in humans. Unmodified electroshock (ECS) is used as a model for epilepsy in the rat. In several seizure models of epilepsy, in particular the dentate hilar somatostatin-containing (SSergic) neurons have been found to undergo degeneration. To assess the potential loss of SSergic hilar neurons after repeated ECS, 10 rats were given 110 ECS, one per day, 5 days a week. One day after the last ECS the rats were anesthesized, perfused, the brains cut on a vibratome and prepared for nonradioactive in situ hybridization for somatostatin along with five control rats. Like rats given 10-36 ECS in earlier studies, the ECS-treated rats displayed a markedly increased neuronal hybridization labeling when compared with control rats. The total number of dentate hilar SSergic neurons of each rat was estimated using the optical disector technique. The mean number of hilar SSergic neurons in the ECS-treated rats was 12,785, compared to 12,392 in the control rats. The total number of hilar SSergic neurons in ECS-treated versus control rats was not significantly different (Student's t test; t value = .35; p = .74). We conclude that repeated ECS treatment does not cause loss of hilar SSergic neurons.

Microglia and macrophages are the major source of tumor necrosis factor in permanent middle cerebral artery occlusion in mice.

The proinflammatory cytokine tumor necrosis factor (TNF) is known to be expressed in brain ischemia; however, its cellular and temporal appearance is not fully settled. In this study, nonradioactive in situ hybridization for murine TNF mRNA was performed on brain sections from adult C57x129 mice at 6 hours, 12 hours, 24 hours, 2 days, 5 days, or 10 days (six to eight mice per group) after induction of permanent focal cerebral ischemia. Cortical infarct volumes were estimated, and TNF mRNA-expressing cells were counted within the infarct and infarct border using Cast-Grid analysis. At 12 hours, a peak of 19.2 +/- 5.1 TNF mRNA-expressing cells/mm2 was counted, contrasting two to three times lower values at 6 and 24 hours (6.4 +/- 4.6 and 9.2 +/- 3.4 cells/mm2, respectively) and <2 cells/mm2 at 48 hours and later stages. The TNF mRNA-expressing cells were distributed along the entire rostrocaudal axis of the cortical infarcts and occasionally within the caudate putamen. At all time points, TNF mRNA colocalized with Mac-1-positive microglia/macrophages but not with Ly-6G (Gr-1)-positive polymorphonuclear leukocytes. Similarly, combined in situ hybridization for TNF mRNA and immunohistochemistry for glial fibrillary acidic protein at 12 and 24 hours revealed no TNF mRNA-expressing astrocytes at these time points. Translation of TNF mRNA into bioactive protein was demonstrated in the neocortex of C57B1/6 mice subjected to permanent middle cerebral artery occlusion. In summary, this study points to a time-restricted microglial/macrophage production of TNF in focal cerebral ischemia in mice.

Microglial and macrophage reactions mark progressive changes and define the penumbra in the rat neocortex and striatum after transient middle cerebral artery occlusion.

Transient middle cerebral artery occlusion in rats leads to infarction of the lateral part of the striatum and adjacent neocortex, with selective neuronal necrosis in the bordering penumbral zones. Administration of glutamate, cytokine, and leukocyte antagonists have rescued mainly neocortical neurons, indicating differences in the degenerative processes. The aim of this study was, therefore, to describe the microglial/macrophage activation and polymorphonuclear leukocyte recruitment patterns and to correlate these with the ischemia-induced degenerative processes. The analysis showed significant differences in the characteristics and timing of the microglial/macrophage responses between the caudate putamen and neocortical infarct zones, the infarct zones and their associated penumbral zones, as well as between the striatal and the neocortical penumbral zone. Infiltrations with polymorphonuclear leukocytes into the infarct zones were limited and shortlasting and confined to the acutely degenerating striatum and piriform cortex. A delayed, massive infiltration with lipid phagocytes into the caudate putamen infarct markedly contrasted an early recruitment and activation of microglia/macrophages in the adjacent penumbra. Within the neocortex, a later onset of degeneration along the insular-parietal axis was marked by neuronal expression of heat shock protein and a progressive microglial activation with induction of the full repertoire of microglial activation markers, including a widespread microglial major histocompatibility complex (MHC) class II antigen expression. We interpret the present results as delineating two differentially progressing penumbral zones, which are likely to reflect differences in the underlying degenerative processes. Differences in the microglial/macrophage activation pattern attract special attention, as these cells may constitute specific targets for therapeutic intervention.

Estimation of the number of somatostatin neurons in the striatum: an in situ hybridization study using the optical fractionator method.

Somatostatin-containing neurons of the striatum constitute fewer than 5% of the total neuronal population. Their involvement in the feedforward inhibition of the spiny projection neurons, the modulation of other interneurons, and the regulation of regional blood flow indicates that this small population of neurons plays an important role in the processing of information in the striatum. As a first step in developing a quantitative structural framework within which a more rigorous analysis can be made of the functional circuitry of the striatum, we used modern unbiased stereological techniques to make estimates of the total number of neurons expressing mRNA for somatostatin in the striatum of rats. The strategy developed involved the application of the optical fractionator technique to relatively thick tissue sections that were hybridized in situ with a relatively short oligonucleotide probe conjugated to a nonradioactive reporter molecule. The approach is generally applicable to other subpopulations of in situ hybridized cells in other parts of the brain and can provide a link between molecular neurobiology and stereology. The mean total number of neurons on one side of the striatum was estimated to be 21,300. An analysis of the sampling scheme indicated that counting no more than 200 neurons in a systematic sample of not more than 15 sections per individual results in an estimate with a precision that is more than sufficient for comparative and experimental studies. The issues that must be considered when analyzing in situ hybridized tissue with modern stereological methods, the interpretive caveats inherent in the resulting data, and the unique perspectives provided by data like that presented here for striatal somatostatin neurons are discussed.

Development of microglia in the prenatal rat hippocampus.

The distribution and appearance of microglia cell precursors in the prenatal hippocampus were examined in embryonic day 14 (E14) to E21 rats by nucleoside diphosphatase histochemistry. For comparison, the differentiation of astroglial cells was analyzed from E17 by vimentin and glial fibrillary acidic protein immunohistochemistry. Based on morphologic features, nucleoside diphosphatase-positive microglial cell precursors were classified as ameboid microglial cells and primitive ramified microglial cells. Ameboid microglia were present in the hippocampal primordium on E14. As the hippocampus developed, however, ameboid microglia gradually transformed into primitive ramified microglia, first recognized at E19. Microglial cell precursors, often related to nucleoside diphosphatase-labeled blood vessels, were particularly observed next to the pial surface on days E14 and E17 and in the highly vascularized area around the hippocampal fissure from E19. Within the brain parenchyma, the microglial cell precursors tended to be located within the differentiating cell and neuropil layers rather than in the germinative zones. The late developing dentate gyrus remained almost devoid of microglial cell precursors before birth. Vimentin-positive astroglial processes with radial orientation were observed throughout the hippocampal subregions from E17. In contrast, glial fibrillary acidic protein-positive, radial processes were barely discernible in the fimbria and the dentate gyrus before E19. The results are discussed in relation to the possible interactive role of microglial cells in central nervous tissue development and histogenesis. Regarding the origin of hippocampal microglial cell precursors, the present observations support the view that these cells may well originate from different mesodermal sources depending on time and localization.

Leukocyte infiltration and glial reactions in xenografts of mouse brain tissue undergoing rejection in the adult rat brain. A light and electron microscopical immunocytochemical study.

Neural mouse xenografts undergoing rejection in the adult recipient rat brain were characterized with regard to infiltrating host leukocytes and reactions of graft and host astro- and microglial cells. Rejection occurred within 35 days with infiltration of the grafts by in particular macrophages and T-cells as well as blood-brain barrier (BBB) leakage for IgG. In the surrounding host brain microglial cells showed increased histochemical staining for nucleoside diphosphatase (NDPase) and increased immunocytochemical expression of complement receptor type 3 (CR3), while astroglial cells displayed an increased immunoreactivity for glial fibrillary acidic protein (GFAP). Light microscopic findings of rat major histocompatibility complex (MHC) antigen class I on microglial cells, endothelial cells and leukocytes were confirmed at the ultrastructural level and extended to include a few astrocytes. Rat and mouse MHC antigen class II was only detected on leukocytes and activated microglia. We suggest that host macrophages and activated host and xenograft microglial cells act in situ as immunostimulatory cells on T-helper cells, and that increased levels of donor MHC antigen class I may further enhance the killer activity exerted by host T-cytotoxic cells.

Reduction of the microglial cell number in rat primary glial cell cultures by exogenous addition of dibutyryl cyclic adenosine monophosphate.

The present work examined the effects induced by dibutyryl cyclic adenosine monophosphate (dB-cAMP) on microglial cells in primary glial cell cultures from newborn rats. Microglial cells were identified by OX42 immunohistochemistry and nucleoside diphosphatase histochemistry. Double staining for astrocytes was carried out by combination with glial fibrillary acidic protein immunolabeling. Addition of 0.25 mM dB-cAMP to the cultures decreased the microglial cell number about sixfold. The findings suggest that the effect of dB-cAMP on the microglial cells might be either a direct action of dB-cAMP on the microglial cells or an indirect effect mediated by the astroglial cells.

Enriched immune-environment of blood-brain barrier deficient areas of normal adult rats.

The circumventricular organs (CVOs) in the brain are without a blood-brain barrier (BBB) and as such directly exposed to blood plasma constituents and blood-borne pathogens. In light of previous studies showing discrepancies regarding the immunocompetence of these organs, we initiated the present study to provide a comprehensive immunohistochemical analysis of the cellular expression of immune-associated antigens within the pineal gland, area postrema and the subfornical organ. In all CVOs, subpopulations of cells morphologically similar to complement receptor type 3 immunoreactive microglial/macrophage cells expressed major histocompatibility complex (MHC) class II antigen, leucocyte common antigen (LCA/CD45), as well as CD4 and ED1 antigen. Based on morphological criteria the MHC class II antigen expressing cells could be grouped into a major population of classical parenchymal and perivascular ramified microglial cells and a minor population presenting itself as scattered or small groups of rounded macrophage-like cells. CD4 and ED1 antigen were expressed by both cell types. CD45 was preferentially expressed by macrophage-like cells. MHC class I antigen was expressed by the vascular endothelium in both BBB-protected and BBB-deficient areas and was additionally present as a lattice-like network throughout the BBB-deficient parenchyma in all CVOs. The results suggest that the BBB-free areas of the brain besides being constantly surveyed by blood-borne macrophages, possess an intrinsic immune surveillance system based on resting and activated microglial cells, which may function as a non-endothelial, cellular barrier against blood-borne pathogens.

Triple immunosuppression protects murine intracerebral, hippocampal xenografts in adult rat hosts: effects on cellular infiltration, major histocompatibility complex antigen induction and blood-brain barrier leakage.

Recently we reported protection of intracerebral mouse to rat hippocampal xenografts upon treatment with a combination of cyclosporin A, prednisolone and azathioprine. These findings are now supported in an extended analysis of graft-infiltrating cells. Host T-cell and macrophage infiltration and the immunocytochemical level of cellular expression of major histocompatibility complex class I and II antigens, measured by densitometric analysis, were compared between recipient rats receiving cyclosporin A alone or cyclosporin A in combination with prednisolone and azathioprine. The combination therapy resulted in a much improved survival of the xenografted hippocampal tissue with preservation of organotypic granule and pyramidal cell layers. Graft infiltration by T-cells and macrophages was significantly lower and the level of major histocompatibility complex class I and II antigen expression by the infiltrating cells markedly reduced. Lower expression of donor-type major histocompatibility complex class I antigen was also found in the xenografts in the trimedicated recipients, together with reduced blood brain barrier leakage and astrogliosis at the host-graft interface. The results demonstrate the benefits of using combined immunosuppressive strategies for protection of histoincompatible brain xenografts in the central nervous system.

Somatostatin and neuropeptide Y in organotypic slice cultures of the rat hippocampus: an immunocytochemical and in situ hybridization study.

The neuronal distributions of somatostatin and neuropeptide Y and their respective mRNAs in hippocampal slice cultures were examined by immunohistochemical staining and in situ hybridization. For the in situ hybridization we used an alkaline phosphatase-labelled oligodeoxynucleotide probe for somatostatin mRNA and an 35S-labelled oligodeoxynucleotide probe for neuropeptide Y mRNA. For both neuropeptides the immunostained and hybridized neurons displayed a comparable, organotypic distribution. Most labelled neurons were located in the dentate hilus and stratum oriens of CA3 and CA1. Additional neurons were found in stratum radiatum and pyramidale of CA3, but very few in the corresponding layers of CA1. In all locations the density of somatostatin- and neuropeptide Y-reactive cells exceeded that observed in vivo. Also, the hybridization signal of the individual neurons appeared enhanced in the slice cultures. Methodologically it was noted that the non-radioactive alkaline phosphatase-labelled oligodeoxynucleotide probe gave excellent in situ hybridization results with detailed cellular resolution and no apparent problems of tissue penetration, even when used on whole-mount explants. These results demonstrate that somatostatin and neuropeptide Y-immunoreactive and mRNA containing neurons retain their organotypic distribution and basic morphological characteristics in the slice cultures. The supernormal density of these neurons and their hybridization signals indicate that a transient developmental increase in neuropeptide expression may persist in vitro.

The relationship between oral dyskinesias produced by long-term haloperidol treatment, the density of striatal preproenkephalin messenger RNA and enkephalin peptide, and the number of striatal neurons expressing preproenkephalin messenger RNA in rats.

Neuroleptic-induced oral dyskinesias in rats, a putative analogue to human tardive dyskinesia, may be due to excitotoxic degeneration within the striatum. Haloperidol treatment for 34 weeks increased the optical density of preproenkephalin messenger RNA in individual striatal neurons and enkephalin peptide in the neuropil, regardless of the level of oral dyskinesia produced. However, using unbiased stereological methods, an increased number of striatal neurons expressing preproenkephalin messenger RNA was observed only in rats that did not develop pronounced oral dyskinesias during haloperidol treatment. Said in another manner, the haloperidol-treated animals that developed pronounced oral dyskinesias, failed to produce an increase in the number of neurons expressing preproenkephalin messenger RNA. These results indicate that the mechanism by which neuroleptics induce oral dyskinesias in rats, and perhaps tardive dyskinesia in humans, involves a functional disturbance or even damage to a subpopulation of enkephalinergic neurons in the striatum.

Microglial reactivity correlates to the density and the myelination of the anterogradely degenerating axons and terminals following perforant path denervation of the mouse fascia dentata.

Transection of the entorhino-dentate perforant path is a well known model for lesion-induced axonal sprouting and glial reactions in the rat. In this study, we have characterized the microglial reaction in the dentate molecular layer of the SJL/J and C57Bl/6 mouse. The morphological transformation of the microglial cells and their densitometrically measured Mac-1 immunoreactivity were correlated with the density of silver-impregnated axonal and terminal degeneration and the myelination of the degenerating medial and lateral perforant pathways. Anterograde axonal and terminal degeneration leads to: (i) altered myelin basic protein immunoreactivity with the appearance of discrete myelin deposits preferentially in the denervated medial and significantly less so in the lateral perforant path zone from day 2 after lesioning; (ii) an increase in number and Mac-1 immunoreactivity of morphologically-changed microglial cells in the denervated perforant path zones with more pronounced morphological transformation of microglia in the medial than in the lateral perforant path zones at day 2 but not day 5 after lesioning; and (iii) a linear correlation between the density of microglial Mac-1 reactivity and axonal degeneration in the medial but not in the lateral perforant path zone at two days postlesion, and a linear correlation in both zones at five days postlesion. We propose that the differentiated microglial response is due to the different densities of axonal and terminal degeneration, as observed in the individual cases. The finding of a potentiated or accelerated microglial activation in the medial as compared to the lateral perforant path zone suggests different kinetics of microglial activation in areas with degenerating myelinated and unmyelinated fibers.

Histochemical demonstration of purine nucleoside phosphorylase (PNPase) in microglial and astroglial cells of adult rat brain.

The histochemical localization of enzymes associated with purine nucleoside metabolism indicates that glial cells might participate in the regulation of these compounds in the central nervous system. In the present study we examined the histochemical localization of purine nucleoside phosphorylase (PNPase) in sections from adult rat brain. Some sections were also sequentially stained immunocytochemically for astroglial or microglial cells utilizing glial fibrillary acidic protein (GFAP) or OX-42 antibodies, respectively. Our observations showed that PNPase was restricted to glial cells, whereas neurons always remained negative. Brain sections stained for both PNPase and GFAP showed that the GFAP-positive astroglial cells were always PNPase positive. Other PNPase-positive but GFAP-negative cells were also observed. These cells resembled microglial cells, and brain sections reacted for both PNPase and OX-42 confirmed this by showing that the major part of OX-42-positive microglial cells were PNPase positive. In these sections, the PNPase-positive but OX-42-negative cells present resembled astroglial cells. From our double staining experiments, we conclude that PNPase is present in both astroglial and microglial cells in normal adult brain.